EPA-670/2-74-054
JULY 1974 Environmental Protection Technology Series
SOURCES OF OIL AND WATER IN
BILGES OF GREAT LAKES SHIPS
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio
-------
EPA-670/2-74-054
July 1974
SOURCES OF OIL AND WATER
IN BILGES OF GREAT LAKES SHIPS
By
John B. Woodward
Department of Naval Architecture and Marine Engineering
University of Michigan
Ann Arbor, Michigan 48104
Program Element No. 1BB038
Project Officer
William Librizzi
Industrial Waste Treatment Research Laboratory
Edison, New Jersey 08817
.,,r.v. WQC
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
For tale by the Superintendent of Documents, U.S. Government
Printing Office, Washington, D.C. 20402
-------
REVIEW NOTICE
The National Environmental Research Center—
Cincinnati has reviewed this report and approved
its publication. Approval does not signify that
the contents necessarily reflect the views and
policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or com-
mercial products constitute endorsement or recom-
mendation for use.
11
-------
FOREWORD
Man and his environment must be protected from the adverse
effects of pesticides, radiation, noise and other forms of
pollution, and the unwise management of solid waste. Efforts
to protect the environment require a focus that recognizes
the interplay between the components of our physical environ-
ment—air, water, and land. The National Environmental
Research Centers provide this multidisciplinary focus through
programs engaged in
O studies on the effects of environmental
contaminants on man and the biosphere, and
O a search for ways to prevent contamination
and to recycle valuable resources.
This report discusses the sources of oily bilge water pollution
from Great Lakes ships, and therefore is intended to be a con-
tribution to the prevention of contamination of water resources.
A.W. Breidenbach, Ph.D.
Director
National Environmental
Research Center, Cincinnati
iii
-------
ABSTRACT
Sources of bilge water and of oil in bilge water were surveyed aboard
five ships of the Cleveland Cliffs Iron Company. The ships included
two powered by steam turbines, one by a uniflow steam engine, one by
a conventional reciprocating steam engine, and one by a diesel engine.
It is found that many sources of bilge water are clean sources.
Although no accurate estimate of the water thus contributed to the
bilges can be offered, it is concluded that diverting these sources
from the bilges could ease the task of separating, storing, and dis-
posing of oil wastes.
Several samples of water were taken from each ship, and analyzed for
total, fixed and volatile non-filterable residue, color, pH, turbidity,
total organic carbon, and oil and grease concentration.
This report was submitted in fulfillment of EPA Grant R802475 by The
University of Michigan under the sponsorship of the Environmental
Protection Agency. Work was completed as of October 1, 1973.
-------
CONTENTS
Page
ABSTRACT iv
LIST OF FIGURES vi
LIST OF TABLES vii
ACKNOWLEDGMENTS viii
SECTIONS
I. CONCLUSIONS 1
II. RECOMMENDATIONS 3
III. INTRODUCTION 5
THE PROBLEM " 5
THE INVESTIGATION 6
DIFFERENCES BETWEEN GREAT LAKES AND SEA PRACTICE 8
IV. SOURCES OF WATER IN MACHINERY SPACE BILGES 11
STEAM TURBINE 11
DIESEL 17
RECIPROCATING STEAM 18
V. SOURCES OF OIL IN MACHINERY SPACE BILGES 20
INTRODUCTION 20
DISCUSSION 20
VI. BILGE WATER SAMPLES 24
INTRODUCTION 24
SAMPLE IDENTIFICATION 24
SAMPLE COLLECTION AND ANALYSIS 26
VII. APPENDIX - CLEVELAND CLIFFS FLEET 33
-------
FIGURES
No. Page
1 TYPICAL BILGE ARRANGEMENT 6
2 BILGE PIPING SCHEMATIC FOR STEAMER
WILLIAM G. MATHER 7
VI
-------
TABLES
No. Page
1 SAMPLE SUMMARY 27
2 TOTAL, FIXED, AND VOLATILE NON-FILTERABLE RESIDUE 29
3 pH, COLOR AND TURBIDITY 30
4 TOTAL ORGANIC CARBON 31
5 OIL AND GREASE 32
VII
-------
ACKNOWLEDGEMENTS
The work reported here was carried out aboard ships of the Cleveland
Cliffs Iron Company, and hence is entirely dependent upon its sincere
cooperation. John L. Horton, assistant manager of the marine department,
and John C. Culbertson, fleet engineer, were the principal liaison. Ship-
board personnel were uniformly cooperative and helpful. The support of
all is acknowledged with sincere thanks.
Vlll
-------
SECTION I
CONCLUSIONS
1. A typical Great Lakes machinery space contains roughly 150 potential
sources of bilge water. About one third of these are open ended pipe
drains that discharge only clean water. These could therefore be
piped to a clean drain tank for disposal overboard without contri-
bution to external pollution. The remaining bilge water, which
presumably would remain oily, would require separation before over-
board disposal, or could be held for shore disposal. The benefit of
diverting the clean drains would therefore be in greatly reducing the
volume of oil waste to be handled. It may be possible to so reduce
the water volume that methods of disposal otherwise impractical, such
as incineration, become feasible. Any method of oil-water separation
is likely to be reduced in cost by a reduction in water throughput.
The cost of additional piping to divert clean water from the bilges
is unknown.
2. Bilge water sources are numerous, diverse, intermittent, and often
in the nature of leakages; in consequence, it does not appear
feasible to measure their flow rates. Some flow estimates, strictly
from visual observation, are quoted in this report, but the conclusions
of the first paragraph are based on the sum of these observations, and
not on hard data.
3. No oil is deliberately discharged to the bilges, except in the oil-
water mixtures to be mentioned in 6.
4. Drips of lubricating oil from reciprocating machinery are a potential
source of oil in the bilge that can be eliminated by careful house-
keeping (Cleveland Cliffs is doing well in this respect).
-------
5. Accidental spills of oil are a possible source of bilge contaminatior
that can be reduced by care in design, building, and operation, but
which doubtless cannot be eliminated. If the bilges are kept dry,
however, it should be feasible to clean many spills by wiping or
absorption, rather than flushing to the bilge sump.
6. Several sources from which oil and water enter the bilge together
exist, these being (1) residue for centrifugal purifiers, (2) con-
taminated heating coil drains (casualty situation only), (3) stern
tube in-leakage, and (4) drainage of condensation and oil from
reciprocating propulsion engines. Of these, only (3) seems to be
subject to ready elimination, so that keeping oil and water totally
separate before they enter the bilge does not appear to be feasible.
But if (3) is eliminated, the remaining sources on non-reciprocating
ships are small enough that it may be feasible to drain the oil-water
to a slop tank, rather than dumping to bilge.
7. The conventional (i.e., non-uniflow) reciprocating steam propulsion
engine is a special problem. Separating oil and water before it
reaches the bilge does not appear feasible, but further investigation
may show that the leakages of oil and water can be significantly
reduced. This, however, is a statement of hope, rather than a con-
clusion.
-------
SECTION II
RECOMMENDATIONS
The following three items are recommendations to Great Lakes ship
operators:
1. Housekeeping recommendations
Operating personnel should be instructed to avoid flushing of
oily wastes into bilges; spills to be wiped up rather than washed
away. Drips and leaks to be caught by portable receptacles.
Running of water into bilges via vents and drains to be kept to
a minimum. In general, a "dry bilge is a good bilge" attitude
should be promoted.
2. Maintenance recommendations
Vigilance should be exercised to keep all pump and valve packings
in good condition.
3. Added gravity separation
Even casual visual inspection shows that the bilge well acts as
a gravity separator to remove gross oil contamination. However,
the small size and awkward location of the well result in all or
part (if the crew takes the trouble to skim the surface) of the
oil being entrained by the bilge pump suction. It is therefore
recommended that a simple gravity separator be placed in the
bilge pump discharge line, this being basically a tank of (say)
100-gallon capacity from which surface oil can be drained to a
slop container. Although such a device will not produce a sheen-
free effluent, it should nonetheless reduce the amount of oil
escaping overboard.
The following measures can be beneficial, but may require expense on
the part of the ship owners. Since the magnitude of this expense has
not been estimated, the measures are offered as suggestions for
-------
consideration, rather than as firm recommendations for action.
4. All piping drains to bilge, unless known to be potential sources of
oil contamination, should be piped to a clean drain tank, for dis-
posal overboard without passing through the bilges.
5. Reduction of bearing oil flow to the reciprocating steam engines
should be investigated. This might be done by addition of a bearing
temperature monitoring system, which would allow careful trial
reductions in oil flow with low risk of bearing damage.
The following steps are recommended for the Environmental Protection
Agency:
6. since the generation of oil bilge water cannot be eliminated, EPA
should support efforts to develop separation processes.
7. Since the water samples taken under this program give only a glimpse
of the bilge water problem presented by the several hundred ships
active on the Great Lakes at any one time, EPA should continue to
collect such samples.
8. The concept of effecting a major reduction in flow of water to the
bilges should be pursued through a more detailed study of a single
ship. This would consist of preparation of piping arrangement
sketches and job specifications for implementation of a clean drain
system. This to be done by, or in cooperation with, a. marine design
firm or shipyard so that a cost estimate could be prepared.
9. A program similar to the one reported here should be attempted
aboard the new Great Lakes ships (i.e., entering service since the
1955-1970 building hiatus). These ships are diesel-propelled, have
oil-lubricated stern tubes, and generally are not well typified by
the old ships studied here.
-------
SECTION III
INTRODUCTION
THE PROBLEM
Bilge water is that water which collects from almost innumerable sources
into the lowest internal part of a vessel—its "bilges." The principal
accumulations of bilge water are found in the machinery spaces; cargo
holds, tanks, and void spaces usually collect only trifling amounts of
such water unless a casualty lets in sea or weather. Within machinery
spaces there are many sources of leakage (e.g., pump glands, stern tube),
and the bilge is used as a sump to receive wastewater from many sources.
Typically these sources offer low flow rates, and are intermittent, so
that piping to recover the water is not justified. The water leaking or
draining to the bilges flows across the tanktop (Figure 1) to a sump,
where it is picked up by a bilge pump for discharge overboard. Oil, and
possibly other contaminants, have also drained to the bilges. Although
there will be some gravity separation of oil and water in the bilge well,
and the water will therefore not be badly contaminated with oil if the
engineering crew takes the trouble to manually skim the well surface,
some oily contamination of bilge-water remains.
Oil draining to the bilges is mostly lubricating oil, but fuel oil may
also be found.
It should be noted that bilgewater serves a useful function in that it
tends to flush oil off the tanktop, thereby preventing the buildup of a
hazardous combustible layer.
On ships powered by steam turbines or by diesel engines, there is usually
an almost negligible deliberate discharge of oil to the bilges in normal
operation. However, the reciprocating steam engine uses once-through
lubrication for many of its rotating and sliding bearings. In the
-------
conventional (i.e., non-uniflow) engine, this oil drains to the tank
top below the engine, whence it flows to the bilge well. Cylinder
lubrication is accomplished by injection of oil into the cylinder,
whence it ultimately appears in the exhaust steam. With the jet-type
condensers used on the Great Lakes, the condensed steam, with oil
entrained, is discharged overboard. This practice is followed by
both conventional and uniflow engines
Figure 2 illustrates a typical Great Lakes bilge piping system.
LEAKAGE AND DRAINS
I
I
I
\V
PUMP
AUTO
, .START- |"
STOP -*•,
o
OVE.RBOAR.D
CD
BOTTOIVl OF SHIP
N \NNERBOTTOM OR TANKTOF>
Figure 1. Typical Bilge Arrangement.
THE INVESTIGATION HERE REPORTED
This investigation was carried out aboard vessels of the Cleveland Cliffs
Iron Company. The purpose was to locate all sources of water and oil
flowing to the bilges, and to determine if it were possible to eliminate
such sources, i.e., provide some means of disposal, other than discharge
to bilge. This was intended to contribute to solution of the oily bilge
water problem by keeping oil and water separate before they come in con-
tact at the bilge well. A second purpose was to collect samples of bilge
water for subsequent laboratory analysis.
-------
AFT PEAK
TANK
FUCU CHL.
STOftA&E TA»4V«.
/ \
\
AUTOIsAATIC
TO WATER
<
-I3"
3"
HAA»N~*
L
1 j
9
% SE.RVIC.E. 1*1
C-TlOKi
\
V
UUBE OIL.
SUMP
/ \
/ \
\l/X
COKIOCMSATC
POMP NA/ELL.
F»RC STATION
2"
UUBC. OIL.
BIL&E SYSTEM
STEAMER WM G. IMATHER
ADAPTED FRONt MANITOW/OC. &H\P-
SUlUDIMCk "DVVS A4-SO
H£ CAOS.S CONKjKCTIOM OP OIU&C. SyiTftN'
WITH PlRt SVSTB.IV\. NORNlAU Btl_
-------
The investigation was carried out by on-board examination of five Cleve-
land Cliffs ships in their normal operation during the 1973 shipping
season. The investigator, who is an engineer having experience with
theory, design, and operation of marine machinery, inspected the bilges
and piping thereto, measured or estimated flows where possible, observed
operating practices, questioned the operating engineers, collected water
samples, and examined piping drawings. The investigation concentrated
on the propulsion machinery spaces, but also covered compartments housing
steering engines and bow thrusters, and the areas in vicinity of deck
machinery.
Details of ship-place-date are these:
PONTIAC Cuyahoga River; Cleveland to Detroit 5/1/73, 5/2/73
CADILLAC Cuyahoga River 5/8/73
WM P SNYDER Cuyahoga River 5/9/73
RAYMOND H REISS Detroit to Cleveland 5/11/73
WM G MATHER Detroit to Cleveland 5/21/73
These ships were chosen to give a variety of power plants; PONTIAC and
MATHER are steam turbine vessels, REISS is diesel, CADILLAC is conven-
tional steam reciprocating, and SNYDER is uniflow steam reciprocating.
DIFFERENCES BETWEEN GREAT LAKES AND SEA PRACTICE
Diesel
There is no essential difference among Great Lakes inland, and sea
practice with respect to diesel machinery, especially if discussion is
limited to U.S. vessels. For example, the popular General Motors (Electro-
motive Division) locomotive engine is being applied to new vessels building
in all three areas with no differences in engines or their auxiliary
systems.
Many foreign ocean-going vessels, however, are powered by low-speed diesel
engines capable of burning "heavy" fuels, usually mixtures of residual oil
-------
and a lighter distillate. Because residual oil contains most of the
impurities present in the original crude oil, these ships are usually
provided with fuel washing systems in which the oil is mixed with fresh
water (dissolves water soluble impurities), then passed through a
centrifuge to remove the water. The removed water is oily and thereby
constitutes a source of oily bilge water. This water is added at a
maximum of about 10 percent of the oil flow. Fuel for a 10,000 hp engine
will therefore produce an oily water flow of roughly one gallon per
minute. Although no ships of the U.S. Great Lakes fleet are so equipped,
many of the foreign vessels entering the Lakes are powered by low-speed
diesel engines, and it seems likely that many are therefore washing fuel
in the manner described.
Steam turbine practice on the Great Lakes is almost identical to ocean
practice. Differences that are apparent are largely of historical nature:
no steamers have been built on the Lakes in almost 20 years, while they
continue to be built for ocean service. A development in this interval
significant to the bilge question is the widespread adoption of the oil-
lubricated stern bearing, which obviates the need for a lubricating water
flow through the bearing and into the bilges. A randomly-chosen ocean
ship is likely to be newer than any steam laker, and therefore less
likely to have the stern tube bilge source. Note, though, that the oil-
lubricated stern tube is not exclusive to steam vessels, but is used as
well on diesel ships.
The greatest difference between ocean and Great Lakes practice occurs in
the case of reciprocating steam ships. The biggest factor is the almost
total absence of these vessels in modern ocean service. In effect,
there is nothing to compare the Great Lakes vessels to. Looking back,
though, a major difference can be seen in the condensate systems.
Typical Great Lakes engines exhaust into a jet condenser, wherein the
steam mixes with lake water, and is discharged directly overboard.
Cylinder oil is thus discharged overboard. Sea practice used a surface
condenser; condensate did not contact sea water, and was continuously
recycled. Cylinder oil was removed by filter sponges that were
-------
periodically washed to renew them. Wash water containing the removed
oil was doubtless dumped into the bilges, whence it found its way over-
board. The end result was therefore much the same.
10
-------
SECTION IV
SOURCES OF WATER IN MACHINERY SPACE BILGES
STEAM TURBINE SHIPS
This discussion is based on inspection of the PONTIAC and WILLIAM G
MATHER. It begins with a simple list of bilge water sources, followed
by comments on each. No significance is implied by the order in which
the sources are listed. Numbers in parentheses are the approximate
numbers of such sources.
Shell condensation (1)
Stern gland leakage (1)
Pump and turbine gland leakage (20)
Gage glass blowdown (4)
Deck drains from auxiliary spaces (2)
Condenser vents (3)
Lube oil purifier (1)
Tank overflows (4)
Tank drains (4)
Pump casing drains and vents (20)
Sea chest drains (6)
Safety valve drains (4)
Relief valves (8)
Contaminated drains (1)
Moisture separators (2)
Deck wash connections (2)
Sample sink (1)
Valve glands (50)
Piping system drains (10)
Engineers' lavatory and washing machine (2)
(In the following discussion, features of the non-turbine ships are
mentioned if they are not unique to the type of power plant.)
11
-------
Shell Condensation
Although some condensation is usually visible on the inner hull surfaces
in the machinery spaces, this is a trivial source.
Stern Gland Leakage
This water leaks along the propeller shaft from the sea, or is injected
into the stern tube bearing area—and subsequently leaks inward—by the
fresh water cooling system (for cooling and lubrication). In either
case, the leakage is sea water, except that it may be contaminated if
grease lubricant is used in the gland (as is done on the MATHER). The
rate is highly variable, depending on the amount of wear that has accumu-
lated, and on the individual chief engineer's favored practice. An esti-
mated 1 gal/min was observed on the PONTIAC, about 5 gal/min on the
MATHER.
Pump and Turbine Glands
Some leakage through the glands of centrifugal pumps is required for
cooling and lubrication, though the rate is highly variable, depending
upon the condition of the gland packing. Rates on operating pumps were
observed from near zero to an estimated 1 gal/min .
Flow through centrifugal feed pumps is essential at all times to keep
them from overheating during periods when demand by the boiler may fall
to zero, and this flow is often obtained by large gland leakage. This
practice was observed on both the PONTIAC and MATHER, with the flows
estimated to be in the neighborhood of 1 gal/min .
The connecting rods of reciprocating pumps generally show a slight
leakage also. At the steam end, the leakage may be partially in the
form of vapor.
12
-------
Propulsion turbine glands are typically fitted with a leakage collection
system that should prevent leakage to the exterior under ordinary circum-
stances. However, drain pockets are also provided, with a pipe leading
to the bilge to direct any leakage that may occur, showing that some
leakage is anticipated. Auxiliary turbines are typically fitted with
carbon packing that reduces leakage to a trifling amount, and that
largely steam.
Gage Glass Slowdown
Boiler gage glasses occasionally need flushing to remove solids that
might interfere with visibility. Amount of water released to bilges
varies according to need and engineers' practice. It is probably less
than 10 gal/day .
Deck Drains from Auxiliary Spaces
This is principally drainage from the steering engine space, which is
typically adjacent to the main machinery space on a somewhat higher level
than the main operating platform. A noticeable flow from this source is
likely only in the case of a steam steering engine (used on all ships
observed except the diesel-propelled REISS), which may have some leakage
around connecting rods and valve stems. A flow on the order of 0.1 gal/
min was observed on the PONTIAC; essentially zero on the other vessels.
Condenser Vents
Circulation of water through condensers may be impeded by air trapped at
high points in the circulating system, these usually being in the conden-
ser heads. Vents from the heads are therefore provided, with pipe leading
directly to bilge. The vent may or may not be used, depending on draft
of the ship (i.e., air binding should be worse at light draft, since this
puts the condenser higher with respect to the waterline), and on the
chief engineer's judgement as to need. It was observed in use only on
the PONTIAC; and there only for the main condenser; flow was an estimated
2 to 3 gal/min .
13
-------
Lube Oil Purifier
Heavy contaminants, mostly water, are centrifuged from the main lube oil
system (serves propulsion turbines, reduction gears, and propulsion shaft
bearings). Rate of water extraction may run from near zero to about
1 gal/day . Although the quantity is small, this is a significant source
because the imperfection of the separation process leaves some oil in the
water. This, then , is one of the identifiable sources of oily bilge
water. The practice observed on the MATHER is to hold the water in a
slop tank, then reprocess it through the purifier to remove as much
remaining oil as possible, before discharge to bilge.
Tank Overflows
Water tanks operating under atmospheric pressure are provided with over-
flows (if above atmospheric, then relief valves) piped to bilge. Typical
are potable water tank, fresh water drain collecting tank, contaminated
drain tank. Nominally, flow issues only during an abnormal operating
condition, such as overfill of the potable water tank. Oil tanks, such
as fuel and lube oil tanks, must be provided with vents or overflows,
but these are not led to the bilge. No overflows were observed.
Tank Drains
Tank drains are necessary to permit draining for repair work and winter
layup. No flow during operation.
Heat Exchanger Drains
Same discussion as tank drains.
Pump Casing Drains and Vents
Same discussion as tank drains. Also, vents opened to bleed air during
initial startups, and to flush casing of pumps taking suction from the
sea after encounters with mud. The latter action was observed on the
14
-------
PONTIAC fire and general service pump in the Cuyahoga River, discharging
about 10 gal/min to the bilges.
Sea Chest Drains
Same discussion as tank drains. Also, mud flushing may be necessary as
discussed just above.
Safety Valve Drains
Condensation is likely on the downstream side of safety valves following
operation, hence a drain line open to bilge is provided for each valve.
Flow to be expected only following an operational aberration (should
never happen) that causes safety valves to lift.
Relief Valves
Relief valves on piping systems, heat exchangers, turbine casings, etc,
generally discharge to the bilge. There should be no flow in normal
operation. Relief valves on fuel oil and lube oil piping systems dis-
charge to the low pressure parts of these systems.
Contaminated Drains
Contaminated drains are the condensed returns from steam heating coils
in fuel tanks. Although the water is normally clean, there is the
possibility of leakage of oil into the coils, hence the name "contami-
nated." The drains are collected in a tank where they can be inspected
through a sight glass for contamination. If oil appears, the incoming
drains can be diverted into the bilges. Although this diversion is
normally zero, it might be on the order of 10 gal/min in the event of
contamination.
This is another identifiable source of oily bilge water, although a zero-
flow source in normal circumstances.
15
-------
Moisture Separators
These remove moisture from the compressed air used for pneumatic controls,
and to operate portable tools. This water drains to the bilge in trivial
amounts.
Priming Pump Discharge
The priming pump is used to draw air from the suction side of ballast
pumps to establish flow when these are started. Some water must appear
toward the end of the priming cycle, and this is discharged to the bilge.
Quantity should be on the order of 10 gallons per voyage.
Deck Wash Connections
Some sea water piping system, such as the firemain, will have one or
more hose connections for use in washing floor plates or equipment in
the machinery spaces. Use depends on the need for such practices.
Sample Sink
Water samples are periodically drawn from the boiler for chemical analysis,
with excess going to bilges via sink drain. The samples must be cooled by
flowing through a small heat exchanger as they are drawn, and the cooling
water also goes into the bilges, either directly or via the sample sink
drain. Flow from this cooler was observed on the SNYDER at about 10 gal/
min . Flow should be intermittent, only for 30 seconds, say, once per
watch while a sample is being drawn.
Valve Glands
Leakage around stems of valves is possible, depending on the condition of
the packing. A large sea valve on the REISS was observed to be leaking
at an estimated 1 gal/min .
16
-------
Piping System Drains
Generally the same discussion as tank drains. However, one case of
operational use was observed; on the MATHER the firemain drain was open
continuously, discharging about 5 gal/min to the bilge. This was said
to be a common practice, intended to keep circulation through the piping
to avoid freezing during early season operation. Drains from steam
piping to remove condensation usually are piped to a drain collecting
tank, but some such drains may instead go to the bilge, with brief
slight flow during warmup.
Engineers' Lavatory and Washing Machine
Engineers' lavatory drains to the bilge if it is below the waterline.
A washing machine in the engine room is likely to be found only on
reciprocating engine ships, there for washing filter sponges used to
filter oil from the main condensate.
DIESEL SHIP
This discussion is based on inspection of the RAYMOND J REISS, the only
diesel ship in the Cleveland Cliffs fleet, and one of the few diesel
ships in the established (i.e., built before 1970) U.S. Great Lakes
fleets.
This ship is a converted steamer, and retains many steam auxiliary
systems, fed by a pair of auxiliary boilers. The list of sources dis-
cussed for the steam turbine ships is generally applicable, with only
the items directly associated with the main engines being different.
The main condenser vent, the turbine gland leakage, and the lube oil
purifier from the preceding list are absent. On the other hand, the
propulsion diesel engine provides no additional source of either bilge
water or bilge oil in its normal operation.
17
-------
Feed water for the REISS auxiliary boilers is treated by an ion-exchange
softener (comparable treatment for the MATHER and PONTIAC main boilers
is provided by distillation). This unit requires occasional regeneration
(not observed, however) which entails a discharge of back-flush water to
the bilge.
My visit to the REISS occurred shortly after a casualty to the engine
lube oil system, an event that dumped a large quantity of oil into the
bilge. The bilges were very oily at the time, as is reflected in the
bilge water samples described in Section VI.
Diesel vessels in general may differ in significant respects from the
REISS. The STEWARD J CORT (Bethlehem Steel), first of the recent new-
buildings in the U.S. Lakes fleet following a moratorium of about 15
years, has no auxiliary steam system, hence reduced piping and no oil-
dripping reciprocating steam pumps. Centrifugal purifiers for fuel oil
and lube oil may or may not be used; neither, for example, is aboard
the CORT. That vessel burns a low-viscosity fuel, which requires only
minimal treatment before use, and the same is true of the REISS.
RECIPROCATING STEAM
With the exception of the main propulsion engine, the situation here is
much the same as aboard the steam turbine ships. That is an important
exception, however, when the engine is the traditional open design
(CADILLAC). Some condensation of steam occurs within the engine cylin-
ders. At least some of the condensation leaks around piston rods, and
some steam leakage condenses immediately outside the engine. The
resulting water drips to the engine room tanktop, whence it flows toward
the bilge sump. Although the flow is not large (difficult to estimate,
but perhaps on the order of 1 gal/min), it is highly significant in
that it flushes oil leakage from the engine toward the bilge sump. Part
of this flow may be cooling water applied externally to bearings.
The uniflow steam engine (SNYDER) does not produce bilgewater, except
perhaps for some condensation during warmup. There is no condensation
18
-------
in normal operation; indeed, the prevention of this condensation is the
major motivation behind the uniflow design. The engine is also provided
with a crankcase oil sump, much in the manner of the diesel engine, so
that there is no oil drainage onto the tanktop requiring flushing water.
Both types of reciprocating engines contribute water to the bilges
indirectly, by way of the treatments required of their boiler feed water.
Exhaust from these engines is condensed in a jet condenser, a device that
mixes its cooling water with the exhaust steam. The resulting mixture is
discharged overboard, but a minor part of it (on the order of 10%) is
returned to the boilers. This feed water, being mostly lake water, must
be treated to remove natural impurities. The processing equipment used
(zeolite process) requires periodic regeneration, including a backflush
which goes into the bilges.
The exhaust steam is oily, since lubricating oil is injected directly into
the engine cylinders. The water taken from the condenser for boiler feed
is therefore oily, and the oil must be removed by filtration. The filter
elements are sponges which are washed periodically in a conventional
household washing machine. If this machine is located low in the ship,
its drain goes to the bilges.
The zeolite backflushing and the sponge washing should together contribute
a quantity on the order of 100 gal/day to the bilges.
19
-------
SECTION V
SOURCES OF OIL IN MACHINERY SPACE BILGES
INTRODUCTION
Since all moving machinery must be lubricated, and since leakage or care-
less handling of lubricants and fuel seemingly must occur occasionally,
it is impossible to pinpoint all sources. Some definite sources were
identified, however, and these are discussed in this section. These
sources are:
stern tube lubricant
lube oil and fuel oil purifiers
contaminated drains
drippage from steering gear and reciprocating pumps
drippage from oil burners
reciprocating propulsion engines
DISCUSSION
Stern Tube Lubricant
The sterntube of the MATHER is lubricated by a grease (Mobilgrease L-3)
injected by a hand pump at a rate of about 0.5 gal/day . Most of the
grease is swept out into the machinery space bilges by the flow of
cooling water inward along the propeller shaft. The other four vessels
observed do not use this method of lubrication.
Lube Oil and Fuel Oil Purifiers
Lube oil purification by centrifuge is used on steam turbine ships, may
be used on diesel ships, and similar fuel oil purification may be used
on diesel ships (only the steam turbine case actually observed here).
The water removed from the oil carries some oil with it as a consequence
of the incompleteness of the process. This water, discharged normally
20
-------
to the bilge, is therefore a source of oil.
Contaminated Drains
Steam heating coils are commonly used in fuel tanks as a means of
reducing the viscosity of the oil. Since oil leakage into the piping
(at a defective joint, say) is possible, and since oil must be kept out
of the boilers, returning condensate is inspected for oil contamination.
This is accomplished by holding the condensate briefly in a tank
arranged for easy visual detection of an oil film on the water surface.
If such a film appears, the return flow of water is manually diverted to
the bilges until the defect is corrected or isolated. If the defect
recurs, this process then becomes a source of oil in the bilges.
Drippage From Steering Gear and Reciprocating Pumps
Steam steering engines (observed on all ships except the REISS, which
has electrohydraulic gear) and reciprocating pumps require lubrication
of exposed parts (e.g., piston rods), and thus experience some drippage
of oil to the deck,and then ultimately to the bilges. This is a
typical source on all the ships observed, but because of careful house-
keeping was contributing nothing to bilge contamination. For example,
all obvious drips were seen to be served by ex-coffee cans that were
presumably emptied (into a slop oil container, into the cargo, etc,
but hopefully not into the bilge sump) periodically by the crew. But
there is obviously the possibility of neglect and of spillage from the
many open containers of oil.
Drippage From Oil Burners
Burners being removed from the furnace and serviced may drip fuel oil
along the firing aisle. However, there was no evidence of bilge contami-
nation from this source on the ships examined, apparently due to careful
housekeeping practices by the crew.
21
-------
Reciprocating Propulsion Engines
The conventional (i.e., non-uniflow) steam engine, as found aboard the
CADILLAC in the present study, is lubricated by once-through supply of
oil to its many sliding and rotating bearings. The oil is allowed to
drain to the tanktop, whence it flows to the bilge sump, aided by water
leakage from the engine. The quantity involved can be estimated from
lube oil consumption rates, as is noted below.
Both the CADILLAC engine and the uniflow SNYDER engine discharge cylinder
lubricant via the exhaust steam. This oil does not appear in the bilges,
but goes directly overboard via the condenser discharge.
An estimate of the oil entering the Lakes from the two reciprocating
engines can be made from consumption figures supplied by the respective
chief engineers, checked against annual purchase by Cleveland Cliffs.
The following rates of consumption were supplied by the chief engineers
of the CADILLAC and SNYDER: CADILLAC, 5 quarts per day cylinder oil,
10 gallons per day bearing oil; SNYDER, 9 gallons per trip cylinder oil.
If it be assumed that the operating season is 270 days per year, and
that the SNYDER's 9 gallons per trip is approximately lh gallons per
day, the annual contributions are easily computed. There is a third
ship powered by reciprocating engine, the CHAMPLAIN, in the Cleveland
Cliffs fleet. Since its engine is four-cylinder, its consumption may
be estimated at 4/3 that of the CADILLAC (3 cylinders). The calculation
can then be completed to give the following results:
Cylinder Oil Bearing Oil
(via condenser) (via bilges)
CADILLAC 340 2700
SNYDER 405
CHAMPLAIN 450
1195 gal/year
Purchased, 1972 HQ5
22
-------
The last line is abstracted from the Cleveland Cliffs purchase summary
for 1972, and shows fair confirmation of the estimate. (It also suggests
that I have overestimated the influence of the additional cylinder on the
CHAMPLAIN.)
A more tenuous estimate can be made of the oil discharged by the total
of reciprocating ships on the Great Lakes. As of 1970 there were
reported (A. T. Kearney Co. report to U.S. Maritime Administration,
Research Prospectus for Marine Pollution Control on the Great Lakes,
1972) to be 190 ships of the U.S. and Canadian Lakes fleets with recipro-
cating engines. If it can be assumed that the ships named above are a
representative sample, then the estimates are
Cylinder oil 1105 x 190/3 = 69,700 gallons/year
Bearing oil 5500 x 190/3 = 348,000 gallons/year
Because of the large uncertainties in these estimates, they should be
regarded as very rough indeed.
23
-------
SECTION VI
BILGE WATER SAMPLES
INTRODUCTION
As an adjunct to the work discussed in preceding sections, samples of
bilge water, and of water contributing to the total bilgewater problem,
were collected, since it was possible to extract only a few samples
from each ship, the samples are intended to be representative of what
may be found, and no statistical implications are intended (e.g.,
samples from bilge pump discharge do not necessarily represent average
conditions.
Several of the samples represent condenser overboard discharge from
reciprocating-engined ships (CADILLAC and SNYDER). Although this is
not truly bilgewater, it is known to contain oil originating from
engine cylinder lubrication. The samples are therefore included.
SAMPLE IDENTIFICATION
Turbine Steamer PONTIAC
1. Engine room bilges just forward of reduction gear. Water in this
area was originating principally from the main condenser vent.
Vessel in Cuyahoga River.
2. shaft alley bilges just forward of bilge sump. Representative of
water as it enters sump. Vessel in Cuyahoga River.
3. Cuyahoga River water. Taken for comparison with other samples,
particularly 1. and 4., and for other vessels sampled in the river.
4. From tank top at near main condensate pumps. Principal source of
water appeared to be vent from nearby cooling water pump. Source
of oil in this water not definitely identified. Vessel in Cuyahoga
River.
24
-------
Reciprocating Steamer CADILLAC
1. Condenser overboard discharge (should contain engine cylinder oil).
Vessel in Cuyahoga River.
2. Bilge pump discharge. Principal source of bilge water at the time
of sampling was propulsion engine drainage. Surface of bilge sump
visually much oilier than this sample. Vessel in Cuyahoga River.
3. Surface of bilge sump. This sample lost.
4. Boiler room tank top. Source of water unknown.
Reciprocating Steamer WILLIAM P_ SNYDER
1. Condenser overboard discharge (should contain cylinder oil). Vessel
in Cuyahoga River.
2. Bilge pump discharge. This sample should be relatively clean, since
operators periodically skim oil from the bilge sump before the gross
film reaches pump suction level. Vessel in Cuyahoga River.
3. Boiler room bilge sump. Vessel in Cuyahoga River.
Diesel Ship RAYMOND J REISS
1. Detroit River water. For comparison with bilge samples. Vessel off
Monroe.
2. Engine room bilge. Taken outboard of aft end of propulsion engine.
Bilges throughout flooded and very oily. Vessel in western end of
Lake Erie.
3. Bilge pump discharge. This sample and 2. also may be contaminated
with detergents because of recent attempts of crew to clean up
spilled oil. Vessel in western end of Lake Erie.
Turbine Steamer WILLIAM MATHER
1. Bilge pump discharge. Vessel in lower Detroit River, off Grosse
Isle.
2. Detroit River, taken from a sea chest drain. Sample may be contami-
nated by sea chest filler remaining from winter layup. Vessel in
25
-------
lower Detroit River, off Grosse Isle.
3. Bilge sump. This sample skimmed from tops of the sump, and so should
be oilier than sample 1. Major contaminent appeared to be stern tube
lubricant. Vessel in western end of Lake Erie.
4. Stern tube leakage. This is lake water, but may be contaminated with
stern tube lubricant. Vessel in western end of Lake Erie.
5. Engine room bilge. Taken near main condensate pump, and principal
water source appeared to be this pump. Vessel in western end of Lake
Erie.
6. Drip from contaminated drain tank drain. Vessel in western end of
Lake Erie.
SAMPLE COLLECTION AND ANALYSIS
All samples were collected in new 32-oz polyethylene laboratory bottles.
Bilge pump discharge samples (Cadillac 2, Snyder 2, Reiss 3, Mather 1)
were drained from bilge sump casing vents. Condenser overboard discharge
samples (Cadillac 1, Snyder 1) samples were drained from condensate pump
vents. River and lake water samples (Pontiac 3, Reiss 1, Mather 2) were
drained from cooling water piping within the engine room. Mather 4 and 6
samples taken by holding bottle into continuously running streams of the
water. All other samples taken by dipping bottle into puddle; in the
case of sump samples (Cadillac 3, Snyder 3, Mather 3), they were skimmed
from the surfaces.
Analyses were performed according to the methods in "Standard Methods
for the Examination of Water and Wastewater," 13th Edition, 1971, A.P.H.A.,
save for the TOC determinations which were done using the sealed ampule
oxidation technique using an Oceanography International Corp. system.
TOC numbers obtained in this way should be as good, if not better than
those obtained by the injection method.
Analytical procedures in "Standard Methods" are quite comparable to,
although not in all cases exactly the same as, those in the EPA manual.
Results of anlayses are summarized in Table 1.
26
-------
Table 1. SAMPLE SUMMARY
Sample
Mather #1
Mather #2
Mather #3
Mather #4
Mather #5
Mather #6
Snyder #1
Snyder #2
Snyder #3
Pontiac #1
Volume/ ml.
Initial
700
700
800
700
700
200
1000
900
700
600
Water
700
650
800
600
650
200
950
800
650
600
Description
Bilge Pump Discharge
Detroit River and
Seachest Filler
Bilge Pump
Stern Tube (Lake
Erie)
Eng. Km. Bilge Near
Condensate Pumps
Cond . Sump
Drip from contain.
drain tank
Condenser Overboard,
Cuyahoga River
Bilge Pump Discharge
Boiler Room Sump
Fwd of Reduction
Gear
Remarks
no apparent oil phase;
some large filamentous
(rope-like) particulate
matter
no apparent oil phase;
low cone. sett, sol;
mod . turbidity ; susp .
sol.
no apparent oil phase;
very high cone. sett.
sol ; partic . matter
sim. Mather #1
no apparent oil phase;
low cone. susp. & sett.
sol; some partic. sim.
Mather #1
no apparent oil; poss.
emulsion; mod. susp.
sol; high turbidity
no apparent oil; mod.
sett, sol; low susp.
sol.
no apparent oil; mod.
sett. & susp. sol;
water colored
no apparent oil; low-
mod, sett/susp sol;
some partic. matter
sim. Mather #1
no apparent oil; sus-
pect presence of oil ;
water colored; low sett.
sol; mod. susp. sol.
no apparent oil; high
sett, sol; mod susp sol;
water colored
27
-------
Table 1 (continued). SAMPLE SUMMARY
Sample
Pontiac #2
Pontiac #3
Pontiac #4
Reiss #1
Reiss #2
Reiss #3
Cadillac #1
Cadillac #2
Cadillac #4
Volume , ml .
Initial
600
800
700
800
900
800
800
800
800
Water
575
800
700
800
900
800
800
800
800
Description
Shaft Alley
Cuyahoga River
Under Mn Condenser
Lake Erie near
Mouth Detroit River
(cool sys.)
Engine Room Bilge
Bilge Pump Discharge
Condenser Discharge
Bilge Pump Discharge
Boiler Room Tank Top
Remarks
oil phase present; mod.
susp. sol; low sett.
sol.
no apparent oil phase;
partic. sim. Mather #1;
mod. sett, sol; high
susp. sol.
oil present; turbid &
highly colored; mainly
w/oil; high susp. &
sett. sol.
no apparent oil; high
susp. sol; mod. sett.
sol.
oil phase apparent;
extremely turbid; oil/
water emulsion/disper-
sion
oil phase apparent;
extremely turbid; oil/
water emulsion/disper-
sion
ext. high sett. S susp.
sol; highly colored &
turbid; black material
present in sample.
no apparent oil; mod.
sett. & susp. sol; some
partic. sim. Mather #1
ext. high sett. & susp.
sol; highly colored and
turbid; no apparent oil
**NOTE: there was no "Cadillac #3" sample.
28
-------
Table 2. TOTAL, FIXED, AND VOLATILE NON-FILTRABLE RESIDUE
Sample
Mather #1
Mather #2
Mather #3
Mather #4
Mather #5
Mather #6
Snyder #1
Snyder #2
Snyder #3
Pontiac #1
Pontiac #2
Pontiac #3
Pontiac #4
Reiss #1
Reiss #2
Reiss #3
Cadillac #1
Cadillac #2
Cadillac #4
Total
(mg/ltr)
1.0
5.5
97.3
7.0
86.0
NR
99.0
15.0
13.0
172.5
14.5
6.0
427.5
12.0
200.0
900.0
127.0
6.8
146.5
Fixed
(mg/ltr)
ND
4.5
12.0
4.0
ND
NR
84.0
10.0
7.0
137.5
1.0
ND
277.5
8.5
ND
ND
97.0
2.0
28.5
Volatile
(mg/ltr)
1.0
1.0
85.3
3.0
86.0
NR
15.0
5.0
6.0
35.0
13.5
86.0
150.0
3.5
200.0
900.0
30.0
4.8
118.0
ND : not detected
NR : not run; insufficient sample
Method: APHA, 1971
29
-------
Table 3. pH, COLOR AND TURBIDITY
Sample
Mather #1
Mather #2
Mather #3
Mather #4
Mather #5
Mather #6
Snyder #1
Snyder #2
Snyder #3
Pontiac #1
Pontiac #2
Pontiac #3
Pontiac #4
Reiss #1
Reiss #2
Reiss #3
Cadillac #1
Cadillac #2
Cadillac #4
pH
6.9
7.0
6.6
7.1
6.4
3.9
6.4
6.2
6.0
6.7
7.2
6.4
10.0
6.5
5.7
NR
6.9
6.6
5.5
Color
5
5
10
10
25a
5
20
10
5
20
20
25
150b
5
125a
125a
10
5
5
Turbidity, FU
4
5
10
4.8
40
6
23
6
8
22
8
9
83
10
48
50
32
8
4.5
Remarks
___
___
___
_ —
color due to oil
___
___
___
___
___
___
___
—
color due to oil
color due to oil
NR : not run
a color by 1:5 dilution
b color by 1:10 dilution
30
-------
Table 4. TOTAL ORGANIC CARBON
Sample
Mather #1
Mather #2
Mather #3
Mather #4
Mather #5
Mather #6
Snyder #1
Snyder #2
Snyder #3
Pontiac #1
Pontiac #2
Pontiac #3
Pontiac #4
Reiss #1
Reiss #2
Reiss #3
Cadillac #1
Cadillac #2
Cadillac #4
(mg/ltr)
3.1
2.2
10.3
2.4
46.7
0.8
7.0
2.9
4.1
10.6
16.0
5.7
77.5
1.6
2.33 x io3
625.0
11.3
4.1
6.3
31
-------
Table 5. OIL AND GREASE
Sample
Oil (mg/ltr) x 10
Mather #5
Mather #3
Pontiac #1
Pontiac #2
Pontiac #4
Reiss #2
Reiss #3
Cadillac #1
4
3
7
0.8
5
131.2
39.2
not detected
32
-------
SECTION VII
APPENDIX
CLEVELAND CLIFFS FLEET 1973
CADILLAC
Champlain
Charles M White*
Cliffs Victory*
Edward B Greene
Frontenac
PONTIAC
RAYMOND H FEISS
Thomas F Patton*
Thomas M Girdler
Walter A Sterling*
WILLIAM G MATHER
WILLIAM P SNYDER JR
Willis B Boyer
TONNAGE
gross/net
9053/6364
8757/5569
9115/5719
11151/7309
11726/8730
7898/5980
7918/5903
8220/6496
9115/5719
9115/5719
13122/9879
8653/6772
8603/6650
8603/6650
YEAR
BUILT
1943
1943
1951
1951
1952
1923
1917
1916
1951
1951
1961
1925
1912
1911
YEAR
ENGINE
1943
1943
1946
1945
1952
1954
1955
1966
1946
1946
1942
1954
1950
1952
POWER
2500
2500
9900
9350
7700
5500
5500
4320
9900
9900
7700
5500
5000
4950
ENGINE TYPE
stm recip(3X)
stm recip(4CC)
stm turbine
stm turbine
stm turbine
stm turbine
stm turbine
diesel
stm turbine
stm turbine
stm turbine
stm turbine
stm recip(U)
stm turbine
Notes: 1. Ships with * by name were converted from ocean service.
Year given in third column is date of conversion.
2. Power is shaft horsepower for turbine ships, indicated
horsepower for reciprocating engine ships, and brake
horsepower for the diesel ship.
3. U = uniflow, 4CC = four-cylinder compound, 3X = triple
expansion.
4. Data on tonnages and powers is taken from the Record,
American Bureau of Shipping, 1972.
33
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2
EPA-670/2-74-054
4. TITLE AND SUBTITLE
•• SOURCES OF OIL AND WATER IN BILGES
OF GREAT LAKES SHIPS
7. AUTHOR(S)
John B . Woodward
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Michigan
Department of Naval Architecture and
Marine Engineering
Ann Arbor, Michigan 48104
12. SPONSORING AGENCY NAME AND ADDRESS
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
3. RECIPIENT'S ACCESSION'NQ.
5. REPORT DATE
July 1974; Issuing Date
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
10. PROGRAM ELEMENT NO.
1BB038;ROAP 21BBU;TASK 03
11. CONTRACT/GRANT NO.
R802475
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
Sources of bilge water and of oil in bilge water were surveyed aboard
five ships of the Cleveland Cliffs Iron Company. The ships included
two powered by steam turbines, one by a uniflow steam engine, one by
a conventional reciprocating steam engine, and one by a diesel engine
It is found that many sources of bilge water are clean sources.
Although no accurate estimate of the water thus contributed to the
bilges can be offered, it is concluded that diverting these sources
from the bilges could ease the task of separating, storing, and
disposing of oil wastes. Several samples of water were taken from
each ship, and analyzed for total, fixed and volatile non-filterable
residue, color, pH, turbidity, total organic carbon, and oil and
grease concentration.
This report was submitted in fulfillment of EPA Grant R802475 by the
University of Michigan under the sponsorship of the Environmental
Protection Agency. Work was completed as of October 1, 1973.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
*Water pollution, *Marine engines,
*Waste water, Water analysis,
Diesel engines, Reciprocating
engines, Ship turbines, *0ils
18. DISTRIBUTION STATEMENT
Release to public
b.lDENTIFIERS/OPEN ENDED TERMS
*0ily wastes, *Bilge
water
19. SECURITY CLASS (This Report)'
UNCLASSIFIED
20. SECURITY CLASS (This page)
UNCLASSIFIED
c. COSATI Field/Group
13J
21. NO. OF PAGES
42
22. PRICE
EPA Form 2220-1 (9-73)
34
•&U.S. GOVERNMENT PRINTING OFFICE: 1974-757-58V5329 Region No. 5-II
------- |