mep803a odd oil leak help

[Ray70]

Definitely not coolant, it behaves totally different when you drip it on the paint.

 

[Guyfang]

Do you still have the top and doors on the gen set? If so, thats the first thing I would do when I am scrutinizing A set for leaks. Yes, its kinda a PITA. but you can see a lot better into dark places.

 

[2Pbfeet]

Do you still have the top and doors on the gen set? If so, thats the first thing I would do when I am scrutinizing A set for leaks. Yes, its kinda a PITA. but you can see a lot better into dark places.

I think given where @jqc99 seems to be at this point, it might be a PITA that might be warranted, too. An impact driver or drill with the right torque settings makes pretty quick work of it, and when you drop the nuts over the air filter for the umpteenth time, you can put in a couple of rivet nuts. If it does turn out to be a fuel leak, a chunk of work is already done.

All the best,

2Pbfeet

 

[Light in the Dark]

He might even be lucky and have one with all captive nuts (like whats on the edge above the fuel fill panel, where access to the backing nut isn't possible). I've had a number of sets come in that were almost entirely captive nuts, which is sooo nice.

 

[Toolslinger]

Yea, but if they're all captive nuts, you don't get to use all the weird shaped 3/8 wrenches you've collected over the years to access some of the nuts....

And then there's the captive nuts that seize and spin, and you get to break things to get in...

 

[jqc99]

Well, I put the dye in Saturday morning and ran it for awhile, shut it down and brought it back into the outbuilding for UV inspection. Nothing found anywhere. I had turned the lights out to make sure it was dark in there too. I checked again yesterday and today. Nothing. Whatever it is it's really slow. I'll keep checking it daily but I won't have any time to start pulling panels for awhile. I may try some dye in the fuel but wonder if I should find another color so I know what's actually leaking if there's a place that could leak either fuel or oil. Frustrating.

 

[2Pbfeet]

Well, I put the dye in Saturday morning and ran it for awhile, shut it down and brought it back into the outbuilding for UV inspection. Nothing found anywhere. I had turned the lights out to make sure it was dark in there too. I checked again yesterday and today. Nothing. Whatever it is it's really slow. I'll keep checking it daily but I won't have any time to start pulling panels for awhile. I may try some dye in the fuel but wonder if I should find another color so I know what's actually leaking if there's a place that could leak either fuel or oil. Frustrating.

On the bright side, you have a data point. No oil leak, yet. That may not be the data point you wanted. Did you double check that you can see the dye, on say the dipstick, or dribbled off the dipstick on a white paper towel? (Control data point, right?)

If it were me, I would have another look tomorrow on the off chance that the leak is caused by some heat cycle (cold / hot / cold). After that, I would be tempted to run one more start, warm up, load run cycle, and then let it cool. If you don't see anything with you UV light, I would be looking at fuel lines, fuel fittings, and fuel filters as a potential source of the leak. You might try blotting with a paper towel at fittings.

I would not use dye in the fuel until you have changed the oil to keep confusion to a minimum.

All the best,

2Pbfeet

 

[Ray70]

On the bright side, you have a data point. No oil leak, yet. That may not be the data point you wanted. Did you double check that you can see the dye, on say the dipstick, or dribbled off the dipstick on a white paper towel? (Control data point, right?)

If it were me, I would have another look tomorrow on the off chance that the leak is caused by some heat cycle (cold / hot / cold). After that, I would be tempted to run one more start, warm up, load run cycle, and then let it cool. If you don't see anything with you UV light, I would be looking at fuel lines, fuel fittings, and fuel filters as a potential source of the leak. You might try blotting with a paper towel at fittings.

I would not use dye in the fuel until you have changed the oil to keep confusion to a minimum.

All the best,

2Pbfeet

Agreed, don't put dye in the fuel yet. at least not that same color.... it will only add to the confusion if you start leaking OIL the day after putting dye in the fuel!

 

[jqc99]

On the bright side, you have a data point. No oil leak, yet. That may not be the data point you wanted. Did you double check that you can see the dye, on say the dipstick, or dribbled off the dipstick on a white paper towel? (Control data point, right?)

If it were me, I would have another look tomorrow on the off chance that the leak is caused by some heat cycle (cold / hot / cold). After that, I would be tempted to run one more start, warm up, load run cycle, and then let it cool. If you don't see anything with you UV light, I would be looking at fuel lines, fuel fittings, and fuel filters as a potential source of the leak. You might try blotting with a paper towel at fittings.

I would not use dye in the fuel until you have changed the oil to keep confusion to a minimum.

All the best,

2Pbfeet

Lots of initial data points. I checked the dye with the UV light while it was still in the bottle first, to make sure it actually fluoresced. After I poured it in and ran it, I checked the dipstick and it was glowing like a glowstick. I also checked to see if the oil fluoresced at all by itself for good measure. I'd rather check too many things than not enough. Good idea about another run. WIth this arctic blast here, the heating/cooling cycle will be more extreme and could help exacerbate the issue.

Agreed, don't put dye in the fuel yet. at least not that same color.... it will only add to the confusion if you start leaking OIL the day after putting dye in the fuel!

I wasn't planning on dying the fuel yet, but if/when I do, I will try to find a different color for the exact reason you mentioned.

 

[2Pbfeet]

Lots of initial data points. I checked the dye with the UV light while it was still in the bottle first, to make sure it actually fluoresced. After I poured it in and ran it, I checked the dipstick and it was glowing like a glowstick. I also checked to see if the oil fluoresced at all by itself for good measure. I'd rather check too many things than not enough. Good idea about another run. WIth this arctic blast here, the heating/cooling cycle will be more extreme and could help exacerbate the issue.



I wasn't planning on dying the fuel yet, but if/when I do, I will try to find a different color for the exact reason you mentioned.

I'm glad you checked before! That makes everything easier.

That dye is detectable at extremely low levels, although I have never tried to see it on green CARC, so I don't know what the cutoff might be there. Did this unit come to you with large oil stains that might be causing slow weeping?

I would run the engine again to be sure, but it sure looks like more and more like a fuel leak from this couch...

All the best,

2Pbfeet

 

[jqc99]

I'm glad you checked before! That makes everything easier.

That dye is detectable at extremely low levels, although I have never tried to see it on green CARC, so I don't know what the cutoff might be there. Did this unit come to you with large oil stains that might be causing slow weeping?

I would run the engine again to be sure, but it sure looks like more and more like a fuel leak from this couch...

All the best,

2Pbfeet

No stains at all when I got it. If there had been, I would have definitely questioned it.


I'm wondering if maybe I spilled some oil in the enclosure at some point and it's just been slowly weeping out all this time. Maybe when I did an oil change. I don't ever recall spilling any, but that doesn't mean I never did and I just forgot because is was no big deal. But are there enough places for the oil to hide and accumulate for it to weep this long? But then how would it be getting on the underside of the block? From the cooling fan? I don't know. I'm just trying to thing of any other possibilities no matter how remote.

 

[2Pbfeet]

No stains at all when I got it. If there had been, I would have definitely questioned it.


I'm wondering if maybe I spilled some oil in the enclosure at some point and it's just been slowly weeping out all this time. Maybe when I did an oil change. I don't ever recall spilling any, but that doesn't mean I never did and I just forgot because is was no big deal. But are there enough places for the oil to hide and accumulate for it to weep this long? But then how would it be getting on the underside of the block? From the cooling fan? I don't know. I'm just trying to thing of any other possibilities no matter how remote.

In my book, oil on the underside of the block/oil pan does not rule out a spill given the oil filter location.

A less than totally tight oil filter, or an over tightened oil filter, or an oil filter gasket with debris on it, or defective in manufacturing, can all allow a weep of oil. Around here, dust is common, so a "weeping" filter tends to accumulate dust. The reason that I asked about the fan blades was to rule out oil spray or misting from somewhere, so I think that what you have is a weep. The next run should help figure out whether it is an oil leak (I would pay special attention to the underside of the oil filter and the right angle oil pan drain fitting / tube) or a fuel leak.

All the best,

2Pbfeet

 

[CallMeColt]

I think it's fuel. Diesel likes to migrate UP CARC from some reason.

I read all three pages.... it will only leak after it has been run & then has sat?

Turn the machine to run & let the fuel pump run for 15 minuets or so. Best to do this with your dye. This will expose any leak at all on the low pressure side with no heat from the motor to burn/evaporate it off. I've had fuel leak from the low pressure hose to the injection pump very slightly and soak the foam at the bottom. Unless the engine didn't get hot, you didn't notice it at all.

It will not smell like fuel because of it's low concentration and all the other grime/much/CARC/oil from 20+ years of being a military generator. Even a low hour, clean generator is dirty.

Just my input. If it was oil, it would find it's way out constantly & you'd see it with the dye.

 

[jqc99]

I think it's fuel. Diesel likes to migrate UP CARC from some reason.

I read all three pages.... it will only leak after it has been run & then has sat?

Turn the machine to run & let the fuel pump run for 15 minuets or so. Best to do this with your dye. This will expose any leak at all on the low pressure side with no heat from the motor to burn/evaporate it off. I've had fuel leak from the low pressure hose to the injection pump very slightly and soak the foam at the bottom. Unless the engine didn't get hot, you didn't notice it at all.

It will not smell like fuel because of it's low concentration and all the other grime/much/CARC/oil from 20+ years of being a military generator. Even a low hour, clean generator is dirty.

Just my input. If it was oil, it would find it's way out constantly & you'd see it with the dye.

The leak is very slow, so I don' t know if it only leaks when it sits. I'll try to find some other colored dye for the fuel and give it a shot. As mentioned before, a different color dye than the oil in case the oil dye happens to show up when I try the fuel test. Then I might not know what's actually leaking. Thank you for the suggestion.

 

[Toolslinger]

I was kinda curious too...
Multiple colors available.

https://tracerproducts.com/uv-fluorescent-dye-multi-colored/