EPA-460/3-73-003
       DESIGN OF RECIPROCATING
    SINGLE  CYLINDER  EXPANDERS
                           FOR STEAM
                       FINAL REPORT
        U.S. ENVIRONMENTAL PROTECTION AGENCY
            Office of Air and Water Programs
         Office of Mobile Source Air Pollution Control
        Alternative Automotive Power Systems Division
               Ann Arbor, Michigan 48105

-------
                               EPA-460/3-73-003
  DESIGN OF RECIPROCATING
SINGLE CYLINDER EXPANDERS
             FOR STEAM
           FINAL REPORT
                 Prepared by

            S . E. Eckard and R. D. Brooks

             Nuclear Systems Programs
                 Space Division
                General Electric
              Cincinnati, Ohio 45215


            Contract Number: 68-01-0408
               EPA Project Officers:

        W. B. Zeber, E. Beyma, and P. L. Sutton



                 Prepared for

       U.S. ENVIRONMENTAL PROTECTION AGENCY
           Office of Air and Water Programs
        Office of Mobile Source Air Pollution Control
       Alternative Automotive Potoer Systems Division
             Ann Arbor, Michigan 48105
                 October 1973

-------
This report is issued by the Office of Mobile Source Air Pollution
Control, Office of Air and Water Programs, Environmental Protection
Agency, to report technical data of interest to a limited number of
readers.  Copies of this report are available free of charge to
Federal employees, current contractors and grantees, and non-profit
organizations - as supplies permit - from the Air Pollution Techni-
cal Information Center, Environmental Protection Agency, Research
Triangle Park, North Carolina 27711 or may be obtained, for a
nominal cost, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia 22151.
This report was furnished to the U.S. Environmental Protection Agency
by General Electric in fulfillment of Contract No. 68-01-0408 and has
been reviewed and approved for publication by the Environmental
Protection Agency.  Approval does not signify that the contents
necessarily reflect the views and policies of the agency.  The material 1
presented in this report may be based on an extrapolation of the
"State-of-the-art".  Each assumption must be carefully analyzed by
the reader to assure that it is acceptable for his purpose.  Results
and conclusions should be viewed correspondingly.  Mention of trade
names or commercial products does not constitute endorsement or
recommendation for use.
                     Publication No. EPA-460/3-73-003
                                    11

-------
                           TABLE OF CONTENTS

                                                                  Page


1.0  INTRODUCTION	    1

2.0  ABSTRACT	    4

3.0  CONCLUSIONS AND RECOMMENDATIONS	    6

4.0  EXPANDER DESIGN 	    9

     4.1  Expander Design Technology Base	    9

          4.1.1  High Speed Steam Engine Valving Configurations
                 and Valve Gear Designs	   10
          4.1.2  Wrist Pin Bearing Design for High Uni-
                 directional Loading	   12
          4.1.3  Oil Free High Temperature Piston Ring Design.  .   13
          4.1.4  Piston Rod Sealing	   14

     4.2  Expander Preliminary Design Studies	   15

          4.2.1  General Cylinder Configuration Selection.  ...   15
          4.2.2  Investigation of Ford Motor Company Variable
                 Cut-Off Mechanism	   23
          4.2.3  Cam Shaft Drive	   24
          4.2.4  Valve and Valve Stem Seal	   24
          4.2.5  Piston Rings	   24

     4.3  Expander Description	   25

          4.3.1  Crosshead Piston Expander 	  	   25

                 4.3.1.1  Expander Crankcase 	   27
                 4.3.1.2  Expander Lube System 	   30
                 4.3.1.3  Expander Component Hardware	   32

          4.3.2  Trunk Piston Expander"	   32

     4.4  Expander Performance Analysis	   45

          4.4.1  Expander Breathing Analysis 	   45
          4.4.2  Steady-State Temperature Distribution 	   53
          4.4.3  Stress Analysis	   64

                 4.4.3.1  Crosshead Piston Expander	   64
                 4.4.3.2  Trunk Piston Expander	   65
            -•                       ^
          4.4.4  Piston Ring Wear Prediction	   65

                 4.4.4.1  Crosshead Power Piston Graphite Rings.   65
                                  iii

-------
                           TABLE OF CONTENTS


                                                                  Page


                 4.4.4.2  Trunk Expander Piston Rider Ring. ...  75

          4.4.5  Recompression Valve Dynamics	78
          4.4.6  Crosshead Expander Power Piston Rod Natural
                 Frequency	79

5.0  EXPANDER MATERIALS	83

     5.1  Materials Technology Base	83

          5.1.1  Cylinder Block, Cylinder Head, Intake Manifold,
                 Exhaust Manifold	84
          5.1.2  Cylinder Liners/Piston Rings 	  88
          5.1.3  Piston	91
          5.1.4  Piston Pin	91
          5.1.5  Piston Pin Bearing	92
          5.1.6  Connecting Rod	94
          5.1.7  Camshaft/Cam/Cam Follower (Tappet)	94
          5.1.8  Inlet Valve	99

     5.2  Expander Materials Selection Study	103

          5.2.1  Material Properties	106

                 5.2.1.1  Low Carbon and Low Alloy Steels .... 106
                 5.2.1.2  Medium Carbon Low Alloy Steels	Ill
                 5.2.1.3  Cast Iron	Ill
                 5.2.1.4  Aluminum Alloys	116
                 5.2.1.5  Case Hardened Low Alloy Steels	120

     5.3  Materials Recommendations	121

6.0  LUBRICATION	125

     6.1  Lubrication Technology Base - Solid Lubricants	125

          6.1.1  Criteria for Solid Lubrication 	 126
          6.1.2  Forms and Types of Solid Lubricants	129

                 6.1.2.1  Self-Lubricating Solids 	 131

                          6.1.2.1.1  Carbon-Graphites 	 131
                          6.1.2.1.2  Polytetrafluoroethylene
                                     (PTFE)	137
                          6.1.2.1.3  Metals	139
                          6.1.2.1.4  Sulfides/Selenides 	 142
                          6.1.2.1.5  Porous Metal Composites. .  . 144
                                   iv

-------
                           TABLE OF CONTENTS

                                                                  Page


                          6.1.2.1.6  Hard Surfacing Materials . . 145
                          6.1.2.1.7  Mixed Composites 	 147

          6.1.3  Bonded Solid Film Lubricants .  .  . . ,	147

     6.2  Lubrication Technology Base - Liquid Lubricants .... 150
     6.3  Lubricant Recommendations	159

          6.3.1  Solid Lubrication	159
          6.3.2  Liquid Lubrication 	 162

7.0  TEST FACILITY	163

     7.1  Facility Description	163
     7.2  Expander Instrumentation	171

8.0  TEST RESULTS-CROSSHEAD PISTON EXPANDER	173

     8.1  Component Performance(s)	173

          8.1.1  Camshaft/Valve Lifter (Tappet)  	 173
          8.1.2  Inlet Steam Valve/Housing	174
          8.1.3  Recompression Valve	175
          8.1.4  Piston Rings/Cylinder Liner	177
          8.1.5  Power Piston Head	177
          8.1.6  Power Piston Rod Seal	182
          8.1.7  Crosshead Piston 	 182
          8.1.8  Wrist Pin Bushing	185
          8.1.9  Other Components	185
          8.1.10 Crankcase Lubricant	185

     8.2  Thermodynamic Performance 	 187

          8.2.1  Thermodynamic Performance - Graphite CC-5A
                 Piston Rings 	  187
          8.2.2  Thermodynamic Performance - Cr3C2 Coated Inconel
                 X-750 Piston Rings	199

9.0  TEST RESULTS - TRUNK PISTON EXPANDER	208

     9.1  Component Performance	208
     9.2  Thermodynamic Performance 	 213

REFERENCES	221

-------
                         LIST OF ILLUSTRATIONS

Figure No.                                                        Page

 4.2-1    Valve - Variable Cut Off Mechanism	16
 4.2-2    Valve - Variable Cut Off Mechanism	   17
 4.2-3    Cam Shaft Drive, Head, Valve, and Cam Details 	   18
 4.2-4    Cam Shaft Drive, Steam Expander Trunk Type Piston ...   19
 4.2-5    Piston Ring Packing - Sleeve Valve Steam Expander -
          Crosshead Type Piston	20
 4.2-6    Valve Sleeve Packing - Cutoff Valve Steam Expander -
          Crosshead Type Piston	21
 4.3.1 .   Crosshead Piston Expander	26
 4.3-2    Waukesha CFR-48 Crankcase Assembly (Source Reference:
          Waukesha Motor Company Bulletin 850M) 	   28
 4.3-3    Waukesha Crankcase Showing Modified Gear Train	29
 4.3-4    Steam Expander Crankcase, Modification Details	31
 4.3-5    Power Piston for Carbon-Graphite Rings	33
 4.3-6    Pressure Balanced Carbon-Graphite Piston Ring 	   34
 4.3-7    Cylinder Liner.	35
 4.3-8    Power Push Rod	36
 4.3-9    Aluminum Crosshead	37
 4.3-10   Wrist Pin and Bushing	38
 4.3-11   Steam Inlet Valve Assembly	39
 4.3-12   Cam Tappet Assembly	40
 4.3-13   Connecting Rod	r .   41
 4.3-14   Camshaft	42
 4.3-15   Assembly of Trunk Piston Expander . . .	 .   43
 4.3-16   Assembled Trunk Piston Expander 	 .   44
 4.3-17   Trunk Piston and Rod Assembly, Cr C  Rings	46
 4.3-18   Piston For Trunk Expander (Upper Portion) 	   47
 4.3-19   Piston For Trunk Expander (Lower Portion) 	   48
 4.3-20   Trunk Piston Oil Exclusion Ring	   49
 4.3-21   Trunk Expander Piston Assembly	50
 4.3-22   Trunk Piston and Rod Assembly (Graphite Ringe). ....   51
 4.4-1    Expander Indicator Diagram, 2000 RPM, 10° Valve Lead. .   54
 4.4-2    Expander Indicator Diagram, 2500 RPM, 20° Valve Lead. .   55
 4.4-3    Theoretical Indicator Diagrams Single Cylinder Expander  56
                                  vi

-------
                         LIST OF ILLUSTRATIONS


Figure No.                                                        Page

 4.4-4    Calculated Temperature Profile for Crosshead Piston
          Expander	   62

 4.4-5    Calculated Temperature Profile for Trunk Piston
          Expander	   63

 4.4-6    Location of Stresses, Crosshead Expander 	   68

 4.4-7    Critical Stress Locations for Trunk Expander 	   70

 4.4-8    Graphite Ring Geometry and Load Components
          (lb/3.5ir in)	   73

 4.4-9    Maximum Allowable Stress Concentration Factor and
          Fractional Area with Zero Wear of Graphite Piston Rings
          in 1000 Hr. Mean PI = 290 psia - upstream pressure on
          piston ring.  AP = pressure drop across piston ring.  .   74

 4.4-10   Maximum Allowable Stress Concentration Factor and
          Fractional Area With Zero Wear of Graphite Trunk
          Piston Rider Ring in 1000 Hours	   77

 4.4-11   Examples of Recompression Valve Dynamic Calculations
          for Inlet Pressure of 1000 PSIA and Different Rotative
          Speeds and Condenser Pressure, Pc.   All Solutions Shown
          are Acceptable Except for 1000 RPM, where Pc = 1000
          PSIA	   80

 4.4-12   Maximum Allowable Condenser Pressure for Inlet Pressure
          of 1000 PSIA and 78 Ib. Recompression Valve Spring
          Preload	   81

 5.2-1    Maximum Allowable Design Stresses for Low Carbon, Low
          Alloy and Austenitic Stainless Steels in Large Steam
          Generating Systems 	  107

 5.2-2    Effect of Temperature on Long Time Steam Corrosion
          Rates	110

 5.2-3    Effect of Long Time Exposure to Steam at 1100°F.  .  .  .  110

 5.2-4    0.2% Yield Strength Medium Carbon Low Alloy Steels .  .  113

 6.1-1    Free Energy of Oxidation Reactions Involving Graphite
          and Steam	133

 7.1-1    Steam Expander Test Facility Schematic 	  164

 7.1-2    Single Cylinder Expander Test Facility 	  167

 7.1-3    Single Cylinder Expander Test Facility 	  168

 7.1-4    Installation of Crosshead Expander	169

 7.1-5    Installation of Crosshead Expander 	  170
                                 vii

-------
                         LIST OF ILLUSTRATIONS
Figure No.                                                        Page

 8.1-1    Inlet Valve Stem Steam Leakage Rage	176
 8.1-2    Piston Ring Wear at Middle of Segments (Carbon Graphite
          Grade CC5A)	179
 8.1-3    Crosshead Expander Carbon Graphite Rings (CC-5A) After
          52.4 Hours on Test	180
 8.1-4    Crosshead Expander Carbon Graphite Rings (CC-5A) After
          241.9 Hours on Test	181
 8.1-5    Oil Concentration in Steam Condensate	183
 8.1-6    Crosshead Expander Cr3C2 Coated Inconel-X Rings After
          Approximately 10 Hours of Testing	188
 8.2-1    Crosshead Expander Performance	190
 8.2-2    Crosshead Expander Performance	191
 8.2-3    Crosshead Expander Performance.	 193
 8.2-4    Crosshead Expander Performance	194
 8.2-5    Crosshead Expander Typical P-V Diagram	195
 8.2-6    Oscilloscope Photos 	 196
 8.2-7    Single Cylinder Crosshead Expander - 1500 RPM 	 197
 8.2-8    Single Cylinder Crosshead Expander - 1500 RPM	198
 8.2-9    Single Cylinder Crosshead Expander - 1500 RPM	200
 8.2-10   Measured Temperature Distribution of Crosshead
          Expander	201
 8.2-11   Comparison of Crosshead Expander Specific Steam Con-
          sumption with Graphite and with Cr C  Coated Inconel-X
          Piston Rings		202
 8.2-12   Comparison of Crosshead Expander Power Production with
          Graphite and with Cr3C_ Coated Inconel-X Piston Rings  . 203
 8.2-13   Comparison of Crosshead Expander BMEP with Graphite
          and with Cr_C  Coated Inconel-X Piston Rings	204
 8.2-14   Comparison of Crosshead Expander P-V Diagrams with
          Graphite and with Cr3C2 Coated Piston Rings at 1500 RPM
          and Similar Steam Inlet Conditions	205
 8.2-15   Comparison of Crosshead Expander P-V Diagrams with
          Graphite and with Cr3C2 Coated Piston Rings at 500 RPM
          and Similar Inlet Steam Conditions	206
 9.1-1    Trunk Piston Liner Following 9.2 Hours Test 	 210
 9.1-2    Trunk Piston Skirt Following 9.2 Hours Test 	 211
                                   viii

-------
                         LIST OF ILLUSTRATIONS
Figure No.
 9.1-3    Cam Lobe From Trunk Piston Expander	212

 9.2-1    Trunk and Crosshead Expander Brake Mean Effective
          Pressure As A Function of Rotational Speed, Inlet
          Steam Conditions, and Piston Ring/Cylinder Liner
          Materials	215
 9.2-2    Trunk and Crosshead Expander Brake Horsepower As A
          Function of Rotational Speed, Inlet Steam Conditions,
          and Piston Ring Material	216

 9.2-3    Trunk and Crosshead Expander Brake Specific Steam
          Consumption As A Function of Rotational Speed and Piston
          Ring Material for Steam Inlet Conditions of Approxi-
          mately 400 psia, 700°F	217

 9.2-4    Trunk and Crosshead Expander Brake Specific Steam
          Consumption As A Function of Rotational Speed and Piston
          Ring Material for Steam Inlet Conditions of Approxi-
          mately 400 psia, 1000°F	218

 9.2-5    Trunk and Crosshead Expander Brake Specific Steam
          Consumption As A Function of Rotational Speed and
          Piston Ring Material for Steam Inlet Conditions  of
          Approximately 700 psia, 700°F 	 219

 9.2-6    Crosshead and Trunk Expander Brake Specific Steam
          Consumption As A Function of Testing Time.   Piston
          Ring Material was CrgC2 Coated Inconel X-750 and
          Cylinder Material was Type 440C Stainless Steel  .... 220
                                   ix

-------
                            LIST OF TABLES ,

Table No.                                                         Page

4.4-1     Expander Parameters for Indicator Diagrams	57
4.4-II    Effect of Speed on Indicated Expander Performance ...  58
4.4-III   Stresses and Factors of Safety at Principal Points Of
          Interest in the Crosshead Expander	66
4.4-IV    Stresses and Factors of Safety at  Principal Points Of
          Interest in the Trunk Expander	69
4.4-V     Fractional Parameters 	  71
4.4.VI    Properties of Grade 107 Carbon Graphite 	  71
5.1-1     Plain Carbon and Low Alloy Steels Used for Pressure
          Containing Components in Steam Generating Plants. ...  86
5.1-II    Cylinder Liner/Piston Ring Material Combinations for
          Heavy Duty Diesel Engines	90
5.1-III   Bearing Materials Used in Internal Combustion Engines .  95
5.1-IV    Materials Used for Inlet and Exhaust Valves in
          Internal Combustion Engines 	 101
5.1-V     Materials Used for Valve Inserts and Hard Facing
          Applications	104
5.2-1     Creep Rupture Properties of Cr-Mo Alloy Steels at
          1000°F	109
5.2-II    Oxidation-Corrosion of Cr-Mo Steels at 1100°F 	112
5.2-III   Dimensional Growth of Cast Irons in High Temperature
          Steam	115
5.2-IV    Nominal Compositions of Aluminum Alloys	117
5.2-V     Room Temperature Tensile Properties of High Silicon
          Aluminum Alloys	118
5.2-VI    Typical Mechanical Properties for Aluminum Alloy
          Forgings (4032, 2014 and 2219 in T6 Condition	119
5.2-VII   Materials Recommendations for Single Cylinder Steam
          Expanders	122
6.1-1     Properties of Carbon-Graphite Grades	135
6.1-II    Catalytic Activity of Oxides in Graphite Oxidation. . . 136
6.1-III   Soft Metals Used in Bearings	140
6.1-IV    Chemical Composition and Hardness of BISHIRALLOY. . . . 141
6.1-V     Properties of Clevite 300	. . 143
6.1-VI    Chemical Composition of LP Alloys	147

-------
                            LIST OF TABLES

Table No.                                                         Page

6.2-1     Properties of Experimental Synthetic Hydrocarbon
          Lubricant	156
6.2-II    Performance of Selected Types  of  Synthetic Oils  ....  158
6.3-1     Candidate Piston Ring/Cylinder Liner Combinations  .  .  .  161
8.1-1     Piston Ring Wear	178
8.1-II    Water Leakage into Crankcase	184
8.1-III   Change in Properties  of XRN-1301C Oil After 187.7  Hours
          of Engine Operation 	  186
8.2-1     Crosshead Expander Test Data - May 10-June 7, 1972.  .  .  189
                                 xi

-------
                    1.0   INTRODUCTION

      The water base Rankine cycle powerplant with a reciprocating ex-
pander has been identified as one of several possible low emission al-
ternatives to the internal combustion engine for automotive application.
To achieve the performance necessary to make the automotive steam engine
a truly viable alternative requires temperatures and pressures in the
order of 1000°F and 1000 psia respectively.   At these elevated operating
conditions conventional methods of lubricating the upper cylinder walls
and inlet valves of a reciprocating expander can pose difficult problems.
In an effort to find a solution to the lubrication problem,  the Advanced
Automotive Power Systems Development Division of the Environmental
Protection Agency sponsored a program with the General Electric Company,
Space Division, to analytically and experimentally evaluate  self-lubricat-
ing materials suitable for the automotive steam engine application.

      This program included the design of practical single cylinder  re-
ciprocating expanders for the purpose of evaluating solid lubricants and
other supporting materials for use in a Rankine cycle automotive pro-
pulsion system utilizing a water base fluid.   This investigation had the
following objectives:

      •  Establish preliminary design of trunk piston and crosshead
         piston single cylinder expanders.
      •  Evaluate expander materials technology for higher temperature
         operation with solid lubricants.
      •  Evaluate solid and liquid lubricants for reciprocating expanders.
      •  Build and test one trunk piston expander and one crosshead
         piston expander at conditions representative of the engine  system.

      Following the design and fabrication of two single cylinder recipro-
cating expanders - a crosshead piston expander and a trunk piston expander
                                   1

-------
these expanders with different combinations of ring, cylinder and valve
stem ring materials were tested at pressures and temperatures up to 1000
psia and 1000°F, respectively.  Piston ring materials that were evaluated
were graphite impregnated with antimony (Carbone Co. Grade CC5A) and a
Cr3C  cermet (Koppers Co. K-1051) coated Inconel X-750.  Both of these
materials were tested with a hardened Type 440-C stainless steel cylinder
liner with no liquid lubrication.  In addition to the power piston, the
expander inlet steam valve required unlubricated valve stem guides and
seals.  Seal rings for the valve stem were made of cold worked Haynes
alloy No. 25 and some were made of 17-7PH coated with DuPont LPA-101
hard facing.  These valve stem rings rubbed against the valve housing
which was a cast Ni-Resist Type 3D.  The valve and valve stem were ni-
trided H-ll tool steel.

      For the trunk piston expander a rider ring was provided to support
the high piston side loads.  The rider ring was a split ring design made
of Pure Carbon Company graphite Grade P5NR.  Also, two oil exclusion
rings and a steam seal ring were provided at the bottom of the trunk
piston.  These were standard rings made of K-35 Ni-Resist, K Iron and
K-6E Iron.

      The expander crankcase was liquid lubricated in a conventional
manner of splash and pressurization.  However, since the expander was
expected to have hot spot temperatures exceeding 400°F, and since some
steam blowby was expected to accumulate in the crankcase, a specially
compounded synthetic hydrocarbon oil (Grade XRN-1301-C) supplied by the
Mobil Research and Development Corporation was used and evaluated.  This
lubricant contained oxidation, rust, wear, and foam inhibitors.

      In order to get meaningful results on piston ring and cylinder
wear and friction, it was necessary to exclude essentially all oil from
the power cylinder.  In  the crosshead expander this was not too difficult
following some minor design changes, but the conventional metallic oil
exclusion rings for the  trunk expander were not very effective in ex-
cluding oil from the steam or steam from the crankcase oil.  Chevron ring
seals made of PTFE impregnated with bronze and MoS_, mounted in a floating
housing to minimize radial loads on the seal, were successful in excluding
oil from the steam to less than 4 ppm In the crosshead expander.  Water

-------
accumulation in the crankcase for the crosshead expander averaged 8.1 m/hr
a possible major leakage path being the inlet valve stem seal.  Testing of
the trunk piston expander was of short duration (approximately 9.2 hours)
and accurate data on the rate of oil migration into the steam or expander
condensate was not obtained.  However, the rate was significantly higher
than for the crosshead expander.

      Of the few materials tested, the best results were obtained with
graphite (CC5A) rings rubbing against a hardened Type 440-C cylinder liner
in the crosshead expander.  For the first 200 hours of testing the total
radial ring wear was approximately 0.040 inches - giving a wear rate of
0.0002 inch/hour.  At 240 hours total wear was sufficient to cause ring
breakage.  Wear performance of the Cr_C_ coated Inconel X-750 rings  in
a Type 440-C liner was much less impressive.   Ring wear was rapid which
resulted in ring breakage in less than ten hours of testing.

      Even with the graphite rings, ring/liner friction and steam blowby
were higher than predicted.  At 1000 psia, 1000°F steam inlet conditions
theoretically indicated efficiency was calculated to be 85%, at 2000 RPM.
Measurements for these conditions gave an engine efficiency of 55%.

      Measurement of cylinder pressure as a function of crankangle proved
to be a challenging task.  Several types and makes of pressure trans-
ducers were tried with only limited success.   Best results were obtained
with a water cooled, inert gas buffered Dynisco Model PT49A pressure
transducer.

-------
                         2.0   ABSTRACT

      A reciprocating expander has been identified as one of several
possible types of expanders which may be applicable to an automotive
Rankine cycle engine.  For high engine efficiency, steam pressure and
temperature up to 1000 psia and 1000°F,respectively, are necessary.  One
approach to steam cylinder lubrication is the application of solid lubri-
cants which are capable of withstanding these high pressures and tempera-
tures without excessive friction and wear.

      Several solid lubricants and wear resistant materials were tested
in a single cylinder steam expander of both the crosshead piston and
trunk piston configuration.  Also a specially compounded water resistant
synthetic hydrocarbon oil was evaluated as a crankcase bearing lubricant.
Both the crosshead piston and trunk piston expanders were fabricated and
tested over a range of conditions depicted as follows:

                  Speed Range, 500-2000 RPM
                  Inlet Steam Temperature, 700-1000°F
                  Inlet Steam Pressure, 400-1000 psia
                  Condenser pressure, *v» 20 psia

      The crosshead piston expander was first tested for a total of 242
hours with antimony impregnated carbon-graphite rings (Grade CC5A) rubbing
against a hardened Type 440-C stainless steel cylinder liner.  Expander
performance was generally as predicted by earlier analysis with the ex-
ception of higher blowby than anticipated.  The carbon-graphite ring
wear was excessive for a projected 3000-hour life.  A second crosshead
piston expander buildup incorporating Cr_C_ coated Inconel^750 rings and
a Type 440-C liner was tested and after six hours of operation at tem-
peratures and pressures of 1000°F and 700 psia respectively, serious
blowby was evident.  Inspection of the ring and liner surfaces revealed
excessive roughness and the test was terminated.

-------
      The trunk piston expander was assembled with Cr-jC- coated Inconel  X-750
sealing rings, a carbon-graphite (Grade P5NR) rider ring, and cast  iron  oil
exclusion rings at the bottom of the piston.  The cylinder liner was
hardened Type 440-C stainless steel.  Performance of the trunk piston
expander was not as good as the crosshead piston expander in terms of
shaft horsepower output, brake specific steam consumption, etc.  After
approximately 9.2 hours of testing, failure of a crankshaft bearing
terminated the test.  Piston ring and cylinder liner scoring and wear
was excessive.

      Generally, the test results indicate that solid lubricants are
potentially applicable as ring, cylinder, and other components in a
high temperature reciprocating expander.   Although those few materials
tested under this program exhibited relatively short life and no
completely satisfactory combination of wear resistant and solid lubricant
materials were identified, the very limited number of tests  conducted
are not sufficient to rule out a non-liquid lubricated reciprocating
steam expander.   The synthetic hydrocarbon crankcase lubricant performed
well through all the tests providing adequate lubrication at bulk oil
temperatures up to 230°F.

-------
 3.0   CONCLUSIONS   AND   RECOMMENDATIONS

      To some extent it is difficult to separate design from material success
or problems.  However, from a pure design point of view both the crosshead
piston and trunk piston expanders were designed and proved to be sturdy test
vehicles for steam expander component testing.  The very rugged Waukesha
CFR-48 crankcase proved to be reliable during all phases of testing except
when a main bearing was starved of oil.  (This was no fault of the crankcase.)

      In both expander types the heavily loaded valve cam was marginal for
long time operation at high speed.  Also, the crosshead piston wrist pin
was somewhat marginal for long time operation.  However, neither component
presented a real operational problem following minor corrections.  The power
piston connecting rod seal for the crosshead expander required early
modifications but later proved to be trouble free.

      Testing of the trunk expander was short and was terminated following a
bearing seizure.  However, during less than ten hours of operation it was
evident that oil was not being excluded from the steam system and vice versa.

      Both expanders were well balanced dynamically as indicated by low
vibration.  However, expander noise level was considered high - mostly valve
gear noise.  From a mechanical design point of view the program objectives
were generally met with good success.  However, testing time was too short
to uncover possible long term deficiencies in the designs.

      Design improvements can be made to any reciprocating steam expander
depending on the ground rules or limitations set forth.  For an unlubricated
piston and cylinder the problem becomes one of identifying the right materials
for sliding components, or of designing away from sliding (non-contacting)
components in regions where lubricants are not viable.  Studies have shown
that a properly designed labyrinth sealed piston, which would not contact

-------
the cylinder wall, might give performance as good or better  than  a piston
fitted with graphite rings.  When properly supported the labyrinth piston
would not be subjected to wear.  Such piston designs have been used  in
steam engines and gas compressors with good success.

      Under the test conditions of 1000°F, 1000 psi steam, the carbon-
graphite piston rings (Grade CC-5A) rubbing against a hardened Type 440C  SS
cylinder liner showed far superior performance to the hard-hard materials
combination of Cr-C- cermet against Type 440C SS.  Although the wear rate
of the upper carbon-graphite ring was too high (0.070 inch in 242 hours)
for a 3000 hour life, the wear rate of the lower ring appears to be acceptable.
It is recommended that further study be carried out to redesign the rings
in an attempt to lower the unit loading and to evaluate other carbon-
graphite materials that may have improved wear resistance.   For example,
it is known that carbon-graphites recently developed for the Wankel engine
tip seal are compatible in steam and have superior wear resistance to the
CC-5A grade.  Also, alternate design configurations could greatly
improve the performance of a solid lubricated system.   For  example,  making
the cylinder of a carbon graphite instead of the piston rings provides
much more solid lubricant surface - thus significantly increasing the life
of the system.

      The performance of the work hardened Haynes alloy No.  25 valve  stem
rings against the cast Ni-Resist D-3 valve housing was  relatively good;  the
LPA-101 hard facing coating on the rings is not suitable for the steam seal
application.  Improvement in wear of the housing and steam  leakage through
the seal may be possible through the use of a wear-in coating to be  applied
either to the ring surfaces or the ID surface of the Ni-Resist housing.
Other material combinations that may provide improved performance are
1) carbon-graphite housing vs ringless stem,  2) Cr^O, coated or solid
cast LPA-101 rings vs nitrided steel housing.

      The 15% bronze + 5% MoS2 filled PTFE piston rod seal  in the crosshead
expander performed flawlessly.  No measurable wear was observed and oil
leakage into the steam condensate was maintained at values  of <4 ppm.

-------
      The crankcase lubricant, Mobil XRN-1301C, provided adequate
lubrication to all the bearing surfaces and the cam lobe/tappet interface
for a period of 187 hours.  Water concentration as high as 1.3% and
bulk oil temperatures of 250°F did not affect the lubricant and there
was no evidence of serious oxidation or degradation of any kind of the
lubricant.
                                    8

-------
                 4.0   EXPANDER   DESIGN

4.1   Expander Design Technology Base

      In conformity with contract guidelines, the reciprocating expander
configuration was selected to be that of a single acting, uniflow engine.
Two single cylinder expanders were to be designed, built, and tested -
one a crosshead piston type and the other a trunk piston type.  Steam in-
let conditions were to be 1000°F, 1000 psia with an exhaust pressure of
20 psia.  These expanders were to be representative of one cylinder of a
four cylinder engine of approximately 150 hp output, suitable for auto-
motive application.  A governing requirement of these designs was that
solid lubricants be employed for the piston rings, piston rod packing
(where employed), and valve stem seals.  Such use of solid lubricants
was intended to initiate the development of an engine free of lubricating
oil contamination in the steam.

      The task of establishing the technology base for the design of the
two expanders was implemented by making the following investigations:

      1.  High speed steam engine valving configurations, and valve gear
          designs.
      2.  Wrist pin bearing design suitable for high unidirectional loading.
      3.  Oil free high temperature piston ring design.
      4.  Oil free valve stem sealing, friction, and wear.
      5.  Piston rod sealing.
      6.  General reciprocating engine mechanical design.

      In close coordination with these investigations corresponding ma-
terials selection investigations were carried on.  These are described
in Section 5.0.

      Salient results and conclusions of these investigations are briefly
discussed below.
                                   9

-------
      4.1.1   High Speed Steam Engine Valving Configurations and Valve
              Gear Designs

      Valving configurations which were identified as being of interest
for preliminary design studies include the following:

      1.  Uniflow engine with spring loaded recompression relief exhaust
          valve, simple exhaust porting and engine actuated inlet valve,
          either in the cylinder head or in a side steam inlet passage.

      2.  "Zero" clearance volume engine with exhaust valve and inlet
          valve in the cylinder head.

      Valve types of interest include:  (a) poppet valves of several
varieties, (b) sleeve valves, and (c) rotary valves.  Sleeve and rotary
valves minimize or eliminate the high pressure loading on the actuating
mechanism.  Rotary valves in particular are of interest because they
offer potential capability for a simple variable cut-off mechanism.  How-
ever, because of the formidable sealing problems of these valve types,
it was decided that such an approach involved too much technical risk
for the present program.  Investigative effort was concentrated on un-
balanced and partially balanced cam actuated poppet valves.  Principal
design problems were identified to be high Hertz stress between cam and
cam follower for an unbalanced poppet valve approach, and that of steam
leakage for various types of partially balanced poppet valves.  It was
recognized that valve train accelerations are quite high because the cam
shaft must operate at engine speed as opposed to half engine speed for
four cycle internal combustion engines.  However, it was found that the
valve train forces, at least for an unbalanced poppet valve, are dominated
by high pressure steam loading forces rather than acceleration forces.
Although steam forces produce a high Hertz stress, dynamic design of the
valve train is somewhat simplified.

      One technique for expander load control for a four cylinder engine
is that of inlet valve variable cut-off which must cover a range from
near zero steam admission to a maximum value determined by the limit of
the steam supply.  The status of high speed variable cut-off technology
was reviewed, and the following possible alternative design approaches
were identified.
                                   10

-------
1.  The Caprotti mechanism, which includes a single poppet valve
    actuated by a yoke bearing on two variably phased cams.   (This
    mechanism is successfully employed in Skinner steam engines.)
    For a high speed application, requiring a wide range of cut-
    off, very high valve train accelerations are involved.

2.  A hydraulic valve actuator operating on the principle of the
    Bosch injector has been successfully employed for achieving
    variable valve period by the Waukesha Motor Company and others.
    For a high speed steam expander requiring a camshaft running
    at full engine speed, however, it appeared doubtful that such
    a system has a fast enough response.   This conclusion was sup-
    ported by a communication with American Bosch Corporation.
    However, it was learned that an advanced electro hydraulic
    variable cut-off design for high speed steam engines is under
    development by American Bosch for Thermo Electron Corporation.

3.  A variable fulcrum type of overhead valve actuation mechanism
    has been invented and developed by the Ford Motor Company.
    Preliminary design studies of several forms of this device
    were made with the cooperation of the Ford Motor Company.
    Various problems relating to overall  expander height,  and
    lubrication of the variable fulcrum mechanism resulted in a
    decision not to pursue this approach.   A further complication
    is the fact that this approach to variable cut-off requires a
    mechanism to vary the phase of the cam relative to the crank-
    shaft.

4.  A rotary valve mechanism utilizing a  rotatable cut-off sleeve
    between the rotating valve and the valve housing was reviewed.
    As mentioned above, the formidable sealing problems of this
    approach required extensive development effort - not consistent
    with the current contract.

5.  Two poppet valves in series, variably phased to provide a change
    in the period of overlap between opening of each valve have the
    following advantages:
                            11

-------
         -a.  Long cam event angles can be employed.  This minimizes the
              problems of cam design - permitting reasonable acceleration,
              high lift, and reasonably small cam base circle diameter.

          b.  Full cut-off range (zero to any desired maximum) can be
              readily achieved.

          Problems are also recognized for this approach as follows:

          a.  Because of the flow resistance of two valves in series the
              valves must be larger or employ higher lift than if a single
              valve is used.  Large valves result in high valve train
              loads unless balanced valves are used.  Very high lift tends
              to require long cam events, large cam base diameter, high
              cam rubbing velocity and high valve train acceleration.

          b.  With two series valves, it is difficult to obtain the de-
              sired 5% clearance volume in a uniflow type expander.

          For two poppet valves in series two alternate cam phasing
          mechanisms were identified:

          a.  One involving the use of helical splines, and

          b.  The other involving a planetary gear train between crank-
              shaft and camshaft with phasing variation achieved by ro-
              tating the planet carrier.

      After evaluating the above alternate approaches to variable cut-
off, it was concluded that the two series poppet valves had the lesser
development risks.  However, for the single cylinder test engines a
single poppet valve with fixed cut-off was chosen for simplicity since
the primary objective was to evaluate solid lubricants.

      4.1.2  Wrist Pin Bearing Design for High Unidirectional Loading

      The high MEP of the steam expander under long cut-off, and the fact
that the wrist pin load does not reverse direction (does not facilitate
      "<-**
squeeze film lubrication of the wrist pin bearing) required that special
attention be given to the wrist pin design.  Applicable technology is

                                  12

-------
found in the design of wrist pin bearings for two cycle Diesel  engines.
Several approaches are as follows:

      1.  Employment of needle type antifriction bearings.

      2.  Employment of a palm type bearing which provides for  relatively
          low pressure loading over a large area.

      3.  Use of a helically grooved bushing in the connecting  rod with
          lubricating oil force fed to the grooves, the wrist pin being
          fixed in the piston.  (Pitch of the grooves must be less than
          the amplitude of the rocking motion so that the motion drags
          oil from the grooves over the lands of the bearing.)

      4.  Use of helically grooved bushings in the piston with the piston
          pin fixed to the connecting rod.

     Consultation with the General Electric Diesel engine design group
resulted in the selection of approach No. 4 above.  Unit loading in the
order of 4000 psi were considered to be safe, since loadings in excess
of 5000 psi have been successfully used in two cycle Diesel engines.

      4.1.3  Oil Free High Temperature Piston Ring Design

      Friction, wear, and sealing effectiveness of oil free piston rings
are the most important questions toward which this experimental program
is directed.  Basic design parameters of interest are the following:

      1.  Number of rings employed for 1000 psi sealing pressure
      2.  Width of rings
      3.  Use of pressure balancing to reduce normal force and resulting
          friction and wear.
      4.  Segmentation of rings and sealing between segments
      5.  Design of springs for preloading segmented rings
      6.  Provision for positive sealing behind the ring
      7.  Rubbing velocity.

      An extensive survey was made in the literature and by contact with
manufacturers and users of carbon and other types of high temperature
oil free rings.  No directly comparable application was identified.  The

                                  13

-------
most pertinent experience was that with high pressure oil free compressor
piston rings.  Recommendations for number of rings, and width of rings
varied widely.  Values selected do not represent the most conservative
practice, but are the result of judgments which balance design for low
friction and wear against the requirements of an acceptably simple and
light weight design.  In references 1, 2, and 3, some bases for analytical
prediction of ring life were found, and the following general design
features were applied:

      1.  For non metallic rings (graphite) two piece segmented construc-
          tion was applied with a one piece spring backup.  For metallic
          rings (coated) one piece construction was used.

      2.  Ring widths were approximately .3" for non metallic and .125"
          for coated metallic rings.  Three non metallic compression
          rings were used for the crosshead design and two for the trunk
          design (because of smaller available space).  For an alternate
          backup ring design, three coated metallic rings were used for
          both crosshead and trunk designs.

      3.  For non metallic rings, in which .060" to .080" radial wear
          was anticipated, holes were provided in the top edge of the
          ring groove to assure access of sealing pressure behind the
          ring, thus preventing ring flutter.

      4.  Pressure balancing was provided with the non metallic rings
          to reduce average normal pressure, friction, and wear.

      4.1.4  Piston Rod Sealing

      "In the crosshead piston design, the steam piston rod must seal the
condenser steam pressure (20 psia) against atmospheric pressure, and also
prevent the transport of oil into the steam.  Metal temperature due to
steam contact was reasonably low (approximately 300°F),  Seal friction,
however, may produce local high temperature.  Some cooling was provided
through the use of a hollow piston rod into which oil was sprayed.  Two
alternate types of seals were identified for such an application:

      1.  A Chevron packing made of bronze filled PTFE
                                  14

-------
      2.  A segmented carbon ring packing held in place by garter springs.

      4.1.5  Oil Free Valve Stem Sealing

      Two approaches were selected for valve stem sealing:

      1.  The use of multiple precision metallic rings for sealing be-
          tween the valve stem and the valve stem guide housing.

      2.  The use of carbon ring packing in a packing gland surrounding
          the valve stem.

      Since assurance of success seemed high with (1), based on General
Electric steam turbine valve sealing experience, this approach was selected
for the test expanders.

4.2   Expander Preliminary Design Studies

      Prior to the selection of detail features for the two single cylinder
test expanders, studies were conducted in the following areas:

      1.  General cylinder configuration, cylinder sizing, and expander
          speed range.
      2.  Investigation of Ford Motor Company variable cut-off mechanism
      3.  Cam shaft drive configuration
      4.  Valve and valve stem seal configuration
      5.  Wrist pin bearing configuration
      6.  Piston rod packing configuration
      7.  Piston ring configuration for crosshead and trunk piston de-
          signs.

      These studies are shown in Figures 4.2-1 through 4.2-6, and are
briefly described below.

      4.2.1  General Cylinder Configuration Selection

      The principal alternatives with regard to the general cylinder con-
figuration are:

      1.  A near zero clearance volume, valve-in-head design, with both
          inlet and exhaust valves located in the cylinder head.

                                   15

-------
Figure 4.2-1. Valve - Variable Cut Off Mechanism.

-------
Figure 4.2-2.  Valve - Variable Cut Off Mechanism.

-------
00
                           Figure 4.2-3. Cam Shaft Drive, Head, Valve, and Cam Details  (GE Dwg.  707E702).

-------
Figure 4.2-4 Cam Shaft Drive, Steam Expander Trunk Type Piston (GE Dwg. 707E719, Sh. 2)




                                          19

-------
NJ
O
                                                                                       WRING PlU
                                                                  66M. DESIfiH
                                - t.Tt*« IW.ET
                    Figure 4.2-5.  Piston Ring Packing
                                    (GE Dwg. 707E705).
- Sleeve  Valve Steam Expander - Crosshead Type Piston

-------
ro
                               Z3TEAM IMU.T
                                                                   5TCAH
Figure 4.2-6. Valve  Sleeve Packing
              (GE Dwg.  707E710).
                                                        - Cutoff Valve Steam Expander - Crosshead Type Piston

-------
      2.   A uniflow engine configuration with exhaust ports opening at
          approximately 90% of stroke and with an inlet valve located
          either in the head or in a side inlet passage.

      Advantages and disadvantages of these alternatives can be summarized
as follows:

      Zero Clearance Volume Design Advantages

      1.   With the exhaust valve open throughout the upward stroke of
          the piston, work of recompression can be virtually eliminated.
          This results in a higher MEP and a smaller cylinder displace-
          ment for a given power output.

      2.   Expansion can extend to the end of the downward stroke of the
          piston - thus increasing the effective expansion ratio.  This
          is a factor favorable to high efficiency.

      Disadvantages

      1.   Cooling of the upper part of the cylinder occurs during steam
                        «
          exhaust.  This results in "condensation loss" during the sub-
          sequent expansion.

      2.   High thermal gradients occur in the cylinder head due to the
          proximity of the inlet and exhaust steam.  This is unfavorable
          from the standpoint of mechanical design reliability.

      3.   The need for an actuated exhaust valve in addition to an actuated
          inlet valve results in undesirable complexity.

      Uniflow Design Advantages

      1.  The elimination of cylinder cooling and related losses associated
          with reverse flow exhaust.  With proper design control of the
          clearance volume, efficiency can be kept equal to that of zero
          clearance volume design.

      2.   Cylinder recompression pressure reduces the unbalanced load on
          the valve train during the inlet valve actuation period.
                                   22

-------
      3.  Mechanical simplicity.

      Pis advantages

      1.  Larger cylinder displacement for a given power output and effi-
          ciency.

      2.  Compression relief valve is required to prevent the recompres-
          sion pressure from exceeding the inlet pressure - thus possibly
          causing premature opening of the inlet valve and subsequent
          damage to the valve train.

      Largely from considerations of  mechanical reliability and simplicity,
the uniflow approach was chosen.

      Thermodynamic calculations for  a 170 IHP, 2500 RPM uniflow engine
indicate an adiabatic efficiency of approximately 80% for an inlet pres-
sure and temperature of 1000 psi and  1000°F at an exhaust pressure of 20
psi.  This is for a clearance volume  of about 5% and exhaust opening at
90% of stroke.  A bore of 3-1/2" and  a stroke of 3-1/2" with approximately
15% cut-off is required.

      4.2.2  Investigation of Ford Motor Company Variable Cut-Off Mechamism

      A survey of alternate variable  cut-off mechanisms led to the con-
clusion that it might be feasible to  make use of the Ford Motor Company
variable cut-off mechanism on the test expanders.  Accordingly, a design
study of this mechanism was made at the Ford Plant.  Results are shown
in Figures 4.2-1 and 4.2-2.  The essential feature of the mechanism is
an overhead rocker arm linkage for which the effective pivot can be varied
by the adjustment of an eccentric bearing.  This changes the amplitude
and event angle of the valve motion.   As shown by Figures 4.2-1 and 4.2-2,
the mechanism can be arranged to permit either inward or outward opening
of the valve.  For inward valve operation it is essential that a balanced
valve be used to prevent steam inlet  pressure from opening the valve.  A
further potential problem is the need for oil lubrication of the contact
interface' between the valve actuating rocker and the adjustable pivot.
Perhaps the greatest disadvantage of  the Ford mechanism for the steam car
application is the increased expander height.

                                  23

-------
      4.2.3  Cam Shaft Drive

      The principal alternatives considered were the use of an overhead
or crankcase mounted camshaft.  The former would be belt driven while the
latter would be driven from the main crankshaft through a gear train de-
signed for a crankshaft to camshaft speed ratio of unity.  These alternate
arrangements are illustrated in Figures 4.2-3 and 4.2-4.

      4.2.4  Valve and Valve Stem Seal

      Considerable attention was given to the selection of the valve con-
figuration, and to the design of the oil free valve stem seal.  Valve
types investigated included:

      1.  The double-beat and balance piston poppet valve
      2.  A ring sealed sleeve type valve
      3.  The simple unbalanced poppet valve.

The advantage of the balanced valves is the reduced loading on the valve
train - specifically upon the cam and cam follower.  The disadvantages
of balanced valves are complexity and leakage.  After detail sizing of a
simple unbalanced poppet valve, and following calculation of the cam-
tappet Hertz stress the unbalanced poppet valve proved to be quite acceptable.

      Alternate types of valve stem seals are illustrated in Figures 4.2-5
and 4.2-6.  Metallic piston ring seals and a graphite packing ring seal
are shown.  The former enjoys a substantial advantage in simplicity.  The
valve stem seal is divided into two parts with a drain to the condenser
between them.  This approach minimizes the pressure differential available
for forcing steam into the expander crankcase.

      4.2.5  Piston Rings

      Design considerations relating to the compression rings have been
discussed above.  Two alternate designs, a graphite ring, and a coated
metallic ring were selected - the second design to be used as a backup
in the event of failure of the first.

      In the trunk piston expander a graphite rider (side load carrying
ring), and two oil scraper rings are provided at the bottom of the piston
                                    24

-------
skirt.  Piston side load is carried by the rider ring and an oil lubricated
land at the extreme bottom of the piston skirt.  Thus, side load bearing
areas are provided both above and below the wrist pin.  Above the oil
scraper rings on the trunk piston is a metallic compression ring to seal
condenser pressure from the crankcase.

4.3   Expander Description
                                      (A)
      4.3.1  Crosshead Piston Expanderv '

      The crosshead expander shown in Figure 4.3-1 consists of an oil
lubricated aluminum crosshead which is connected to the steel power piston
by a rigid connecting rod.  Isolation of steam and oil is accomplished
by a rod seal between the crosshead and the power piston.  The rod seal
shown is a bronze impregnated Teflon chevron seal mounted in a floating
housing.  The power piston connecting rod is tubular for reduced weight,
and to permit internal oil cooling.  An oil jet located in the top of
the crosshead wrist pin sprays oil into the power piston connecting rod
primarily to remove heat generated by the connecting rod seal.

      The wrist pin and sleeve bearing design is quite similar to current
2-cycle Diesel engine practice.  Oil is supplied to the sleeve bearing
under pressure through the wrist pin journal.   The sleeve bearing con-
tains a circumferential groove and closely spaced helical grooves to per-
mit the free flow of oil to all surfaces of the wrist pin.

      Figure 4.3-1 shows three relatively wide rings which are pressure
balanced segmented graphite rings backed up by an Inconel ring spring.
The expander has a removable cylinder liner so that appropriate combina-
tions of piston ring and cylinder materials can be evaluated with minimum
difficulty.

      The L-head poppet valve configuration insures that lubricating oil
is well isolated from the high temperature parts of the expander - thus
eliminating the need for thermal insulation to prevent high oil tempera-
tures.  Also the orientation of the valve gear minimizes the risk of oil
entering the steam system by way of the steam inlet valve.
                                   25

-------
                                                   RKOMPRES9ON
                                                  VALVE VENT LINE-I
                                 RECOMPRESSION
                                 RELIEF VALVE
ALTERNATE ROD SEAL
  ARRANGEMENT
                                                                            STEAM INLET VALVE
                             CYLINDER LINER WITH
                               EXHAUST PORTS
                                      OIL JET

                                    ROD SEAL-OIL

                                  \  CROSSHEAD
                                  ^-WRISTPINAND
                                     BEARING INSERT
         Figure  4.3-1.   Crosshead  Piston Expander (Dwg.  707E742)
                                                26

-------
      The inlet valve assembly is designed so that the repair of  critical
components such as valve stem seals and valve seat can be accomplished
with minimum rework.  In addition, different material combinations for
the valve and valve seat, and for the valve stem bearing and seals can
be evaluated merely by replacing the valve and/or valve housing subassembly
without the necessity of remachining major components.

      The design of the valve gearing minimizes mass and linkage flexi-
bility - thus providing fast valve response and minimal inertia forces
throughout the valve gear system.  The valve gearing is oil lubricated
except for the top portion of the valve stem-which is provided with solid
lubrication.  The valve cam located within the crankcase receives some
oil by splash lubrication, but an oil jet is provided to spray oil directly
on the valve cam to insure sufficient lubrication between the cam and
cam follower.

      A recompression relief valve is provided in the cylinder head.
This valve is designed to relieve the cylinder pressure under all condi-
tions of expander operation, including motoring, preventing the steam in-
let valve from being forced open due to excessive cylinder pressure.

      4.3.1.1  Expander Crankcase

      Figure A.3-2 is a cross section of the Waukesha Motor Company Model
CFR-48 crankcase assembly used in the crosshead piston steam expander
assembly.  Figure 4.3-3 illustrates the basic cam gear arrangement after
being modified for a 1 to 1 speed ratio between crankshaft and camshaft.

      The CFR-48 crankshaft has a 3.625" stroke, 3.00" diameter main
journals' and 2.50" diameter crankpin.   All journals are nitrided and
ground to an 8 EMS finish.  Main bearings, bronze sleeve type,  are pres-
sure lubricated through oil galleys in the crankcase.   Oil is supplied
to the crankpin through holes in the crankshaft from the front main bearing.
Crankshaft thrust loads are restrained by slotted face type bronze bearings
on each side of the left main bearing (refer to Figure 4.3-2).   The right
end of the.shaft is provided with,a 16" diameter steel flywheel, and
provides for power take-off.
                                  27

-------
                                          SECTION THRU
                                        TACHOMETER DRIVE
                                                                      SECTION THRU
                                                                     OIL RELIEF VALVE
           SECTION THRU
             CAMSHAFT
SECTION THRU OIL PUMP
       SECTION THRU
      BALANCING SHAFT
   Figure 4.3-2.  Waukesha CFR-48 Crankcase Assembly
                   (Source Reference: Waukesha Motor Company Bulletin 850M)
                                        28

-------
Figure 4.3-3.   Waukesha Crankcase Showing Modified Gear Train (P72-2-5B),
                                   29

-------
      A gear case contains the gearing for the camshaft, balance shafts,
oil pump, and other external accessories.  Counter rotating balance shafts
are driven at engine speed through an idler gear.  Two camshaft positions
are available in the basic CFR crankcase.  The standard camshaft position
for internal combustion engine application is on the right side (flywheel
end looking forward).  However, it was more logical to use the left side
camshaft position for the steam expanders since more space was available
for installation of an idler gear between the crankshaft and camshaft.
Sefer to Figure 4.3-4 for modification details.

      4.3.1.2  Expander Lube System

      The basic CFR lubrication system consists of a gear pump, a strainer,
and a pressure regulating valve.  Oil galleys bored in the casing lead
to all bearing bores providing full pressure lubrication to all bearings.
The basic lube oil system of the crankcase was modified to include a
cartridge type oil filter, a water cooled oil cooler, and a turbine type
oil flowmeter.

      The lube system for the crosshead and trunk piston expanders are
identical.  Oil enters the pump through a screen strainer located in the
crankcase.  The gear type pump is located on the accessory case, and is
driven off the end of the idler gear at engine speed.  The pump discharge
was piped directly to a cartridge type filter mounted on the expander
base plate.  A pressure gage at the filter inlet was provided.  From the
filter,  oil was piped to the oil relief valve which was bolted directly
to the crankcase and ported to the main oil galley.  Excessive pressure
above the set point, resulted in some oil being by-passed back into the
crankcase sump.  Oil being supplied to the engine bearings flowed through
an oil cooler and a flow meter before entering the main oil galley.

      Oil supply temperature was regulated by water flow to the cooler.
An oil sump temperature of approximately 250"F was held during expander
operation.  An electric "Calrod" type heater mounted under the crankcase
was utilized to preheat the oil prior to startup.  A 250°F temperature
level vaporized water that entered the sump as a result of steam leakage.
The crankcase breather pipe located on the side of the engine was con-
nected to a small water cooled heat exchanger.  Water vapor being expelled

                                   30

-------
/*
                         WOS ?U»4 IV
                         40LC AMD IWJTA
                         LOCKIU6
                         .*« -.-*! DlA 6«lf TO
                         Of^n* W KXtlTIWO
                         CUT o«r, ----- "
                         TAP  M*
                       EXISTIWC HOLI IV CC«(
                       C'ffft S.7X- *.» KH. TO
                       O'PTH 9HOW/NI.
'. ORlli fc   /*"
: ^ttSctlW  /

Ut-HUC »*'!* -*
Figure 4.3-4.  Steam Expander  Crankcase,  Modification  Details  (GE Dwg.  221R901).

-------
from the crankcase passed thorugh the heat exchanger where it was con-
densed, collected, and measured.  This provided an indication of seal
leakage rate.

      4.3.1.3  Expander Component Hardware

      Figures 4.3-5 through 4.3-14 show the major hardware components
used in the crosshead expander.
                                  (A)
      4.3.2  Trunk Piston Expander^ '

      Figure 4.3-15 depicts the assembly of the trunk piston expander.
Except for the power piston the trunk piston expander is quite similar
to the crosshead  expander design.  For example, there is no difference
in basic design of the valve gearing or recompression valve.  Also, the
connecting rod wrist pin and wrist pin bearing design is essentially the
same.  Two major  differences between the two expanders are as follows:
(1) the trunk piston expander is shorter in overall height, and  (2) the
power piston and  connecting rod assemblies are quite dissimilar.  However,
the reciprocating weights for the two expanders are almost the same -
approximately 8.2 pounds.  The assembled expander is shown in Figure
4.3-16.

      The trunk piston is a two piece design which permits the following:

      1.  The use of the same wrist pin design as used in the crosshead
          expander

       2.  The use of dissimilar materials for the top and bottom portions
          of the  piston for weight reduction

       3.  Provides for positive oil/steam isolation in the region of the
          wrist pin while permitting adequate oil lubrication of the
          wrist pin without subjecting the oil to high temperature

       4.  Provides for oil cooling of the lower portion of the piston

       5.  Provides for the interchange of the upper portion of the piston
          to accommodate alternate sealing ring configuration and materials.
                                   32

-------
                                       Top
u>
                               Pressure
                                 Vent
                                 Ports
                                             itj'i'pfi'TT'm
                                             nKluijiil
zJ
                                 Figure 4.3-5.  power Piston for Carbon-Graphite Rings  (P72-3-4U)

-------
                                         Pressure
                                         Balancing
                                          Grooves
Figure 4.3-6.  Pressure Balanced Carbon-Graphite Piston Ring (P72-3-2AD)
                                   34

-------
OJ
                                                                                                       Seal
                                                                                                      Groove
                                            Figure 4.3-7.   Cylinder  Liner (P72-3-4C).

-------
OJ
                                                                                 Hard Chromium Plate
                                                                                 2 RMS Finish
                                               Figure 4.3-8.   Power  Push Rod (P72-3-4A).

-------
  ;

         Push Rod
         Mounting
          Face
            \
Recompresslon
  Vent Hole
                                                Wrist Pin'
                                                  Bore
I.

           fj^^                               ;
   Figure 4.3-9.   Aluminum Crosshead (P72-3-4K)

-------

UJ
00

                                    Wrist Pin
                                     Bushing
                                    (1 of 2)
                                                                    • Retainer
                                                                    Screw and
                                                                     Oil Set
                                                                              Wrist Pin
                                                    I
pirate ii— —•si   r~s|   r~*|r~5i  r~5|   r~7t   r
IniiljlliijJiilUUiUL™
                                          Figure 4.3-10.   Wrist Pin and  Bushing (P72-3-4G).

-------
l*J
-o

                                                       Seal Ring
                                                        Grooves
                                                 •Inlet
                                                  Valve
                                                                Hard Facing

 Valve
Housing
                                                                                             L
                                            Figure  4.3-11.  Steam  Inlet Valve Assembly (P72-3-AF),

-------
Push Rod
 Socket
         Tappet
                                           Tappet
                                           Housing
            ,)
                                                   I11


     Figure 4.3-12. Cam Tappet Assembly (P72-3-4D).

-------
Figure 4.3-13.  Connecting Rod (P72-3-4H)

-------
                                                 Cam Lobe
Figure 4.3-14.   Camshaft (P72-3-4S),

-------
 -WRIST PIN AND
  BEARING INSERT
                REGOMPRESSION
                 REUEF VALVE
                POWER PISTON
                 AND SKIRT
                 iSOBORE
                   ILJET
                 rEXHAUST
                   STEAM
CYLINDER LINER WITH
  EXHAUST PORTS
                                    RECOMPRESSION
                                    VALVE VENT 1:1 NE
                                                                              SIGHT PORT


                                                                            CAMSHAFT
                                                                            CRANKSHAFT
                                                                            3.625 STROKE
                                                                                   CAM-TAPPET
                                                                                    OIL JET
       Figure  4.3-15.  Assembly of Trunk Piston Expander (Dwg. 707E743).

                                         43

-------
Figure 4.3-16.  Assembled Trunk Piston Expander (P72-6-3S)

-------
      To keep the trunk piston as short as possible, the basic expander
design shown in Figure 4.3-15 shows two non-metallic sealing rings com-
pared with the crosshead expander which contains three non-metallic seal-
ing rings.  However, the alternate trunk piston design contains three
metallic sealing rings for the same piston length.   A wider third ring,
called a rider ring, is required in the top portion of the trunk piston
to carry piston side loads and is not intended to function as a seal ring.

      The trunk piston contains three rings near the bottom of the piston.
The top ring of the three is a seal ring to restrict the flow of steam
into the expander crankcase.  The two bottom rings  are oil exclusion rings
to prevent oil from entering the steam system.  The lower circumferential
surface of the trunk piston fuctions as a piston side load support or
second rider ring.  Since the lower part of the trunk piston is oil lubri-
cated a separate solid lubricant rider ring is not  required below the
wrist pin.

      The trunk piston expander also has a removable cylinder liner so
that different ring/liner material combinations can be evaluated with
minimum expander rework.

      The crankcase assembly including crankshaft,  gear train,  counter
rotating balance shafts,  oil pump, etc.  are identical to the crosshead
expander.  Due to the shorter overall cylinder height,  the valve axis  is
inclined at a slightly different angle than the crosshead.   The lubrica-
tion system for the trunk piston expander is identical  to  the crosshead
expander.

      Figures 4.3-17 through 4.3-22  show the major  trunk piston expander
components which are different from  the  crosshead expander.

4.4   Expander Performance Analysis

      4.4.1  Expander Breathing Analysis

      A computer program  was written and used to perform expander breath-
ing calculations.   The purpose of this analysis was the following:
                                  45

-------
HiiiliHliiiiliiliiiiliiiiliiiiliiiiliiiiliiiilniilniinniliiiiliiii!!
Figure 4.3-17.   Trunk Piston and Rod Assembly,
                     (P72-6-3B)
                             Rings

-------
Figure 4.3-18.  Piston For Trunk Expander (Upper Portion)
                (P72-6-3C)
                           47

-------
Figure 4.3-19.  Piston for Trunk Expander (Lower Portion)
                (P72-5-4H)

-------

Figure 4.3-20.  Trunk Piston Oil Exclusion Ring (P72-5-2B)

-------
Figure 4.3-21.  Trunk Expander Piston Assembly (P72-6-3A)
                           50

-------
                     i'l't'l'I'I'ITI'I'l'l'I'I'ITlTITI1)1)1!
                    fftCHt* '	\\	<	21	. . '  	 V



                          *l TTI  
-------
      1.  To determine the inlet valve diameter, maximum lift, total
          event angle, cam lead angle, cam base circle radius, and the
          cam nose radius in such a manner that the steam flow rate is
          approximately 400 pounds per hour for an indicated horsepower
          and indicated efficiency near an optimum level.

      2.  To insure that the cam and cam follower Hertz stresses, pres-
          sure times velocity parameter, and the valve train accelerations
          are within limits.

      3.  To obtain predicted engine indicator diagrams for specified
          test conditions over a range of operating speeds.

      The computer program calculates the steam flow into the cylinder
which takes place during discrete time intervals of the valve operating
cycle, and also calculates the co-occurring changes in cylinder pressure
which result from the combined effects of steam inflow and cylinder volume
change.  The valve and cam geometry are inputs to the program.  For portions
of the expander cycle during which the valve is closed an expansion or
                                             V
compression process following the equation FV = constant was calculated.
Flow through the cylinder exhaust ports was calculated in a manner similar
to that of steam inflow.  The computer program was based upon the assump-
tions of perfect gas properties for the superheated steam, isentropic
flow from the engine inlet plenum to the minimum valve orifice cross sec-
tion, and upon no recovery of valve orifice velocity head within the
cylinder (valve orifice static pressure equals cylinder pressure).  An
empirical orifice flow coefficient obtained from data in reference 5 was
applied.  The effective ratio of specific heat, y» f°r superheated steam
was treated as a variable function of pressure and temperature through
the cycle.  However,  this was found to be an unnecessary refinement, at
least insofar as steam flow prediction was concerned.  Calculated flow
variations corresponding to a change in y from 1,2 to 1.3 were found to
be approximately 5%.  This assumed variation in y exceeds the physical
variation which occurs as the steam goes through the expander cycle, and
the resulting calculated flow variation was substantially less than the
expected flow variation resulting from the effects of piston blowby and
cylinder wall heat transfer.  The latter effects were neglected in order
to finalize the valve and cam design within the short time available,
                                   52

-------
and also because it was felt that an accurate prediction of these effects,
particularly that of piston blowby, could not be made prior to testing.
An estimated flow margin allowance for these factors was made in the final
design selection.

      Engine indicator diagrams calculated by the breathing program are
shown in Figures 4.4-1, 4.4-2 and 4.4-3.  These diagrams all correspond
to the selected valve and cam geometry.  For operation at 2500 RPM the
cam shaft would be shifted so that the inlet valve would close 10° earlier
than for operation at 2000 RPM.  This shift changes the valve lead angle
from 10° to 20° and limits the steam flow to an estimated value which is
within the test facility capacity at the higher RPM.  Figure 4.4-1 shows
the calculated indicator diagram for operation at 2000 RPM, 1000°F inlet
temperature, 1000 psia inlet pressure.  Valve lead is 10°.   Figure 4.4-2
shows the calculated indicator diagram for operation at 2500 RPM with the
same inlet conditions as for Figure 4.4-1.  The valve lead is 20°.   Fig-
ure 4.4-3 shows the effect of expander speed on the indicator diagram.
As speed is reduced the cylinder pressure drop occurring during the steam
admission period becomes less, and the steam flow per revolution increases.
This occurs from the fact that the admission time period for the steam
flowing through the valve restriction increases with reduction in speed.
The steam flow per revolution is reduced by increasing valve lead since
the average cylinder pressure during the admission period increases with
increasing valve lead (reduced average AP across the valve orifice).

      Table 4.4-1 gives a summary of the analysis of the indicator diagrams
shown in Figures 4.4-1 and 4.4-2.  This table also indicates the selected
geometry of the valve and cam.  Table 4.4-II gives a summary of the analysis
of the indicator diagrams of Figure 4.4-3.

      4.4.2  Steady-State Temperature Distribution

      A broad heat transfer analysis was conducted to determine an ap-
proximate temperature map for the two single-cylinder steam expanders
employing the existing THT-D Transient Heat Transfer computer program.
This program is capable of analyzing general three-dimensional heat trans-
fer systems employing a finite difference method with appropriate speci-
fied boundary conditions.  A variety of modes of heat exchange may be

                                   53

-------
           1000
in
                                                                     I     I     I     I
                                                                   Inlet Temperature - 1000°F
                                                                   Inlet Pressure - 1000 PSIA
            100 —
                                               12    14   16    18    20   22   24
                                                   CYLINDER VOLUME - CU. IN.
34   36
                            Figure 4.4-1.  Expander  Indicator Diagram, 2000 RPM, 10° Valve Lead.

-------
         1000
Ul
                                                              Inlet  Temperature - 1000 F
                                                              Inlet  Pressure - 1000 PSIA
                                            12   14    16    18   20   22    24
                                                CYLINDER VOLUME - CU. IN.
                         Figure 4.4-2.  Expander Indicator Diagram,  2500 RPM, 20° Valve Lead.

-------
                   1200
U1
                                                                 Inlet Condition* 1000'P, 1000 p»i«
                                                2000 8PM
                                                1500 RPM
                                                1000 BPM
                                                 500 RPM
                                                   15        20       25

                                                 Cylinder VoluM, cu.  in.
                             Figure 4.4-3.   Theoretical Indicator Diagrams Single Cylinder Expander

-------
                              TABLE 4.4-1
              EXPANDER PARAMETERS FOR INDICATOR DIAGRAMS
Indicator Diagram                   Figure 4.4-1      Figure 4.4-2
Valve Event                         70°               70°
Valve Lead                          10°               20°
Valve Throat Diameter               .75"              .75"
Maximum Lift                        .10"              .10"
Cam Base Radius                     1.125"            1.125"
Cam Nose Radius                     .45"              .45"
RPM                                 2000              2500
Steam Flow Rate                     380 Ib/hr         374 Ib/hr
(Before correction for piston
blowby and cylinder wall heat
transfer)
Indicated Horsepower                48.6              46.2
Indicated Expander Efficiency       82%               79%
                                  57

-------
                  TABLE 4.4-II




EFFECT OF SPEED ON INDICATED EXPANDER PERFORMANCE
      (Reference Diagrams of Figure 4.4-3)
Speed
500
1000
1500
2000
Flow, Ib/hr
139
255
319
380
EMP (psi)
388
352
310
276
IPH (hp)
17.1
31.0
40.9
48.6
Indicated
Efficiency
0.791
0.785
0.825
0.823
                        58

-------
analyzed including heat transfer by conduction, convection, and radiation
with or without internal heat generation or surface heat flux.

      Special attention was given to piston ring, piston and cylinder
temperature as a result of steam temperature, fractional heat generation,
and the boundary conditions which are expected to exist during expander
operation.

      Complex geometrical areas in the expanders were replaced with simple
geometry to facilitate the node assignment for the computer calculations,
without changing the basic design.  After dividing the geometry into a
series of numbered nodes, measurements were made of full scale drawings
and the necessary areas and volumes for each node were calculated and
served as input to the computer program.  Assumptions and boundary condi-
tions for the computer analysis used were:

      1.  Due to symmetry, it was assumed that no heat flowed across the
          centerline or through the circumferential faces.

      2.  Thermal insulation on the outside of the expander was simulated
          with an appropriate convection coefficient.

      3.  The ambient air temperature was taken as 80°F, and the film
          coefficient for the surface exposed to air was estimated by
          the appropriate equation for natural convection over a vertical
          cylindrical surface.

      4.  Steam temperatures were assumed to be 1000°F in the intake valve
          region and 540°F in the piston/cylinder region.  The exhaust
          steam temperature used was 250°F.

      5.  The convective heat transfer coefficient for steam was estimated
          from conventional equations, such as the Dittus-Boelter equation
          for flowing fluid.  In the region enclosed by the cylinder and
          the top of the piston, where some steam condensation might
          occur, the film coefficient was estimated by equations appropriate
          for film condensation over a flat plate.

      6.  The cooling oil temperature used was 250°F, and the film coef-
                                  59

-------
          ficients were estimated from equations appropriate for fluid
          flowing through tubes or over vertical surfaces.
                                                                      o
      7.  All contact coefficients were assumed to be 20,000 Btu/hr-ft -°F.

      There are five significant places where sliding metal parts generate
heat through friction:

      1.  Graphite ring with cylinder lining (graphite-steel)
      2.  Crosshead with cylinder (aluminum-steel)
      3.  Teflon packing with piston rod (teflon-steel)
      4.  Rider ring with cylinder lining (graphite-steel)
      5.  Intake valve ring with valve guide (steel-steel).

      The heat generated by friction is calculated from:
)  i
                                              (F U)  d8
where f = Sliding coefficient of friction
      J = 778 ft-lb/Btu
      F = Normal force acting on the sliding surface, a function of 6
      u = Sliding velocity, a function of 6
      6 = Crank angle

      In general the friction heat generated by a variable normal force
and a variable sliding velocity is a function of crank angle.  For steady
state calculations, however, the above equation can be used  to calculate
the frictional heat, and  can be considered as the mean effective fric-
tional heat generated by  two sliding parts.

      Typical values of the sliding coefficient' of friction, f, from the
Mechanical Engineers' Handbook - 7th edition,  (6)  are as follows:
                                       Dynamic Coefficient of Friction
          Materials                    Dry           Grease-Lubricated
      1.  Steel on steel               0.42              0.03-0.1
      2.  Steel on graphite            0.21                0.09
      3.  Teflon on steel             0.04                	

                                   60

-------
      To a first approximation, the frictional heat generated at  an  expander
speed of 2500 RPM are as follows:

                                       Total Heat Generated by Friction,
          Sliding Parts                	Btu/hr	
      1.  Graphite Ring - Cylinder                  305
             (f = 0.1 and
            *F = 3 Ibf/in of
                 circumference) (59)

      2.  Crosshead - Cylinder                     1980
             (f = 0.07)

      3.  Teflon packing - piston rod                26.2
             (f - 0.04 and
             F = 3 Ibf/in of
                 circumference)
             4,
      4.  Rider Ring - Cylinder                    2840
             (f = 0.1)

      5.  Intake valve - valve guide                170
             (f = 0.4 and
             F = 3 Ibf/in of
                 circumference)

      The total frictional heat was assumed to be distributed evenly be-
tween the stationary and the moving parts.  For example, for the cross-
head-cylinder case above, 990 Btu/hr goes into the crosshead piston and
990 Btu/hr goes into the cylinder.

      Steady state temperature distributions were obtained from THT-D
program computations, and are shown in Figure 4.4-4 for the crosshead
expander and Figure 4.4-5 for the trunk expander.  The temperature levels
shown are only approximate values since a rigorous thermal analysis was
not considered essential.  Measured temperature distribution is presented
in Section 8.0.
*Ring tension by design.  Excludes gas pressure loading.
                                  61

-------
                + CALCULATED TEMP-F*
                O ASSUMED TEMP-F*
Figure  4.4-4.  Calculated Temperature Profile  for Crosshead Piston Expander
                                       62

-------
                                                                 885
                +CALCULATED TEMP-T
                o ASSUMED TEMP-F*
Figure  4.4-5.  Calculated Temperature Profile for Trunk  Piston Expander
                                     63

-------
      4.4.3  Stress Analysis

      4.4.3.1  Crosshead Piston Expander

      Before detailed cycle calculations had been made on the crosshead
expander, a. set of reasonable but conservative estimations of critical
loading parameters were made in order to estimate component part stresses.
The loading parameters were calculated in sufficient detail to allow
their estimation at any crank-angle throughout the cycle, so that they
could be combined to determine their worst-case summations.

      The piston position, velocity, and acceleration; cylinder swept
volume; and inertia force were calculated for 1000 RPM and a 6.146 Ib.
reciprocating weight which loads the wrist pin.  These parameters could
be corrected for different rotative speeds and reciprocating weights.
The piston pressure forces were calculated using the pressures calculated
from an assumed cycle with 15 percent cut-off(^ 40° ATDC) and isentropic
expansion from 40° to 135° (exhaust port opening) according to:

                   PV1*28 = 1000 (2.747)1'28 = 3645

The pressure remained at the condenser pressure of 20 psia until the ex-
haust port closed at 225°, after which isentropic compression occurred
according to:

                   PV1'28 = 20 (32.535)1'28 = 1725

This model gave a mean pressure of 290 psi and temperature of 575°F.

      The vertical pressure and inertia forces on the wrist pin were com-
bined to give the total vertical forces.  Side pressure and inertia forces
on the wrist pin and crosshead piston were calculated and combined with
the total vertical forces to determine the total wrist pin loads.  The
inertia forces due to angular acceleration of the connecting rod were
included.  The combined forces at 2500 RPM varied from 250 - 7800 Ib. or
110 - 3470 psi projected pressures on the wrist pin bearing area of
2.25 in2.
                                  64

-------
      Table 4.4-III and Figure 4.4-6 identify the stresses and factors
of safety (FS) at the principal points of interest in the crosshead ex-
pander.

      4.4.3.2  Trunk Piston Expander

      There are virtually no differences between many of the components
of the trunk expander and the corresponding ones of the crosshead expander.
The principal changes are in the piston assembly and the connecting rod.
The same type of cycle modeling assumptions were made in order to predict
internal pressures and loads.

      Figure 4.4-7 and Table 4.4-IV identify critical stress locations,
stresses, and factors of safety.  Those which are the same as for the
crosshead expander are omitted.

      4.4.4  Piston Ring Wear Prediction

      4.4.4.1  Crosshead Power Piston Graphite Rings

      Zero wear is defined as wear so slight that the surface finish of
the wear track is not significantly different from the finish of the
              (2)
virgin surface   .  The wear scar is, then, roughly equal to one-half of
the peak-to-peak value of the surface finish.

      The engineering model states that the wear between two loaded sliding
surfaces can be controlled by limiting the maximum shear stress,  T   ,
                                                                  msx
in the contact region.  Zero wear can be obtained for a specific number
of passes if:

                             T    <  p  T                         psi
                              max —      y
      The fractional parameter p is dependent upon the number of passes,
the materials and lubricant used, and the lubricating conditions, as
shown in Table 4.4-V.  For 2000 passes, p is designated as p .
                                   65

-------
TABLE 4.4-1II
STRESSES AND FACTORS OF SAFETY AT PRINCIPAL POINTS
OF INTEREST IN THE CROSSHEAD EXPANDER

Location
A
B
C
E
F
6
H
I
J
K
L
M
N
0
P
Q
R
S
T
U
V
W
X
Y
Z
AA
AB
AC
AD
(Figure 4.4-6 Identifies
Effective Stress
Cycle, ksi
12.3 + 12.3
0.9 + 0.9
3.1+ 3.1
22.6 + 13.2
14.2 + 52.0
20.5 + 20.5
2.7 + 0.5
44.7 + 12.1
2.1+ 2.1
36.5 + 24.3
86.0 + 15.6
17.6
42.4 + 42.4
17.7 + 17.7
78.0 _ 88.1
12.2 + 10.3
42.6 + 7.6
75.2 + 29.1
30.1 + 13.7
9.6+ 0.8
,53.0 + 20.5
57.3+ 8.2
67.8
1.1+ 1.1
30.7+27.3
41.6
43.9
27.0
5.5 + 5.5
Location)
Allowable
Stress, ksi
+ 42.0
+ 42.0
+ 42.0
+ 47.0
+ 64.8
+ 50.0
+ 4.5
+ 48.8
+ 4.5
+ 61.8
+ 31.5
43.0
+ 75.0
+ 42.0
125.0
+ 51.5
+ 40.0
+ 55.0
+ 34.0
+ 50.5
+ 31.0
+ 35.0
81.0
+ 18.5
+ 44.8
84.0
120.0
32.0
+ 16.0

Factor of
Safety, -
3.41
47.3
13.4
3.57
1.25
2.44
9.09
4.05
2.14
2.55
2.03
2.44
1.77
2.37
1.42(b
5.00
5.26
1.88
2.48
11.63
1.51
4.27
1.19
17.1
1.64
2.02
2.82
1.19
2.94
      66

-------
                        TABLE 4.4-III (Cont'd.)

          STRESSES AND FACTORS OF SAFETY AT PRINCIPAL POINTS
          	OF. INTEREST IN THE CROSSHEAD EXPANDER	
                  (Figure 4.4-6 Identifies Location)
Location
AE
AF
A6
AH
AI
AJ
Effective Stress
Cycle, ksi
12.3
2460 Ib
10.8 + 10.8
57.7 + 20.9
77.2 + 16.8
15.1 + 15.1
Allowable*- '
Stress, ksi
32.0
3590 Ib
+ 20.0
+ 44.5
+ 36.0
+ 20.0
Factor of
Safety, -
2.60
1.46
1.86
2.13
2.14
1.33
NOTES:  (a)  The allowable stresses shown as + are the allowable alternating
             stresses when the mean stress if the same as shown in the
             effective stress tabulation, operating temperature for
             1000 hrs. considered.

        (b)  Spring - Maximum allowable stress is 125.0 ksi for initial
             stress of 78.0 ksi.  Maximum imposed stress is 88.1 ksi.

        (c)  FS = critical load/applied load.
                                   67

-------
                          AA
Figure 4.4-6.  Location of Stresses, Crosshead Expander (Ref. to Table 4.4-III)
                                          68

-------
                             TABLE 4.4-IV
STRESSES AND FACTORS OF SAFETY AT PRINCIPAL POINTS
OF INTEREST IN THE TRUNK EXPANDER

Location
A
B
C
G
H
I
J
K
L
M
Q
R
S
T
U
V
W
(Figure
4.4-7
Identifies Location)
(a)
Effective Stress Allowable v
Cycle, ksi Stress, ksi
70.2 +
10.0 +
3.7 +
4.9
4.6 +
51.3 +
2.0 +
34.3 +
82.5 +
17.6
4.2 +
71.0 +
36.8 +
2.9 +
4.5 +
14.3 +
65.6 +
3.2
10.0
3.7

2.7
22.3
2.0
23.5
6.2

3.8
12.2
36.8
2.9
4.5
14.3
3.9
+ 31.5
+ 47.5
+ 47.5
11.6
+ 5.0
+ 47.0
+ 6.7
+ 61.8
+ 33.5
39.6
+ 46.6
+ 39.3
+ 44.3
± 29-6
+ 44.3
+ 31.0
+ 41.8

Factor of
Safety, -
9.90
4.75
12.9
2.38
1.86
2.10
3.36
2.63
5.41
2.25
12.2
3.21
1.20
10.1
9.88
2.17
10.7
NOTES:  (a)  The allowable stresses shown as + are the allowable alternating
             stresses when the mean stress is the same as shown in the
             effective stress tabulation, operating temperature for
             1000 hr. considered.
        (b)  Stresses for locations not shown are the same as for the
             crosshead expander.

-------
K-
                                                                            AF
                 Figure 4.4-7.  Critical Stress Locations for Trunk Expander
                                (Refer to Table 4.4-IV)
                                         70

-------
                              TABLE 4.4-V
                         FRACTIONAL PARAMETER
    Type of Lubrication         p_        Material/Lubricant System
                                 K.
Quasi-hydrodynamic             1.00    Fully hydrohynamic lubrication
                               0.54    Partial hydrodynamic lubrication

Dry or boundary lubrication    0.54    Low susceptibility to transfer
                               0.20 '   High susceptibility to transfer

      Since no published data exists for the exact material combinations
to be used in the single cylinder expanders, it is estimated that pR = 0.20
for the unlubricated graphite ring in contact with a steel cylinder.
This estimate was thought to be conservative, and is the value found for
                                       (2)
dry 52100 steel vs. 7 stainless steels.  '

      The properties of the piston ring material, U.S. Graphite Co.
Grade 107, are shown in Table 4.4-VI.  In the absence of specific informa-
tion concerning its allowable shear yield strength, it is assumed that
T  = 7500 psi, the RT tensile strength.  The ring material is considered
so brittle that the yield and fracture strengths are equal.   (Another
method of estimating T  fron the microhardness gives a larger allowable
v>

                             TABLE 4.4-VI
                PROPERTIES OF GRADE 107 CARBpN GRAPHITE

        Composition     -  Carbon graphite impregnated with Sb
        Strength        -  at RT, will increase with temperature
                                Compression         - 38,000 psi
                                Transverse fracture - 10,000 psi
                                Tensile             -  7,500 psi
        Elastic Modulus -  E = 3.5 ,x 10  psi
        Hardness        -  Sclersocope      92
                           DPH (converted) 865
                                   71

-------
      The pressure, and therefore, frictional loads on the ring/cylinder
interface will not be constant throughout the cycle.  The pressure dif-
ferential will decrease toward zero as the steam expands in the cylinder
and the exhaust ports are uncovered.  Using an idealized representation
of the pressure vs. crank angle relationship a mean effective cylinder
pressure of 290 psi will occur for the 15% cut-off condition.

      The contact load, F1 at the ring/cylinder interface will be a func-
tion of the spring load, F1, the pressure load, F1, the frictional load,
                          s                      p
F*, and the ring geometry as shown by Figure 4.4-8.

      Assuming that all radially outward adjustments in ring position
which are required for wear compensation are made at bottom dead center,
where all forces are zero except for the spring force, the horizontal
frictional force is zero during the rest of the cycle.  The contact stress
equation reduces to:
                    '  pl   4 + 0.0622 AP - 0.0296 ?1
                 qo = IT	672415	               psi

      To determine conditions which will ensure zero wear for a desired
life of N passes for a given mechanism, the following relation has been
derived:
             Tmax = K qo   '°'5  + v   -  <"*?—'     P»T"         psi
       Solutions are shown in Figure 4.4-9 for the graphite piston rings,
for  the  average value of P- =  290 psia, AP equal to three fractions of
P-,  and  coefficients of friction from 0.05 to 0.50 as functions of rota-
tive speed.

       By combining equations the fractional area of the piston ring with
a  stress concentration less than the allowable for zero wear in 1000 hours
is determined.  These calculations are also plotted in Figure 4.4-9 as
fractional area with zero wear.  The significance of this figure is to
show that, according to analytical predictions, the graphite piston rings
should experience only minor wear near the edges under the ideal condi-
tions  assumed  in the calculations.
                                   72

-------
2.885 Dia
3.500 Dia
                     —H  h—0.065
AP
                                                                        AP
                                                  + 0.140242 AP
        4 + 0.062250 AP - 0.029612P
          Figure 4.4-8.  Graphite Ring Geometry and Load Components  (lb/3.5Tr  in.)
                                          73

-------
  200
 o
 4J

 O
 c
 o
a


                                                                                           03
                                                                                           Of

                                                                                           0)
                                                                                           (0


                                                                                           §

                                                                                           Tl
                                                                                           4J

                                                                                           U

                                                                                           efl
                                                                                    -499
                                                       2000
                                                                                3000
                                  Rotative Speed  - RPM
     Figure  4.4-9. Maximum Allowable Stress Concentration Factor and Fractional Area

                    with Zero Wear of Graphite Piston Rings in 1000 Hr.

                    Mean P. s 290 psia »  upstream pressure on piston ring.

                    AP = pressure drop across piston ring.
                                           74

-------
      This analysis  assumes  certain ideal  conditions which  are  not  ex-
pected to exist in the actual single  cylinder expander.   For  example,  the
analysis makes no provisions for wear due  to foreign abrasive particles,
wear due to ring misalignment or rocking in the piston ring groove,  or
wear caused by rough surfaces such as scored cylinder walls.

      4.4.4.2  Trunk Expander Piston  Rider Ring

      In the trunk expander rider ring wear analysis, it was  assumed that
there were no steam  pressure forces acting on the rider ring  since the
ring was notched on  its rubbing face  to prevent it from becoming a pres-
sure sealing ring.   Therefore, the only load on the rider ring  is the
side force imposed by steam pressure  on top of the piston and the opposing
force from the connecting rod.  Piston inertia forces are also  included.
The trunk piston was designed so that the  graphite rider ring supported
only half of the total side forces.  The bottom skirt of the piston  sup-
ported the other half of the piston total  side load.

      The cyclic average of all positive side loads on the graphite  rider
ring is 210 Ibs., and the cyclic average of all negative side loads  (on
the opposite side of the piston) is only 20 Ibs.   Only the 210 Ib. load
will be considered in the following discussion.   The width  (axial length)
of the graphite rider ring to be considered is 0.7 inches.

      For a ring outside diameter which is less  than the cylinder diameter,
the Hertz formula    for the contact stress is:
             Max Sc = qQ = 0. 798  / _	^-^	T_ , psi            (a)
where D- = Cylinder dia. =3.5 inches
      D,, = Ring dia., inches
      EI = 3.5 x 106 lb/in2 (graphite)                      psi
      E2 - 30 x 106 lb/in2 (steel)                          psi
      v., = v» = Poisson's ratio =0.3
      P  = load/linear inch = 210 lb/0.7 inch

                                   75

-------
      Equation (a) reduces to:
                      qo = 13711  /jp- -1 ,  psi
                                 /   2
      For conforming geometries, the maximum shear stress,  Tmgx,  is a
function of q , a corner or edge stress concentration factor K,  and the
             °
coefficient of friction y.

                                                     , '
      The latter half of the above expression are the conditions required
to ensure zero wear for a life of N passes for a given mechanism with
the fractional parameter p  = 0.20 and shear yield strength of 7500 psi.
                          R
      Solutions are shown in Figure 4.4-10 for the graphite rider ring
for expander rotative speeds of 500 - 3000 RPM, for coefficients of
friction 0.05 <_ y <_ 0.50, and for diametral differences (D..-D-) = AD of
0.1, 1, and 10 mils.

      The fractional area of the rider ring with a stress concentration
factor less than the allowable for zero wear in 1000 hours is shown
plotted in Figure 4.4-10.

      It may be concluded that the higher contact stresses generated by
a large difference in ring and cylinder diameters will cause greater-
than-zero wear over a significant fraction of the ring face, while only
very small regions near the edges of the ring will have greater-than-
zero wear when the diameter difference is small.  At least two conditions
will tend to reduce an initially large diameter difference:  (1) mechanical
or thermal loads could spring the ring ends outward and increase the
effective diameter of the ring, and (2) the ring will wear toward exactly
the same diameter as the cylinder with time.  Either should decrease
the initial wear rate significantly.  Also, as the ring edges wear from
sharp corners to circular arcs, stress concentration factors and wear
rates should decrease.
                                  76

-------
 40,
 20
                                                                                     99.9
 10
0.8
0.6
0.4
                                                                         0.5
                                    AD = 0.5 Mil
                                    AD = 5 Mils
                                    AD = 10 Mils
                                                                                    99.5
                                                                                    99
                                    AD = 50 Mils
                                                                                    95
                                                                                    90
                                                                        y = 0.1   J
                                                 80


                                                 70

                                                 60

                                                 50

                                                 10

                                                10
                                                       O
                                                       o
                                                                                          t-l
                                                                                          as
                                                                                          
-------
      Using calculated operating temperatures and manufacturing drawing
tolerances, the average AD was 11 mils for the Type 440C stainless steel
cylinder liner and graphite rider ring.  Initially, a large wear rate
can be expected to occur, but the wear rate should decrease significantly
as the wear process laps the ring and cylinder toward the same diameter
and the ring edges toward circular arcs.

      A satisfactory wear life is predicted  by the above simplified and
ideal model.

      4.4.5  Recompression Valve Dynamics            i

      A finite time•increment analysis was made of the recompression valve
dynamics.  A constant increment of one crank-angle degree was used which
gave time increments of 1/3000 sec at 500 RPM to 1/15000 sec at 2500 RPM.
The computation method was as follows:

      -  Assume a condenser pressure and rotative speed, and let inlet
         pressure = 1000 psia.

      -  Start calculations at the crank angle at which the calculated
         cylinder isentropic recompression pressure reached 1200 psia,
         producing a valve lifting force greater than the 78 Ib. recom-
         pression spring preload.

      -  Determine recompression valve acceleration from F = ma, or
         a = F/1618.76 Ibm sec2/in.

      -  Determine valve initial, average, and final velocities upward
         for each time period.

      -  Determine valve initial, average, and final lifts upward for
         each time period.

      -  Determine average valve flow area for each time period.
      -  Determine average valve flow  coefficient  for each time period.
      -  Determine average steam flow  through valve for each time period.
      -  Deduct steam flow through valve from cylinder volume at the be-
         ginning of the next time period and repeat calculations for the
         new time period.
                                   78

-------
      -  Continue calculations until the inlet valve opened at either
         -20° for 2500 RPM or -10° for <_ 2000 RPM by cam action  (acceptable
         solution) or until the recompression pressure reached 1670 psia
         and produced a net lifting force on the inlet valve  (unacceptable
         solution).

      Samples of such calculations are shown on Figure 4.4-11.  All of
the conditions shown except for 1000 RPM, P  = 100 psia, are acceptable
since the inlet valve will open from cam action before 1670 psia recom-
pression pressure is reached.

      The results of all calculations are summarized on Figure 4.4-12
which show the allowable variation of condenser pressure with RPM.  Less
favorable results occur when the inlet pressure is less than 1000 psia.

      4.4.6  Crosshead Expander Power Piston Rod Natural Frequency

      The transverse bending natural frequency of the crosshead power
piston and its rod as a cantilevered beam with the piston mass at its
                                          /Q\
free end was calculated using the formula:
                                   3EI           ,2.   2
                                               rad /sec
                        9o
                         3m   L3(M+0.23m)

where  E = 28 x 106 psi (rod at 300°F)
       I = 0.01508 in4 (rod)
       L = 5.75 in.
      mg = 0.6217 Ib (rod-calculated)
      Mg = 3.043 Ib (piston-measured and calculated)

Then   u> = 899 rad/sec, and f = (-r^) 60 = 8580 cpm.
                                 ZTT
      Therefore, the rod/piston system was considered satisfactory for
the first three harmonics of the maximum operating speed of 2500 RPM,
and for higher harmonics for lower operating speeds.

      The load deflection characteristics of the piston rod were calculated
using the formula for a cantilever beam with an intermediate load:
                                   79

-------
   1700
   1650
   1600
                                                         r-™—'£.j\j\j fu. ri  .-^ *.<-<
         — EF = 0 on Inlet Valve	1-	
co
PH
(0
co

-------
  120
  no
in
(X.
  1670 PSIA,
                                                      RE  ~
                                        £ Forces on  Inlet Valve = 0 and
                                        Inlet Valve  Will be Forced Open
                                        by Cylinder  Steam Pressure

                                        PD_ = Recompression Pressure
                                         KE
                                         For  Inlet  Valve Opening
                            Pc = 49 PSIA Acceptable
                            Even With No
                            Recompression Valve
  40
                            I
                                      I
I
               500
                        1000        1500        2000
                            Rotative Speed - RPM
          2500
                                                                           3000
 Figure 4.4-12.
               Maximum Allowable Condenser Pressure for Inlet Pressure
               of 1000 PSIA and 78 Ib. Recompression Valve Spring
               Preload.
                                     81

-------
              y = - fer (3 a2£-a3) = -1.924 x 10~4 W in.
                    obJ.
where  a = 5.75 in.
       H = 6.83 in. (fixed end of rod to top of piston)
       W = applied load, Ib.
       E and  I same as  previous values

      The top of the piston will be deflected 0.010 in. radially by a
side load of  52 Ib. applied at the free end of the rod  (approximately
the bottom of the piston).
                                   82

-------
            5.0   EXPANDER   MATERIALS
5.1   Materials Technology Base

      In recently designed steam reciprocating engines for automotive
            (12}
applications   ', maximum steam temperatures of about 700°F have been utilized.
Temperatures have been restricted to this general range by materials
limitations, in one form or another.  For example, the GM SE-101 engine
had a maximum steam temperature of 700°F in recognition of the possible
thermal degradation of the organic lubricant used to prevent excessive
wear at the steam piston cylinder interface.  A design pressure of 800 psia
arose from considerations of expander bearing loads.  These operating
conditions (700°F, 800 psia) might be considered to be current practice
in the design of steam engines for automotive applications.  Engine thermal
efficiency is quite limited, under these conditions, as compared with
modern central steam power stations operating at 1000° to 1050°F and
2400-3500 psia.  These advanced conditions have been reached by increasing
technology over the last 40 years, the same period in which there has been
essentially no increase in the cycle conditions for the steam car engine.

      The higher temperatures associated with the more efficient automotive
steam engine place constraints on materials due to mechanical strength
and corrosion considerations.  To insure the long life (*»• 3000 hours) re-
quired for satisfactory use in the reciprocating steam expanders, the
materials must be resistant to oxidation, corrosion by steam and have
adequate mechanical properties (yield strength, resistance to creep and
cyclic loading, hardness, etc.) at the anticipated operating conditions.
Surface properties can be favorably altered by the application of a sur-
face coating;  use of surface coatings are considered where appropriate,
         " *                       -v
i.e., wear surfaces.  In addition to the constraints imposed on materials
by the engine operating conditions,  economic constraints can limit the
utilization of existing technology in the solution to materials problems.
                                  83

-------
Cost is a major factor in large quantity production of automotive com-
ponents.  Economy dictates that the lowest cost alloy or material, with
properties suitable to the service conditions, be used.

      Probably, the most limiting factor in the successful operation of
highly efficient steam engines will be lubrication.  The dry conditions
which superheated steam imposes on a system at the operating condi-
tions can cause severe galling and excessive wear problems in those com-
ponents whose surfaces are in relative motion, i.e., the cylinder liner/
piston ring interface and the inlet valve face/seat and inlet valve stem/
guide interfaces.  Lubrication of these components are discussed in
Section 6.0.

      A literature search and industrial survey was made to identify and
evaluate the current materials technology in support of the expander de-
signs described in Section 4.0.  Personal visits were made to the Ford
Motor Company Automotive Research and Engineering Center and the General
Electric Company Diesel Engine Department, Large Steam Turbine and
Generator Department and Medium Steam Turbine and Generator Department
to  identify materials used in current reciprocating and steam turbine
components.  To supplement the information obtained from these visits,
library searches were made covering automotive components and oxidation
and wear resistant materials for use in dry steam environment through
the General Electric Technical Information Center Library and Automatic
Information Retrieval System, the SAE and ATME-ASM Transactions and the
Engineering Index.

      A summary of the results obtained from  the technology review follows:

      5.1.1  Cylinder Block, Cylinder Head, Intake Manifold. Exhaust Manifold

      Experience with materials in  large  central station  steam power  station
has shown  that  the useful  life of steel is essentially unlimited  if the  proper
steel is selected and the  service conditions  and water chemistry  are  properly
controlled.  Some steels have been  in service for  forty to  fifty  years.
Invariably when a failure  occurs the cause can be  attributed  to impurities in
the steam and the problem  usually is solved by correction of  the  service
conditions.  Plain carbon  and low alloy steels are commonly used  for  the
                                     84

-------
fabrication of pressure-containing components such as feedwater piping,
steam piping, valve bodies, flanges, discs, etc.,that are regularly  found
in steam generating systems.  The most commonly used steels are listed in
Table 5.1-1 together with the appropriate ASTM designations and maximum
service temperatures.

      The use of these steels up to the maximum temperature listed in
Table 5.1-1 will vary with the anticipated service conditions and from user
to user.  For example, one manufacturer limits the use of 1.25% Cr-0.5% Mo
steel to 800°F.  In actual practice carbon steels have been used success-
fully in steam service at temperatures to 750°F for 25 to 30 years.  Simi-
larly, the 0.5% Mo, the 1.25% Cr-0.5% Mo and the 2.25% Cr - 1.0% Mo alloy
steels have proven to be satisfactory for service at temperatures up to
850°, 1000°, and 1050°F, respectively.  However, there is a major restriction
in the use of 0.5% Mo steel in that it is not recommended for use in welded
structures.  Welded 0.5% Mo steel has a strong tendency to graphitize in
high temperature service.  All of these steels can be considered for use
in the major static components for automotive  steam engines.

      In contrast to the high pressure-containing components in steam generating
systems, the materials currently being utilized in the production of the
cylinder blocks, cylinder heads and intake manifolds of gasoline and diesel
fueled internal combustion engines for passenger cars,  trucks and locomo-
tives are gray cast iron and cast steel.   The most common gray cast iron
being used for the production of these components in passenger cars is
SAE G4000 (ultimate strength - 40,000 psi).  SAE G4000 contains 3 - 3.3% C,
1.8 - 2.1% Si, 0.6 - 0.9% Mn and consists of a lamellar pearlite matrix
with Type A flake graphite.  For parts subjected to higher pressure or
heavy duty engines, SAE G6000 (ultimate strength 60,000 psi) gray cast
iron is used.   In one large diesel engine the cylinder head is produced
from a cast,  weldable Mn-Mo steel (0.20% C - 1.35% Mn - 0.25% Mo).   The
steel casting is normalized at 1650°F (during the hardening of the valve
seats) to a hardness of 180 - 225 BHN.   This treatment produces an ultimate
strength of 90,000 psi,  a yield strength of 60,000 psi, tensile elongation
of 10% and reduction-in-area of 25%.   Special heat resistant parts are
frequently made from Ni-Mo alloy steel.  Intake manifolds in some internal
combustion engines are also being produced from cast aluminum alloy 355
                                  85

-------
Table 5.1-1

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.

Plain Carbon
Pressure Containing
Alloy
Carbon Steel
(0.35% C max)
Carbon Steel
(0. 25% C max)
Carbon Steel
(0.35% C max)
Carbon Steel
(0. 25-35% C max)
Carbon Steel
(0.06-0.18% C -
0.35% Si max)
Carbon Steel
(0. 27% C max)
0.25% C - 0.5% Mo
0.25% C - 0.5% Mo
0.15% C - 0.5% Mo
0. 20% C - 0. 5% Mo
0.15% - 1.0% Cr -
0. 5% Mo
0.15% C - 1.0% Cr -
0. 5% Mo
0.20% C - 1.25% Cr -
0.5% Mo
0.15% C - 1.25% Cr -
0.5% Mo
0.15% C - 1.25% Cr -
0. 5% Mo
0.15% C - 1.25% Cr -
0.5% Mo
0.18% C - 2.5% Cr -
1.0% Mo
0.15% C - 2.25% Cr -
1.0% Mo

and Low Alloy Steels Used for
Components in Steam Generating Plants
ASTM
Designation
A216 (Grade WCB)
A216 (Grade WCA)
A105 (Grades 1,
ID
A106 (Grades A,
B,C)
A192
A210
A217 (Grade WC1)
A182 (Grade Fl)
A335 (Grade PI)
A209 (Grade Tia)
A182 (Grade F12)
A225 (Grade P12)
A217 (Grade WC6)
A182 (Grade Fll)
A225 (Grade Pll)
A213 (Grade Til)
A217 (Grade WC9)
A182 (Grade F22)
86
(13)
Max. Service
Form Temp., °F
Casting
Casting
Forging
Pipe
Boiler Tube -
High Pressure
Superheater
Tube
Casting
Forging
Pipe
Boiler and
Superheater
Tubes
Forging
Pipe
Casting
Forging
Pipe
Boiler & Super-
heater Tubes
Casting
Forging

850
850
850
850
850
850
850
850
850
950
1050
1000
1050
1050
1050
1100
1050
1050


-------
                  Table 5.1-1  (Cont'd.)
       Plain Carbon and Low Alloy Steels Used for
Pressure Containing Components in Steam Generating Plants
Alloy
19.
20.
21.
22.
23.
24.
25.
26.
0.15% C -
1.0% Mo
0.15% C -
0.5% Mo
0.15% C -
1.0% Mo
0.15% C -
1.0% Mo
0.15% C -
0.5% Mo
0.15% C -
1.0% Mo
Type
(18%
Type
(18%
304
Cr -
321
Cr -
2.
2.
2.
3.
5.
9.
25% Cr -
0%
25%
0%
0%
0%
Cr -
Cr -
Cr -
Cr -
Cr -
SS
10% Hi)
SS
10% Ni-Ti)
ASTM
Designation
A225
A213
A213
A213
A213
A213
A213
A213
(Grade
(Grade
(Grade
(Grade
(Grade
(Grade
(Grade
(Grade
P22)
T3b)
T22)
T21)
T5)
T9)
TP304)
TP321)
Form
Pipe
Boiler
heater
Boiler
heater
Boiler
heater
Boiler
heater
Boiler
heater
Boiler
heater
Boiler
heater
\
Max. Service
Temt).. 8F

& Super-
Tubes
& Super-
Tubes
& Super-
Tubes
& Super-
Tubes
& Super-
Tubes
& Super-
Tubes
& Super-
Tubes
1050
1200
1200
1200
1200
1200
1500
1500
                           87

-------
(Al-5% Si - 1.25% Cu - 0.5% Mg - 0.1% Zn - 0.2% Ti - 0.1% Mn - 0.2% Fe).

      Materials used for the exhaust manifold in internal combustion
engines are subjected to the highest temperatures in the engine and would
correspond to the intake manifold of the steam expander.  Alloyed cast
iron is most commonly used for the exhaust manifold in the internal com-
bustion engine.  The SAE G4000d, a Cr-Mo alloyed cast iron and SAE G4000e
a Cr-Mo-Ni alloyed cast iron are typical.  The range of operating tempera-
tures that have been reported for the higher temperature components of
the internal combustion engine that require the use of alloyed gray cast
iron are:

           Component                     Operating Temperature Range. °F
      Cylinder Heads                           450 - 1000 (locally)
      Exhaust Manifolds                        300 - 1200
      Exhaust Valve Seat                       800 - 1300

      The alloyed cast irons can be considered for components of the steam
expander that operate at lower temperatures and pressures.  Where strength
is critical and greater reliability (safety) and oxidation resistance is
required, as is encountered in high pressure steam systems, nodular cast
iron is utilized.  Typical nodular cast irons are SAE D5506, D4512 and
D7003.  SAE D5506 has a yield strength (0.2% offset) of 55,000 psi and
a tensile elongation  (^ 2 inches) of 6%; the structure is ferritic-pearlitic.
SAE D4512 has a yield strength of 45,000 psi, a tensile elongation of 12%
and has a ferritic structure.  SAE D7003 has a yield strength of 70,000
psi, a tensile elongation of 3% and a pearlitic structure.

       5.1.2  Cylinder Liners/Piston Kings

      As stated previously, the high-speed relative motion between the
cylinder liner and the power piston rings in combination with the high
temperatures and pressures that exist within the cylinder chamber make
these components two of the most critical components in the engine.  Al-
though the materials currently in use as cylinder liners and piston rings
in gasoline and diesel fueled internal combustion engines were reviewed,
the technology is not directly applicable to the reciprocating steam
expanders being designed to operate with dry solid lubricants.  However,
                                   88

-------
 cylinder  liner/piston ring material  combinations  used in oil free air
 compressors  are  applicable and will  be  discussed  in Section 6.0.

      The materials most  commonly used  for  cylinder liners  under  oil
 lubricated conditions are alloyed gray  cast iron,  the Meehanite cast
 irons,  chromium  plated gray  cast iron and nitrided gray  cast iron.   Typical
 of  the  alloyed gray cast  iron is a composition  of  3.35%  C - 2.15% Si -
 0.65% Mn  - 0.4%  Cr; the liner is hardened by quenching from 1600°F into
 salt at 475°F followed by cooling in air.   The  resulting hardness is
 Rockwell  C 45 minimum.  The  Meehanite cast  irons are high strength,  fine
 grained castings combining the properties of cast  iron and  steel.  A
 commonly  used Meehanite,  Grade GA, has  an ultimate strength of  50,000
 psi.  Liners that are chromium plated or nitrided  are produced  from  centri-
 fugally cast gray cast iron.  The liners are centrifugal cast in  order to
 achieve the bore surface  integrity (minimum porosity)  that  is required
 for the hard chromium plating and nitriding processes.   A typical centri-
 fugally cast gray cast iron  used in  liners  for  large diesel engines  has
 an  ultimate strength  of 35,000 psi and  a hardness  of ^ 250  BHN.

      Gray cast  iron  also is the most common material used  for  oil lubricated
 compression and  oil piston rings.  Occasionally AISI 1070 and 52100  bearing
 steels  are used.  However, the steel  segment or steel rail  types  of  oil
 rings are almost always made of AISI  1070 to 1095  steels.   The  chromium
 plating of piston rings has  been found  to reduce wear  of  the rings and
 cylinders by •*• 75%.   However, chromium  cannot be run against itself  so
 that the  rings cannot be  chromium plated if  the liner  is  chromium plated.
 Chromium  plating is used primarily on compression  rings  in  heavy  duty
 engines and on oil rings of  the segmented type.  Since chromium plate
 reduces the fatigue endurance limit of  cast  iron,  higher  strength cast
 irons must be specified for  chromium  plating applications.  Two examples
 of  cylinder liner/piston ring combinations are  shown in Table 5.1-II.

      The K-28 nodular cast  iron (3.3% C - 2.2% Si - 0.5% Mo - 0.5%  Cu -
 0.05% Mg) is a centrifugally cast martensitic ductile  iron  and has a
 hardness  of 372  - 437 BHN.   The K-27  ring also  is  a  centrifugally cast
martensitic nodular cast iron with a  hardness of 250 - 283  BHN.   The K-6E
 ring is a statically  cast gray iron  (3.7% C  - 2.85%  Si -  0.45% Mo -  0.3% Cr)
 consisting of free graphite  flake in  a  fine  pearlite matrix and has  a

                                   H'J

-------
                             Table 5.1-11

                  Cylinder Liner/Piston Ring Material
               Combinations for Heavy Duty Diesel Engines
                                            Piston Rings
	Cylinder Liner	Type	Material	

1.  Hard Chromium Plated      Top Compression   K-28 Nodular Cast Iron
    Centrifugally Cast Gray   _      _-       _„_„    _    T
    Cast Iro                  Bottom Seal       K-6E Gray Cast Iron
                              Oil               K-Iron

2.  Nitrided Centrifugally    Top Compression   Hard Chromium Plated
    Cast Gray Iron                      _
                              Bottom Seal       K-Iron

                              Oil               K-Iron
                                   90

-------
hardness of 230 - 297 BHN.  The K-iron  (3.7% C - 2.65% Si) is a statically
cast unalloyed gray cast iron consisting of free graphite flake in a
matrix of pearlite and with an ultimate strength of 30,000 psi.

      5.1.3  Piston

      The pistons for automotive internal combustion engines are produced
primarily from aluminum die castings.  The light weight and excellent
thermal conductivity of aluminum alloys made them very attractive as the
trend in engines went toward higher speeds and higher compression ratios.
Most pistons are made from SAE 328 (Al, 11 - 12.5% Si, 0.9% Fe, 1-2% Cu,
0.5 - 0.9% Mn, 0.4 - 1.0% Mg, 0.05% Ni, 1.0% Zn, 0.25% Ti) and SAE 332
(Al, 8.5 - 10.5% Si, 1.2% Fe, 2-4% Cu, 0.5% Mn, 0.5 - 1.5% Mg, 0.5% Ni,
1.0% Zn, 0.25% Ti).  These alloys are usually heat treated (T5) to a
hardness of approximately 100 BHN and are tin plated for scuff resistance.

      For heavy duty engines, the piston head usually is produced from
alloyed (Cr-Mo) cast iron.  However, in order to achieve greater reliability,
at least one manufacturer has changed to a cast Mn-Mo steel (0.20% C -
1.35% Mn - 0.5% Si - 0.25% Mo) in the production of power piston heads.
The Mn-Mo steel has a minimum yield strength of 60,000 psi and a hardness
of 180 BHN.  In order to reduce weight of the cast steel power piston,
the lower portion is constructed from a forged aluminum alloy - usually
4032 alloy (11 - 13.5% Si, 0.8 - 1.3% Mg, 0.5 - 1.3% Cu, 0.5 - 1.3% Ni)
in the T-6 condition.  The 4032-T6 alloy has a minimum yield strength of
42,000 psi and a hardness of 115 BHN.

      One of the major reasons that the high silicon content aluminum
alloys are used in the fabrication of aluminum pistons is to minimize
the difference in thermal expansion between the cylinder wall material
and the piston material.  The coefficient of thermal expansion for 4032
alloy between RT and 572°F is reduced to 11.7 x 10   in./in./°F from
13.1 x 10~  in./in./°F for unalloyed aluminum.

      5.1.4  Piston Pin
          •*     " ; ~    - - -          ^
      Failure of piston pins rarely occurs.   However,  0.001 - 0.002 inch
wear of the piston pin results in a noisy engine so that selection of
materials to minimize wear is very important.  Piston pins art: normally

                                  91

-------
produced from AISI 1117 (1.15% Mn), 1016, 5115 or 5120 (0.8% Cr) steel
for automotive engines and AISI 8620, 8640 (0.55% Ni - 0.5% Cr - 0.2% Mo)
and 5046 (0.4% Cr) alloy steel for heavy duty engines.  The low carbon
steels are carburized and the higher carbon steels are case hardened by
induction to a surface hardness of Rockwell C 60.  Usually the surface
is polished to a high finish, i.e., on the order of 2 RMS.  It has been
determined experimentally that the load that the piston pin can withstand
without scoring is directly proportional to the surface finish - the
better the finish the higher the permissible load.  Under severe condi-
tions of loading, the surface of the pin can be roughened by shot peening
(followed by lapping to remove the displaced metal) to form depressions
in the surface in 'order to better hold the lubricant.

      5.1.5  Piston Pin Bearing

      There are a relatively large number of metals and metal alloys that
are used in sleeve bearings.  The bearing alloys can be grouped in the
following classes:   tin-base alloys, lead-base alloys, copper-lead alloys,
tin-bronze, silver,  aluminum alloys, zinc-base alloys, gray cast irons,
non-metallic materials  (PTFE) and overlays.  The proper selection of the
bearing material will depend upon  the  type of load, i.e., steady state or
cyclic, amount of load, speed,  temperature, corrosive  conditions, oil
supply, dirt contamination, and  shaft  hardness.

      Type of load - the  load carrying capacity of cyclically loaded
bearings, as in piston  pin bearings, is  usually determined by the bearing
fatigue strength of  the material.  This  property is of minor Importance
for bearings under constant load.

      Amount of load -  cyclic loads  up to 1500 psi are considered low and
can be supported by  ordinary babbitt bearings  (tin or  lead base alloys).
Thin babbitt overlays can operate up to  2500 psi.  The limit of alternating
loads for  the copper-lead bearing material is approximately 4000 psi.  A
load of 5000 psi can be supported by a trilayer bearing consisting of a
mild steel  (AISI 1010)  back, a  25% Pb-Cu alloy intermediate layer and a
plated babbitt overlay.   The latter  bearing is used in some heavy duty
engines.
                                   92

-------
      Speed - babbitt alloys are best suited for high speed operation.

      Temperature - strength and corrosion rates are affected by the
service temperature.  Certain excellant bearing materials, such as babbitts,
are restricted by their low melting points.  The tin-bronzes such as
SAE 62 (86.7% Cu - 10% Sn - 10.3% Pb - 2.0% Zn - 1.0% Ni) and phosphor
bronzes such as SAE 64 (78% Cu - 10% Sn - 10% Pb - 0.75% Zn - 0.5% Ni -
0.5% Sb) can be used as bearings at temperatures approaching 600°F.  Maxi-
mum load  for these materials is *• 4000 psi.  Both materials have excellant
fatigue properties but poor anti-seizure, conformability and embeddability
properties; anti-seizure can be improved by silver plating.

      Corrosion Resistance - the tin-base babbitt alloys are the most
resistant to the corrosive action of acids that are formed in lubricating
oils.

      Oil Supply - a common cause of failure in bearings is the loss of
the oil supply or breakdown in the oil film.  This results in direct
contact between the pin and the bearing surfaces and a large increase in
friction.  When this happens, the anti-seizure characteristics of the
bearing are most important.  The rating of materials for this property
in order of decreasing anti-seizure qualities are:  tin babbitt, lead
babbitt, aluminum alloys, copper-lead alloys, leaded bronzes, silver and
bronze.

      Dirt Contamination - in bearing applications where there is a high
level of contamination by particulate matter, the bearing material must
possess good embeddability.  Dirt can be embedded in soft babbitt ma-
terials without doing harm.

      Shaft Hardness - the harder the bearing, the harder the shaft must
be to prevent damage to the shaft.  In general applications, soft babbitt
bearings can be used satisfactorily with steels as soft as 130 - 165 BHN;
for copper lead (to 25% Pb) the minimum hardness of the steel shaft must
be 165 - ZOO BHN; for aluminum alloys the minimum hardness of the steel
shaft must be 200 - 300 BHN; and for leaded bronze the minimum hardness
of the steel shaft must be 300 - 400 BHN.
                                   93

-------
      A listing of recommended bearing materials for various bearing loads
in internal combustion engines is given in Table 5.1-III.  The materials that
are currently being used for piston pin bearings in the two leading large
diesel engines for locomotives are (1) a leaded tin bronze (80% Cu -
10% Sn - 10% Pb), both as a solid bushing and with a AISI 1010 steel
backing (SAE 792) and (2) a tri-metal bearing consisting of a AISI 1010
steel backing, a 0.015 in. thick silver intermediate layer and a 0.0003
in. thick lead overlay.  These bearings require a good finish and many
axial grooves for pressurized oil lubrication.  A crankshaft bearing
material that is currently in use in automotive engines is a 0.030 in.
thick SAE 780 (8280) aluminum alloy sheet (6% Sn - 1% Cu - 0.5% Ni -
1.5% Si) bonded to a 0.001 in. thick Ni plated AISI 1010 steel backing.
A recommendation received from one of the leading aluminum producers
also called for the use of a leaded tin bronze bearing for the piston
pin bearing and a AISI 1010 steel backed aluminum alloy bearing for the
crankshaft.

      5.1.6  Connecting Rod

      The high alternating loads imposed on the connecting rod in re-
ciprocating engines require the use of relatively high strength steels
of high quality.  For this reason connecting rods are generally produced
from contour forgings made from vacuum degassed ingots.  The most common
steels used for the production of connecting rods are AISI 1041 for use
in automotive engines and AISI 8640 (0.55% Ni - 0.5% Cr - 0.2% Mo), 4140
(0.95% Cr - 0.2% Mo) and  4340  (1.8% Ni - 0.8% Cr -  0.25% Mo) for heavy
duty and diesel engines.  One automotive engine manufacturer utilizes
SAE 80002 malleable iron  castings for the production of connecting rods.

      5.1.7  Camshaft/Cam/Cam Follower  (Tappet)

      The cyclic loading  and high localized surface stresses that occur
as a result of the action between the cam surface and cam follower  (tappet)
makes the materials selection for these components  very  important.  In
some engines, the localized Hertzian stress on the  cam surface approaches
200,000 psi.  Numerous combinations of camshaft/tappet materials are  in
common use.  Hardenable gray cast iron, such  as SAE C4000  is most widely
used for automotive camshafts.  A typical composition IH 3.3% C -  2.25% SI -
0.6% Mn - 0.95% Cr - 0.5% Mo.  As cast, the camshaft has a  hardness of
                                  94

-------
                                                      Table 5.1-III
                                 Bearing Materials Used in Internal Combustion Engines
       Bearing Load, psi
   Lowest Cost Material
              Alternate Materials
                                     Advantages of
                                  Alternate Materials
          1200 max
          1200 - 2200
vo
          2200 - 2800
          1200 max
                                            (Small and Medium Size Engines)
Pb Babbitt on steel
(0.015-0.030 in.)
Pb Microbabbitt on steel
(0.002-0.005 in.)
Babbitt-impregnated sin-
tered Cu-M on steel
                            Babbitt-impregnated sin-
                            tered bronze on steel
                            Cu-25% Pb on steel
                            Al alloy on steel
Pb Babbitt on steel or
bronze (0.015-0.030 in.)
Zr or cast Al alloy
Sn Babbitt on steel
(0.015-0.030 in.)

Babbi tt-impregnated
sintered Cu-Ni on steel

Babbitt-impregnated
sintered bronze on steel

Sn Microbabbitt on steel
Al alloy on steel or solid

Cu-35% Pb on steel
Solid Al alloy or Al alloy
on steel

Above materials + 0.001 in.
Pb alloy overlay

Cu-25% Pb on steel + 0.001 in.
Pb alloy overlay

Al alloy on steel + 0.001 in.
Pb alloy overlay
(Large Engines)

         Sn Babbitt on steel or bronze
         Solid Al alloy
                                           (a) (b) (c)
                                           (d)
                                           (a) (b) (d)


                                           (a) (b) (d)


                                           (d)
                                           (a) (b) (d)

                                           (a) (b)
                                           (a) (b) (d)


                                           (b) (e)

                                           (d) (e) (f) (g)


                                           (d) (e) (f) (g)
                                   (d)
                                   (c)  (d)

-------
                                                 Table 5.1-III (Cont'd.)
                                 Bearing Materials Used in Internal Combustion Engines
       Bearing Load, psi
   Lowest Cost Material
           Alternate Materials
   Advantages of
Alternate Materials
\o
          1200 - 1500
          1500 - 2000
          2000 min.
Pb Babbitt on steel or
bronze (0.020 in. max)
(Large Engines)
      Pb Microbabbitt on steel or
      bronze
Pb Microbabbitt on steel
or bronze (0.002-0.005 in.)
Solid Al alloy
      Aolid Al alloy

      Cu-Pb (30-40% Pb) on steel
      Solid Al alloy or Al alloy on
      steel
      Above materials + 0.001 in.
      Pb alloy overlay

      Cu-25% Pb. on steel + 0.0005-
      0.002 in. Pb alloy overlay
      Al alloy on steel + 0.0005-
      0.002 in. Pb alloy overlay
(a) (b)

(c) (d)

(a) (b)
(c) (d)

(b) (e)


(a) (b)  (c)


(d) (g)
       (a)   Yield strength
       (b)   Fatigue strength
       (c)   Possible cost savings
       (d)   Corrosion resistance
       (e)   Conformability
       (f)   Embeddability
       (g)   Inherent lubricity

-------
248 -  311 BHN; the cam surfaces are  flame or induction hardened  to  a Rockwell
C hardness of 54 min.  For heavy duty engines the camshafts are  forged  and
are usually produced from water quenched carbon steels and low alloy steels
of 0.5 - 0.7% C, AISI 4340, or a curburizing grade such as AISI  8620 steel.
Heat treatment of the high carbon steels to achieve the necessary hardness
of the cam surfaces is accomplished by either conventional through-hardening
or case hardening processes using flame or induction.  One diesel engine
manufacturer uses a forged AISI 1080 shaft which is quenched and tempered
to a hardness of 235 BHN.  The cam and bearing surfaces are induction case
hardened to a hardness of Rockwell C 60 - 65 to a depth of 0.060 in.  Care
must be taken to insure sufficient depth of case as thin, brittle surface
hardened cases tend to spall in service.

      Tappets usually fail by scuffing or rapid loss of surface.  The most
common automotive tappet material is a gray cast iron of a composition similar
to 3.2% C - 2.25% Si - 0.8% Mh - 1.1% Cr - 0.6% Mo - 0.55% Ni.  The  cast iron
tappet is hardened by quenching in oil from 1550°F and tempered to a hardness
of Rockwell C 55 - 60.  Steel tappets also are used and are produced from
hardenable steel such as 52100 bearing steel or high-carbon molybdenum steels
of the AISI 4000 series or from carburizing grades of steel such as AISI 51200
or 8620.

      In some heavy duty engines, where tappet wear has been a severe problem,
it was found that the utilization of material containing varying amounts of
carbides in the microstructure improved the service life.   In one case the
successful use of hardened and tempered D-2 tool steel in solving the wear
problem was attributed to free carbides in the microstructure.  The D-2 alloy
has a high carbon content in conjunction with stable carbide formers.  The
composition is 1.5% C - 12% Cr - 0.8% V - 1.0% Mo.   A higher volume carbide
content in the microstructure can be obtained with alloys such as Stellite
Star J (2.5% C - 32.5% Cr - 17.5% W - Bal Co)  and this material at Rockwell
C 60 hardness is being used for tappets in some heavy duty engines.   Experience
has shown that the Star J alloy shows superior performance against a nitrogen
rich surface in comparison to a carburized surface.   It should be noted that
          -*
if a nitrided surface is to be used for the cam,  a carbonitride process should
be utilized in order to obtain the desired case depth to carry the high
loads.   The core depth of a straight nitrided steel surface is too thin for
application on cam or tappet surfaces.   A solid tungsten carbide cermet also
                                    97

-------
is being considered for use as a tappet material.  In some very heavily loaded
cam/tappet designs a roller cam follower is employed.  An AISI 8620 steel
carburized to a surface hardness of Rockwell C 55 min. and a case depth of
0.060 inch has performed satisfactorily in roller type cam followers.  Again
a surface finish of 2 - 4 RMS is recommended.

      Tappet faces are usually finished to about 6 RMS for the same reasons
a good finish is required for the piston pin, i.e., improved load carrying
ability.  Shot peening of the tappet surface is sometimes done to improve
the retention of the lubricant.  Coating of the highly finished tappet surface
to aid in the wearing-in process and improve frictional characteristics is
common practice.  An oxide coating (Fe,~0,) is generally applied to chilled
cast iron and a phosphate coating is generally applied to hardened steels or
gray cast irons.  The phosphates can be manganese phosphate, zinc phosphate or
iron-manganese phosphate.  Other coatings that have been used are oxidates,
manganates, and sulfides.

      The following material combinations were considered as candidates for
the cam and tappet in the single cylinder engines:
               Cam

      1.  AISI 8620/cairburized
          (Re 58 min.)

      2.  AISI 8620/carburized
          (Re 58 min.)

      3.  AISI 8620/carbonltrided
          (Re 60-65)

      4.  Hardenable Cast Iron
          (Re 54 min.)
      5.  Hardenable Cast Iron
          (Re 54 min.)
        Tappet

Chilled Cast Iron (Re 54)
Ferrox Coating (Fe-O,)
Carboloy 883 (WC + 6% Co)
(RA 92)

Star J (Re 61)
Hardenable Cast Iron
(Re 55-60)/Mh-Fe Phosphate
Coating

D-2 Tool Steel (Re 58-60)/
Mh-Fe Phosphate Coating
                                    98

-------
       Bearing materials used for camshafts in heavy duty diesel engines are
 generally cast aluminum alloys.   Aluminum alloy 750 (6.25% Sn - 1.0% Ni -
 1.0%  Cu)  is  commonly used.

       5.1.8    Inlet  Valve

       Operating conditions  for valves  in large central  steam power stations
 vary  considerably from the  operating conditions of valves for automotive
 and diesel internal  combustion engines.   The operating  conditions  for steam
 valves for central power stations are  much less severe  than for the valves
 in internal  combustion engines.   Actuation of many steam valves is on an
 infrequent basis so  that fatigue and wear of the valve  face and seat is not
 a severe  problem.  Service  temperatures  of the steam valve are significantly
 lower than those for the exhaust valve of an internal combustion engine and
 they  do not  fluctuate as they do in  the  internal combustion engine;  thus
 creep and thermal fatigue problems are less severe.  In addition,  corrosion
 as a  result  of high  temperature  combustion products  is  not a problem in steam
 valves.

       In  sumnary,  the years of experience in the operation of steam valves
 or exhaust valves  in internal combustion engines are not  directly  applicable
 to the selection of  materials for inlet  steam valves in reciprocating
 steam expanders.   Selection of materials for the various  components  of  the
 steam inlet  will have to be based on the knowledge of the basic properties
 of the materials with respect to corrosion in steam, mechanical properties
 and wear  behavior  under conditions  of no lubrication and as  supplemented
 with  valve experience in reciprocating internal  combustion engines.

      Inlet steam valves in large central power steam systems are generally
fabricated from a martensitic 12% Cr stainless steel (AISI 410) or Crucible
422 alloy  (12% Cr - 1% Mo - 1% W - 0.8% Ni - 0.25% V);  the stem and face
are nitrided.  Both the valve seat and  valve guide are made from Stellite
6B.   The Stellite 6B on the valve seat  usually is applied as a weld over-
lay on a low alloy steel (2.25% Cr - 1.0% Mo alloy).
                                     99

-------
      The materials in use in valve components for internal combustion
engines will vary considerably with the operating temperature of the valve.
The lower strength and increased corrosion and wear rates that accompany
increased service temperatures necessitates the use of more highly alloyed
materials as the temperature of the valve increases.  Since the inlet
valves in internal combustion engines operate at temperature under 600°F
and exhaust valves operate at temperatures as high as 1550°F, it is ob-
vious that the most severe materials problems are associated with the
exhaust valve.  The development of improved valve materials over the years
has been primarily by empirical methods.  This stems from the fact that
it is not possible•to simultaneously reproduce all of the service condi-
tions of the valve by any other means than in an actual engine.

      Materials that are recommended for use in inlet and exhaust valves
in various types of engines are listed in Table 5.1-IV.  The steels in Group
A of Table 5.1-IV are used for light duty inlet valves that operate at low
temperature or for short times.  They also are used for stem materials
in two-piece valve construction where they can be welded to higher alloy
heads and used for heavier duty inlet and exhaust valves.

      Steels in Group B through E  are especially made for valve applica-
tions with the Group B steels being the least expensive and the Group E
steels being the most expensive.   The sigma phase forming steels in Group
D have good hot hardness and superior resistance to wear than the austenitic
steels of Group E.  However, they  are more brittle and have less resistance
to creep than the Group E steels.  The nickel-rbase super alloys listed
in Group F are only used where valve temperatures are very high because
of their high cost.

      Short operating lives of valves can usually be traced to permanent
dimensional changes that take place in the valve components during service
as a result of inadequate creep strength or poor wear and/or hot corrosion
characteristics.  The terms used to describe the dimensional changes are:
      a.  elongation,
      b.  projection of the stem,
      c.  face runout, and
      d.  tip recession.

                                    100

-------
                                                Table 5.1-IV

                Materials Used for Inlet and Exhaust Valves in Internal Combustion Engines
          Application
                    Valve Materials
 Approx. Max.
Exhaust Valve
  Temp. °F
A.  Intake Valve-Light Duty
B.  Intake Valve-Heavy Duty
C.  Intake Valve-Heavy Duty
    Exhaust Valve-Light Duty
D.  Exhaust Valve-Heavy Duty
                    Carbon Steel

    SAE NV-1 (AISI 1041)
(a)  SAE NV-2 (AISI 1047)
             (AISI 1050)

                  Low Alloy Steels

    SAE NV-4 (AISI 3140)
             (AISI 4150)
    SAE NV-6 (AISI 5150)
             (AISI 6145)
    SAE NV-5 (AISI 8645)
                Martensitic Steels

    SAE HNV-2 (Sil F) 4 Si-2.25 Cr (0.4C)
                      4 Si-2.25 Cr 1.5 Ni-0.85 Mo (0.4C)
                Martensitic Steels

(b)  SAE HNV-3 (Sil 1) 3.25 Si-8.5 Cr (0.45C)
                      2.75 Si-7.5 Cr-1.5 Ni-0.85 Mo (0.3C)
(b)  SAE HNV-6 (XB)    2.25 Si-20.0 Cr-1.5 Ni (0.8C)
            Austenitic-Sigma Phase Alloys

    SAE EV-1 (SCR)  23.5  Cr-4.75 Ni-2.75 Mo (0.45C)
    SAE EV-2 (TXCR) 24.0  Cr-3.75 Ni-1.35 Mo-3.75 Mn (0.4C)
     1350
     1550

-------
                                                 Table 5.1-IV (Cont'd.)

                     Materials Used  for  Inlet  and Exhaust Valves  in  Internal  Combustion Engines
               Application
                    Valve Materials
 Approx. Max.
Exhaust Valve
  Temp. °F
     E.  Exhaust Valve-Heavy Duty
o
NJ
     F.  Exhaust Valve-Heavy Duty
               Austenitic Steel Alloys                              1550

    SAE EV-7 (2155N)  21 Cr-5 Ni-5.5 Mn (0.2C,  0.25N)
                     21 Cr-4 Ni-7 Mn (0.4C,  0.1N,  0.22P)
(a)  SAE EV-8 (21-4N)  21 Cr-4 Ni-9 Mn (0.4C,  0.4N)
    SAE EV-5 (Sil 10)19 Cr-8 Ni-3 Si (0.4C)
    SAE EV-4 (21-12N)21 Cr-12 Ni (0.2C,  0.2N)
             (Cast)   25 Cr-12 Ni (0.2C)
    SAE EV-9 (TPA)    14 Cr-14 Ni-2.4 W-0.35  Mo (0.45C)
             (Cast)   15 Cr-15 Ni-3.5 Si-0.4  Mo (l.OC,  0.25 Cu)
               Austenitic Ni Alloys                                1650
                                        SAE HEV-2  (Inconel M)
                                        SAE HEV-3  (Inconel X-750)

                                    (c) SAE HEV-5  (Nimonic 80A)

                                        SAE HEV-6  (Nimonic 90)
                               16 Cr-Bal Ni-3 Ti-0.5 Al (0.03C)
                               15 Cr-Bal Ni-1 Cb-2.5 Ti-0.9 Al
                               (0.04C)
                               20 Cr-Bal Ni-2.5 Ti-1.2 Al
                               (0.05C)
                               20 Cr-Bal Ni-18 Co-2.5 Ti-2.2 Al
                               (0.05C)
     (a)
        Used in current automotive engines

        Used in current high performance automotive engines - intake valves
     (c)
        Used in current heavy duty engines

-------
Elongation  ("wire  drawing") is  the permanent increase in  length  as  measured
from  the valve  face  to  the tip.  This  causes a  decrease in  the "lash"
(clearance  between valve  tip and tappet) so that  the valve  face  does not
seat  properly and  results in "blowby"  of the exhaust gases.   This can  be
corrected by using a material with greater resistance to  creep.  Face
runout is caused by  hot corrosion which also allows leakage of the  ex-
haust gases.  Stem projection is similar to elongation in that it reduces
"lash"; it  is the  total elongation as measured  through the valve guide
and includes any wear of  the seat and valve face  and elongation of  the
stem.  Tip  recession results in an increase "lash" and leads  to fracture
of the valve.   Face  runout and  stem projection, due to corrosion and wear
of the seat and valve face and  tip recession, can usually be  corrected by
appropriate use of valve  seat inserts  and/or corrosion resistance and
hard  facing materials.  Commonly used  insert and  facing materials are
listed in Table 5.1-V.  The Group D and E materials are used most often
with  the cobalt base alloys, Group E,  are preferred for improved re-
sistance to corrosion.

      Examples  of  material combinations used in exhaust valve components
for current automotive  and large diesel internal  combustion engines are
given in Table  5.1-VI.

5.2   Expander  Materials  Selection Study

      The structural materials  used in the containment of high tempera-
ture  (1000° - 1050°F) and high  pressure (2400-3500 pai)  steam as  well as ma-
terials used for specialized components such as steam valves have been
studied in  depth for years and  their properties are well  documented.
Similarily, materials in  use in heavy  duty and  high performance recipro-
cating engine components  are well established and most of these materials
are adaptable to use in corresponding  components  of the reciprocating
steam expander.  For this reason,..there was no  need to conduct experi-
mental investigations of  candidate expander materials for this program.
                                   103

-------
                                               Table 5.1-V
                      Materials Used for Valve Inserts and Hard Facing Applications
    Material
Nominal Composition, %
                                                                                    SAE Alloy
                                                                                  Designation
C
A.
B.
C.

D.

E.




Cr-Mo and Cr-Mo-W Steels 0.
(Inserts) (a) 1.
1.
0.
1.
Cr-Mo Cast Iron 2.
(Inserts) 2.
W Steels 0.
(Inserts)
Ni Alloys
(Facings)
Co-Cr-W-Ni Alloys
(Facings)



0.
2.
0.
1.
(b) 1.
1.
2.
(c) 2.
65
00
00
65
35
50
25
50
55
00
20
00
25
60
50
40
Mn Si Cr
0.6 1.00 3.
0.6 2.50 4.
0.6 0.25 5.
0.35 0.25 5.
0.35 0.45 3.
0.60 2.00 3.
1.00 1.00 2.
4.00 -
0.25 3.
0.40 0.30 25.
1.00 1.00 19.
28.
- 28.
- 24.
- 30.
- 29.
0
0
0
0
5
0
5
5
0
5
0
0
0
0
0
Ni Mo
W
- 5.00 -
8.50 -
- 1.25 -
3.00 3.5
1.0 6.50 5.5
5.00 -
7.50 -
9.5
- -
60.0 -
78.0 -
_ _
3.0 -
24.0 -
- -
39.0 -
13.0
8.5
—
4.0
4.5
12.5
11.0
15.0
C.o
0.
1.
0.
25 (V)
00 (V)
20 (Cu)
HNV-7
_
—
67.
55.
37.
52.
10.


0
0
0
0
0
VF-4
VF-1

VF-2
VF-5

VF-3
(X-782)
(80-20 NiCr)

(Stellite 6)
(Stellite F)

(Eatonite)
(a)
(b)
(c)
Light duty on large gasoline and diesel engines.
Strongest and toughest of the Co base alloys.
High hot hardness and resistance to shock and pitting.

-------
                                               Table 5.1-VI
    Exhaust
Valve Component
                         Material Combinations Used in Exhaust Valve Components
                                     for Internal Combustion Engines
                                                     Material
                                                              (a)
                     Light Duty
                   Spark Ignition
                                 Heavy Duty
                               Spark Ignition
  Heavy Duty
    Diesel
Head

Face (overlay)

Stem

Seat (Insert)



Guide
SAE EV-8 (21-4N           SAE HEV-5 (Nimonic)

SAE VF-5 (Stellite F)     SAE VF-3 (Eatonite)

SAE EV-8 (21-*4N)(b^       SAE HEV-5 (Nimonic) 80A

        (Eatonite or L.E. Jones Alloy)
                                                                   (b)
                                -(Gray Cast Iron>
                                                                           Inconel 751
                                                                                         (c)
SAE VF-2  (Stellite 6)

AISI 3140(c)

AISI 410 SS
(Re 40-49)
Weld overlay

Gray Cast Iron
(Type A Flake in
Pearlite)
(a)

(b)

(c)
Material compositions are listed in Tables  5.1-IV and 5.1-V

One piece construction.

Two piece construction - flash welded.

-------
      However, prior to the selection of the structural materials for
the two single cylinder steam expanders, their strength properties and
compatibility in a steam-air environment were reviewed.  In addition,
material and process specifications were established for each component
in the steam expanders.

      5.2.1   Material Properties

      5.2.1.1 Low Carbon and Low Alloy Steels

      The use of any one steel for structural applications and contain-
ment of high pressure steam primarily depends on the sustained metal
temperatures.  Low carbon steel is used for steam generation tubes and
the low temperature regions of the superheater where the metal tempera-
tures do not exceed 750° - 800°F.  As the temperatures in the steam
generating system increases, steels of higher alloy content are employed.
The C-0.5% Mo steel can be used up to about 850°F; the low chromium-
molybdenum steels (1.0 - 2.25% Cr) up to 1000° - 1050°F; the intermediate
chromium-molybdenum steels (3 - 9% Cr) up to 1100°F; and the austenitic
stainless steels above 1100°?.

      The temperature limits imposed on the various steels are the result
of loss in elevated temperature strength (yield and creep), structural
changes in the microstructure of the steel that may detrimentally affect
the properties, i.e., ductility, and oxidation-steam corrosion considera-
tions.  The maximum allowable design stresses for the low carbon, low
alloy and austenitic stainless steels used in boiler construction as a
function of temperature are shown in Figure 5.2-1.  The maximum allowable
stress is based on the ASME Boiler and Pressure Vessel Code which limits
the allowable stress to the level at which the creep rate is 0.01% in
1000 hours at the design metal temperature.  Additional restrictions on
the allowable stresses are imposed by the code and they are based on the
stress-rupture strength of the material as modified by suitable allowances
for loss in metal cross-section due to oxidation and corrosion.  From
Figure 5.2-1, the allowable stress for low carbon steels begins to drop
above 700°F and at 850°F is only half the value at room temperature.  The
maximum allowable stresses for the low alloy Cr-Mo steels are sustained
up to about 800°F whereupon they start to drop and reach 50% of the room

                                  106

-------
                                                Carbon steel
                                                Tube  SA-192-A
                                                Carbon steel
                                                Tube  SA-210
                                                Pipe  SA-106-B
                                                Plate  SA-201-B
                                                C-0.5Mo
                                                Tube  SA-209-T1
                                                1.25Cr-0.5Mo
                                                Tube  SA-213-T11
                                                Pipe  SA-335-P11
                                                Plate  SA-387-C
                                                2.25Cr-lMo
                                                Tube  SA-213-T22
                                                Pipe  SA-335-P22
                                                Plate  SA-387-D
                                                18Cr-10Ni-Ti
                                                Tube  SA-213-TP321H
                                                     SA-312-TP321H
                                                     SA-376-TP321H
                                                     SA-240-321
                                                16Cr-13Ni-3Mo
                                                Tube  SA-213-TP316H-
                                                Pipe  SA-312-TP316H
                                                     SA-376-TP316H
                                                     SA-240-316
                              Pipe
                              Pipe
                              Plate
                             Pipe
                             Plate
                                                   I
   200    300  600  800  1000   1200  1400  1600

                Metal Temperature,  °F
Figure 5.2-1.
Maximum  Allowable  Design Stresses  for
Low Carbon, Low Alloy and Austenitic
Stainless Steels in Large Steam
Generating Systems.   Reference 14
                               107

-------
temperature values at 1000°F.  The 50% loss in the maximum allowable de-
sign stresses for the austenitic stainless steels occurs at about 1200°F.
Of the low alloy, Cr-Mo steels, the 1.25% Cr-0.5% Mo alloy has superior
creep rupture properties at temperatures on the order of 1000°F, Table 5.2-1
As chromium is added to achieve superior oxidation resistance, the creep-
rupture strengths are reduced.  Care must be exercised in the fabrication
of components from these steels, i.e., amount of cold work, heat treat-
ment, etc., such that grain size and phase morphology are not adversely
affected with respect to their influence on strength and ductility.

      The  thermal stability of the low alloy steels also restricts the
temperature at which the steels can be used.  Phase changes, such as the
graphitization of carbon steels above 750°F and C-Mo steels above 850°F,
can result in drastic reduction in strength and ductility of the steel.
Chromium additions of more than 0.5% in the steel eliminate this problem.

      Compatibility with the  surrounding environment is another important
consideration in the selection of the structural material.  In the steam
expander,  as is  the case for  the boiler, the exterior surfaces of the
containment materials are subject to oxidation by oxygen in the air and
the internal surfaces are oxidized by the oxygen in the steam.  Each
steel has  a threshold temperature above which rapid oxidation or corrosion
takes place generally due to  the formation of a thick porous scale.  As
a result the load carrying ability is reduced and, in the case of boiler
tubes,  restricted flow  and reduced heat transfer efficiency occurs.  Suf^
ficient oxidation and corrosion resistance can be achieved in the steels
by the  addition  of small amounts of chromium usually in excess of 1%.  The
chromium in the  steel improves the oxidation resistance by promoting the
formation  of a tightly  adhering scale which  inhibits further oxidation.
The degree to which chromium additions improve the oxidation resistance
of low  alloy steels in  steam  can be seen in Figures 5.2-2     and 5.2-3^    .
In Figure  5.2-2, the data are plotted as penetration in mils/yr. vs. tem-
perature and in  Figure  5.2-3, the data are plotted as penetration in mils
as function of time at  1100°F.  It is important to point out that these
data (Figure 5.2-2) were obtained under actual service conditions from
samples periodically cut from a large steam plant  (Detroit Edison) and
show higher corrosion rates  than in steady state laboratory tests by a
factor  of *v/ 2.
                                   108

-------
                          Table 5.2-1



  Creep Rupture Properties of Cr-Mo Alloy Steels at
                Stress  (ksi) to Produce    „       -   _       . ,  .
                           7               Stress  for Rupture, ksi
Alloy
1
1.
2.
3
5
7
9
Cr
25
25
Cr
Cr
Cr
Cr
- 0.
5
Cr -
Cr -
- 1.
- 0.
- 0.
- 1.
0
5
5
0
(a)
Mo
0.5 Mo
1.0 Mo
Mo
Mo
Mo
Mo
a iiireep nani uj. j./0 in
10,000 hours
11
12
12
10
9
.8
.0
.0
.5
.0
8.0
12
.0
L 	
1000 hrs.
28.
27.
20.
18.
19.
18.
22.
4
0
4
0
0
8
8
10,
18.
21.
15.
14.
14.
14.
19.
000 hrs
8
0
4
7
6
0
1
Annealed condition.
                              109

-------
   70
   60
   50
40
   30
t-l
CO
Q>
10
01
U
CO
OS

C
O
•H
to
o
o
o

e
3  20

en

u 10
>
     950
                                                              lCr-0.5Mo

                                                       u2.25Cr-0.5Mo
                                                          C-O.SMo
                                           Carbon Steel
             1000
                                                                 5Cr-0.5Mo_


                                                                 	—"

                                                                \   9Cr-lMo
1200
                       1050        1100        1150

                              Temperature, °F


Figure 5.2-2.  Effect of Temperature on Long Time Steam Corrosion Rates.
                                                                             1250
   0.20
   0.15  —
§
0)
c
•i
   0.05
      0
                                                                            7.5
           Figure 5.2-3.
                              Time, hours x 1


                        Effect  of  hong Time Exposure to Steam at 1100°F.


                                      110

-------
      Similar long time oxidation-corrosion  tests in full size  commercial
tubing  (2 in. OD x 0.5 in. wall) were  conducted at  the Philip Sporn  plant
of the American Electric Power Company.  In  these tests, scaling  data
were obtained on the internal and external surfaces after varying periods
of exposure at 1100°, 1200°, 1350°, 1500°F and 2000 psi steam pressure.
Data for the 1100°F exposures are given in Table 5.2-II.  There is little
difference in scale" thickness between  the two surfaces   ' .

      5.2.1.2 Medium Carbon Low Alloy  Steels

      A number of dynamic components in the steam expander require the
use of high quality relatively high strength materials.  For these com-
ponents, i.e., power piston head, piston rod, connecting rod and bolts,
the following medium-carbon, low alloy steels were reviewed:

                                      Nominal Composition, %
                  Alloy	C     Ni   Cr   Mo    V
                  AISI 4140          0.4    -   1.0  0.2
                  AISI 4340          0.4   1.8  0.3  0.25
                  17-22-A            0.45   -   1.0  0.55  0.3
                  H-ll               0.35   -   5.0  1.5   0.4

      The 0.2% yield strength of these alloys are compared in Figure  5.2-4.
H-ll steel has an exceptional combination of high strength and toughness
and is used for critical and highly stressed components.   When the alter-
nating stress endurance strength is the limiting criteria, vacuum melted
grades of the steels are specified.   For example,  the endurance limits
for vacuum melted and air melted AISI 4340 alloy steel heat treated to
a strength level of 200,000 psi are 105,000 psi and 90,000 psi^respectively,
at room temperature.

      The oxidation-steam corrosion characteristics of the medium-carbon,
low alloy Cr-Mo-V steels are similar to the low-carbon, Cr-Mo alloy steels.
Data for a Cr-Mo-V steel are given in Table 5.2-II.

      5.2.1.3* Cast Iron

      For reasons of cost and in certain wear applications,  cast iron
should be specified wherever possible.   Cast iron grades usually are

                                  111

-------
                            Table 5.2-II
fa")
Oxidation-Corrosion of Cr-Mo Steels at 1100°FV '
Alloy 
-------
     280
 S
    240
    200
co
P.
CO
CO
0)
CO
160
    120
     80
     40
                             Note: All  Steels in Hardened and
                                   Tempered Condition
                               H-ll
                               17-22-A
                               4340
                               4140
                 510-560 BHN
                 311-363 BHN
                     340 BHN
                     360 BHN
                                                               .H-ll
                                                           4340
                                I
                                        I
       0
              200
400         600

 Temperature, °F
                                                       800
                                                              1000
     Figure  5.2-4.   0.2% Yield Strength Medium Carbon Low Alloy  Steels,
                                   113

-------
specified to a specific strength level rather than to a chemical composi-
tion, i.e., SAE G4000, and numerous grades of cast iron are available.
The ultimate strengths and endurance limits of gray cast iron generally
are not affected by temperature until above 800°F.  For example, a 2.84C -
1.5 Si gray cast iron with an ultimate strength of 48,400 psi has a fatigue
endurance limit of 20,000 psi.  With the exception of a slight dip in
strength levels at the intermediate temperatures, to 42,000 psi ultimate
strength and 18,000 psi endurance limit, the initial strength levels are
retained to a temperature of approximately 800°F after which they decrease to
corresponding levels of 35,000 psi and 14,000 psi.  Notched fatigue strength
usually are within 10% - 20% of the unnotched strength level.

      Although improvements in the high temperature strength of gray cast
irons are possible by alloying with molybednum, where reliability is re-
quired in critical components of high pressure systems, the use of nodular
cast iron is preferred over gray cast iron.  Nodular cast iron can be
made to significantly higher strength levels than gray cast iron in addi-
tion to being able to deform plastically.  Measurable tensile elongation
values of 3 - 30% are possible with the nodular irons.  They also have
good resistance to mechanical shock.  Typical room temperature properties
for a 80-60-03 Pearlitic Grade of nodular iron are as follows:
                                                                       (a)
                                                  Endurnace Limit, ksi
      Ultimate, psi    Yield, psi    Elong., %    Unnotched	Notched
          95-130         55-80          3-9          40             24
 ^a'For casting with 100 psi ultimate strength.

      One of the major concerns in the use of cast irons in steam  (or air)
at elevated temperatures is their dimensional instability due to oxida-
tion.  This growth in dimensions can be significantly large in gray cast
iron because of the continuous network of graphite.  However, additions
of chromium and the use of nodular cast iron or, for more severe condi-
tions of corrosion, the use of the Ni-Resist cast irons can eliminate or
minimize the dimensional changes due to oxidation.  Data on dimensional
growth for various cast irons as a result of exposure to steam at  900°F
are given in Table  5.2-III.
                                   114

-------
                      Table  5.2-III

           Dimensional Growth of  Cast  Irons
             in High  Temperature  SteamQ-8,19)
     Cast Iron
                             Growth at  900°F. in./in.

                        500 Hrs.   1000 Hrs.    2500  Hrs.
Gray Cast Iron
Gray Cast Iron
+ 0.46% Cr

Gray Cast Iron
+ 0.6% Cr

Ni-Resist Type 2
Ni-Resist Type 3
              (b)
              (b)
Ni-Resist Type 2D
(Nodular)

Ni-Resist Type 3D
(Nodular)
                  (b)
               (b)
                         0.0023
                         0.0003
0.0052
0.0000
0.014
0.001
                                                     (a)
                                               0.0005
                                                      (a)
0.0005
0.0003
0.0003
0.0010
0.00045
0.0005
0.0015
0.00048
0. 0005
0.0000
(a)

(b)
1000°F/2000 hours.

Ni-Resist Type 2, 2D:
Ni-Resist Type 3, 3D:
                          O.OC max-2.25 Si-20 Ni-2.1 Cr.
                          2.6C max-2.1 Si-30 Ni-3.0 Cr.
                           115

-------
      5.2.1.4 Aluminum Alloys

      The use of aluminum alloys for dynamic components and cylinder walls
in internal combustion engines is highly desirable because of the low
density and high thermal conductivity of aluminum alloys.  However, the
calculated temperatures in the crown of the power piston head, particularly
for the crosshead piston expander, and the calculated stresses in the
connecting rods and the piston rod of the crosshead piston expander pre-
cludes the use of aluminum alloys for these applications.  At this stage
in the expander designs, only the skirt portion of the trunk piston and
the crosshead piston itself warrant consideration.  For these applications,
it is desirable to utilize alloys with as low a coefficient of thermal
expansion as possible that is consistent with suitable mechanical prop-
erties.

      As previously stated, the following alloys  (high silicon contents)
are used in piston heads and skirts in automotive spark ignition and large
diesel engines:  SAE 328, SAE 332 and 4032.  The nominal compositions and
the coefficients of thermal expansion are given in Table 5.2-IV.  Two other
high strength forging alloys (2014 and 2219) that have been recommended
for high temperature engine components also are listed.  For comparative
purposes, the room temperature minimum tensile properties of these alloys
are given in Table 5.2-V.

      Typical properties of the forged 4032 alloy that was selected for
the piston components in this program together with typical properties
for 2014 and 2219 alloys are listed in Table 5.2-VI.  From these data it
is apparent that the 4032 and 2014 alloys have poor thermal stability
at the higher temperatures in contrast to the excellant thermal stability
of 2219 alloy.  Significant degradation in the tensile properties is not
observed in the 2219 alloy until a temperature in excess of 600°F is
reached or after very long time exposures at 600°F (10,000 hrs.).

      Careful consideration must be given to the  use of aluminum alloys
in steam.  Although aluminum alloys should perform satisfactorily in dry
steam at elevated temperatures, i.e., 600° - 700°F, erosion may be a
problem in wet steam.   Further, it is known that  conventional commercial
aluminum alloys will corrode very rapidly in high purity water at tem-
peratures over 400°F.   However, aluminum alloys have been used as cladding
                                   116

-------
                                             Table 5.2-IV
                                 Nominal Compositions of Aluminum Alloys
                                  Nominal Composition,  %
Alloy Si Fe Cu Mn Mg Ni Zn Ti Other
SAE 328(a^ 11.75 0.9 1.5 0.7 0.7 0.0005 1.0 0.25
SAE 332^a) 9.5 1.2 3.0 0.5 1.0 0.5 1.0 0.25
4032^ 12.25 1.0 (c) 0.9 - 1.1 0.9 0.25(c) - 0.1 Cr(c)
2014(b) < 0.8 - 4.4 0.8 0.4 -
2219(b) 0.2(c) 0.3(c) 6.3 0.3 0.02(c) - 0.1(c) 0.06 °*^0V,
U. -Lo Zr
VjU
in.
11.
12.
11.
13.
13.
ej. . rjtpciiis j-uti
/in./°F x 10~6
5 (68-392°F)
0 (68-392°F)
7 (68-572°F)
6 (68-572°F)
6 (68-572°F)
(a)



(b)



(c)
Casting - automotive engines.



Forging - large diesel engines,
   Maximum.

-------
                           Table 5.2-V
                Room Temperature Tensile Properties
                 of High Silicon Aluminum Alloys


                                      Minimum Tensile Properties	

                                  Ultimate,    0.2% Yield,    Elong.,
Alloy      Condition    Temper       psi          psi            %
SAE 328    Cast           T5         32           26           low

                          T65        42           37           low

SAE 332    Cast           T5         31           -            low

4032       Forged         T6         52           42             5

2014       Forged         T6         65           55            10
                                 118

-------
                                                    Table 5.2-VI

                              Typical Mechanical Properties for Aluminum Alloy Forgings
                                        (4032,  2014 and 2219 in J6 Condition)
vo
Test
Temp . ,
Alloy °F
4032 75

300

500
*
2014 75
300

500

2219 75


300

500

600

700

Exposure
Time Prior
to Test,
hours


1/2
4.000
1/2
1000
-
1/2
1000
1/2
1000
w


1/2
1000
1/2
1000
1/2
1000
1/2
1000
Tensile Properties
Ult. , ksi
55

46
44
23
10
70
51
34
25
11
64


51
49
30
30
22
18
12
5
Yield, ksi Elong. , %(4D)
46

41
40
22
7
63
49
30
24
9.5
45


39
37
22
22
16
14
10
4.5
9

9
9
12
45
15
14
20
18
43
10


20
20
25
25
26
28
30
80
oung s Fatigue Properties
ksi x 10^ Cycles Stress, ksi
11.3 1 x 10® 18.0
5 x 10 16.5
10.5 1 x 10® 13.0
5 x 10 11.5
9.3 1 x 10® 5.5
5 x 108 5.0
10.5
9.6
9.6
8.5
8.5
10.6 1 x 10® 16.5
X
5 x 10° 15.0
9.9
9.0

8.6

7.6

6.4

-------
materials in pressurized water reactors operating in excess of 300°F.
In fact, an Al-Ni-Fe alloy (X8001) has been developed specifically for
this purpose and has shown outstanding resistance to corrosion in high
purity water at elevated temperatures.  Corrosion tests on the X8001 alloy
containing 0.5% Fe and 1.0% Ni in high purity water exhibited the following
corrosion rates:

            Exposure
          Temperature, °F                 Corrosion Sate, mils/yr
               550                                   1.7
               600                                   3,1
               680                                   8.0
 (a)
   Test  duration - approximately 1500 hours.

       5.2.1.5  Case Hardened Low Alloy Steels

       Dynamic  components  that are relatively  low stressed but  are  subject
 to wear  in service require that their surfaces be hardened.  An  excellent
 example  is the camshaft where cyclic Hertzian stresses on the  surface
 approach 200,000 psi.  In these applications, induction hardening  of high
 carbon steels  (AISI  1080), carburizing  or nitriding of low-medium  carbon
 low  alloy steels  (AISI 8620, 4340)  are  employed.  Since fatigue  fractures
 generally initiate at  the surface,  it is extremely important that  the
 surfaces of case hardened components be free  of defects or  irregularities
 that will provide stress-risers.  Assuming  defect free surfaces, the
 surface  hardening treatments will substantially increase the fatigue en-
 durance  limit, partly  because of the induced  compressive stresses  in the
 surface.  As an example,  where the  fatigue  endurance  limit  of  heat treated
 AISI 4340 is approximately 75,000 - 80,000  psi, shot  peening the surface
 will increase  the endurance  limit to approximately 90,000 - 100,000 psi
 and  nitriding  the surface will increase the endurance limit to 120,000
 to 135,000 psi.  In  a  crankshaft application, designed for  a minimum
                            /
 115,000  psi tensile  strength, a AISI 4140 steel heat  treated to  a  85,000
 psi  tensile strength and  nitrided on the wear surfaces, provided the
 necessary wear and fatigue resistance.
                                    120

-------
5.3   Materials Recommendations

      The materials recommendations for the single cylinder steam expanders
were based on the materials technology survey (Section 5.1) and the
materials selection study (Section 5.2).  In general, considerable conservatism
was exercised in the selection of materials for the engines in this program.
The conservatism in materials selection was based on the fact that the engines
were to be used as test vehicles.  The selection of materials was further
compromised by the fact that only one or at most two components of any design
were required.  For this reason no castings were utilized for major components
of the expander other than the valve guide and tappet.  A listing of the
materials selected for each component in the single cylinder steam expanders
together with the recommended materials specification and heat treatment
condition are given in Table 5.2-VII.  These materials were selected on the
basis of a high probability of success rather than cost.   Future studies
would be useful, directed toward materials optimization with respect to
cost.  For example, the addition of approximately 1% Cr + 0.20% Mo (AISI 4140)
to AISI 1045 steel represents an increase in price of the finished mill
product to about $0.05/lb for quantities in excess of 10,000 pounds.   In
the final selection of materials, mechanical properties and environmental
compatibility must be balanced against cost.  Materials must be used up to
the limit of their acceptable life before specifying a more costly material.
                                   121

-------
                                                   Table 5.2-VII
                          Materials Recommendations for Single Cylinder Steam Expanders
     Expander
     Component
1. Inlet Manifold
   Cylinder Head
   Valve Body
2. Exhaust Manifold

3. Cylinder Block

4. Cylinder Block-
   Lower (Crosshead
   Piston Expander
   Only)
5. Cylinder Liner (A)
                  (B)
6. Piston Compression
   Rings          (A)
                  (B)

(Trunk Only)
Rider Ring
Lower Seal Ring
Scraper Seal Ring
Oil Cutter Ring
      Recommended
       Material
2.25% Cr-1% Mo Steel
1.0% Cr-0.5% Mo Steel

1.25% Cr-0.5% Mo Steel

AISI 304 SS/Gas Nitride
Bore
17-22-A/Hard Cr Plate
Bore

AISI 440C
Sb Impregnated
Carbon-Graphite
Koppers K-1051/over
Inconel X-750

Carbon-Graphite

Koppers K-35 Ni-Resist
Koppers K-Iron
Koppers K-6E
        Material
     Specification
         Condition
ASTM A387 Grade D, Plate
ASTM A335 Grade P22, Pipe


ASTM A387 Grade B, Plate
ASTM A335 Grade P12, Pipe
ASTM A387 Grade C, Plate
ASTMA335 Grade Pll,  Pipe
ASTM A240, Plate
ASTM A376, Pipe
GE/P11BYA11, Nitride


AMS 6304C/GE B5F5, Forging
GE/P16DYA10, Chromium Plate


AMS 5630C, Forging
See Section 6.0 of Report
Normalize 1700°F, Temper 1250°F
Normalize 1700°F, Temper 1250°F
Normalize 1700°F, Temper 1250°F
Anneal 1900°F/Rapid Cool
Normalize 1750°F/l-l/2 Hrs.
Mar Quench 650°F/10-15 min.
Temper 318 - 363 BHN
Austenitize 1900°F/Oil Quench
Double Temper 675°F/4 Hrs.
Hardness Re 52-56

-------
                                                     Table 5.2-VII(Cont'd)
                             Materials Recommendations for Single Cylinder Steam Expanders
U)
         Expander
         Component
   7. Power Piston
   8. Piston Rod
        (Crosshead Only)

   9. Rod Seal
        (Crosshead Only)
  10. Crosshead Piston
  11. Connecting Rod
  12. Piston Pin
   13. Piston Pin Bushing
      Bushing

   14. Camshaft/Cam,
      Bearing  Surfaces

   15. Tappet


   16. Tappet Housing
      Recommended
        Material
17-22-A (1% Cr-0.55% Mo-
0.3% V)


H-ll Steel
PTFE-Bronze, MoS_
Filled
Al 4032-T6 Alloy
AISI 4140
AISI 8620/Carburized
80 Cu-10 Sn-10 Pb/Mild
Steel Backing

AISI 8620 Carburized
Chilled Cast
Antichafing Coating

Nitralloy 135 Mod/
Gas Nitride Bore
       Material
    Specification
AMS 6304C/GE B5F5, Forging
AMS 6487C, Bars & Forgings
GE/P16DYA10,Chromium Plate
ASTM B247, Forging
AMS 6382G, Forging
AMS 6276C,Bars & Forging
GE/P11BYA11,Carburize


SAE 792
AMS 6276C, Bars & Forgings
GE/P11BYA11, Carburize

Eaton EMS 91 or EMS 80
ASTM A355 Class A, Bar
Nitride - Drawing Control
        Condition
Normalize 1750°F/l-l/2 Hrs.
Mar Quench 650°F/10-15 min.
Temper 311-363 BHN

Austenitize 1850°F/Air Cool
Double Temper 1000°F/2 Hrs.
Hardness Re 52-55
Solution Treat
   955 ± 15°F/2 Hrs.
   Water Quench 150-212°F
Age 340°F/10 Hrs.

Austenitize 1550°F/Oil Quench
Double Temper 1000°F/2 Hrs.
Hardness 320-360 BHN

Core Hardness Re 58 min.
Core Depth 0.040-0.070 in.
2 RMS Finish

Stabilize 1000°F/4 Hrs.
Core Hardness Re 58 min.
Core Depth 0.040-0.070  in.
Austenitize 1750°F/Oil  Quench
Temper 1100°F/2 Hrs.
Core Hardness  330-370 BHN
Case Hardness  RC  65  (converted)
Case Depth 0.015-0.020  in.

-------
                                                    Table 5.2-VII  (Cont'd)

                           Materials Recommendations for Single Cylinder Steam Expanders
       Expander
       Component
      Recommended
        Material
17. Push Rod
18. Valve Head.Facing
       Valve  Stem

NJ  19. Valve  Seat
**        (Facing)
   20. Valve  Guide
21. Valve Rings
22.  High Temp.
    Static Seals
AISI 8620/Carburize
H-ll Steel/Stellite 6
                              H-ll Steel/Gas Nitride

                              Stellite 1

                              Ni-Resist D-3
                               (Nodular)
H.S.25/LPA-101 Coat
H.S.25
                                                                  Material
                                                                Specification
        Condition
                                                            AMS  6276C,  Bars  &  Forgings
                                                            GE/P11BYA11,  Carburize

                                                            AMS  6487C,  Bars  &  Forgings
                                                            ASTM.A399 Type R Co-Cr-A
                                                            Hard Surface  Rod
                           Nitride - Drawing Control


                           ASTM A399 Type R Co-Cr-C
                           Hard Facing Rod
                           ASTM A439, Ductile Iron
                           Casting
                                                            AMS 5759
                                                            Coating - Drawing Control

                                                            AMS 5759
Core Hardness Re 58 min.
Core Depth 0.040-0.070 in.

Austenitize 1850°F/Air Cool
Double Temper 1000°F/2 Hrs.
Hardness Re 52-55
Coating Thickness - 0.125 in.
Case Hardness Re 65 (Converted)
Core Depth 0.010-0.015 in.
Coating Depth 0.125 in.
Coating Hardness Re 48 min.

Stabilize 1600°F/2 Hrs.
Furnace Cool to 1000°F
    @ 100°F/Hr.
Hold 1000°F/1 Hr.
Slow Cool
Cold Worked & Aged @ 1100°F/5 Hr.
Hardness Re 45-50
Coating Thickness 0.003 in.
Material Cold Reduced to
   Re 45-50

-------
                    6.0   LUBRICATION

6.1   Lubrication Technology Base - Solid Lubricants

      One of the most critical problems associated with the design of an
efficient steam engine is that of lubrication within the cylinders and
the inlet steam valve.  In order to fully utilize the potential thermal
efficiency of the steam cycle, it is necessary to operate the expander
with high inlet steam temperature.  With increased steam temperature,
there is a corresponding increase in temperature of the materials within
the cylinders and valves.  The vast majority of steam expander designs,
from the early Stanleys to the most modern engines, obtain upper cylinder
lubrication by means of hydrocarbon lubricants pumped in with the steam.
Due to thermal degradation, oxidation, and sludging of these lubricants,
their use generally limits steam temperatures to about 750°F.   In recog-
nition of this limitation, the General Motors SE-101 engine was limited
to a maximum steam temperature of 700°F.

      The Pritchard engine     is lubricated in a similar way.  A slow
speed mechanical pump forces small quantities of oil into the steam to
act as an upper cylinder lubricant.   Approximately one pint per 500 miles
is used in this fashion.  After 10,000 miles of operation with a closed
steam cycle, 20 pints (2.5 gallons)  of oil will have been pumped into
the steam system.  A number of other more recent steam engine designs
also have expanders with the lubricant added to the working fluid side
of the engine.

      Even though lubrication technology for steam engines has been es-
tablished for large, low-speed engines operating at moderate temperatures
and pressures,  an adequate lubrication system has not been identified
for highly efficient steam engines that operate at high inlet steam tem-
peratures and at high inlet pressures.  There are two approaches which
                                  125

-------
appear most feasible for lubrication within the cylinders of high tem-
perature, high pressure steam engines:  (a)  to inject water-miscible
fluids, which have been extensively developed for cutting and grinding
applications, into the steam; or (b)  to use solid lubricants which have
been developed for extreme condition applications.  With the injection
of a water-miscible fluid, it would be necessary to add a separating system
to separate the lubricant from the water to avoid fouling of the steam
generator tubes; also, this approach would result in a high consumption
of water-miscible fluid.  For this reason, use of solid lubricants appears
to be the most feasible method of lubricating the cylinder walls in order
to permit the utilization of higher steam temperatures.

      A literature search and industrial survey was made to identify and
evaluate the current technology in solid lubricants and wear resistant
materials.  Searches were made through the following resources:  General
Electric Company Automatic Retrieval System, SAE Transactions, Engineering
Index, NASA Scientific and Technical Information, AGAKD and the Defense
Documentation Center.  The industrial survey was made to determine those
solid lubricants that either are in commercial production or in the stage
of advanced development.

      6.1.1    Criteria for Solid Lubrication

      The most important function of any lubricant is to keep the rubbing
surfaces separated, both to prevent wear of the moving parts and to
minimize the high coefficients of friction associated with metal-to-metal
contact.  A solid lubricant keeps the moving surfaces separated solely
through its own strength.  However, to do so, the solid lubricant must
remain in the clearance between the moving surfaces for the entire de-
sign life of the engine.  It follows then, that in order for a material
to be an effective solid lubricant, it must possess most or all of the
following properties and characteristics depending upon the form of the
lubricant:

      1.  Good bonding properties or high tendency to adhere to rubbing
          surfaces
      2.  Low shear strength
      3.  High compressive strength
                                   126

-------
       4.  Laminar  crystal  structure
       5.  High bulk  fracture strength
       6.  Good thermal  stability
       7.  Chemical compatibility with surrounding environment  (including
          structural materials)
       8.  Low wear
       9.  Good resistance  to thermal shock
     10.  Good fabricability.

       Although all of the  above listed properties are important with
respect to achieving satisfactory performance by solid lubrication, the
ability of the lubricant to adhere to the rubbing surfaces is  absolutely
essential in order to protect the surfaces and prevent scoring, galling,
gross  welding and  resultant metal transfer.  This corresponds  to wetting
in liquid lubricated systems. For self-lubricating composites, it is
necessary that the lubricant be transferred to the mating surface to form
an adherent thin film.  For example, the ability to form a transfer film
is what makes graphite  an  excellant lubricant.

       Low shear strength is Important from the standpoint of achieving
a low  coefficient  of friction.  The coefficient of friction for solid
lubricants in general is high in comparison to oils, as shown in the
following tabulation:
                                                 Typical
              Lubricant                   Coefficient of Friction
        Oil                                      0.001
        Solid Lubricant-Bonded                   0.01
        Solid Lubricant-Composite                0.1
        Metal-Metal  (Steel)                Approaching 1.0

From these data it can be seen that it is desirable to utilize materials
with as low a shear  strength as possible.

      A high compressive strength is desirable from the standpoint that
the higher the compressive strength of the lubricant, the higher is the
load that can be applied before metal-to-metal contact is made.  In
effect, the lubricant should have a low shear strength/compressive strength
ratio  for best performance.
                                  127

-------
      The laminar crystal structure of the lubricant is associated with
low shear strengths and low coefficients of friction.  Although effective
lubrication can be achieved with solids that do not have a laminar struc-
ture, the most effective solid lubricants are those that do have the
laminar structure and have good surface adherence properties.  The laminar
structure consists of alternating layers of atoms in which there are
strong bonds between atoms within the layer (covalant or ionic forces)
and weak bonds between atoms in adjacent layers (Van der Waals' forces).
The Van der Waals'forces are easily broken in the "layer-lattice" struc-
ture resulting in the successive atomic layers readily sliding over each
other such as basal plane slip in the hexagonal structure of graphite.

      Solid lubricants that have a laminar crystal structure are highly
anisotropic with respect to mechanical properties.  In cases where the
lubricant is fabricated into a self-lubricating composite, the lubricant
matrix must have sufficient strength to withstand high pressure differ-
entials and frictional drag loads at high temperatures without changing
shape or fracturing.  To endure the start-up cycle of a steam engine,
the self lubricating composite also must have adequate resistance to thermal
shock.

      Thermal stability of the lubricants is required to assure acceptable
performance over the entire temperature range and throughout the design
life of the engine.  Since the total surface temperature of the lubricant
includes the heat of the environment and the heat of friction, the in-
fluence of both must be considered.  Thermal stability includes resistance
to thermal decomposition and crystallographic changes as well as being
resistant to oxidation in air and steam.  Crystallographic changes will
result in changes in frictional characteristics.  However, the effects
of surface heat are not always detrimental since some solid lubricants
are designed to work as a liquid phase.

      Another important consideration is compatibility with all engine
materials and small quantities of organic lubricants.  Consideration of
free energy exchange between steam and the candidate materials indicates
that a large number of candidate solid lubricants are available for
evaluation.  However, free energy exchange is somewhat complicated by
the addition of cylinder and piston materials.

                                  128

-------
      Finally, an optimum solid lubricant is one that has low wear and
the ability to replenish the transfer film such that long life is ob-
tained.  Also, for self-lubricating applications, the solid lubricant
must be readily fabricable into usable shapes.

      6.1.2    Forms and Types of Solid Lubricants

      The lubricating solids can be used in several different forms:
(a) as loose powder dusted or rubbed on the mating surfaces or supplied
to the parts in a controlled stream of a carrier gas; (b) as bonded ma-
terials attached to the surfaces with an adhesive binder; (c) as self-
lubricating bearing materials where the lubricating solid is dispersed
throughout a metal or plastic bearing or where the lubricating solid
forms the matrix of a fabricated bearings, and (d) as additives to oils
and greases to replace the chemical extreme pressure (EP) additives
normally used.  The latter application of solid lubricants is not appli-
cable to the lubrication of cylinder walls and will not be discussed
further.  Of the remaining three forms in which a solid lubricants can
be applied, each form has its own advantages and disadvantages.

      Loose powders, which have the advantage of minimizing fabrication
or assembly problems5have the disadvantage of providing a very thin film
of limited life; the film either has to be replenished by external means
in order to achieve the desired life or be used only to assist in the
wear-in period.

      Bonded solid film lubricants generally provide longer life than
powders with the performance of the bonded solid film varying with the
specific lubricant, the bonding agent and the method of application.
Some dry films depend solely on small particle size for bonding.   The
lubricant particles are held together only by the attraction arising
from the extremely fine particle size and there is little bonding to the
substrate.   A limited amount of dispersing agent is sufficient to bring
about bonding.  Particle*-bonded dry films can be dispensed from aerosol
containers dr dispersed in water,  organic solvent or other volatile
carrier.  Most of the particle-bonded lubricants are air-drying and re-
quire no baking.  However,  the particle-bonded lubricants have the same
problem as the loose powders in that they have very limited life unless

                                  129

-------
they can be replenished.  An organic resin (phenolic, epoxy, polyimide)
may be added to increase binding between particles and bonding to the
substrate.  The resin produces an adherent paint-like film that offers
the best wear life and widest range of use.  The lubricant particles also
can be bonded together by a silicone, a suitable ceramic or metal salt
(sodium silicate, aluminum phosphate) that is hardened upon removal of
the solvent.  This type of film can be very hard and is more temperature-
resistant than the organic resin bonded films.  The type of binder that
is used depends upon the adhesion required, the anticipated service
temperature and environment, and the desired wear life.  This type of
bonded film can be applied by spraying, dipping or brushing.  Newer
application techniques include electrophoretic deposition, plasma spraying
and sputtering.

      One of the major problems associated in the bonded films is that
the lubricant is usually limited by the binder itself.  For example, the
resin bonded solid lubricants are limited to use temperatures of up to
about 400°F; silicones and silicates have permitted service temperatures
to increase to about 600°F at the expense of higher friction and lower
wear life.  The organic resin and metal salt bonded solid  lubricants also
have a limited wear life that can be attributed to the very small volume
of lubricant available.  This becomes evident when one considers that  the
film is only 0.0002 - 0.0005 in. thick.

      Because of the relatively short life of the bonded solid lubricants,
self-lubricating components containing solid lubricants have been de-
veloped for longer life bearing and seal applications in order to take
full advantage of the good lubricating qualities of solid  lubricants.
The solid  lubricants are incorporated into the self-lubricating composites
by impregnation, sintering, and various forms of high-temperature, high-
pressure  compaction.  Self-lubricating bearings generally  fall into two
major classes:   (a) solid lubricant matrix utilizing graphite or the
lubricant  plastics such as nylon, polytetrafluoroethylene  (PTFE), etc.,
with reinforcing materials as fiberglass, graphite, molybdenum disulfide,
and soft metals; and  (b) metal matrix in which lubricating solids such
as graphite, PTFE, metal disulfides, selenides, or tellurides and low
melting point metals are mixed with metal/alloy powders of copper, iron,
                                  130

-------
 nickel,  molybdenum,  etc., and sintered under pressure or impregnated
 into a porous metal  matrix skeleton.

       In the application of solid lubrication for  cylinder  walls in high
 performance reciprocating steam expanders where steam temperatures  approach
 1000°F,  the peak steam pressure is 1000 psi and peak piston ring side
 loads are on the order of 200 - 300 psi at  an average linear velocity of
 1200 -1600 feet/minute, it was assumed that self-lubricating composites
 would be required to achieve the necessary  life.   Therefore, in  the
 survey of solid lubricants,  emphasis was placed on the self-lubricating
 solids or composites.   Bonded solid film lubricants were considered
 primarily as an auxiliary lubricant during  the wear-in process or as  an
 aid in establishing  a transfer film.   The following materials were  re-
 viewed as lubricants to minimize wear  in the cylinder wall/power piston
 interface and valve  stem/guide interface:

       Self-Lubricating Solids
            Carbon-graphites
            Polytetrafluoroethylene (PTFE)
            Sulfides/Selenides
            Metals
            Porous metal composites
            Hard surfacing materials

       Bonded Solid Films

            Sulfides/Selenides
            PTFE
            Graphite
            Soft Oxides

       6.1.2.1   Self-Lubricating  Solids

       6.1.2.1.1 Carbon-Graphites

      Graphite was found to show promise of  being a very  effective solid
lubricant for steam engine applications.  Graphite has been  used  as  a lubricant
for many years and is used extensively  for brushes  in electric  motors.
In fact, it was excessive wear of motor brushes in  high altitude  aircraft
                                                                  s.
                                  131

-------
that led to the detailed study of the mechanism of graphite lubrication.
                                                                   (21 22  23)
In a detailed study of this problem by the General Electric Company   '  '
it was discovered that the low friction and wear normally exhibited by
graphite is due not to any lubricant quality inherent in the graphite,
but to adsorption on the surface of substances from the atmospheric en-
vironment.  The laminar structure of graphite undoubtedly is essential,
but alone it is not sufficient.  It was found that water vapor, adsorbed
on the surface, caused reduction in friction and wear rate.  The minimum
pressure of water vapor that is necessary to achieve minimum wear rates
is on the order of 3 - 5 torr.  Oxygen showed a similar effect but at
much higher pressures, i.e., 300 - 500 torr.  It has been proposed that
only the edges of the graphite crystals need be covered with the adsorbed
film to achieve the low friction and wear rates by permitting the laminar
layers of the hexagonal lattice to slip.  In the absence of water vapor
or oxygen the edges of the laminar layers of atoms are locked by free
radicals formed by the evaporation of volatile oxides of carbon.  Graphite
then is expected to maintain good lubricity in the cylinder wall because
of the presence of high pressure steam.

      In extremely high temperature applications, graphite tends to oxidize
and its uses are restricted to neutral or reducing atmospheres.  However,
graphite has been successfully used as seals for steam turbine driven
electrical generating equipment which tends to verify its stability in
high temperature steam.  The threshold oxidation temperature of graphite
                                        (24)
in steam has been listed as about 1300°F     and good resistance to
oxidation by steam is expected at the maximum proposed cycle temperature
of 900°F.  This may be further verified by examination of the free energy
changes for the two oxidation reactions:
                          C + H20 = CO +
and
The free energies of  these reactions are plotted against temperature in
Figure 6.1-1.  As may be seen on this plot, the driving force for the
reactions  (negative free energy) does not become appreciable until tem-
peratures  greater than about 1200°F.  This is in essential agreement
with the temperature  limit cited above.  In air, graphite cannot be used
                                   132

-------
   20
   15
 8
 0)
 P.
   10
4-1
U
cfl
H-l
O
00
S-i
0)
0)
1-1
    0
                          500      -            1000

                                Temperature, °F
                                                     1500
 FiRiire 6.1-1.
Free Energy of  Oxidation Reactions  Involving Graphite
and Steam.
                   133

-------
much above 750°F without severe oxidation reactions.
                                                     I
      Even though graphite displays excellent properties from the stand-
point of thermal stability, thermal conductivity, low elastic modulus,
low shear strength and thermal shock resistance, it has poor strength
properties in tension.  Higher strengths are possible with the carbon-
graphite grades of which there are an infinite number to choose; lower
wear rates also are possible with the carbon-graphite materials.  These
materials are available commercially and, although it is an oversimplifica-
tion, are produced in four general classes:  (a) straight carbon-graphites
with varying carbon/graphite ratios, (b) resin impregnated, (c) metal
impregnated  (Cu, Cu-Pb, Cu-Sn, Pb, Ag, Babbitt) and (d) high temperature
oxidation inhibited grades.  The maximum service temperatures for these
materials in air are reported to be up to 500°F for the resin Impregnated
grades, 700"F for most of the straight carbon-graphites and metal im-
pregnated grades, and 1000° - 1200°F for the high temperature grades.
Properties of candidate carbon-graphite grades that are believed to be
suitable for application as piston compression rings in a steam environ-
ment are listed in Table  6.1-1.  Care must be exercised in the use of the
oxidation-inhibited, high temperature grades and some of the resin im-
pregnated grades in steam and water because of the tendency for some of
the resins and oxidation inhibitors to exude to varying degrees in these
environments. Consideration also must be given to the compatibility of
graphite with metals used for the impregnation process as well as mating
metals/alloys that are in contact with the graphite  in service.  Studies
         /25^
by McGeev   ' at General Electric have shown that metal oxides that can
be reduced to lower oxides or to the metallic state by graphite in the
presence of  oxygen appear to function as catalysts in the oxidation of
the graphite.  The relative activity of various metal oxides with respect
to their effect on the oxidation of graphite is shown in Table 6.1-II
and is  based on the change in ignition temperature of pure polycrystalline
graphite.  The data show that lead oxide is particularily effective in
lowering the ignition  temperature of graphite.

      Although there are numerous influencing factors, the coefficient
of friction  for the various grades of carbon-graphite will generally
range between 0.1 to 0.25.  Wear rates also will vary significantly with
                                  134

-------
                                                                          Table 6.1-1
u>
in
Properties of Carbon-Graphite Grades
'

Vendor Grade
VS Graphite 86
Div' 110
2980
102
103
107
Pure Carbon P5NR
Co. pQ3
P658RC
X3310
Graphite- Bronze
Metallizing Graphalloy
Corp.
UCC CDJN
CJP
Carbone 5890
Corp. 589Q
JP-500

, ..
TypeW
CG-R
CG
CG-HT
CG-Cu
CG-Ag
CG-Sb
CG
G
CG
CG-HT
CG-Cu- Sn


CG-HT
CG-HT
G
G-HT
CG
(a) G Graphite
CG Carbon-Graphite
CG-R Resin Impregnated
Hardness
Shore

Compr
Transv.
Fracture

Tensile.
(5cleto*cope) Str.,ksi Str.,ksi Str.,ksi
100
92
67
88
89
92
80
75
90
85
-


105
65



CG-HT
CG-Sb,
Cu,Ag,
Cu-Sn
36
32
15
38
35
38
30
20
38
29
-25


36
26



12.5 '
10.0
4.5
9.5
10.5
10.0
8.5
8.0
11.0
9.4
_


8.8
8.0



High temperature/oxidation
Metal
Impregnated
9.5
8.5
4.0
7.5
7.5
7.5
7.0
5.5
8.0
8.5
~8.0


7.0
7.0



inhibited


E
106 psi
3.0
3.3
1.8
2.6
2.8
3.5
2.7
1.7
3.1
2.6
-


3.2
2.1





Coef .Th.
Exp.,
in/in/°F
£
x 10~6
(RT-500°F)
2.3
2.3
2.3
3.1
3.1
2.4
2.2
1.9(b)
2.2
2.3(c)
2.3


2.3
2.3






Th.Cond.
Btu/F.t/Hr
"F
7.6
8.0
30.0
8.7
9.3
7.0
_
-
_
-
_


4.4
13.3






Apparent
Density
gm/cc
1.90
1.90
1.85
2.35
2.70
2.20
1.70
1.82
1.80
2.05
_


1.76
1.77







Borosity
Vol. %
0.5
1.0
4.0
4.0
3.0
1.0
20.0
10.0
2.0
0.4
_


-
-

	 	 .




Max.
Service
Temp. ,Air
500
700
1000
700
700
700
500
900
500
1000
750


875
1100





     (b) RT to 900CF


     (c) RT to 1000'F

-------
Table 6.1-II
Catalytic Activity of
Catalyst
(acetate or oxide)
Pb
V
Mn
Co
Cr
Cu
Mo
Ag
Cd
Fe
Pt
Ni
Ir
Rh
Ru
Pd
Ce
Zn
W
Hg
Sn
Uncatalyzed
Oxides in Graphite
w/o as Metal
0.15
0.20
0.45
0.33
0.95
0.20
0.15
0.16
0.21
0.13
0.03
0.45
0.40
0.20
0.30
0.30
0.72
50.00
0.02
0.10
0.10

Oxidation
Ignition
Temp., °C
384 (738°F)
490
523
525
540
570
572
585
590
593
602
613
638
622
640
659
692
700
718
720
738
740 (1364°F)
      136

-------
the grade of carbons-graphite.  Overall, the metal Impregnated carbon-
graphite appears to be the most suitable for use as compression piston
rings in the steam expander.  Data indicate that after an initial high
wear rate during the wear-in period, the metal impregnated carbon-
graphites have superior wear rates for long time service; however, the
coefficients of friction for the metal impregnated carbon-graphites
generally are higher than the other grades.  In addition, the metal im-
pregnated carbon-graphites have low porosities and permeabilities, are
strong and are compatible with steam.  With the exception of their tendency
to exude in the presence of water, some of the oxidation inhibited, high
temperature grades also are attractive.  They have good friction and wear
characteristics and low values of porosity and permeability.

      Additional possibilities involve composites of graphite with rela-
tively soft oxides.  A cadmium oxide-graphite mixture has friction coef-
ficients in the order of 0.1 or less over most of the temperature range
                   ( 26^
from 100 to 1000°F v   '.  The cadmium oxide is presumed to improve the
adherence of graphite to the surface and, hence, to improve the lubricity
effectiveness.

      Materials that would be satisfactory as mating materials in contact
with the carbon-graphites under dynamic conditions are AISI 440C or hard
chromium plate with the latter being limited to service temperatures of
less than 700°F.  A high hardness is desirable, i.e.,  greater than
Rockwell C 45, and the surface finish should be smooth but not too smooth
so as not to be able to hold the transfer film.  A finish on the order
of 8 - 12 RMS appears satisfactory.

      6.1.2.1.2 Polytetrafluoroethylene (PTFE)

      Plastic bearings made from polytetrafluoroethylene (PTFE)  are
available for nearly every type of application.  They  have been  used
successfully as inserts to plain bearings,  as reinforced thin sheets on
plain spherical bearings, and as the retainer material for ball  bearings.
PTFE is one of those materials that does not have a laminar layer-lattice
structure and yet has an exceptionally low coefficient of friction.  The
low friction has been attributed to its low surface energy.   This results
in very weak adhesion and shearing takes place primarily at the inter-
                                     (27)
face rather than in the bulk material    .   It also has the necessary
                                  .137

-------
characteristic of establishing an adherent transfer film on most rubbing
metal surfaces.  The major disadvantage of PTFE is its low compressive
strength which results in cold flowing under relatively low loads and
limits its use to temperatures on the order of 400 - 450°F (it also begins
to thermally decompose at about 540°F).  Other disadvantages of PTFE are
its poor thermal conductivity and high coefficient of thermal expansion.

      To improve the flow characteristics of PTFE, PTFE composites have
been developed.  The three most common materials that have been used to
reinforce the PTFE are bronze, glass and carbon-graphite.  In one recent
                                                         (28)
symposium of PTFE seals for reciprocating air compressors    , it was
generally considered that no one filler material has demonstrated superior
performance over the others and that lack of consistency in  the composites
was a major factor in the erratic performance.  Three typical compositions
that are in use for seals in reciprocating compressors are:

      1.  PTFE + (15% Bronze + 5% MoS,2)
      2.  PTFE + 20% Glass
      3.  PTFE + 10% Carbon
               (% by volume)
                    (29)
      G. L. Griffin     has suggested that non-uniformities  in the struc-
ture as a result of molding practice of filled polymers could account for
differences in performance.  In particular, the surface of molded com-
posites is usually deficient in filler material and this could result in
varying friction and wear behavior.

      Other developments to improve the poor creep properties of PTFE
are the use of fibers of PTFE.  Compacts can be made from PTFE fibers
which have 25  times the' tensile strength of conventional PTFE compacts
made from powders.

      Halliwell of the U.S. Navy Marine Engineering Laboratory has con-
ducted tests with PTFE reinforced with metallic filament windings
in an attempt  to develop a piston seal for high-pressure air compressors.
The filament winding approach has resulted in a superior reinforced ma-
trix as compared to randomly dispersed particles and fibers  previously
reported.  This new approach to the problem has resulted in  extended
compressor life with low rates of wear and leakage.  Halliwell found that

                                   138

-------
reliable compressor operation could be obtained at 5000 psig for periods
beyond 1000 hours using such seals in lieu of conventional split rings.
The operational life of PTFE rings in a reciprocating steam expander will
be affected by the quality (dryness) of the steam.  Wear rates of PTFE
decrease with dryness until a dewpoint of ^ -40°F is reached whereupon
the wear rate increases drastically.  However, small amounts of condensed
water will cause an increase in wear rates and should be avoided.

      Polyimides are the next generation of lubricating plastics pushing
the use temperature from the 400° - 450°F for PTFE to 500° to 600°F.
The polyimides also have superior friction and wear properties.  Unfor-
tunately the polyimides cannot be used in high temperature (> 212°F)
steam or water because they will hydrolize.

      6.1.2.1.3 Metals

      In some applications, metals are used on bearing surfaces to achieve
improved friction and wear characteristics.  The metals (and alloys) used
in these applications can be categorized in three classes:  (a) classical
layer-lattice crystal, (b) low shear strength and (c) soft oxide formers.
                   (31)
Buckley and Johnson^  ' investigated the influence of crystal structure
of metals on their friction and wear behavior and found that relatively
low coefficients of friction could be obtained for cobalt in which the
hexagonal crystal structure is stabilized.  A coefficient of friction of
0.3 was measured for an oxide-free 25% Mo-Co alloy at speeds of 2000 ft/
                          —9
min (750°F; pressure of 10   torr).  The coefficient of friction of cubic
cobalt under similar conditions is on the order of 0.7.

      A number of metals that do not have the layer lattice structure
are still widely used in bearing applications.  These metals are listed
                                                (32)
in Table 6.1-III;  they are soft  and shear  easily    .   For example,  the inetals
lead, tin, indium and silver are used in many common plain bearings, i.e.,
piston pin bearing.  However, the coefficients of friction of these
metals below their melting points are relatively high (> 0.3).   When
used above their melting points (liquid film) friction coefficients are
considerably lower, on the order of 0.17 for lead.
                                  139

-------
                              Table 6.1-III

                     Soft Metals Used in Bearings
              Metal          Moh hardness      Melting point,
                                                    °C
Indium
Thallium
Lead
Tin
Cadmium
Gold
Silver
Platinum
Rhodium
1
1.2
1.5
1.8
2
2.5
2.5-3
4.3
4.5-5
155
304
328
232
321
1063
961
1755
1955
      A commercial material incorporating a soft, low shear strength metal
                                   (33)
for lubrication is the Bishiralloysv   .  These alloys are produced by
pressing and sintering alloy powders, followed by impregnation with lead
and finally rolling to size.  The compositions and hardnesses of these
materials are given in Table 6.1-IV.  Alloy C is reported to have the best
wear resistance; the coefficient of friction is on the order of 0.2.  A
continuous film of lead is maintained on the surface by flow of the lead
in the pores of the alloy.  However, because of the low melting point
of lead (621°F) the maximum ambient service temperature for the Bishiralloys
is about 575°F.

      The metal gallium is a liquid at 86°F and effective boundry lubrica-
tion is obtained with gallium rich films on AISI 440C stainless steel.
The coefficient of friction of gallium coated AISI 440C vs. AISI 440C at
500°F in air was measured to be 0.104 (sliding speed 390 ft/min, load
1000 grams)(34) .

      Oxidation of metals will result in significant reductions in the
                                                             (35)
coefficients of friction and protection of the metal surfaces    .  The
transition temperatures for improved friction and wear behavior were
determined for iron, copper, nickel, molybdenum and chromium to be 100° -
200°F, 400° -  500°F, 1200° - 1400°F, 800° - 900°F, 800° - 1100°F, re-
spectively.  Coefficient of friction values of 0.2 - 0.3 were obtained

                                  140

-------
                                         Table 6.1-IV
                       Chemical Composition and Hardness of BISHISALLOY
                   Basic Composition of BISHIEALLOY Matrix        _ „. „      ,.
                                     wt  «                        wt % Range of

                                       '

BISHIRALLOY-A
BISHIRALLOY-B
/
Fe Cu Ni Mo Cr C
Bal 5.0 — — — less than
0.2
Bal 2.5 1.5 2.5 — less than
0.2
.impregnated x.eaa naraness
in BISHIRALLOY Brinell
15 ~ 30 75 - 95
15 ^ 30 105 - 135
BISHIRALLOY-C      Bal   2.5   1.0   1.0   5.0   less than

                                                 0.2                15 ^ 30           65 - 95

-------
upon the formation of the oxides in comparison to values of 0.5 - 1.0
for clean surfaces at lower temperatures.  The formation of so-called
soft oxides MoO , WO , Cu 0, ZnO, Co 0 , PbO, CdO were the most effective
in preventing surface damage.  The lowest friction coefficient was ob-
tained with PbO.

      A commercial product is available which reportedly provides low
friction and wear characteristics as the result of the properties of its
oxide surface film.  The material is Clevite 300, an iron base alloy con-
taining cobalt and molybdenum.  The alloy is produced by powder metallurgy
techniques and its properties are listed in Table 6.1-V.

      6.1.2.1.4 Sulfides/Selenides

      The sulfides and selenides of molybdenum, tungsten, tantalum and
columbium have the layer-lattice structure similar to graphite and have
extremely low shear strengths.  Further, they do not require water vapor
or oxygen for their lubricating ability.  Another difference between the
sulfides or selenides and graphite is that the weak easily sheared bonds
occur only between every third layer of atoms instead of every layer as
is the case for graphite.  In MoS_, for example, each layer of molybdenum
atoms is strongly bonded to the adjacent layers of sulfur atoms on each
side but the layer of sulfur atoms is weakly bonded to the sulfur atoms
in the next adjacent layer so that shear occurs through the weak sulfur-
sulfur bonds.  Recent studies on the lubricating mechanism of MoS^ were
                                                      (29)
reported at the solid lubrication conference in Denver

      Since MoS» occurs in nature, it is considerably less expensive than
the other sulfides and the selenides or tellirides which are synthetically
produced.  For this reason most of the common solid lubricant composites
based on this group are made from MoS_.  The effectiveness of MoS? in air
is limited to about 750°F where oxidation begins  to effect its lubricity.
The oxidation products of MoS. are MoO  and a sulfur compound which can
result in abrasion and corrosion and an increase  in the coefficient of
friction at the  lower temperatures.   (It should be noted that at tempera-
tures above 1300°F - 1400°F, Mo03 can be an effective lubricant; the melting
point of Mo(>3 is 1463°F).
                                   142

-------
         MECHANICAL PROPERTIES
                                                    Table 6.1-V
                                              Properties of Clevite 300
•e-
co
Temp. (F)
75
600
900
1200
FRICTION
Yield
Strength, psi
70,500
-
62,250
48,000
AND WEAR PROPERTIES
Ultimate
Strength, psi
106,000
107,000
95,800
71,000

Elongation, %
-
2.8
2.7
6.6

Elastic
Modulus
psi x 106
28.4
25.8
24.4
22.5

Compression
Strength, psi
210,000
-
-
-

Hardness
RA
66 (31
57
-
35


Re)




         (Based on face-seal tests of Clevite 300 against itself at 18 psi load and 150 fps surface speed,

         in air.)


                       Temp.. °F	Coefficient .of Friction, "f"	Wear Rate, in/hr
500
750
1000
1200
0.072
0.063
0.052
0.047
0.00073
0. 00062
0. 00035
0. 00030

-------
      Composites based on the solid MoS7 lubricant have been developed by
                                   (37)
Boeing      and described by Devine    .  The composites developed by Boeing
were quite brittle and even the most successful composite, 90% MoS2~8% Fe-2%-Pt,
would have strength problems because of the brittle iron sulfides that form at
the sintering temperatures.  Piston rings have been produced from iron bonded
MoS_ and evaluated by Midwest Research(38) .  The solid composite called "Navy
     (37)
Lube"v    is a MoS2-Graphite-Sodium Silicate (NaOSi02) and is not only weak
but the NaOSiOp is hygroscopic and would not be suitable for use in the steam
environment.

      Another series of composites based on MoS_ and WS0 are described
          (39)
by Hopkins   ' .  -Three of these composites have shown low wear rates and
low friction in air.  The compositions of these materials are:

      1.  53% WS2 - 12% Co - 35% Ag
      2.  80% MoS2 - 20% Ta
      3.  50% MoS2 - 38% Ta - 12% Fe

      Overall, the use of solid lubricant composites with MoS_, or other
disulfides or diselenides, as the base do not appear attractive for use
as piston rings in steam expanders primarily because of their low strengths
and fragile nature.

      6.1.2.1.5 Porous Metal Composites

      In order to utilize the good lubricating qualities of some of the
better  solid lubricants that are too weak to be used as the base of
self- lubricating composites, porous metal composites have been developed.
These composites are generally produced by powder metallurgy techniques
in which the metal or metal alloy and the lubricant are blended, pressed
and sintered to form a metal matrix throughout which are discrete pockets
of solid lubricant.  In one investigation, it was found that a minimum
concentration of 5% of the solid lubricant was required to achieve a
transfer film and thus low wear rates.  Friction coefficients of 0. 2 were
measured*-   '.  On the other hand, too high a concentration of the lubricant
results in a weak structure.  Examples of this type of porous metal com-
posite  are:  (a) 5-15% MoS2 in a 95% Ag - 5% Cu matrix, (b) 5-40% MoS2
in a nickel or 80% Ni - 20% Cr matrix, (c) the commercial Molalloy - MoS2
in a refractory metal (Ta/Mo) matrix.

                                  144

-------
      Another method  of producing  the porous metal composites is  to  prepare
a metal skeleton with controlled porosity with respect to volume  and
distribution and impregnate the pores with the lubricant by application
of pressure and heat.  Metal skeletons with up to 65% void volume have
been produced with Inconel 600, Inconel X-750, Hastelloy X, Nichrome V
alloys in which PTFE,  graphite, MoS  have been impregnated.  A Inconel
600 (60 - 65%voids) impregnated with PTFE has performed well in seal
applications at temperatures up to 500°F.

      The metal fluorides have shown promise for use as lubricants for
high temperature application in air.  Sliney at NASA Lewis Research
Center has investigated the use of fluorides for a number of years    '   '   .
Coefficients of friction of self-lubricating porous metal composites (40%
void volume in Inconel X-750), vacuum-impregnated with fused fluoride
eutectic of 62% BaF  - 38% CaF« were measured to be less than 0.1 at
1000°F and at a sliding velocity of 2000 ft/min while under load.   The
coefficient of friction increases as the temperature and speed decrease;
at 500°F and a speed of 1000 ft/min the coefficient of friction is approxi-
mately 0.3.  More recent studies have shown that coefficients of friction
of 0.2 + 0.05 could be obtained at speeds as low as 500 ft/min over the
temperature range of RT to 1700°F.  These data were obtained with a porous
nickel alloy impregnated with a high temperature enamel (NBS-418)  and
an overlay of the BaF« - CaF  eutectic.

      Finally, this approach is compatible with classical piston ring
materials technology where gray cast irons are used because of the ex-
cellent lubricity imparted by flakes of graphite.   Similarly,  the nodular
Ni-Resist alloys of 20 - 30% nickel content with its spheroids of  graphite
has been shown to have good friction and wear characteristics in conjunc-
tion with good corrosion resistance in superheated steam.   It is believed
that the use of strong porous metal composites impregnated with a solid
lubricant of low shear strength and ability to establish a transfer film
is one of the most attractive materials for application as piston rings
in reciprocating steam expanders.
         *
      6.1.2.1.6 Hard Surfacing Materials

      In applying the  adhesion theory in friction and wear, relatively
low coefficients of friction are achieved between two sliding hard ma-
                                  145

-------
terials that have no affinity for each other, i.e., materials with little
or no mutual solubility or tendency to form intermetallic compounds.  The
hard, wear resistant surfacing materials fall into this class of materials
and warrant investigation.

      Plasma sprayed coatings are of interest because they can be readily
applied.  Koppers has plasma sprayed rings with their K-1051 coating
(metal bonded Cr«C2)  for use in lubricated,  reciprocating
steam expanders running against nitrided steel or AISI 440C stainless steel
which was heat treated to a hardness of Rockwell C 50-55.  Since the
hard metal carbides are metallic in nature with respect to their atomic
bonding, a more suitable materials combination with respect to "alloying
tendency" might be an oxide coated cylinder liner running against a
carbide coated ring.  Possible combinations are Union Carbide Corp. LC-19
(Cr20  + A1o°3^ sPrayed on the cylinder wall and either LW-1 (Cobalt
bonded WC) or LC-1  (Nichrome bonded Cr C ) sprayed on the ring surface.
In dry rubbing tests between LC-1 (Cr3C_-15% Ni-Cr) and LA-2(A12C>3),
coefficients of friction of 0.15 - 0.27 were measured in the temperature
range of RT to 1400°F.

      E*I. du Pont de Nemours and Co. have recently made available a
series of anti-friction and anti-wear LP alloys offering outstanding ad-
vantages.  Metallurgically, the LP Alloys consist of hard grains of an
intermetallic compound with a Laves Phase structure dispersed in a softer
matrix, that provides good embeddability characteristics in contrast to
other hard facing materials.  Currently preferred compositions contain
Laves-phase hexagonal close-packed intermetallic compounds of cobalt,
molybdenum and silicon  (Co Mo~Si and CoMoSi) in a cobalt-rich matrix.
The Laves phase, depending upon its composition has a micro-hardness of
1000 to 1500 DPH and the matrix has a hardness of 200 to 800 DPH.  Over-
all, the bulk hardness ranges from R  30 to R  60.  Tests have shown
                                    c        c
that, under extreme loading and boundary lubrication conditions, LP Alloys
exhibit very little wear.

    •  LP Alloys are produced as medium and fine powders for plasma  spraying
and powder metallurgy applications; they also can be cast as semi-finished
parts.  The compositions  (in wt. %) of LP Alloys are shown in Table 6.1-VI.

                                  146

-------
                             Table 6.1-VI
                    Chemical Composition of IP Alloys


                    Co       Mo        Si      Cr     Vol. % Laves
LPA 100
LPA 200
LPA 300
LPA 400
35
70
45
62
35
28
48
23
10
2 	
7
2 8
65
20
98
50
In one application, the sleeve and roller bearings and the piston skirt
and rings of a 2-cycle engine were coated with LPA-100 and the engine
operated without oil in the fuel for 50 hours with satisfactory results.

      Another new material of interest that is available from E.I. du Pont
de Nemours is the compound Ni B.  The material is applied to the surface
by electro-chemical methods and has a microhardness of 900 - 1500 knoop
(R  60-70).  Friction data of Ni_B against hardened steel show a coef-
ficient of friction of 0.1 at 600°F in air, unlubricated.

      Other materials in this class that could be considered are:  various
grades of SiC, Si.N, and TiC in the form of pressed and sintered compacts
or applied as coatings.

      6.1.2.1.7 Mixed Composites

      Multicomponent self-lubricating composites have been developed that
contain a metal matrix, a film former, and a load-carrying component ^^'.
Examples of these types of composites are Ag-PTFE-WSe2,Cu-PTFE-WSe2,
Ag-Bronze-PTFE-WSe2,Ag*Hg-PTFE-MoSe2. The latter material has shown ex-
cellant friction and wear characteristics at 600°F ^   \  The modified
PTFE melts at 590°F and is thought to provide the excellent friction and
wear characteristics.

      Other mixed component self-lubricating composites that have been
investigated are:  WSe2-Ga/In ^45\and. Graphite-WS -NaF ^6\

      6.1.3    Bonded jSolid Film Lubricants

      The primary disadvantage of the bonded solid film lubricants is
                                  147

-------
that there is little stored lubricant in the film and as a result,  service
lives are limited.  The bonding agents are either organic thermoplastic
or thermosetting resins (phenolic, epoxy, polyimide) or inorganic metal
salt (sodium silicate) and fused ceramic (aluminum phosphate).   Because
the bonded films have so little strength, the films are kept as thin as
possible, usually on the order of 0.0005 inch or less.  For these reasons,
the use of bonded films is generally limited to high load/low sliding
velocities or low load/moderate velocities.  Most commercially available
bonded solid film lubricants are limited in temperature by the bonding
agent.  For example, at temperatures above 400°F, the organic bonding
agents tend to thermally decompose causing failure of the film because
of lack of adhesion.  Commercial products of this type incorporating
various solid lubricants are:

               Solid
             Lubricant       Bonding Agent       Service Temp.. °F
               MoS2            Epoxy                 275 - 300
       90% MoS2-10% Graphite   Phenolic                 300
               PTFE            Polyimide             450 - 500

      A newer bonded film based on the polyimide for use at higher tem-
                                               (47)
peratures was reported by Campbell and Hopkins    .  It consists of MoS
and Sb-O- dispersed in the polyimide binder  (MLR-2) and it shows excellent
wear properties.  However, it should be pointed out again that the
polyimide resins  cannot be used in high temperature steam and water.

      The metal salt and fused ceramic bonded films were developed for
use at temperatures generally above 600°F.  Examples of these  types  of
bonded films were reported by Hopkins ™8^:  MLF-5, is a sodium silicate
bonded MoS--graphite - Au film and MLF-9 is a aluminum phosphate bonded
MoS -graphite - Bi film.  Ceramic bonded MoS -graphite films have been
used in excess of 700°F in air0  As mentioned previously sodium silicate
is hygroscopic and cannot be used on steam or water.
                                   148

-------
      The solid lubricants of interest that are being developed at Midwest
Research are the FSAL 28 (AlPO^ bonded Ba2F + Ca2F) and FSAL 29 (AlPO^
bonded Ba2F + Ca-F + Mg-F).  These lubricants are applied as bonded films
and can be sintered at temperatures below the fluoride eutectic temperatures.
About 6 vol.% A1PO, has been found to be the optimum binder content.
Coefficients of friction of the FSAL 28 versus Inconel X in air are on the
order of 0.1-0.15 at 500°F and 0.10-0.12 at 1000°F.  Loads have varied
between 6000 - 25,000 psi at speeds of "180 ft/min.

      At the higher temperatures, >1000°F, PbO has been shown to exhibit
good friction and wear properties.  Because PbO oxidizes to PbJ3,  at
temperatures below 1000°F resulting in loss in lubricity, additions of SiO~
                                                  (49)
are made to PbO to inhibit the oxidation reactions    .  Very low friction
characteristics (0.1 coefficient) have been achieved with 10% Si02-PbO bonded
film at temperatures of 500° to 1200°F.

      Another very promising new material that is being studied is graphite
fluoride, (CF )     .  Coefficients of friction equal or superior to MoS9
             3c n                                                        &-
and graphite were reported at temperatures up to 750°F in air.   Superior
wear properties are attributed to greater adhesive qualities.  Although
moisture is beneficial,  it is not a prerequisite for low shear strength.

      The use of bonded solid film lubricants does not appear to be as
promising for the lubrication of the cylinder walls in a steam reciprocating
steam expander as do the self-lubricating composites.  It would appear
that if the bonded solid films were to be evaluated for this application,
they should be applied to the cylinder wall rather than the piston rings
in order to provide a greater amount of available lubricant for increased
life.
                                    149

-------
6.2   Lubrication Technology Base - Liquid Lubricants

      Lubrication of the large, slow speed reciprocating steam expander
operating under steam at relatively low temperatures and pressures has
been well established for years.  On the other hand, lubrication of the
small, high speed and high performance steam expander operating at high
steam temperatures and pressures is considerably more complex and may be
more complicated than the lubrication of internal combustion engines.
Using conventional techniques of lubricating steam cylinders and valves
by injecting lubricants directly on the parts and/or by injecting the
lubricants into the inlet steam, two separate lubricating systems have
been used, one for lubricating the components in the crankcase and one
for lubricating the cylinder walls and valves that are operating in the
high temperature steam.  This is accomplished through the use of a cross-
head piston design; both the General Motors SE-101 and SE-124 steam
engines employed this concept.  A further complication is the fact that
the cylinder walls cannot be cooled without decreasing the efficiency of
the engine cycle so that the fluid lubricating the hot cylinder walls and
valves will come in contact with metal surfaces that are at temperatures
of 600° - 800°F.  Thermal decomposition products of the lubricant resulting
from being in contact with metals at these temperatures will tend to foul
up the steam generator tubes unless the lubricant is separated from the
condensate water.  Frequent lubricant changes would be expected.
                                   150

-------
      Successful use of solid lubricants to lubricate the cylinder wall/
piston ring and valve stem/valve guide interfaces will eliminate or greatly
minimize many of the problems encountered with liquid lubricants.  In fact,
it may be possible to utilize a standard trunk piston design.  However,
due to the fact that the engine cylinders cannot be cooled, as stated
previously, even the bulk temperature of the crankcase lubricant may be
exceptionally high in comparison to the crankcase temperature in con-
ventional internal combustion engines.  Crankcase lubricant bulk tempera-
ture may be as high as 300°F unless a special cooling system is employed
to cool the lubricant.  Even so, the limiting temperature may be the hot
spot temperature that the lubricant will see on the bottom surface of
the power piston head of the trunk piston design or the internal surface
of the piston rod in the crosshead piston design.  The average  bulk
lubricant temperature and maximum hot spot temperatures for the two engine
designs in this program are:

                                         Max. Lubricant Temp..  °F
         Engine Design                   Bulk           Hot Spot
         Trunk Piston                    250              540
         Crosshead Piston                250              425

Although these temperatures are not unusually high for some
conventional petroleum oils, it may be desirable to select a lubricant
capable of operating at higher temperature in order to achieve  a longer
service life or some specific property.   It may be possible to  provide
a lubricant that only requires changing every two years  or even the life
of the engine as is the case for oils in large steam turbines.   A trade-
off can be made between the cost of the lubricant and the service life.

      Hydrocarbon oils that are extracted from petroleum sources and
synthetic lubricants both were considered for use as the crankcase lubricant
in the steam expanders.  In the selection of a suitable lubricant,  it is
necessary to:  first, select a suitable base and second  to select the
proper chemical additives that are compatible with the base and the en-
vironment..  To be considered a good lubricant for the intended  application,
the lubricant must have the following important characteristics:
                                  151

-------
      1.   Good oxidation resistance
      2.   High thermal stability
      3.   Low volatility
      4.   Suitable viscosity
      5.   Low pour point
      6.   Adequate lubricity
      7.   High ignition temperature
      8.   Good hydrolytic stability

      The last characteristic is particularly important with respect to
steam expander applications in that some contamination of the crankcase
with water is highly probable, particularly during start-up and shutdown.
For this reason, the lubricant needs better rust protection and much
better demulsibility than do current motor oils.  Additives similar to
those used in hydraulic oils, i.e., calcium sulfonates could be used
effectively for rust protection.  The partial organic acid esters and
phosphorus containing acid esters also are used for this purpose. Rust
inhibitors usually have a high polar attraction toward metal surfaces
and form a continuous protective film over the surfaces.  Care must be
exercised in the selection of rust inhibitors to avoid corrosion reactions
with nonferrous metals or emulsions with water.  Chemical additives are
rarely effective in improving the hydrolytic stability of lubricants and
are seldom used to improve this property; hydrolytic stability is an
inherent quality of the base lubricant.  Hydrocarbons have excellent
hydrolytic stability and major concern is with the additives and with
synthetic lubricants.  The straight alkyl-chain compounds generally are
more easily attacked by water than the highly branched structures.  For
example, the esters have long chains and generally have poor hydrolytic
stability at temperatures of 400°F and for this reason the esters will
not be considered as a base for this application.  Hydrolysis and water
contamination  can result in the following reactions:  (a) change in physical
properties, (b) generation of sludges and other insoluble compounds,
(c) decrease in solubility of essential additives, (d) corrosive attack,
(e) release of volatile compounds due to chemical breakdown.  Obviously
all of these changes are detrimental and must be kept to a minimum.  It
is expected that a complete reformulation of the crankcase lubricant will
be required because most of the extreme pressure agents and inhibitors
currently in use have little  tolerance to water.
                                   152

-------
      Because the bulk temperature of the crankcase lubricant is expected
to be higher than the temperatures in the current internal combustion
engines, oxidation stability of the lubricant must be superior to the
current premium-grade motor oils.  Oxidation of the hydrocarbons results
in the formation of oil-soluble acid compounds which increases the vis-
cosity of the oil and can be corrosive; oxidation of the synthetics are
more complex depending upon the type of structure, i.e., some form volatile
gases and do not change the viscosity.  Additives to inhibit oxidation
include sulfur and phosphorus compounds and the amine and phenolic com-
pounds.

      Other additives which are believed to be required in the lubricant
are defoamants and anti-wear additives.  The most common defoamant is the
organo silicon oxide polymer and only small concentrations are required
(1 - 20 ppm) to inhibit foam formation caused by the action of the engine
crankshaft.  Anti-wear additives are required to minimize friction and
wear in highly loaded components such as the cam/tappet interface where
Hertzian stresses can be as high as 200,000 psi.  Extreme-pressure (EP)
additives are used and they react with the rubbing metal surfaces to form
a lubricating film which protects the metal surface when the lubricating
oil film is lost.  The most common EP additives are zinc, phosphorus,
sulfur and chlorine compounds.

      In the survey for a suitable crankcase lubricant for the steam ex-
panders to be tested in this program, one of the more attractive solutions
appeared to be tne use °f a highly refined turbine oil.  Many of these oils
are the result of extensive development work at the Pennsylvania State
University, where processes have been developed for "super-refining"
mineral oils to achieve lubricants which have lower pour points, better
viscosity-temperature characteristics and higher thermal and oxidative
stability.  Within the family of super-refined mineral oils, it is possible
to use either paraffinic or aromatic-base oils.  There are distinct dif-
ferences in the physical properties of these two types of hydrocarbons.
Because of their intermolecular spacing, the paraffinic  oils are more  com-
pressible,' show less change in vfscosity with temperature, have lower
densities and lower pour points.
                                  153

-------
      The production of super-refined mineral oils is essentially a clean-
up operation where the oils are carefully dewaxed, many of the polar im-
purities are removed, and selective catalytic hydrogenation is used to
minimize unsaturation.  This process removes not only undesirable im-
purities, but also naturally-occurring oxidation inhibitors and polar
compounds, which are essential for boundary lubrication.  Thus, it is
necessary to compound these oils with oxidation inhibitors and anti-wear
additives.  Since thermal decomposition will also proceed by the forma-
tion of free radicals, a small percentage of an aromatic disulphide was
found to be effective in protecting the oil.  The use of the aromatic
disulphide gives an added bonus since it will also act as an anti-wear
agent by surface reaction between the metal and the sulphide.

      The use of blending techniques opens up many possibilities as far
as the control of properties such as viscosity and pour point.  A wide
range of properties can be obtained by blending suitable base stocks to
achieve the bulk viscosity and pour point required for the application.

      A common turbine oil that has proven to have satisfactory service
(*>» 20 years) at temperatures < 250°F in steam turbines and generators is
produced by a number of refining companies and is known as Teresso 65,
Industrial Oil 61, Tellus 69 or DTE Extra Heavy.  The oil has oxidation
and rust inhibitors but no EP anti-wear additive which would have to be
formulated.  Characteristics of the oil are as follows:

      Flash point, deg F, min	420
      Viscosity, Saybolt Universal, sees 100 deg F, min 	 540
                                                    max	700
      Viscosity index, min	•	  85
      Pour point, deg F, max	+25
      Neutralization value  (total acid number)
         mg KOH/g, max	0.80
      Oxidation stability test, hours, min  (a)	1000
      Sludge, %, max  (b)	0.10
      Rust prevention test	Shall Pass
       (at)
          Test  carried out  to  total acid number of  2.0 mg KOH/g.
          In oil  taken from oxidation  test.
                                   154

-------
       The major  disadvantages  to  the use of  this  oil  are  the  relatively
 high pour point  and  the  low hot spot temperature  capability.  Although a
 modified turbine oil would probably be  satisfactory for the crosshead
 piston design, it may not be satisfactory for the trunk piston  design  be-
 cause of the anticipated 540°F hot spot temperature.

       The synthetic  hydrocarbon lubricants have about a 100°F advantage
 over the mineral oil based turbine oils and  appear to have the  necessary
 properties for the application.   For relatively long time service, the
 following generalized maximum  service temperatures can be stated:

       Base         Max.  Bulk Oil  Temp.. °F    Max. Hot Spot Temp.. °F
 Mineral Oil        250°  - 275° (possibly              * 500
                                300°F)
 Synthetic Hydro-  350°  - 375° (possibly              *> 600
 carbon                         400°F

       A special  synthetic hydrocarbon formulation  (XKN-1301-C) was pre-
 pared  by Mobil   Research and Development Corp. for evaluation in the
 program.  The lubricant  has suitable oxidation,  rust and foam inhibitors
 and an anti-wear  additive.  The properties are given in Table 6.2-1.  The
 lubricant has a  low pour point (< -65°F) and a viscosity in the SAE 30
 range.  However, its high viscosity index (149)  puts it in the SAE 40
 range  at elevated temperature.  The lubricant appears to have good anti-
 foam performance and good rust protection and although it forms a small
 amount of emulsion, no emulsion problem would be expected in the engine.

      A general  comparison of various other synthetic lubricants is made
 in Table 6.2-II    .   Known disadvantages  of some  of these synthetic lubri-
 cants  include poor hydrolytic stability of the esters and poor viscosities
and high pour points of  the polyphenyl ethers (in addition to their high
 cost).  Projected costs of the synthetic hydrocarbons are not expected
 to exceed current premium grade motor oils.

      The ESSO Research and Engineering Co.  (Govt. Research Lab) is under
contract to Steam Engine Systems Corp.  to develop  a suitable lubricant
for the lubrication of cylinder walls in a reciprocating steam expander
                    (52)
operating at 1000°F v  '.   In their work they found  the inhibited blends
of hydrorefined paraffinic distillates and residua  to be the most suitable

                                  155

-------
                           Table  6.2-1
                  Properties of Experimental
               Synthetic Hydrocarbon Lubricant

                           XRN 1301 C
Gravity, °API (ASTM D-287)                            37.4
Specific Gravity                                      0.837
Pour Point, °F (ASTM D-97)                             <-65
Flash Point, °F (ASTM D-92-1)                         460
Viscosity: cs @ 210°F                                 11.05
           SUS @ 210°F                                63
           cs @ 100°F (ASTM D-445)                    74.5
           StS @ 100°F                                345
           cs @ -40°F                                 34,000
Viscosity Index                                       149
Acid Number (ASTM D-664-1)                            0.08
Base Number (ASTM D-664-3)                            0.11
Foam (ASTM D-892)
     Sequence I   Tendency, ml                        330
                  Stability, ml                       0
     Sequence II  Tendencey, ml                       40
                  Stability,                          0
     Sequence III Tendency, ml                        380
                  Stability, ml                       0
Rust Test, 48 Hrs, Distilled Water  (ASTM D-665-2)     Pass
           48 Hrs, Syn.Sea Water  (ASTM D-665-4)       Pass
Emulsion Test  (ASTM D-1401)
     Total Emulsion, ml                               5-10
     Time for 3 ml, Minutes                           31-33
     Time for Complete Break, Minutes                 34-37
Water Trace, ppm                                      40
Surface Tension, Dynes/cm                            30.7
Panel Coker, 24 Hrs @ 600°F, Deposit, mgs             27
Mobil B-10 Catalytic Oxidation, 40  Hrs @ 260°F
      %  Viscosity Increase  @ 210°F                  1
     Lead Loss, mg                                    1.4
     Sludge                                           Nil
     Neutralization Number Increase                  0.2
                              156

-------
                         Table 6.2-1 (Cont'd)
                   Properties of Experimental
               SynthetIc Hydrocarbon Lubricant
(a)
Mobil B-10 Catalytic Oxidation, 40 Hrs @ 325°F
      %  Viscosity Increase @ 210°F                   17
     Lead Loss, mg                                    17.3
     Sludge                                           Nil
     Neutralization Number Increase                   2.8
Mobil B-10 Catalytic Oxidation, 72 Hrs @ 325°F
      %  Viscosity Increase @ 210°F                   230
     Lead Loss, mg                                    225
     Sludge                                           Trace
     Neutralization Number Increase                   12.4
SAE Wear Test, Steel-on-Steel, 30 Min @ 150 Ib Load,
 250°F Oil Temperature
     Total Weight Loss                                0.018 mg
Almen Load, psi                                       8000
Mobil Thin Film Oxidation Test -600°F
  (100 = Clean)
Mobil Thin Film Oxidation Test -625°F
  (100 = Clean)
        93
        69
(a) Mobil Research and Development Corporation
                               157

-------
                             Table 6.2-II

                     Performance of Selected^ Types
                         of Synthetic
                               Probable
                           Maximum Thermal     Resistance     Extreme
                           Stability Limit,     to Water      Pressure
   Class of Compound	        °F	Degradation    Lubricity
Polyglycols                      600            Excellent     Good
Phosphate esters                 800            Good          Excellent
Dibasic acid esters              600            Good          Good
Chlorofluorohydrocarbons         600            Excellent     Excellent
Silicones                        900            Poor          Poor (Steel)
Silicate esters                  800            Poor          Fair
Fluoroesters                     600            Good          Poor
Neopentyl polyol esters          600            Good          Good
Polyphenyl ethers                600             —           Good
Silanes                          700             —           Poor
                                   158

-------
of the petroleum base oils.  A combination tricresyl phosphate (TCP) and
ethyl-702 appears to offer the best anti-wear and anti-oxidant properties
in the selected bases.  The candidate lubricant (designated 2415-50-1) has
a higher viscosity (SUS) rating at the standard temperatures of 100°F
and 210°F, 663 and 74 respectively, but a lower viscosity index (102) than
the Mobil XRN-1301-C lubricant.  However, as discussed previously, there is
no requirement to lubricate the cylinder walls or valve parts in this program.

6.3   Lubricant Recommendations

      6.3.1   Solid Lubrication
      In the steam expander design, there are three areas requiring solid
lubrication; these include the inlet valve,  the piston/cylinder and the shaft
connecting the power piston with the crosshead piston.   The inlet valve stem
will slide in a guide that is exposed to steam and that is sealed by means of
compression rings.  The shaft connecting the power piston to the crosshead
piston will slide in a seal that is exposed to steam on one side and oil on
the other.  The most severe lubrication problem is the piston/cylinder wall,
and either the power piston rings or the liner can serve as the lubricant.

      Although the qualitative correlation of the coefficient of friction and
wear rate is generally recognized, the exact correlating function between the
two is difficult to obtain.  Theoretical relationships may be developed which
are useful in ranking material and lubricant combinations.  However, the
reduction of the theoretical relationships to useful design parameters by
testing at the expected service conditions is necessary.

      Friction effects arise from the tangential forces transmitted across
the interface of contact between two bodies.  The wear phenomena consists of
the removal of material from the contacting surfaces.   Adhesion,  the ability
of contacting bodies to withstand tensile forces after being pressed together,
is the primary interacting phenomena.  The concept of surface energy makes
it possible to develop parameters for making rudimentary predictions of the
performance of specific materials under sliding conditions.
                                  159

-------
      While 90% or more of the resistance to sliding arises from the need
to shear strongly adherent surface atoms of the contacting materials, other
factors have to be considered.  A roughness component arises from the necessity
of lifting one surface over the other's asperities, which usually contribute
about 0.05 to the friction coefficient.  Hydrodynamic lubrication can eliminate
this component (liquid phase lubrication).  A plowing component is present
when a hard, sharp surface digs into a softer surface and produces a groove.
For really rough surfaces this term may be large, but otherwise it is usually
negligible.

      Adhesive wear is best combatted with hard materials with low interaction
tendencies (or one member of the pair a non-metal), or by use of two metals
with a high tendency for interaction reduced by a good boundary lubricant.
The use of hard materials does not in itself produce significantly lower wear
rates, but the probable low alloying tendency may produce a factor-of-10
change in the wear rate.

      Although adhesive wear  (sometimes called galling and scuffing) resulting
in welding of the surface asperities is usually associated with metallic
materials, the adhesion theory is also applied to solid polymers.  The
difference between the metals and plastics is that the deformation in plastics
may be partly elastic over a wide range of load.  Also, the flowing term
and surface roughness are of greater importance in plastics.  The mechanism
of wear with carbon-graphites is primarily abrasion which occurs when asperities
of two moving surfaces touch and wear fragments are formed from one or both
surfaces.  Grooves are generally plowed in the "softer" carbon graphite.

      Based on the current technology of solid lubrication and wear resistant
materials that was discussed in Section 6.1 and in the preceding paragraphs,
a list of candidate material combinations for use in sliding contact was
compiled for the piston ring/cylinder application and is tabulated in Table
6.3-1.  The primary selections were a metal impregnated carbon-graphite
material for the piston rings rubbing against an AISI 440C SS or hard
chromium plated steel cylinder liner.  The same material combination was
selected for the back-up design for the piston rod seal in the crosshead piston
design.  The primary selection for the rod seal was a 15% bronze + 5% MoS,,
filled PTFE material rubbing against a hard chromium plated H-ll alloy rod.
                                   160

-------
                               Table  6.3-1
           CANDIDATE PISTON RING/CYLINDER LINER COMBINATIONS
              Ring
                                              Liner
                                    Hard Cr Plate    or AISI 440C SS
                                    Carbon-Graphite, Sb Metal Impregnated
1. Carbon-Graphite, Sb Metal
   Impregnated ^a'
2. LC-1C (Cr3C2 + 15% Ni Cr) or
   K-1051 (Cr3C2 Cermet)
3. Graphite 5890                   Graphite 5890
4. Hast X Porous Structure (35 v/o)Hard Cr Plate(b)
    + Graphite Filled
5. Hast X Porous Structure(35 v/o) Hard Cr Plate^
    + NBS418 + (BaF2 + CaF2)
    Eutectic Filled
6. LPA-100 (Laves Phase - 65 v/o)  LC-19 (Cr203 + A1203) or Nitrided
                                                                     (a)
     + 20% Ni
 7. Ni3 B

 8. LW-1 (WC + 9% Co)

 9. LC-1C (Cr3C2 + 15% Ni Cr) or
     K-1051 (Cr3C2 Cermet)
10. LSR-1 (TiC) over AISI 440C

11. K-35 (Ni-Resist) or H.S.31
12. K-35 (Ni-Resist)
13. K-35 (Ni-Resist)

14. LW-1 (WC + 9% Co)
                                    Steel or Ni-Resist D3
                                    LC-19 (Cr203 + A1203) or
                                    AISI 440C SS
                                    LC-19 (Cr203 + A1203) or Nitrided
                                    Steel or Meehanite
                                    LC-19 (Cr203 + A1203)
                                    LC-19 (Cr203 + A1203) or Nitrided
                                    Steel
                                    LC-19 (Cr203 + A1203)
                                    Hard Cr Plate + MoS2 Bonded Film
                                    Hast X Porous Structure (35 v/o)
                                    + Graphite Filled
                                    Hast X "Porous Structure (35 v/o)
                                    + (BaF2 + CaF2) Eutectic Filled
                                                                    (b)
(a)  Primary recommendation for first engine.
(b)  Limited to temperature below 700°; for temperatures in excess of
    700°F,, -one of the following liner coatings can be used:
              • LC-19 (Cr203 + A1203) or LC-4 (Cr203)
              • LC-K (Cr3C2 + 15 Ni Cr)
              • Chromize
                                   161

-------
A hard surfacing material (LPA-100 + 20% Ni) and Ni-Resist (D-3 type)
cast iron materials combination was selected for the valve stem ring/valve
guide components with cold worked Haynes alloy 25 as a back-up material
for the valve stem ring.

      6.3.2   Liquid Lubrication

      The Mobil synthetic hydrocarbon oil designated XRN 1301-C was
selected for use in the first single cylinder steam expander test.  This
oil was specially formulated by Mobil for use in this program.  The synthetic
hydrocarbon base was selected because of its excellent thermal stability,
temperature-viscosity characteristics and low pour point.  Appropriate
non-metallic inhibitors have been added to the base to provide the necessary
oxidation resistance, rust protection, anti-wear properties and defoatnant
characteristics.  Properties of the oil are shown in Table 6.2-1.  In
general, the XRN 1301-C oil should have approximately a 100°F advantage in
use temperature (bulk oil temperature and hot spot temperature) over the
oils based on the petroleum base stocks.  Probably the greatest disadvantage
of the XRN 1301-C oil over the natural hydrocarbon oils is its poorer
demulsibility characteristics.  However, it is believed the demulslbility
of the XRN 1301-C is satisfactory for use in the steam expander.

      One of the major reasons for not selecting the inhibited hydrofined
paraffinitic oil (2415-50-1) being developed for SES by ESSO Research and
Engineering Company for use in the program was the fact that it was not
fully developed.  Further changes were anticipated to be made in the anti-
rust inhibitor; also, a pour point depressant was to be added later to lower
the relatively high pour point of + 20°F.  Upon completion and evaluation
of the final formulation, the 2415-50-1 type oil should be reconsidered
for evaluation as the crankcase lubricant.  No existing commercial liquid
lubricant was found to satisfy the crankcase lubrication segments of the
reciprocating steam expander.
                                   162

-------
                   7.0   TEST   FACILITY


7.1  Facility Description

     A schematic representation of the steam expander test facility is
shown in Figure 7.1-1.

     The steam generator was an electrically heated once-through
boiler-superheater which was designed to provide 1000 psi, 1000°F,
superheated steam at a rate of 400 Ib/hr.  Three hundred and eight (308)
feet of 3/8" OD x .031" wall, Type 316 stainless steel tubing was
coiled on a two (2) ft. diameter with a pitch of I'1 to form the once-
through boiler and superheater section.  Calculated pressure drop
through the steam generator at rated output was 80 psi.  The coil
terminates in a small vapor drum which discharged steam through a 1"
schedule 40 pipe to a throttle valve and then to the expander.  The
steam generator was equipped with a pressure relief valve, and was code
stamped in accordance with the ASME Power Boiler Code.  Design pressure
was 1100 psi, and design temperature was 1200°F.  Two separate three
phase, saturable core reactors (180 KWe) controlled power to the steam
generator coil.  Automatic and manual modes of power control were
provided.

     A by-pass control valve and desuperheater permitted checkout of
the test facility prior to installation of the test expander.  The by-
pass valve and expander throttle valve also permitted inlet steam flow
control to the expander with a constant boiler heating rate.  This form
of control was used during startup and shutdown of the expander.

     A General Electric Company Model TLC-65/50 HP, 2500/750 RPM D.C.
type dynamometer was used to load (up to 65 HP) the steam expander.  It
was capable of motoring (operating as a DC motor) up to 50 HP.  Speed
regulation of ± 1% of rated speed was maintained during both modes of
                                 163

-------
                                                                                                    STEAM EXPANDER
                                                                                                     TEST STAND
                                                                                        ^.atil-a
                                                                                   TEMP
                                                                                     BECMW»TEST ENGINE
  I            -""• 'i         — i«
i.4	^	i	n-yxyyQ
I l         :       i11!    «1-»|-:^:
             •        MINI i
             ?]  _*""i|KK>
             V  r TT'U«Bff*»=
                  .Ai-TT--'
                   Tin*    mr m.',.*!,—.
                                         ,o.ors,.  V  ^P^OPS, TyC-lT''"^  ,, PUMP*   ,f
                                                                                                     RESIN OUTLET
                                                                                          DEMINERALIZEO
                                                                                          H2O STORAGE  TANK
                                                                                    ^-PACKAGED STEAM SYSTEM
           Figure 7.1-1.  Steam  Expander  Test Facility Schematic  (GE  Dwg.  7O7E676).

-------
operation.  The dynamometer was capable of delivering up to 500 ft.lbs
of torque, and was used to rotate the expander during checkout of the
expander and during startup on steam operation.

     The steam expander was coupled to the dynamometer through a flexible
Spicer shaft with universal joints at each end and a splined telescoping
center section.  A shear pin adapter at the dynamometer shaft connection
limited inertial torque loading of the expander by the massive dyna-
mometer in the event of "freeze-up" or other failure  which might
suddenly stop the expander.  The dynamometer was mounted on a separate
foundation from the expander, and was isolated from the floor of the
test cell for minimum vibration.  An electronic (strain gage type load
cell) system was used for torque measurement.

     The expander exhaust was piped to a water cooled condenser which
was capable of 400 Ib/hr throughput of steam, and which could maintain
20 psia discharge pressure.  Steam discharge from the desuperheater
also entered the condenser during by-pass operation.

     A water storage tank equipped with a liquid level indicator,  fill-
ing system and oil removal system received the condensate from the con-
denser.  Demineralized makeup water was supplied to the tank directly
from the demineralizer as required during operation.   Open loop opera-
tion could be conducted by discarding the condensate and by supplying
demineralized water for boiler feed.  Most of the expander testing was
done in that manner.

     A plunger type motor driven pump with automatic stroke adjustment
supplied the boiler with water.  A small heat exchanger ahead of the
pump provided cooling to prevent pump cavitation.   The pump was capable
of providing up to 1.0 gpm at 1200 psi to the boiler.  A built-in pres-
sure relief valve prevented an overpressure of the pump.  Pump discharge
pressure and flowrate were controlled automatically by sensing boiler
discharge pressure.  Steam flowrate was determined by measuring boiler
feed water flow.  A turbine type meter was provided for water flow
measurement.
                                  165

-------
     A control console located in the control room contained all the
operating controls and instrumentation for both the facility and steam
expander.  All critical parameters were recorded.  Protective circuits
for overspeed, overtemperature, and overpressure were provided.  Tem-
perature, pressure or expander speed above set limits would result in
an automatic shutdown of the facility.

     Water and steam samples were taken from the test loop for analysis
at several locations.  An oil/steam sampling valve was provided at the
outlet of the expander.  An oil/water removal valve was provided near
the top of the pump head tank, and a water sampling valve was provided
near the inlet to the boiler.  All of these valves provided different
methods of removing fluid samples to be analyzed for oil or other
contaminants.
                             2
     Figure 7.1-2 shows the I R boiler with a portion of its outer
insulated casing removed.  Figure 7.1-3 shows the high pressure plunger
type water pump in the foreground with condenser and water supply tank
toward the rear.

     Figures 7.1-4 and 7.1-5 show the crosshead expander installed and
instrumented ready for testing.

     Considerable difficulty was experienced with the Spicer shaft and
couplings (reference Figure 7.1-4) between the expander and dynamometer.
Several shaft failures were encountered during initial testing of the
crosshead expander.

     An analytical study of the expander-dynamometer rotating system
indicated that the peak torque (for non-resonant condition) on the drive
shaft was 219 ft/lb.  When assuming that the stiffness or torsional
spring constant of the "Spicer" shaft was approximately that of a rigid
tube, the calculated natural frequency was above 2800 RPM.  However,
the exact stiffness of the shaft which included two universal joints
was not known, and apparently the stiffness was much lower than the
value assumed.
                                  166

-------
                                              I R ELECTRIC BOILER
                                              AND SUPERHEATER
                    PUMP WATER   ,
                    SUPPLY TANK !
Figure  7.1-2.  Single Cylinder  Expander Test  Facility  (P72-2-5F)
                               167

-------
Figure 7.1-3.  Single Cylinder Expander Test Facility (72-2-5G)
                              168

-------
o
-
                                   Figure  7.1-4.   Installation  of  Crosshead  Expander  (P72-4-4M)

-------
                                             Cooler
Figure 7.1-5. Installation  of  Crosshead Expander (P72-4-4N).
                            170

-------
     Numerous discussions were held with engineering personnel at the
Dana Corporation, Waukesha Motor Company, T.B.Woods Company, and others
in an attempt to resolve the shaft breakage problem.  It was concluded
that the rotating mass system had a critical torsional speed within the
0 - 2500 RPM operating range.  As a result, a new shaft was designed
which included provisions for torsional damping and reduction of shock
loading.  Also, instrumentation was installed to detect torsional
critical speeds.  Calculations indicated the torsional natural frequency
of the "resilient" shaft to be 610 RPM.  Newly installed instrumentation
indicated the natural frequency to be in the vicinity of 700 RPM, and
therefore operation between 600 and 800 RPM was avoided.  Operation at
500 RPM with high inlet steam pressure resulted in excessive temperature
rise of the shaft.  This was due to oscillations caused by cyclic torque
of the expander as it went from power to recompression during each
revolution.  Therefore, only test points of low cylinder pressure were
obtained at 500 RPM.

7.2  Expander Instrumentation

     Each expander was well instrumented with 33 thermocouples,  4 pres-
sure transducers, 2 vibration pickups, 2 speed pickups, 2 steam flow
sensors, a shaft torque meter, cam push rod force sensor, plus other
miscellaneous sensors.

     Measurement of steam pressure in the expander cylinder as a func-
tion of crank angle was a challenging task.  It was generally concluded
that the difficulties which were encountered with the cylinder pressure
transducer were related to overtemperature or poor temperature distribu-
tion throughout the transducer.

     Three different types of pressure transducers were tried for
cylinder pressure measurement.  All three transducers required water
cooling.  Two of the transducers were strain gage types, and one utilized
a piezoelectric crystal.  All of the transducers had a temperature limit
in the range of 300 - 400°F.  If the diaphragm of the transducer
exceeded'its temperature limit,  transducer output response became
erratic and sensitivity decreased rapidly.  Temperature of the diaphragm
                                 171

-------
for the strain gage transducers was determined from gage resistance
measurements during testing.  Calibration curves; were then used to
determine gage sensitivity.  Zero shift with temperature change was
significant.  Therefore, only peak pressure measurements were con-
sidered to be reliable.  To help reduce the diaphragm temperature, a
small amount of argon gas was continuously injected into the vicinity
of the gage diaphragm.  The argon, a noncondensable gas, reduced the
condensation rate of steam on the water cooled diaphragm - thus reducing
the temperature of the diaphragm.  All cylinder pressure data were
obtained by using this technique.  The piezoelectric transducer was
very temperature sensitive and no valid pressure measurements were
obtained with this unit.  The best cylinder pressure data was obtained
with a Dynisco Model PT49A transducer.  This unit contained water
cooling passages in the diaphragm.
                                  172

-------
                     8.0  TEST  RESULTS
           CROSSHEAD  PISTON  EXPANDER

8.1  Component Performance(s)
     The crosshead piston expander was first rotated by motoring with the
dynamometer on March 30, 1972.  The expander was first operated on steam
on April 26.  Problems arose due to failure of the expander-dynamometer
coupling shaft because of shaft torsional vibratory fatigue.  A temporary
fix allowed the first test point data to be taken on April 27-28, May 2-3,
and May 10-11 while awaiting the delivery of a torsionally-damped shaft.

     Performance testing of the crosshead expander with graphite rings and
with the inlet valve set to open 10° BTDC was completed on June 8.  The
total operating time of the expander on that date was 87.2 hours.  An addi-
tional 152.3-hour endurance test at 1000 psia, 1000°F, 1500 RPM was com-
pleted on June 16.  The total operating time with the original graphite
rings was 241.9 hours.

     Visual examination and dimensional measurements of the expander com-
ponents were made at various times during the 87.2 hours of performance and
152.3 hours of endurance testing of the first engine build-up with carbon
graphite piston rings.  Major inspection of component parts occurred after
motoring checkout and after 7.5, 31.3, 52.4, 152.6 and 241.9 hours of test
under steam.  Following is a summary of the findings of the post test in-
spections performed on crosshead expanders.

     8.1.1  Camshaft/Valve Lifter (Tappet)

     The AISI 8620 camshaft exhibited light pitting on the carburized lobe
adjacent to the contact area.  The noncritical areas of the chill-cast
Cr-Mo cast iron tappet also showed evidence of pitting which appears to be
caused by a form of pitting corrosion.  However, the cam lobe/tappet con-
tact surfaces were highly polished and no evidence of pitting or other
forms of damage were observed.
                                   173

-------
     Early'in the test program during the checkout of the expander,
severe damage to the tappet/cam lobe surfaces was observed after only 2
hours of motoring without steam.  In this case the materials combination
was a cobalt base cast Stellite tappet against a carburized AISI 8620 cam
lobe.  However, inadvertently a cast Stellite 6B alloy tappet with a hard-
ness of Re 38-39 rather than the intended Stellite Star J alloy with a
hardness of Re 60-61 was used.  The failed Stellite 6B tappet had the
classical appearance of spalling (pitting), i.e., irregular shaped holes
apparently caused by rupture of the metal below the surface, and scuffing
due to a galling action between the tappet and cam surfaces.  Visual
examination of the carburized AISI 8620 steel cam surface revealed a build-
up of metal on the nose and on each side of the lobe of the cam surface.
The metal transfer to the cam surface is believed to be caused by a
galling action between the tappet and cam surface as a result of a com-
bination of a high tendency for adhesion between the two surfaces and a low
bulk fracture strength of the Stellite 6B.

     The chill-cast Cr-Mo cast iron tappet with a steam tempered black
oxide (Fe-0,) antichafing coating was selected for the second engine build-
up in preparation for performance testing.  The chill-cast tappet was*
selected on the basis that experience  has shown that the chill-cast iron
tappet generally is superior to hardenable cast iron against hardened steel
             (53)
cam surfaces,      The major disadvantage of chill-cast cast iron tappets
is its greater tendency to failure by spalling (pitting) especially in the
presence of certain EP additives in the oil^  »  »  '   .   However, since
the tappet material was changed to the chill-cast Cr-Mo cast iron and an
auxiliary oil pump was installed to insure pre-startup lubrication no
further cam lobe/tappet damage has occurred other than the slight pitting
corrosion of non-critical areas.

     8.1.2  Inlet Steam Valve/Housing

     The H-ll alloy inlet steam valve stem assembly required a light force
in order to remove it from the cast Ni-Resist Type D-3 alloy housing.  The
cold worked (Re 49) Haynes alloy No. 25 upper stem compression seal rings
had worn a shallow groove (^0.002") in the ID surface of the housing; no
significant wear of the housing was observed at the location of the bottom
                                  174

-------
seal rings.  Possibly the application of a wear-in coating on the Ni-
Resist housing (or rings) would reduce the observed wear by providing
lubrication until the Ni-Resist surface has a chance to work harden and
achieve a harder and more wear resistant surface.  Evidence of some
pitting also was observed on all components of the housing.  However, in
spite of the slight pitting and wear in the housing, the steam leakage
                                             \
past the inlet valve stem seal was on the order of only 0.5% of the total
steam flow at steam conditions of 1000°F/1000 psi, Figure 8.1-1.

     Other components of the inlet steam valve assembly were in excellent
condition.  The nitrided spherical end of the H-ll alloy push rod and the
spherical Stellite 6B seat were unpitted and highly polished.  Both the
Stellite 1 and Stellite 6 alloy valve seat and valve facing respectively
were in good condition and were reuseable.  Although the low alloy steel
inlet valve spring relaxed approximately 1/8" in the first 52.4 hours of
testing, no further problems were encountered with spring relaxation
after the addition of oil jets to the housing to improve cooling of the
spring.

     LPA 101 plasma spray-coated 17-7 PH alloy compression seal rings were
incorporated in the inlet steam valve assembly for the first 52.4 hours of
engine testing.  Difficulties were encountered with the LPA 101 coating
during fabrication of the rings and during the period of testing due to
frequent chipping of the coating.  An increasing rate of steam leakage
through the inlet valve stem (Figure 8.1-1) was observed over the initial
52.4 hours of testing as a result of continued chipping of the coating.
Upon replacing the LPA 101 coated 17-7 PH rings with the cold worked
Haynes Alloy No.  25 rings, no further increase in the leakage rate occurred
and after a period of time the leakage rate decreased to the ^0.5% value.
The LPA type material (laves phase in cobalt matrix) may have potential
as a ring material but not as a coating.   Possibly a cast ring of solid
LPA alloy may warrant evaluation.

     8.1.3  Recompression Valve ,

     All the materials of construction of the recompression valve assembly
are identical to the materials used for the fabrication of the inlet steam
valve with the exception of the spring.  After (100 hours) of operation

                                   175

-------
    2400 -
    2000 -
    1600 -
ti

X^



- §§§


(400-700 psi
,\^j < 700-900 °F
v^v^ (500-2000 RPM
\(400 psi
WOO°F
(500-20QO RPM ,
0 25 50
1
1
A
\
v '









Haynes-25
1000 psi
1000 °F
1000-2000 RPM





^§$
^"X-^^V. ^X^^^V^


1
75
I i i 1 1 '
Ring
/



l\s\ss^s
{400-700 psi
700- 1000 °F
500-2000 RPM



i I l i i i
100 125 150 175 200 225
.





-
_
— -


25
1.2
1.0
0.8


0.6


0.4


0.2

0
                                                                                                                            0)
                                                                                                                            •U
                                                Expander* Operating Time, Hours
                                     Figure 8.1-1  Inlet Valve Stem Steam Leakage Rate

-------
the low alloy steel recompression valve spring relaxed nearly 0.100"  and
was replaced with a spring fabricated from Inconel X-750.  Although even
the Inconel X-750 recompression valve spring had relaxed 0.040" at the
end of the final 141.9 hours of test, it still was reuseable.

     All other components of the recompression valve were in excellent
condition.

     8.1.4  Piston Rings/Cylinder Liner

     At the end of the 241.9 hours of testing, the Sb impregnated carbon-
graphite power piston rings (CC-5A material) exhibited severe wear and
damage.  Each ring segment was minus at least one lap joint and approxi-
mately one half of each ring segment located in the bottom ring groove
was missing.  The pressure balancing circumferential grooves had all but
disappeared from the rubbing surfaces of those segments in the upper ring
groove.  From probe measurements of the ring groove depth with the rings
installed in the expander and actual micrometer measurements at the time the
expander was disassembled after 52.4 hours and at the completion of the test-
ing of 241.9 hours, it appears that the wear of the carbon-graphite rings
is linear out to about 153 hours.   Beyond 153 hours, the wear rate of the
top ring increased and the wear rate of the middle and lower rings appeared
to decrease.  The wear data are shown in Table 8.1-1 and Figure 8.1-2.
Photographs of the rings after 52.4 and 241.9 hours on test are presented
in Figures 8.1-3 and 8.1-4.

     The Inconel X-750 ring spring of the power piston bottom ring groove
exhibited substantial wear at the end joint.   This ring spring has shifted
radially outward due to the missing graphite segments and had scored the
I.D. of the Type 440C SS cylinder liner, particularly near the exhaust
ports.  Visual examination of the cylinder liner after 152.6 hours of test-
ing revealed no scratches or damage of any kind.

     8.1.5  Power Piston Head

     The Cr-Mo-V alloy power piston exhibited little, if any, wear except
for light scratches in the lower ring groove surfaces caused by the Inconel
X-750 ring spring.
                                   177

-------
                                                    Table 8.1-1
00
                                                 PISTON RING WEAR
                                            (Carbon-Graphite Grade CC5A)
Engine Hours
On Steam
0
7.5
31.3
52.4
152.6


241.9
• Ring Groove »
Depth, In.w
0.071
-------
0.100
0.080
0.060
0.040
0.020
 87.2 Hours
 Performance
    Tests
400-1000 psi
700°-1000°F
500-2000 RPM
                                                152.3  Hours
                                              Endurance  Tests
                                        1000°F,  1000 psi,  1500 RPM
                                                                             6 Top
                                                               X
                                                              __ () Middle

                                                              — — Q Lower
                                                                 Code
                                                                     Micrometer

                                                                     Probe
              O'
                                 100
                                                     200
243
                                                TIME,  HOURS
                       Figure 8.1-2  Piston Ring Wear at Middle of Segments
                                     (Carbon Graphite Grade CC5A).
300

-------
                                                       SEGMENTED PRESSURE BALANCED
                                                       GRAPHITE RING

Figure 8.1-3.   Crosshead Expander Carbon-Graphite Rings (CC-5A)
                after 52.4 hours  on test (P72-5-4A)
                                 180

-------

                                  .  .,..-.,
                       1
Figure 8.1-4.
Crosshead Expander Carbon-Graphite  Rings  (CC-5A)
after 241.9 hours on Test  (P72-6-3L)
                               181

-------
     8.1.6  Power Piston Rod Seal

     The power piston push rod chevron seal was in excellent condition.
The seal was fabricated from 15% bronze + 5% MoS- filled PTFE and in-
spection of the seal after completion of the 241.9 hours of test revealed
no measurable wear.  The mating surface was a hard chromium plated H-ll
steel.

     During the initial stages of checkout testing without steam, traces of
oil entered the expander steam chamber through piston rod seals.  It was
concluded that the oil was bypassing the seal by a hydrodynamic pumping
action.  By modifying the mounting of the chevron seal to take advantage
of the hydrodynamic action of the seal, the problem was solved.  The chevron
seal was installed with a floating housing.  An additional wave spring was
added to the seal assembly and all wave springs were placed on the oil side
of the seal.  With this modification to the seal, oil leakage into the steam
expander condensate was maintained at less than 4 parts per million as shown
by Figure 8.1-5.

     Leakage of water past the piston rod seal into the crankcase also was
monitored at intervals during the performance and endurance testing.  Total
water leakage was determined by 1) analyzing the % water content in the oil
(emulsion), 2) measuring the quantity of water (demulsified, condensed
steam) in ml that accumulated in the lower section of the crankcase, and 3)
measuring the quantity of water that' boiled off through the crankcase vent.
These data are given in Table 8.1-II.  Examination of the data show that
relatively little water has leaked into the crankcase.  The water content
of the oil after the final 89.3 hours endurance test analyzed only 0.16%;
however, even the VL.0% water content in the oil after the first 63.5 hours
of endurance testing would not be expected to be detrimental to the lubri-<
                                                                          ;
eating properties of the oil.

     8.1.7  Crosshead Piston

     Both the aluminum alloy crosshead piston and the mating nitrided
Type 304SS lower cylinder wall exhibited little, if any, wear.  Only very
shallow pitting was observed on the lower position of the nitrided ID sur-
face of the cylinder.

                                  182

-------
             12
oo
          C
         5
4J
g
O
I
         •H
         O
             10
              0
                             Seal Failure
                    700-1000°F
                    400-1000  psi
                    500-2000  RPM
                                                        1000°F, 1000 psi
                                                        1500 RPM
                         25
 Modified
 Seal
,Instailed
                         50
       75       100       125      150

          Expander Running Time, Hours
175
                                                                                            200
225
250
                             Figure 8.1-5   Oil Concentration in Steam Condensate .

-------
                              Table 8.1-II
                      Water Leakage into Crankcase
   Total                            Water Drained from      Vapor
Engine Hours     .  Water Content    Crankcase before      Boil-Off
  on Steam         Crankcase Oil,   Analytical Sample   Crankcase Vent
	       	%	   	ml	       ml/hr	

     0                 0.0008 (8ppm)
  12.9                 0.028              -                   nil
  31.3(a)              0.60               5                   4.5
 152.6(b)              1.00 (1 ml)      113                   6.3
 241.9(c)              0.16              50                   1.1
   Piston rod seal failure; crankcase oil drained 2050g oil, 12.3g water.
   63.5 hours on endurance test, all water drained prior to start of
   63.5-hr run.
 (c)
   89.3 hours on endurance run, all water drained prior to start of 89.3-hr run.
                                    184

-------
     8.1.8  Wrist Pin Bushing

     The diametral clearance of the wrist pin/bushing assembly  increased
approximately 0.004".  The high load areas of the wrist pin exhibited
burnished markings but to no discernable depth.

     8.1.9  Other Components

     All other components of the expander were in excellent condition and
reuseable for further testing.

     8.1.10 Crankcase Lubricant

     A sample of the XRN 1301C oil which had accumulated 187.7 hours of
operation was drained from the crankcase after the completion of the 152.3-
hour endurance test and sent to Mobil Research and Development Corporation
for analysis.  The results of the analyses are presented in Table 8.1-III.
The data show relatively little change of the used oil from new oil and no
serious oxidation or degradation.  The slight viscosity increase is possibly
due to the significantly higher water content of the used oil.  The increase
in neutralization number reflects some slight oil oxidation that was con-
firmed by differential infrared analysis.  The ash content is very low and
shows expected traces of iron wear metal and possibly dirt as indicated by
the silicon content.  Overall, the XRN-1301C oil performed well and pro-
vided satisfactory lubrication with little evidence of degradation over
187.7 hours of operation.

     Buildup of the crosshead expander with Cr_C9 coated Inconel-X piston
                                              J *•
rings was completed on June 23.  The expander was run for about one hour at
400 psig and 700° Finlet steam conditions on June 23 for preliminary check-
out.  No difficulty was encountered during the checkout test, and visual
inspection of the piston rings by observation through the exhaust port re-
vealed no ring or cylinder liner damage.

     Performance testing of the crosshead expander with the Inconel-X rings
began ort June 26, with initial^performance being satisfactory.  However,
after approximately six hours, expander performance began to decay, as in-
dicated by poor cylinder recompression and increased steam consumption.
The expander was shutdown and the piston rings were inspected by viewing

                                   185

-------
                              Table 8.1-III
                 Change in Properties of XRN-1301C Oil
                 After 187.7 Hours of Engine Operation
          Property                                New          Used

Viscosity @ 100 cs                               75.45         80.20
          @ 210 cs                               11.07         11.60
Viscosity Index                                  147           147
Neutralization Number                            0.06          0.5
Water Content, ppm                               38            431
Ash, wt %                                        0.001         0.006
Metal Contents, ppm by Emission Spectrograph
    Al                                            -(a)    --"- 0
    Cr                                                         0
    Cu                                                         0
    Fe                                            -            78(b)
    Mg                                                         0
    Ba                                                         0
    Si                                                         4

Differential infrared analysis of new oil vs. used oil shows loss of about
10 - 20% of antioxidant and slight oil oxidation.
   Insufficient ash for determination
   Computer extrapolation value might be slightly high
                                 186

-------
through the exhaust port.  Extensive ring wear and cylinder wall  rough-
ness were observed.  Upon disassembly, the Cr_C2 coated Inconel-X rings
were found to be worn approximately 60 mils which included the  3  to  6 mil
Cr-Cj outer coating.  The hardened Type 440-C liner contained a number  of
axial grooves in the order of 1 to 3 mils deep, and there were  also  axial
streaks of metal buildup on the cylinder wall 2 to 3 mils thick.   The
stepcut tabs on all rings were broken off, and 20 to 30% of each  ring was
missing as shown by Figure 8.1-6.  All other expander components  were
virtually undamaged.

     Oil leakage into the steam condensate was high initially,  i.e., 55 ppm
after 3.2 hours; however the oil concentration rapidly decayed  to  25.2 ppm
after 4.2 hours and 4.0 ppm after 5.4 hours.  It is assumed that  some oil
had been trapped in the system prior to engine start-up which resulted in
the high initial oil content in the condensate.  The piston rod seal was
functioning as expected and further testing would have resulted in oil
concentration of <4.0 ppm.

8.2  Thermodynamic Performance

     8.2.1   Thermodynamic Performance - Graphite CC-5A Piston Rings

     Test results for the crosshead expander are presented in Table 8.2-1.

     Figure 8.2-1 shows the increase in brake horsepower with speed and
with increasing capacity of the steam to perform work as pressure increases.
More efficient breathing and.lower friction power losses at low speeds
cause the deviations from straight lines.  The effects of greater blowby
at lower speeds and higher pressures are also present.  A reference line
of theoretical indicated power at 1000 psia and 1000°F is shown (assumes no
blowby).  The brake data compare favorably with the prediction,  since the
brake power would be expected to be less than the indicated power due to
blowby and friction losses.

     The brake mean effective pressure is shown in Figure 8.2-2  as a
function- of speed and inlet steam conditions.  It reflects the same effects
which the power curves show.  The theoretical mean effective pressure at
1000 psia and 1000°F is shown as a dashed line.  The difference between the
                                  187

-------
Figure 8.1-6.
Crosshead Expander Cr-$C2 Coated Inconel-X Rings
After Approximately 10 Hours of Testing (P72-6-3N)
                                  188

-------
                                                  Table 8,2-1
                                         Crosshead Expander Test Data
                                            May 10 - June 7. 1972
H
00
Test
Point
101E
101G
10 2E
102G
102H
103H
104H
105E'
105H
106E
106H
10 7H
108H
11 7H
119H
122H
123H
124H
127F
12 7H
128H
129G
129H
129HA
Brake
Torque
Ib-in
636
509
530
551
509
403
288
636
594
1356
1272
1060
933
594
254
1102
933
721
1590
1823
1442
1102
1208
1187
BMEP
psi
115
92
95
99
92
73
52
115
107
244
229 -
191
168
107
46
199
168
130
286
328
260
199
218
214
Speed
rpm
475
500
850
1000
1000
1500
1995
546
500
590
500
1008
1495
1001
1992
999
1500
1995
1000
1001
1500
2000
2000
2000
Brake
H.P.
4.79
4.04
7.15
9.15
8.08
9.59
9.13
5.51
4.71
12.69
10.09
16.95
22.12
9.43
8.04
17.47
22.20
22.82
25.23
28.95
34.31
34.97
38.35
37.67
Cond.
Press.
psia
21.0
19.8
22.0
20.4
19.5
19.5
19.5
22.0
19.5
21.6
19.8
19.8
20.0
19.7
19.6
19.6
19.9
20.0
23.0
19.5
19.5
23.7
20.0
19.8
Inlet
Press.
psig
380
403
400
435
408
405
412
400
405
700
700
700
698
440
400
705
700
690
912
1020
990
950
990
1000
Inlet
Temp.
°F
604
700
640
700
694
700
706
634
694
682
712
688
688
1012
988
988
988
1000
1020
1010
1008
940
1035
1040
Steam
Rate
Ib/hr
92
110
102
-
150
170
167
74
91
176
175
280
340
105
127
187
235
282
415
340
402
415
430
432
Specific Operating
Rate Time
Ib/hp-hr hr
19.3
27.2
14.3
-
18.6
17.7
18.3
13.3
19.4
13.9
17.3
16.5
15.4
11.1
15.9
10.7
10.6
12.4
16.4
11.7
11.7
11.9
11.2
11.5 87.2

-------
    60
    50
M
V

g
(X
QJ
to
0)
    20
    10
70° valve event
10° valve lead


Graphite CC-5A Piston Rings
                                                                in
        in
     0
                  500
                                                            1000 psia, 1000°F
                                                            Theoretical IHP
                                                      1015, 1025°F~
                                                  BHP  945, 980°FA
                                                                 715, 990°F O
                                                                 715, 690°F D
                                                 BHP
435, 1000 °FO
420, 675°F O
                   1000        1500       2000

                       Rotational Speed - RPM
 2500
3000
                    Figure 8.2-1. Crosshead Expander Performance
                                          190

-------
    400
co
0.
0)
t-1
en
co
    350
    300
    250
    200
o
tu
14-1
w   150
    100
    50
                      70° valve event
                      10° valve lead

                Graphite CC-5A Piston Rings
                  500
                                    in
        T.
         in
                                 1000 psia, 1000°F
                                 Calculated IMEP

                                  Measured IMEP
                                  1015,  1025°F "  '
                                     1015, 1025°F
                                                                                100
                                                            BMEP
                                                                  945,  980°F
                                                A
                                                            BMEP
                                     715,  990°F
                                     715,  690°FQ_
                               BMEP
435, 1000°FO-
420, 675°F O
                                                                                175
                                                                                150
                                                                                125
                                                    75
                                                                                 50
                                                                                 25
1000         1500        2000
   Rotational Speed - RPM
 2500
3000
                                                        S
                                                        cr
                    Figure 8.2-2. Crosshead Expander Performance
                                         191

-------
measured and theoretical IMEP is chargeable to the theoretical assumption
of no blowby.  The difference amounts to about 6 hp or 13 percent.

     The difference between the IMEP and BMEP curves at 1015 psia and
1025°F represent frictional power losses of 3.9 hp at 1500 rpm and 4.7 hp
at 2000 rpm.  These losses are almost exactly what is expected from two-
stroke diesel engines operating at the sam
mechanical efficiencies are 0.90 and 0.89.
stroke diesel engines operating at the same IMEP    .  The corresponding
     The specific steam consumption is shown in Figure 8.2-3 as a function
of speed and inlet conditions.  The influence of speed is strongest at
500 rpm,but the SSC is practically invariant at the higher speeds.  A
strong influence of the steam enthalpy is also seen.  The influence of
other parameters upon the brake SSC curves is not apparent.  Friction and
blowby would contribute to the difference between the 1005 psia, 1005°F
curve and the theoretical indicated SSC.

     The engine efficiency as a function of speed and inlet conditions is
shown in Figure 8.2-4.   The definition of engine efficiency as shown by
the figure is:

                              o Q°Ut = BHP
                           ne   Q±n    wAh

The theoretical indicated efficiency is shown for reference.

     One cylinder pressure-volume plot is shown by Figure 8.2-5.   Since
this curve is drawn from a small oscilloscope photograph, such as shown
by Figure 8.2-6,  absolute accuracy is not expected, but the correlation
between predicted and actual pressures appear good.  The recompression
curves match very closely until shortly before TDC.  The match is also
                    3
quite good from 7 in  (approximate point of inlet valve closure) through
expansion.  The variation through this interval is probably primarily due
to blowby.

     Following completion of approximately 87 hours of performance testing,
the crosshead expander was held at 1500 rpm with 1000°F and 1000 psia
steam inlet conditions for an additional 150 hours.  Figure 8.2-7
shows the decay of shaft horsepower, and Figure 8.2-8  shows a steady

                                   192

-------
t
o
CO


I
u
•H
14-,
•H
O
0)
o.
en
      30
      25.
      20
      15
      10
       0
        0
                70° valve event
                10° valve lead


                Graphite CC-5A Piston Rings
                                                                  p.
                                                                    in
                                            T.
                                             in  -
                   D
                                           u
                                                            BSSC  420  psia,  674°]



                                                            -BSSC  715,  690°F
                                                             •BSSC 1005,  1005 °F
                  	1000, 1000°F

                                                               Theoretical ISSC
                   500
1000        1500        2000

  Rotational Speed - RPM
                                                                  2500
                                                                              3000
                    Figure 8.2-3. Crosshead Expander Performance
                                      193

-------
   1.0
   0.9
    0.8
u




I   0-7

•H
(4-1
VU

W
01
c
    0.6
    0.5
    0.4
    0.3
       0
70° valve event

10° valve lead


Graphite CC-5A Piston Rings
  500
1000       1500        2000


  Rotational Speed - RPM
                                              Theoretical

                                              Indicated

                                              Efficiency

                                              P.  = 1000 psia
                                                               ,
                                                               in
                                                                  = 1000° F
                                                                Brake

                                                                Efficiency
                                                 .,
                                                 In

                                                 ,
                                                 in
                                        1015 psi«


                                      = 1025°F  "
2500
3000
                    Figure 8.2-4. Crosshead Expander Performance
                                         194

-------
VO
in
            1200
            1000 -
                                                                     2000 BPM

                                                                     70° valve event
                                                                     10° valve lead

                                                            Graphite CC-5A Piston Rings
CO
p.
I
a
         (0
         CO
5
iH
£
             800
             600 -
                                                 1000 psia, 1000°F
                                                          Measured
                                                          1015 psia, 1040°F
             400 -
             200 -
                                                                            'Dynisco  Model PT 49A
                                                                            pressure transducer
                                                   15          20

                                                      Cylinder Volume
                    Figure 8.2-5. Crosshead Expander Typical P-V Diagram

-------
c
-
                                               INL! '  VALVE OPENING
                                EXHAUST PORT UNCOVERED
                                                                                      INLET VALVE PUSH ROD FORCE
                                                                                      EXPANDER SPEED - 2000 RPM
                                                                                      INLET PRESS. - 1000 PSIG
                                                                                      INLET TEMP. - 1000°F
                                                                                      VALVE EVENT - 70°
                                                                                      VALVE OPEN5 - 10° BTDC
                                                                                      FORCE - 215 LB/CM
CYLINDER PRESSURE

EXPANDER SPEED - 2000 RPM
INLET PRESS. - 1000 PSIG
INLET TEMP. - 1000°F
PRESSURE - 155 PSI/CM
                                  CRANK ANGLE                360C

                                              Figure 8.2-6.  Oscilloscope Photos

-------
M
CO
M
O
EC
CO
    60
    50
    40
    30
    20
    10
      0
               Grade CC-5A Graphite Rings
50
100
150
                                                     200
                                                250
                         Accumulated Test Hours

                         150 Hour Endurance Test




      Figure  8.2-7.   Single Cylinder  Crosshead Expander - 1500 RPM
                                  197

-------
,n
iH


 I



 §
•H
CO
c
o
cj
a)
-p
CO

u
•H
u
(U
o.
co
    30
    25
    20
    15
     0
       0
             Grade CC-5A Graphite  Rings
                   50
100
150
200
250
                          Accumulated Test Hours


                          150 Hour  Endurance Test





       Figure 8.2-8.  Single Cylinder Crosshead Expander - 1500 RPM
                                   198

-------
increase in SSC with time.  Both figures reflect an increase  in  steam
blowby as ring wear progressed.  Figure 8.2-9  indicates a decline of
engine efficiency, but there is a strong indication that the  decline  is
asymptotic.   This would be expected since the expander would continue to
produce power with no rings.

     All test data shown by Figures 8.2-1 through 8.2-9  were taken from
the single cylinder crosshead expander which contained antimony  impregnated
graphite (grade CC-5A) piston rings and a Type 440C stainless steel
cylinder liner.  Figure 8.2-10 shows the measured temperature distribution
for the crosshead expander.

     8.2.2   Thermodynamic Performance - Cr-C» Coated Inconel X-750
             Piston Rings	f	

     The most meaningful comparison of the performance of the crosshead
expander with graphite and with Cr,C~ coated Inconel X-750 power piston
rings is the specific steam consumption (SSC) of each assembly.   Figure 8.2-11
shows that the Cr.jC2 coated rings operating at an average inlet  condition
of 396 psia, 687°F had a SSC which was 17 percent higher at 500  rpm to
4 percent higher at 2000 rpm than that of the graphite rings operating at
an average inlet condition of 420 psia, 675°F.  The Cr~C2 coated rings
operating at 695 psia, 702°F had a SSC which was approximately 16 percent
higher than that of the graphite rings operating at 715 psia,  690°F
from 1000 - 2000 rpm.

     The greater SSC of the Cr-C2 coated rings could be caused by increased
steam blowby past the rings and/or by greater friction power losses.   The
brake horsepower (BHP) and brake mean effective pressure (BMEP)  were
greater for the Cr,^ coated rings than they were for the graphite rings,
as shown by Figures 8.2-12 and 8.2-13.

     The greater SSC of the C^oCo coated rings despite higher BHP and
BMEP appears to be due to higher ring friction as indicated by Figures
8.2-14 and 8.2-15 which compare pressure-volume for similar test conditions
for the two ring types.  These .figures fail to show any strong evidence of
greater blowby for the Cr-C^ coated rings since the expansion and recompression
portions of the diagrams are very similar to those for the graphite rings.
                                   199

-------
   1.0
   0.9
Grade CC-5A Graphite Rings
   0.8
§  0.7
W
0)
•H  0.6
oo
   0.5
   0.4
   0.3
      0
       50           100          150
             Accumulated Test  Hours
             150 Hour  Endurance  Test
200
250
       Figure  8.2-9.   Single Cylinder Crosshead Expander - 1500 RPM
                                   200

-------
                                  -Ii50	
STEAM INLET  TEMP   - 10**0°F
STEAM INLET  PRESS  - 1000 PSI
EXPANDER SPEED     - 2000 RPM
STEAM EXHAUST-
                               RECQMPHESSION
                              VALVE VENT LINE-1
                                                                STEAM INLET
           RECOMPRESSION
            RELIEF VALVE
                                                       	STEAM INLET VALVE
        CYLINDER LINER WITH
         EXHAUST PORTS
                OIL JET

              ROD SEAL-OIL

                CROSSHEA
                WRIST PIN AND
                BEARING INSERT  ,
                                                              L-—CRANKCASE



                                                OIL TEMP - 238°F
                                                                         CAM-TAPPET
                                                                           OIL JET
                                                                  CRANKSHAFT
                                                                  1625STROKE
   Figure 8.2-10.  Measured  Temperature Distribution of Crosshead Expander
                                       201

-------
M
JS
PQ
.0
i-t
4J
(X
(0
g
u
 0)
 4J
 w
 o
 0)
    30
    25
     20
15
10
      0
       0
                                                    Cr3C2/Inconel X Rings
                              396 psia, 687°F =•_
                              695 psia, 702°F =f
                              Graphite Rings
                              420, 675°F
                              715, 690°F
              500
1000        1500        2000
   Rotational Speed - RPM
2500
3000
      Figure 8.2-ll.< Comparison  of Crosshead Expander Specific Steam Consumption
                     with Graphite and with Cr.C- Coated Inconel-X Piston Rings
                                         202

-------
     60
0)
I
fr
(0
n
o
PC

I
pq
     50
     40
30
20
     10
                                Cr3C2/Inconel X R
                                                     — —   695 psia, 702
                                                           Graphite Rings
                                                           715, 690°F
                                                         Cr_C9/Inconel X
                                                   .^^v -^^""   J *•
                                                           405, 1004°F =0
                                                           396, 687°F =<

                                                    Graphite Rings

                                                      435, 1000°F
                                                      420, 675°F
                                      ngs
                                      F
                                                                               lings
       0
             500
1000
1500
2000
2500
3000
                               Rotational Speed - RPM
   Figure 8.2-12. Comparison of Crosshead Expander Power Production with
                  Graphite and with Cr3C_ Coated Inconel-X Piston Rings
                                      203

-------
    400
•H
CO
O,
    350
    300
0)
S   25°
CO
2
£   200

Q)
MH
H
 0)
    100
     50
      0
        0
                                                        Cr C /In cone 1 X Rings
                                                            695 psia, 702°F
                                                                Graphite Rings
                                                                715, 690°F
                                                         Cr-C2/Inconel X Rings

                                                        =~ __ 405, 1004°F =0
                                                        0       396, 687°F  =•
                                                                Graphite Rings
                                                                435, 1000 °F
                                                                420, 675°F
                   500
1000        1500        2000

    Rotational Speed - RPM
2500
3000
     Figure 8.2-13.  Comparison of Crosshead Expander BMEP with Graphite and
                     with Cr_C9 Coated Inconel-X Piston Rings
                            J £
                                         204

-------
600
 0
                        •Cr-CL  Rings, P,   -  690 psia,  T.   =  704°F
                           o £.          in               in


                   \     (TP 108,  6/26/72, SSC =  18.0  Ib/BHP-hr,  BMEP

                    \
191 psi, nen  = 0.466)
                                  -Graphite  Rings,  P    = 698 psia,  T   = 688°F



                                       (TP 108,  6/7/72, SSC = 15.A  Ib/BHP-hr,
                                                       BMEP  = 168 psi,  n
                                                                        eng
      0.544)
                                         Cylinder Volume - In'
Figure 8.2-14.  Comparison of Crosshe-ad Expander P-V Diagrams with Graphite and with

                Cr C_ Coated Piston Rings at 1500 RPM and Similar Steam Inlet Conditions

-------
600
500
3
D.

I
400
	p
           in
                      •Cr3C2 Rings, P±n =400 psia, Tin - 704°F



                           (TP 101A, 6/26/72, SSC = 21.8 Ib/BHP-hr,  BMEP = 115 psi, n

                                                                                        0.431)
                                -Graphite Rings, P   = 405 psia,  T   = 712°F



                                            (TP 105,  6/7/72,  SSC = 19.4 Ib/BHP-hr,


                                                                 BMEP = 107 psi, n
                                                                                  eng
                                                                                         = 0.480)
   0
                                           Cylinder Volume - IJT
Figure  8.2-15.   Comparison of Crosshead Expander P-V Diagrams  with Graphite and with

                 Cr_C0  Coated Piston Rings at 500 RPM and Similar Inlet Steam Conditions
                   3 f.

-------
The areas within the P-V diagrams are very nearly the same for each type
of ring - thus the expander indicated horsepower would be very similar.
Also measured engine efficiencies (BHP/wAh.       .  ) are lower for the
                0                         isentropic
^r3^2 coated rings.  It is therefore concluded that  the principal cause
of the increased SSC of the crosshead expander with  Cr,C_ coated rings
was due to higher ring friction.
                                  207

-------
                  9.0   TEST   RESULTS
              TRUNK   PISTON   EXPANDER

9.1   Component Performance   '

      The trunk piston expander was assembled with Cr_C. coated Inconel
X-750 piston rings and with a carbon-graphite (grade P5NR) piston rider
ring.  Oil exclusion rings in the aluminum piston skirt were made of
Type K6E and Type K (cast iron).

      A 1.5 hour motoring checkout test was conducted on July 6.  The
speed was taken to 2000 RPM for approximately 15 minutes.  The expander
performance during checkout was quite satisfactory with the exception of
the presence of more oil than anticipated in the cylinder.  However, a
condenser pressure of 20 psia during steam testing was expected to cor-
rect this.
                                                V
      Performance testing began July 7.  Seven test points (6.75 hrs.
testing time) were taken before the expander was shut down for inspection.
Borescope inspection of the piston and cylinder liner revealed minor scoring
of the liner but the compression rings appeared to be in good condition.
The oil exclusion rings and aluminum piston skirt were not visible for
inspection.  A small amount of water (< 50 ml) was drained from the crank-
case during this shutdown.  Valve stem leakage during the 6.75 hour test
period was approximately 940 ml/hr or approximately 0.5% of total steam
flow.  An analysis of the steam condensate from the condenser for oil
after 3.4 hours of testing indicated an oil concentration of 192 ppm.
The high oil concentration in the condensate probably due to residual oil
being in the system or a result of the high speed motoring checkout test.
Because of the short duration of the test it was not possible to compare
the relative oil leakage between the crosshead piston and the trunk piston.

      The expander was restarted but 2.5 hours later the recorded speed

                                  208

-------
trace indicated a drop in speed from 2000 RPM to 0 KPM in approximately
4 seconds.  An attempt to rotate the expander was unsuccessful.  The ex-
pander was removed from the test facility and dismantled.

      Examination of the expander during and after disassembly revealed
the following:

      1.  The cylinder liner was scored in the area contacted by com-
          pression rings as shown by Figure 9.1-1.  Oil was present in
          the piston ring grooves and on the cylinder wall at shutdown.

      2.  The aluminum piston skirt was badly worn and scored as shown
          in Figure 9.1-2.  It appeared that the piston skirt to liner
          clearance was insufficient at sometime during testing.  Review
          of temperature recorder charts revealed that the crankcase oil
          temperature inadvertently exceeded the maximum allowable by
          about 50°F.   The high oil temperature resulted in excessive
          thermal expansion of the aluminum skirt - thus causing inter-
          ference with the cylinder liner.

      3.  The rear main crankshaft bearing  was seized to the crankshaft.
          Particles of aluminum were present in the main bearing oil
          ports.

      4.  The oil filter appeared completely plugged with a dark slurry
          of what appeared to be carburized oil. .  Filter plugging results
          in the opening a filter bypass if the filter pressure drop
          exceeds 12 psi.   The bypass could allow oil with aluminum chips
          from the piston skirt to enter the main crankcase oil gallies
          and reach the rear main bearing - thus restricting oil flow to
          the bearing.  No damage to other  bearings, gears,  etc.  was
          found.   Emission spectrographic chemical analysis  of  the dark
          deposit that had plugged the filter shows high concentrations
          of iron, copper lead,  zinc and chromium,  plus trace amounts  of
          tin, aluminum and nickel that would be expected to come from
          wear of engine components.   A high concentration of sodium
          indicates the possibility of contamination with the treated
          water.   This combination of oil,  water,  metal fines,  plus the
                                  209

-------
Figure 9.1-1.  Trunk Piston Liner Following 9.2 Hours Test (P72-6-3Z)
                                 210

-------

Figure 9.1-2.  Trunk Piston Skirt Following 9.2 Hours Test (P72-6-3W)
                                 211

-------
Figure 9.1-3.  Cam Lobe From Trunk Piston Expander (P72-6-3U)
                             212

-------
          presence of sodium ions at high temperatures,  could  react  to
          form the grease-like deposit.  Analysis of oil from  the  crank-
          case for water content indicated a concentration of  1.3% in the
          first 500 ml sample and 0.13% in the second 500 ml sample.

      5.  The camshaft lobe, Figure 9.1-3, and cam tappet appeared worn
          or burnished over the contact areas.

      Just prior to bearing seizure, a large amount of condensate  and
oil mixture was observed coming from the crankcase vent.  Bearing  failure
and excessive cam wear may have resulted from poor lubrication:  (1) ex-
cessive oil temperature, (2) too much water in the oil,  (3) restricted
oil flow, and (4) a combination of the above three.

9.2   Thermodynamic Performance

      The brake mean effective pressure (BMEP) of the trunk expander is
compared on Figure 9.2-1 to those of the crosshead expander with both
carbon-graphite (Grade CC5A) rings and a Type 440C stainless steel liner,
and Koppers K1051 (Cr_C  coated Inconel X-750) rings and a Type 440C
stainless steel liner.  Three sets of inlet steam conditions are shown.

      The first test series (400 psia, 712°F) of the trunk expander shows
BMEP which are initially considerably lower than those of the crosshead
expander, but which later in time are only slightly lower.   A strong in-
fluence of "wear-in" time is suggested.  Both ring friction and leakage
losses are involved, but insufficient data is available to determine
which is the more influential.

      The second test series (417 psia, 998°F) shows BMEP which are some-
what lower than those of the crosshead expander but which decrease with
increasing speed as do the crosshead data.   This suggests that "wear-in"
has been completed, but that a higher friction coefficient exists.

      The third test series (690 psia, 703°F) shows its first point having
a BMEP very close to that of the crosshead expander with graphite rings,
with its final point very close to that of the crosshead expander with
Cr^C^ coated Inconel rings.  Similar conclusions relating to "wear-in"
as a function of time may be drawn from the brake horsepower versus speed
                                  213

-------
curves of Figure 9.2-2 and from the brake specific steam consumption
curves of Figures 9.2-3 through 9.2-5.

      The change of brake specific steam consumption (BSSC) with time is
more clearly shown in Figure 9.2-6 for the trunk and crosshead expanders
with Cf_C_ coated rings versus Type 440C liners.  In any pressure-tem-
perature sequence of testing, speed was varied from 500 to 2000 BFM.
Therefore, speed effect could distort the performance versus time trend.
However, it appears from Figure 9.2-6 that testing time had a stronger
influence on improving BSSC (up to the time of ring or expander failure)
than operating speed, pressure, or temperature.
                                   214

-------
    300
U>
p.
0)
w

(0
CO
0)
1-1
u
0)
w
0)
•a
V)
pq
    250
    200
    150
    100
50
      0
                                           End Test
          Trunk

      690 psia,  703°F
                                                         I
                                                              695 psia,  702°F
                                                        	Crosshead 	
                                                        715 psia,  690°F Graphite
                                                                 405 psia,  1004°F
 417 psia,  998°F
-Trunk
                                                         !C"396_£sia,  687°F
                                                                                   Crosshead
                                                                                         t
                                                                     psia,  675 F
       0
             500
                    1000        1500        2000

                       Rotational Speed - EPM
2500
3000
     Figure  9.2-1  Trunk  and  Crosshead Expander  Brake Mean Effective Pressure
                   As A Function  of Rotational Speed, Inlet Steam Conditions,
                   and  Piston Ring/Cylinder Liner Materials
                   Expander
                   Trunk


                   Crosshead
                   Crosshead
                           Piston Ring Material
                           Cr3c2~Coate<*
                           Inconel X-750
                           Carbon-Graphite Grade CC5A
                                                    Cylinder Liner Material
                                                    Type 440C stainless steel

                                                    Type 440C stainless steel
                                       215

-------
                                              J690 psia, 702°F, Trunk I
                                                         695 psia, 702 °F
                                                         -- Crosshead
                                                             psia, 690°F Graphite
                                                          LAOS psia, 100A°F_!Cr C
                                                          - 396 psia, 687°F_j  3 2
                                                           A35 psia,_10000FlCrosshead
                                                           A20 psia, 675°F  [Graphite

                                                                 ;
                                                          I 417 psia,
                                                  i
             500
         1000        1500        2000

               Rotational Speed - EPM
2500
3000
Figure 9.2-2
Trunk and Crosshead Expander Brake Horsepower As A Function
of Rotational Speed, Inlet Steam Conditions, and Piston
Ring Material
                                   216

-------
g
o>
g
o
0)
4J
CO
U
0)
P.
CO
    30
    25
    20
    10
                                                  Trunk

                                         400  psia, 712°F
                                                                   Crosshead
                                                                                 coated
                                                           j
                                    __,  396 psia,  687°F Cr^ coated

                                    —  420 psia,  675°F Graphite
0
                500
1000        1500        2000

   Rotational Speed - RPM
                                              2500
3000
    Figure 9.2-3
Trunk and Crosshead Expander Brake Specific Steam Consumption
As A Function of Rotational Speed and Piston Ring Material
for Steam Inlet Conditions of Approximately 400 psia, 700°F
                                        217

-------
    30
f   25
                                                                                i
 g
     20  -
  .
 §
 CO
 s   15
 0)
 4J
 CO
 o   10
 *rt
 «M
 •d
 0)
 £
 o)    5

 1
 PQ
                                                            i
                                                    Trunk  J

                                           417 psia,998°F Cr    coated
                                                     Crosshead

                                         405 psia, 1004°F Cr     coated
                                                            I
                                                           -I
                           	    420 psia, 1000°F Graphite
                                                                    I
                   500
          1000        1500        2000
             Rotational Speed - BPM
2500
3000
     Figure 9.2-4
Trunk and Crosshead Expander Brake Specific Steam Consumption
As A Function of Rotational Speed and Piston Ring Material
for Steam Inlet Conditions of Approximately 400 psia,  1000°F
                                         218

-------
PQ
 I


 g
•rl
 •U


 I


 1
 o
 0)

 4J

 CO
0

0)

(X
CO


0)


•3
U

pa
     30
     25
     20
     15
     10
                                          T
                                       End Test
                                                  Trunk



                                  690 psia, 703°F Cr3C2  coated

                                                  Crgsshead

                                  ~	 695 psia, /u2cr CrC   coated
                                     	 	 715 psia, 690°F Graphite
                               I
                                           I
                  500
          1000        1500        2000


             Rotational Speed - RPM
                                                                 2500
3000
   Figure 9.2-5
Trunk and Crosshead Expander Brake Specific Steam  Consumption

As A Function of Rotational Speed and Piston Ring  Material

for Steam Inlet Conditions of Approximately 700 psia,  700°F
                                      219

-------
t-l
43

§3
 C
 o
 CO
 §
 o
0)
4J
CO

O
•H
 O
 <0
 p.
CO
 rt
       70
           72
                         Test Hours  (Crosshead + Graphite)

                         74        76       78        80        82
     60
50
     40
     30
     20
     10
      0
                                                       T
                                                            T
               MOO psia
               700°F
               500-2000 RPM
        500-2000
     700 psia
     700°F
   - 1000-2000 RPM
                     ^400 psia,  1000°F
                     1000-2000 RPM


                      A

                            Crosshead
                                         •x-700 psia,  700°F
                                         1000-2000 RPM
                                                     Trunk Piston
                                       ^^*
                                      400 psia,  1000°F
                                      1000-2000  RPM
                                     6          8

                                   Test Hours
                                                  10
                                                                12
14
       Figure 9.2-6
                 Crosshead and Trunk Expander Brake Specific Steam
                 Consumption As A Function of Testing Time.  Piston
                 Ring Material was CroC2 Coated Inconel X-750 and
                 Cylinder Material was Type 440C Stainless Steel
                                        220

-------
                          REFERENCES


 1.   Rabinowcz, E., Friction and Wear of Materials, John Wiley  &  Sons,
      Inc., New York,  (1965).

 2.   MacGregor, C.W., et al., Handbook of Analytical Design for Wear,
      Plenum Press, New York, (1964).

 3.   Summers-Smith, D., "Operating Experience with Filled Piston  Rings",
      (Personal Communication).

 4.   Brooks, R.D., "Design of Reciprocating Single Cylinder Expanders
      for Rankine Cycle Engines", Quarterly Report, Jan. 4, 1972 to
      April 4, 1972, EPA Contract No. 68-01-0408.

 5.   Liston, J.E., Aircraft Engine Design, McGraw-Hill, 1942, p.  199.

 6.   Marks, L.S., Mechanical Engineers Handbook, 7th edition, McGraw-
      Hill Book Co.

 7.   Roark, R.J., Formulas for Stress and Strain, 3rd ed., McGraw-Hill
      Book Co., New York, (1954).

 8.   Rathbart, H.A., et al., Mechanical Design and Systems Handbook,
      McGraw-Hill Co., Inc., New York, (1964), pp. 6-54 and 6-67.

 9.   Brooks, R.D., "Design of Reciprocating Single Cylinder Expanders
      for Rankine Cycle Engines", Quarterly Report, Oct. 4, 1971 to
      Jan. 4, 1972, EPA Contract No. 68-01-0408.

10.   Taylor, C.F., The Internal Combustion Engine in Theory and Practice,
      M.I.T. Press, Cambridge, Mass., (1966), Vol. I, p. 354.

11.   Brooks, R.D., "Design of Reciprocating Single Cylinder Expanders
      for Rankine Cycle Engines", Monthly Progress Report No.  7, July 4
      to Aug. 4, 1972, EPA Contract No. 6^-01-0408.

12.   Vickers, P.T., et al., "The Design Features of the GM SE-101-A
      Vapor Cycle Powerplant", SAR Paper 700163 (Jan. 1970).

13.   Lyman, Taylor, (Ed.) Metals Handbook, Vol. I, Properties and Selec-
      tion of Metals, ASM, (1961).

14.   Lorentz, R.E. and Harding, W.L., "Selection of Materials for Boilers
      and Nuclear Reactors", Metals Progress, (April 1967).

15.   Steels for Elevated Temperatures, U.S. Steel Corp., (1965).

16.   Uhlig, H.H., (Ed.) Corrosion Handbook, Wiley, (1958), p. 578.

17.   Eberle, F., and Kitterman, J.H., "Scale Formation on Superheater
      Alloys Exposed to High Temperature Steam", Behavior of Superheater
      Alloys in High Temperature, High Pressure Steam, ASME, (1968).

                                  221

-------
18.  Engineering Alloys Digest, Cl-28,  (April 1960).

19.  "Properties of Cast Iron at Elevated Temperatures", ASTM STP 248.

20.  Testimony of E. Pritchard, "Automobile Steam Engine and Other External
     Combustion Engines", Joint Hearings of Committee on Commerce, U.S.
     Senate, May 28, 1968 (USGPO Ser. No. 90-82).

21.  Van Brunt, C. and Savage, R.H., "Carbon Brush Contact Films", General
     Electric Review 47, p. 16 (July 1944).

22.  Savage, R.H., "Physically and Chemically Adsorbed Films in the
     Lubrication of Graphite Sliding Contacts", Annals of the New York
     Academy of Sciences, Vol. 53, Article 4,  (June 1951).

23.  Savage, R.H. and Schaefer, D.L., "Vapor Lubrication of Graphite
     Sliding Contacts", Journal of Applied Physics, Vol. 27, No. 2,
     (Feb. 1956).

24.  Campbell, I.E., ed., High Temperature Technology, p. 110, John Wiley
     & sons, Inc., New York  (1956).

25.  McKee, D.W., "Metals Oxides as  Catalyst for the Oxidation of Graphite",
     Carbon, Vol. 8, Pergamon,  (1970).

26.  Wilson, D.S., et al., "The Development of Lubricants for High-Speed
     Rolling Contact Bearings Operating at 1200°F", TR 60-732, WADD
     (Jan. 1961).

27.  Bowers, R.C. and Murphy, C.M.,  "Status of Research on Lubricants,
     Friction and Wear", NLR Report  6466 (Jan, 19, 1967), p. 10.

28.  Summers-Smith, D., "A Review  of the Symposium on TFE Seals in
     Reciprocating Compressors", ASME Annual Meeting, New York, (Dec. 1970).

29.  Lancaster, J.K., "Solid Lubrication", ASME Conference, Denver, Aug.
     1971, Tribology, Vol. 4, No.  4.,  (Nov. 1971), p. 234.

30.  Halliwell, H., et al.,  "An Application of Self-Lubricated Composite
     Materials", presented at the  Annual ASLE Meeting, Toronto, Canada,
     May  1-4, 1967, preprint 67 AM 8A-5.

31.  Buckley, D.H. and Johnson, R.L.,  "Marked  Influence of Crystal Structure
     on Friction and Wear Characteristics of Cobalt and Cobalt Base Alloys
     in Vacuum to 10~9 mm of Mercury",  NASA TN D-2524  (Dec. 1964).

32.  Peterson, M.B., Florek, J.J.,  and Murray, S.F., "Consideration of
     Lubricants for Temperatures Above 1000°F", ASLE Trans., Vol. 2,  No.  2,
     (May 1960), p. 225-234.

33.  Mitsubishi Metal Mining Co.,  Ltd., (Nov.  23, 1971).

34.  Buckley, D.H. and Johnson, R.L.,  "Gallium-Rich Rilms as Boundary
     Lubricants in Air and in Vacuum to 10~8 mm Hg", ASLE/ASME Lubrication
     Conference, Pittsburgh  (Oct.  12-18, 1962).

                                   222

-------
35.  Peterson, M.B., Florek, J.J. and Lee, R.E., "Sliding Characteristics
     of Metals at High Temperatures", ASLE Conference, New York  (Ovt.  1959).

36.  Campbell, M.E. and Van Wyck, J.W., "Development of Design Criteria
     for a Dry Film Lubricated Bearing System", ASD-TDR-62-1057,  (March
     1963).

37.  Devine, M.J., Lawson, E.R., and Bower, J.R., Jr., "Anti-Friction
     Bearing Design Considerations for Solid Lubrication", ASME Preprint
     63-MD-43 (May 1963).

38.  McDonnell, R.D., (Ed.), "Proceedings AFML-MRI Conference on  Solid
     Lubricants", AFML-TR-70-127, (July 1970).

39.  Hopkins, Vern, et al., "Development of New and Improved High Tempera-
     ture Solid Film Lubricants", ML-TDR-64-37, Part II (April 1965);
     Part III (August 1966).

40.  Bisson, E.E., "Non-conventional Lubricants", Advance Bearing Technology,
     NASA SP-38 (1964), p. 217.

41.  Sliney, H.E., Strom, T.N., and Allen, G.P., "Fused Fluoride Coatings
     as Solid Lubricants in Liquid Sodium, Hydrogen, Vacuum and Air",
     NASA TND-2348 (Aug. 1964).

42.  Sliney, H.E., "Self-Lubricating Composites of Porous Nickel and
     Nickel-Chromium Alloy Impregnated with Barium Fluoride-Calcium
     Fluoride Eutectic", NASA TN-D-3484, (July 1966).

43.  Sliney, H.E., "An Investigation of Oxidation-Resistant Solid Lubricant
     Materials", NASA TM-X-6785,  (Aug. 1971).

44.  Bowers, R.C., and Murphy, C.M., "Status of Research and Lubricants,
     Friction and Wear", NLR Report 6466, (Jan. 1967), p. 29.

45.  Chaseman, M.R., "Solid Lubrication for Aero Propulsion Systems",
     AGARD LP-84-71.
                         >>_
46.  Abe, W., et al., "Friction and Wear Characteristics of Solid-Lubricants
     Embedded Plain Bearing at High Temperature", Giles Industries Co.,
     Ltd., Japan.

47.  Campbell, M.E. and Hopkins,  V., "Development of Polyimide Bonded
     Solid Lubricants", ASLE Conference, Toronto, Preprint No. 67-A-7A-1
     (May 1967).

48.  Hopkins, Vern, et al., "MLF-5,  An Inorganic Solid Lubricant Film",
     USAF-SWRI Aerospace Bearing  Conference,  (May 1964).

49.  Sliney, H.E. and Johnson, R.L., "Bonded Lead Monoxide Films as Solid
     Lubricants for Temperatures  up to 1250°F, NASA RM E57B15 (1957).

50.  Fusaro, R.L. and Sliney, H.E.,  "Graphite Fluoride (CFX)  A New Solid
     Lubricant", ASLE/ASME Lubrication Conference, Houston, (Oct. 1969).
                                   223

-------
51.  Wagner, T.O., "Fuels and Lubricants for Steam Propulsion Systems",
     SAE 700662, National Conference, (Aug. 1970).

52.  Nixon, J. and Gagabrant, A.R., "Steam Expander Lubrication Program",
     Quarterly Report No. 3, CTR No. 68-04-0004, (Sept.-Nov. 1971).

53.  Personal Communication, Eaton Manufacturing Co.,  (April 4, 1972).

54.  Jarret, M,P., "Material Considerations for Automotive Camshafts",
     SAE PP 710545, (June 7-11, 1971).

55.  Ambrose, H.A. and Taylor, J.E., "Wear, Scuffing and Spalling in
     Passenger Car Engines", SAE Trans., Vol. 63, (1955).

56.  Havely, T.W., Phalen, C.A. and Bunnell, D.6., "Influence of Lubricant
     and Material Variables on Cam and Tappet Surface  Distress", SAE
     Trans., Vol. -63, (1955).

57.  Ayres, V., Bidwell, J.B., Pilger, A.C., Jr., and  Williams, R.K.,
     "Valve Train Wear as Affected by Metallurgy, Driving Conditions
     and Lubricants", SAE Trans., Vol. 66,  (1958).

58.  Brooks, R.D., "Design of Reciprocating Single Cylinder Expanders
     for Rankine Cycle Engines", Quarterly Report, April 4, 1972 to
     July  4, 1972, EPA Contract No. 68-01-0408.

59.  Engineer's Handbook of Piston Rings, Seal Rings,  Mechanical Shaft
     Seals, p. 10, Koppers Company, Inc., Baltimore, (1967).
                                   224

-------