Onan MDKAW oil leak

[Gilhooley]

I have a long-latency (days) high-flow oil leak on my Onan MDKAW.

Preliminary diagnosis... It leaks all the way down to the bottom of the pan -------> it's the pan

Facts...
○ This began after MONTHS (6... more?) on the hard.
○ I had to get thru some leak-test training from the AF in 1981 but never had to use any of it.
○ The leak demonstrated that when it begins, it is significant... across days. Big mess... PITA cleaning it up.
○ Then I got more serious about diagnosis... I mixed up some UV tracer dye with a quart of oil and poured it in to the fill cap near the non-generator end of the motor.
§ I waited 6 hr... no joy
§ 2 or 3d later, well, the first thing you know old Jed's a millionaire, etc. Catch pan overflowed. Lots of cleanup.
§ But NO obvious leaks glowing anywhere.

○ New experiment... months later..
§ No oil showing on the dipstick
§ Measured how much actual oil was left with a welding rod down the dipstick hole... slightly < 0.25"
§ The pan has only two holes manufactured into it. One in back - plugged, one in front where the Reverso pump connects. No glowing at either, but I need to take a second more caredul look at the one in the back (I'm ancient and used up and am limited how much time I can spend in that position),.
§ I think I should have stopped here..
§ Mixed up another quart of oil/tracer - added it directly to the pan.
§ Days of delay.. nothing
§ Then... it leaked.
§ Nothing glowing found.
§ The ability to see into the space under the oil pan and the built in drip tray is low... mostly the design, partly me. Is there a way to know where the first drops emerge from the pan (assuming it is the pan of course) if it takes a variable number of days for the leak to begin.. but once started, coats the entire drip tray?

○ I have to conclude it's a pan failure.

My questions...
1. Has anyone ever seen a pan in a protected space just fail? I've worked on a large number of engines on tractors, trucks, cars and motorcycles and something just rotting away is 100% unfamiliar.

2. This is an Onan MDKAW... supposedly engine by Kubota. Curious if anyone knows....
i. Is this an aluminum or SS pan?
ii. Is there a known issue with this Onan re: oil pans?
iii. How big a pita is it to pull/swap or repair the pan? The vessel is a Nordic Tug 42.
iv. Is there a reliable repair procedure to the pan? I'm a fairly good machinist... weak welder but know many good ones.
v. Are pans available? I can't find one.
vi. Or.. replace the generator? And repair/sell the Onan.

Any insights into this situation would be appreciated.

 

[twistedtree]

How much are you running the generator, if at all? Does it leak just sitting there, or are you running it from time to time?

 

[O C Diver]

Have you looked at and around the oil filter? What you describe to me sounds like a siphon leak. You fill the oil up and eventually it pushes through the oil pump and comes out at a lower point (the oil filter sits pretty low). The leak stops when the siphon is broken (oil pump intake is above the oil in the pan).

Ted

 

[Blissboat]

Something similar happened to me with a small (5Kw) Onan that was leaving copious amounts of oil in bottom of the genset compartment. It was a mess, all right, and a stubborn mystery that ultimately turned out to be a leaking seal at the end of the power shaft between the engine and the generator. The boat had spent the winter on the hard, not being run, and apparently inactivity had somehow compromised the seal. The leak only appeared as the engine heated up.

 

[Gdavid]

The 9MDKWB that I owned had a cast aluminum oil pan and I am pretty sure that 2 of the motor mounts were build into the pan. I have heard of them corroding through and leaking oil. I don't know my Onan model numbers very well but assume mine would have been older than yours. Mine had a kobota D722 engine. The clearance between the oil pan the the stainless base of the genset was very small and hard to inspect.

 

[Gilhooley]

How much are you running the generator, if at all? Does it leak just sitting there, or are you running it from time to time?

Hi and thanks... the gen was winterized and sitting on the hard for months before the leak began... possibly almost 7 months, so not running.

G

 

[Gilhooley]

Have you looked at and around the oil filter? What you describe to me sounds like a siphon leak. You fill the oil up and eventually it pushes through the oil pump and comes out at a lower point (the oil filter sits pretty low). The leak stops when the siphon is broken (oil pump intake is above the oil in the pan).

Ted

Hi Ted... interesting you bring that up. I just pulled the oil filter a couple of days ago which was motivated by an odd encounter earlier in the month... I was going down the gap between the filter and the wall of the generator box with an inspection mirror.. shining my UV light into the mirror right next to my eye to see if there were any leaks. No reason for me to be looking in there other than completeness.

As the mirror entered that space, three large drops of brightly fluorescing yellow drops fell one after another between my eye and the mirror on the end of a telescoping rod. I could not find anyplace above the splashdown point to emit that! Not just nothing above it fluorescing. Nothing above it to drop it. And lest I seem "touched" a bit, this was witnessed by a life long commercial and fighter pilot who also used to fly Ike around

Further examination of that space revealed zero fluorescing areas but my view was limited by the filter and the tight space. So a few days ago, I pulled the filter. First thing I noticed when it dropped was zero presence of tracer dye in the filter oil (which, BTW, I paid to have changed when it was winterized). Just normally dirty oil.

Since the last round of oil was added via the dipstick hole, and as the engine hadn't been run, and as the oil pan appeared to drain down to near empty (<0.25") I'd be at a loss for this to be a problem up ner the filter mount. Do you see a way for that siphon path to activate without running? I had a problem with a 4 cylinder motorcycle I used to own where, I believe, a worn piston ring allowed pressurization of the sump and shot oil froth anyplace not sealed up really snug, but again, that happened when it was running.

I think you had a great idea there.. appreciate the suggestion.

G

 

[Gilhooley]

Something similar happened to me with a small (5Kw) Onan that was leaving copious amounts of oil in bottom of the genset compartment. It was a mess, all right, and a stubborn mystery that ultimately turned out to be a leaking seal at the end of the power shaft between the engine and the generator. The boat had spent the winter on the hard, not being run, and apparently inactivity had somehow compromised the seal. The leak only appeared as the engine heated up.

Thanks Bliss... appreciated. THAT was exactly my expectation for the source of the leak before I began... an o-ring or a gasket, because that had been my experience for decades. Things don't just OPEN UP without moving. For gaskets and o-rings, I have no qualms at all about using AT-205 because of multiple instances of success. Had the oil pan gasket start leaking pretty strong about 10 years ago and went with 205 again. The leak petered out over hours(?) once I began running it. The gasket apparently re expanded. The good part was that I've changed the oil 2 ro 3 times and still, no leaks.

I regularly turn my main engine 1/3 of a turn or so to keep the gaskets happy.

G

 

[Gilhooley]

The 9MDKWB that I owned had a cast aluminum oil pan and I am pretty sure that 2 of the motor mounts were build into the pan. I have heard of them corroding through and leaking oil. I don't know my Onan model numbers very well but assume mine would have been older than yours. Mine had a kobota D722 engine. The clearance between the oil pan the the stainless base of the genset was very small and hard to inspect.

Man my heart sank when I thought I missed my motor mounts going into the pan! The parts book shows the mounts attach to the block at 4 places. mounting a big vibrating thing to a cast aluminum pan sounds like the designer was asleep on that one.

I do agree that clearance is lousy though. I had a thought about using my borescope to look in there though it only has a white light source... not UV. That would be perfect.

I'm probably wired up to a hard belief my pan failed in some way... getting the pan off the stupid thing sounds like a pushup and a half... hope someone can chime in with a workaround. I'd format out in a second if I felt there was a solid marine services guy to use but my and other's experiences don't let me think too long that will happen. Thanks for your reply.. good context.

G

 

[Lepke]

If the leak is heavy after adding oil but w/o running the engine, then it may be somewhere where the higher oil lever can reach a hole and drips until the oil level is below the hole. If it's an aluminum pan, water in the bottom of the pan can corrode a hole just like diesel fuel tanks. The waste from the organisms that live in the water that creates an acid. Draining the oil with a pump thru the dipstick tube leaves some or all of the water. Draining from the drain plug gets everything.
It also could be a leak from the timing cover, pan gasket, dipstick tube, or rear main seal if the engine is not sitting flat. The rear main seal usually only leaks while running. And there will be some sign of oil at the bottom of the housing where the generator bolts on.
Put some white paper under the whole engine and wait.

 

[O C Diver]

Hi Ted... interesting you bring that up. I just pulled the oil filter a couple of days ago which was motivated by an odd encounter earlier in the month... I was going down the gap between the filter and the wall of the generator box with an inspection mirror.. shining my UV light into the mirror right next to my eye to see if there were any leaks. No reason for me to be looking in there other than completeness.

As the mirror entered that space, three large drops of brightly fluorescing yellow drops fell one after another between my eye and the mirror on the end of a telescoping rod. I could not find anyplace above the splashdown point to emit that! Not just nothing above it fluorescing. Nothing above it to drop it. And lest I seem "touched" a bit, this was witnessed by a life long commercial and fighter pilot who also used to fly Ike around

Further examination of that space revealed zero fluorescing areas but my view was limited by the filter and the tight space. So a few days ago, I pulled the filter. First thing I noticed when it dropped was zero presence of tracer dye in the filter oil (which, BTW, I paid to have changed when it was winterized). Just normally dirty oil.

Since the last round of oil was added via the dipstick hole, and as the engine hadn't been run, and as the oil pan appeared to drain down to near empty (<0.25") I'd be at a loss for this to be a problem up ner the filter mount. Do you see a way for that siphon path to activate without running? I had a problem with a 4 cylinder motorcycle I used to own where, I believe, a worn piston ring allowed pressurization of the sump and shot oil froth anyplace not sealed up really snug, but again, that happened when it was running.

I think you had a great idea there.. appreciate the suggestion.

G

G, I have a MDKAL and did a little looking.

Did you ever fill the oil above the full mark on the dip stick when adding the dye? Short of doing that, I don't think you could generate the siphon effect.

On my generator, all the oil sits in the pan if it's not running. My pan is steel and has only a drain fitting for a protrusion. The motor mounts attach to the engine block. Short of a leak in the pan or over filling the oil level (siphoning through the oil pump), I don't see how the oil could drain down to 0.25" and stop.

Ted

 

[twistedtree]

I think I would start taping blue towels to strategic areas on the engine to try to narrow down the exit point.

 

[Gilhooley]

If the leak is heavy after adding oil but w/o running the engine, then it may be somewhere where the higher oil lever can reach a hole and drips until the oil level is below the hole. If it's an aluminum pan, water in the bottom of the pan can corrode a hole just like diesel fuel tanks. The waste from the organisms that live in the water that creates an acid. Draining the oil with a pump thru the dipstick tube leaves some or all of the water. Draining from the drain plug gets everything.
It also could be a leak from the timing cover, pan gasket, dipstick tube, or rear main seal if the engine is not sitting flat. The rear main seal usually only leaks while running. And there will be some sign of oil at the bottom of the housing where the generator bolts on.
Put some white paper under the whole engine and wait.

Thank you Sir... good stuff. The traditional in-service oil changes always use the reverso unit but during the diagnostic phase, it just comes out without any special prodding. As literally everything does come out, the leak is obviously at the drain hole level, or... a failure somewhere in the lower pan.

The paper suggestion does get at one basic issue... the "exit point" will not be identifiable for more than a short time as the oil will cover the entire leak tray within hours... high-latency (days) high-flow hours. The uv tracer dye serves about the same role as the white paper but in three dimensions. It's been a staple of what sealing parts to change on two of my 4-cyl motorcycles. Actually a bit of a money saver.

Appreciate your comments!

 

[Gilhooley]

G, I have a MDKAL and did a little looking.

Did you ever fill the oil above the full mark on the dip stick when adding the dye? Short of doing that, I don't think you could generate the siphon effect.

On my generator, all the oil sits in the pan if it's not running. My pan is steel and has only a drain fitting for a protrusion. The motor mounts attach to the engine block. Short of a leak in the pan or over filling the oil level (siphoning through the oil pump), I don't see how the oil could drain down to 0.25" and stop.

Ted

Hi Ted.. Never knowingly filled above spec but at this point with all the leaking that's occurred, I don't believe it would matter now.

Since your Kubota engine looks pretty similar to my kubota engine, I think it's reasonable to assume that my oil pan is also steel, so maybe I can get one step closer to which Kubota engine that it is to locate a part more readily than seeking out the Onan equivalent. I'd love to find an accurate cross reference for that. And ANYTHING that provides a recommended procedure/tips for getting at the pan while it'a in the engine room.

As to stopping at 0.25, two things. That's only the result of 2 experiments and given my lack of rigor for that measuring operation, I'd say 0 is within the margin of error for those 0.25" measurements.

Thanks for your interest in my problem Ted... I thought this was going to be an easy/low-frustration operation given my history. "Boat" changes everything.

 

[Gilhooley]

Locating the Kubota engine model for the MDKAW

maybe I can get one step closer to which Kubota engine that it is to locate a part more readily than seeking out the Onan equivalent. I'd love to find an accurate cross reference for that.

My current tack... I've got a detailed list of the engine (kubota) specs for the MDKAW engine. That's seemingly a big fingerprint to zone in on what kubota engine also had those specs in 2004. No luck yet but I ain't that bright and usually take a while to figure just the precise search.

THEN... how do I swap that IN the engine room. Specs say it's a 1000# generator.

 

[luna]

On my 8 & 20 kW Onan gensets, the Kabota engine Serial Number is stamped into that portion of the block that provides support for, but protrudes out past the base of the high pressure fuel injection pump.
Onan may well have applied a little extra paint in this area in an effort to mask the Kabota number.

In both instances the Kabota Serial Number gives me the engines Model Number. A D950 & V2200 respectively, followed by likely a sequence or date code number.

Once you have the engine Model Number you may find like I did, that your local Kabota parts department has no record of this engine, possibly due to an agreement with the guy who likes to apply a heavy coat of paint.

Undeterred, the friendly parts counter guy simply looked up a Kabota tractor that used an identical base engine and sold me the parts I was looking for at about half of the Onan price.

You cannot expect to get the parts to marinize the engine (raw water pump, heat exchanger, etc.) from Kabota as they did not marinize the engine, Onan did.

Your oil pan is likely the same as that used on any terrestrial used engine, so try the Kabota dealer first.

 

[Gilhooley]

Hi Luna and thank you very much. I had stumbled upon the custom that Kubota imprints the model and SN on the valve cover by speaking with one of their distributors, but not that they also add it to the block, which I clearly should have known. The number on the valve cover is about a quarter inch away from the top of the case and is unreadable.

The data on the block will likely be easier though Onan's service and parts catalogs appear to contradict each other. Though maybe the location of the dipstick port is on both sides and one can be plugged based upon installation. Though the fact they're both Onan manuals calls that idea into question. Doesn't matter as much as the reality actually in my engine room though The photo is mine and since my unit is midpoint between yours, I should assume similar placement. Crudely... just to the left of where the oil filter shows in the photo and above the motor mount if I'm reading your direction correctly.

Constant pondering of his has me going back/forth between a pan failure and a break I've failed to detect in the Reverso fill/drain hose. This being the full set of possibilities because of the oil level dropping to <0.25" and there not being anything else down there to fail. Today... I'm on the side of the hose since I just can't see a steel pan failing... unstimulated.. at the 0.25" level and so rapidly that you dump the whole pan in a day or two. I've seen three large home heating oil tanks progress from a first drop to a significant leak and it took literally years.

This process has really eaten my body way faster and resulted in 2 short hospitalizations over the past summer. That position/that engine room is the worst thing I can be doing. If I can all of a sudden magically discover an ethical and proficient diesel mechanic down by my boat, I'll hire then in a flash else screw it... I'll make it my estate's problem. Not holding my breath. Thank you again.

On my 8 & 20 kW Onan gensets, the Kabota engine Serial Number is stamped into that portion of the block that provides support for, but protrudes out past the base of the high pressure fuel injection pump.
Onan may well have applied a little extra paint in this area in an effort to mask the Kabota number.

In both instances the Kabota Serial Number gives me the engines Model Number. A D950 & V2200 respectively, followed by likely a sequence or date code number.

Once you have the engine Model Number you may find like I did, that your local Kabota parts department has no record of this engine, possibly due to an agreement with the guy who likes to apply a heavy coat of paint.

Undeterred, the friendly parts counter guy simply looked up a Kabota tractor that used an identical base engine and sold me the parts I was looking for at about half of the Onan price.

You cannot expect to get the parts to marinize the engine (raw water pump, heat exchanger, etc.) from Kabota as they did not marinize the engine, Onan did.

Your oil pan is likely the same as that used on any terrestrial used engine, so try the Kabota dealer first.

 

[luna]

On your photo, the serial number could be stamped on the horizontal surface of the block just below the air bleed thumb screw at the base of the high pressure pump.

Good luck.