xc/EPA
                                 United States
                                 Environmental Protection
                                 Agency
                                 Industrial Environmental Research
                                 Laboratory
                                 Research Triangle Park NC 27711
                                 Research and Development
                                 EPA-600/S2-82-066  August 1982
Project  Summary
                                 Development and
                                 Demonstration of Concepts for
                                 Improving  Coke-Oven
                                 Door Seals:  Final  Report
                                 A. 0. Hoffman, A. T. Hopper, and R. L Paul
                                  The report discusses the design,
                                 laboratory scale tests, construction, and
                                 field tests of an improved metal-to-metal
                                 seal for coke-oven end doors. Basic fea-
                                 tures of the seal are: high-strength tem-
                                 perature-resistant steel capable of 3
                                 times the deflection of current seals
                                 without  permanent  deformation; no
                                 backup springs and plungers and the at-
                                 tendant requirement for manual inser-
                                 vice adjustments,' seal installed to con-
                                 form to the jamb profile; seal lip height
                                 reduced to give 8 times the inplane flexi-
                                 bility; and compatibility with existing
                                 coke  batteries  and  door handling
                                 machines. Field tests on operating 4 and
                                 6m batteries proved the soundness of
                                 the concept along the straight vertical
                                 sides of the door. However, an unfore-
                                 seen force combination in the four cor-
                                 ners resulted in a net force acting to lift
                                 the seal corners away from the jamb, re-
                                 sulting in unacceptable leakage at each
                                 seal corner. Various schemes were eval-
                                 uated empirically in an attempt to under-
                                 stand and solve the problem. A modified
                                 design to eliminate the problem and re-
                                 duce fabrication cost is proposed. Inland
                                 Steel Company plans to build and test
                                 the modified design at its own expense.
                                 This was a jointly funded EPA/American
                                 Iron and Steel Institute project.
                                  This Project Summary was developed
                                 by  EPA's Industrial Environmental Re-
                                 search Laboratory,  Research Triangle
                                 Park, NC, to announce key findings of
                                 the research project that is fully docu-
                                 mented In a separate report of the same
                                 title (see Project Report ordering infor-
                                 mation at back/.

                                 Project Origin
                                   In 1975 Battelle completed a project
                                 entitled "Study of Concepts for Minimiz-
                                 ing Emissions  from Coke-Oven Door
                                 Seals." This project was jointly spon-
                                 sored by the Industrial Environmental
                                 Research Laboratory of EPA and the
                                 American Iron and Steel Institute (AISI).
                                   The results and recommendations of
                                 the 1975 study were accepted by both
                                 sponsoring organizations and  late in
                                 1976 Battelle  was awarded research
                                 contracts (with EPA and the AISI) deal-
                                 ing with development/engineering/lab-
                                 oratory evaluation of the recommended,
                                 upgraded metal seal. The results of this
                                 effort were accepted by the sponsors
                                 and funding was made available to com-
                                 plete Phase III of the program which con-
                                 sisted of fabrication and field demonstra-
                                 tion of the recommended seal  design/
                                 material.

                                 This Report
                                   This report summarizes the entire ef-
                                 fort to develop an upgraded, retrof ittable,
                                 metal coke-door seal. "Upgraded" here
                                 is defined  as having significantly im-
                                 proved performance in emission control
                                 and operation.

-------
  Not included in this report are results
of a major effort to prevent/minimize
warpage of coke-oven jambs by changes
in the jamb design and materials. This
work was, however, reported in the EPA
report, "Development and Demonstra-
tion of  Concepts for Improving  Coke-
oven Door Seals: Interim Report," EPA-
600/2-78-189 (NTIS PB No. 286 628),
August  1978.
The Recommended Seal Design/
Material
  The recommended seal design was
the result of finding a best "fit" as eval-
uated by a long list of criteria and specifi-
cations. In a systematic approach, each
criterion had to be considered, but the
two major influences on the design were
(a)  a general  specification by the AISI
and EPA, and (b) a criterion strongly de-
sired by Battelle researchers.
  The major specification was that the
new seal be retrofittable, i.e., the new
seal would be a replacement for existing
seals without modification of the doors
or the door-handling equipment. This
also meant that the recommended seal
had to "handle" or accommodate all (or
nearly all) of the jamb and the door warp-
age problems that  exist at  operational
plants. Seals are attached to doors and
the profile congruency of the door with
the oven jamb can  be poor due to past
warpage and heat effects.
  The criterion desired by Battelle per-
sonnel was that the recommended design/
material should eliminate the need for
numerous, manual  seal-adjustment de-
vices associated with all existing seals.
With about 25,000 coke-oven doors in
operation, each with about 20 manual
seal-adjustment devices, the  theory is
that at times workers climbing ladders or
scaffolds would adjust  500,000 de-
vices to improve emission control. Bat-
telle researchers were skeptical  about
the performance of the adjustment de-
vices and the availability of the time and
skill to manipulate them.
  The general approach to an upgraded
seal was to increase the seal flexibility in
every way possible while also increasing
its strength and heat resistance. This ap-
proach resulted in the recommended de-
sign shown in Figure 1.
  The design/material elements leading
to increased flexibility and strength are:
  1. The height of the contacting edge
     [(A) in Figure 1 ] was lowered to a
     9.5-mm height (from 19 mm) to in-
     crease the flexibility of  the edge
     along the jamb by a factor of 8. This
     increased flexibility allowed the
                                   Sealing Ring (B)
Figure 1.     Cross section of a suggested coke-oven seal.
    vertical seal edge to maintain con-
    tact with the jamb in spite of curva-
    tures in  the jamb surface and the
    uneven surface of the seal edge.
   . The main spring element [(B) in Fig-
    ure  1]  was  switched from  304
    stainless or a corrosion-resistant,
    low alloy steel to a high-strength,
    high-temperature spring alloy. Bat-
    telle recommended that the first
    demonstration seals be  made  of
    age-hardened Inconel  X-750.  In
    laboratory tests at 426 °C (800 °F),
    Inconel X-750 tolerated three times
    the amount of seal deflection dur-
    ing simulated door latching before
    plastic distortion by creeping was
    encountered. Increased seal deflec-
    tion is desired to (a) absorb minor
     changes in door and jamb profiles
     without the need for adjustments,
     and (b) keep an evenly distributed
     force  pressing  the  seal  edge
     against the jamb.
  3. Changing the width of the spacers
     [(C)  in Figure  1]  along  the door
     made it possible  (during original
     seal mounting) to bend the entire
     seal to make the profile of the seal
     edge match the general  profile of
     the jamb. This element of flexibility
     was introduced to make it possible
     to  realign the seal edge and jamb
     profile in instances where the door
     and jamb  profiles are particularly
     incongruent.
This design eliminated the need for seal-
adjustment devices (and backup springs)

-------
 pressing in point loading against the out-
 board side of the seal.

 Demonstration Seal Fabrication
   Two of the criteria for judging seal
 designs were (a) whether the design was
 amenable to mass production methods,
 and (b) whether the machining during
 fabrication could be eliminated. For the
 design presented, Battelle recommend-
 ed (1) roll forming the seal shape rather
 than one-at-a-time brake forming, and
 (2) hydroforming the corners rather than
 the cut-fit-weld-machine fabrication me-
 thods now being used.
   Further complications  were  intro-
 duced  by (1) the  requirement that de-
 monstration seals  should be  obtained
 from experienced seal fabricators, rather
 than jobbing  fabrication to  different
 shops; and (2) introduction of a seal-
 corner design that could be fabricated
 only in a hydroforming operation.
   Eleven months after sending out re-
 quest for quotation  seals,  four  seals
 were delivered to the Bethlehem  Steel
 coke plant in Lackawanna, New York.

 Results of the Seal
 Demonstration Efforts
  The recommended seal design/material
 was tested at  Bethlehem Steel's coke
 plant in Lackawanna, New  York  (6-m
 Wilputte battery) and at Republic Steel's
 coke plant in  Youngstown,  Ohio  (4-m
 Koppers batteries).
  At Lackawanna, the usefulness was
 demonstrated of taking simple measure-
 ments of the  variation in distance be-
 tween  the  jamb and  door  (along its
 length) and then installing spacers under
 the new seal to bring the seal edge into
 congruency with badly warped jambs. In
 one instance it was necessary to put a
 1 9-mm (%-in.) steel bar under the top
 horizontal portion of the seal to bring the
 profile of the seal into alignment.
  Zero leakage, increased flexibility, and
 no  manual adjustments were demon-
 strated (for months) along the 6-m ver-
 tical sides of the seal, as expected.
  However, all four seals put into opera-
 tion unexpectedly  released  emissions
 through gaps at each corner. This prob-
 lem persisted even after reworking one
 seal to eliminate variations in  the flat-
 ness of the seal edge. It was deduced
 that the corners were "lifting off" the
jamb; i.e., the corners were reacting to
 forces that act to cause a slight gap at
 the corners during seal deflection. This
 lift off effect was confirmed at Republic
 Steel during tests using a 4-m seal forced
against a flat machinist table.
   The fact that corners on seals can re-
 act counter to a desired pattern was not
 new—U.S. Steel research personnel had
 told Battelle about "corner lifting" they
 had encountered on standard seals. What
 was  obvious  was that  (a) the design
 changes made to avoid this  potential
 problem were not successful in full-scale
 operation, and (b) that Battelle's labora-
 tory work  on  parts of the seal did not
 model the  reactions in complete seals.
 Experiments at Republic Steel  indicated
 that the seal lip was not part of the prob-
 lem and that  some  method had to be
 found  to  relieve the "bunch-up of
 metal" that occurs  during seal deflec-
 tion in the flat portion  of the seal in the
 corners.

 The Modified Seal  Design
   Project originators did not anticipate
 that a considerable amount of problem
 solving would be required after getting
 the first demonstration  results.  Although
 continued analytical work was consid-
 ered (modelling and strain gauge testing
 on a full-scale seal), the last emphasis in
the project was to consider the range of
design changes that could provide ah
empirical solution to the corner problem.
Out of this approach came a modified
seal design (not yet tested). This modi-
fied design is shown in Figure 2.

  The major elements are:
  A— a flat strip, cantilever-spring re-
      placement for the formed S-shape
      used in the original design.
  B— a  heavy-section,  carbon-steel,
      angle-iron seal holder.
  C— a  demountable seal-edge angle
      bolted or welded to the seal ele-
      ment (no machining required).
It is recommended that the spring  por-
tion of the seal be made of 1 7-4 PH
stainless  steel,   rather  than  Inconel
X-750. This lower priced alloy was not
included for consideration in the original
evaluation because it is difficult to form,
but all  forming is eliminated  in  this
design. It is expected that this seal can
be built by steel companies.
  For an empirical approach to a solution
of the corner problem, it is  suggested
                   Door Back
Figure 2.    Modified seal design.
« U.S.aOVEBNMINTWIINTIM30FFIM:1»U-S59-Ol7/0784

-------
     that the strips making up the main por-
     tion of the seal only be butted at the cor-
     ners, not welded. This leaves a space or
     separation  in the corner that (a) makes
     the corner relatively flexible, (b) permits
     consideration of corner backup (springs)
     outboard of the corner to force the cor-
     ner against the jambs, (c) may minimize
     some of the stress generation in the cor-
     ners during deflection, and (d) could in-
     troduce a built-in emission problem in
     the   corners.  To  prevent emissions
     through the spacings, it is suggested
     that this space be closed with a cover
     strip  (or foil) attached outboard of the
     corner.

     Present Status of the Modified
     Seal
       Inland Steel has volunteered to engi-
     neer/fabricate/test  and oven-demon-
     strate one seal of the modified design. At
     this time, engineering is complete and
     fabrication is in progress.
           A. 0. Hoffman, A. T. Hopper, and R. L. Paul are with Battelle-Columbus Labora-
             tories, Columbus, OH 43201.'
           Robert C. McCrillis is the EPA Project Officer (see below).
           The complete report, entitled "Development and Demonstration of Concepts for
             Improving Coke-Oven Door Seals: Final Report," (Order No. PB 82-230913;
             Cost: $12.00, subject to change} will be available only from:
                   National Technical Information Service
                   5285 Port Royal Road
                   Springfield. VA 22161
                   Telephone: 703-487-4650
           The EPA Project Officer can be contacted at:
                   Industrial Environmental Research Laboratory
                   U. S. Environmental Protection Agency
                   Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
          US  tNVIR  PROTECTION
          REGiON  5  U8RAKYT
          230  S  DEARBORN  STREET
          CHICAGO  1L  606U4

-------