Mystery fuel leak. Yanmar 3GMD Engine.

[Ziggler]

1986 Morgan 323. Absorbant pads under original engine saturate after 10 days whether at rest or engine in use. Inspection of fuel line all the way to the injectors shows no leakage. But looking underneath engine, at the 'sump pan(?)' (bottom most part of engine) reveals fresh fuel dripping from the bolts that secure the pan to the engine.

QUESTIONS:

1) Is it even theoretically possible that fresh fuel would be present in the sump pan and therefore able to leak from there? Trying to determine if leak HAS to be higher up and flowing down to the lowest point?

2) If dripping from the pan what horrors will I encounter removing the pan?

I can changed filters, impellers, anodes... but removal of the bottom of the engine seems bigger than my skillset.

 

[MikeHoncho]

Don’t take the oil pan off. It’s coming from somewhere else. Get the dye kits for tracing leaks that uses a black light. It will be pretty obvious with it

 

[jssailem]

Mike is spot on Ziggler. Focus on the top of the engine, along the path of the fuel lines from the tank thru the filters to the engine injectors. Clean and wipe dry all of the engine with lots of rags. If you have a seeping line as much as you say it will become obvious. Dispose of used rags properly. Fuel soaked rags are a smelly hazard.

Leaks often are a loose fitting or a worn washer or compression fitting. When you service your fuel system it is advised to throw the old washers and compression fittings away and replace with new ones. In the rush to get the boat back in use, many owners forget this essential maintenance step and just put the "it still looks good" washer back into the system. This means the "next Guy" will find the seeping leak.

Compression fittings


and copper Banjo bolt washers...

 

[kloudie1]

May be coming from the mechanical fuel pump. Carefully tighten the screws holding the top half onto the bottom and then check again.. a small 90 degree phillips driver is helpful for that.

 

[capta]

It is possible, though unlikely, that you have diesel getting into the oil and therefor the pan, but if so, you should notice it when you check your oil level. I had it happen once on an engine that used a bit of oil. For some reason, I stopped needing to add oil to the engine. Eventually, the oil level rose above full and I got suspicious.
However, you really need to go through the fuel system a lot more carefully as the leak is probably right there in front of you. Hose clamps can cut fuel lines, flair fittings can get tiny cracks in the copper lines, fuel pumps and injector pumps have gaskets, etc.
However, if you are having no starting problems, it would indicate the problem may be in the return line because if the fuel was leaking out of the feed line, you might have air in the line, meaning the engine might stutter a few times before starting.
Good luck.

 

[dlochner]

Look very carefully at the hose fittings. The hose ages and fuel seeps out, a drop at a time. Pay particular attention to the return fuel line, it seems this fitting can allow fuel to siphon out.

Also check the bleed screws. It seems some Yanmars can run even with the bleed screw a little loose.

 

[MitchM]

1/ when i was trying to track a pesky oil line leak on a yan 3JH4E,w engine-off i wiped clean every oil-carrying line , then used blue tape to tape kleenex at every suspicious joint along the oil-bearing line. within an hour a tiny tiny leak was obvious by discoloration of the kleenex, right where the oil line flared to a greater diameter to lead into the oil cooler. the line had developed a leak at the poorly-designed flare joint (engine had only 200 hours/ 6 years on it.) (and of course the part was special order 3 weeks... ) 2/ in the olden days we'd anneal the copper washers , and let them cool then refit them to the filter to save buying new .

 

[Ziggler]

Gentlemen: You are all too kind. I'm sure you've saved me gads of time and effort. What I've learned from what you shared:

1) There is no way the engine pan can drip fresh (pink) fuel since the pan is full of black oil. I feel silly about this now that I realize my mistake. Thank you for that insight from your experience, capta.

2) The fact that the engine starts no matter how much fuel (and there is a half-cup or so) that is ending up in the oil diapers does seem to indicate that the leak may be on the return side of the fuel line. Otherwise, one would expect that the fuel line wouldn't be primed and would need to be bled to start. I will start my next level of intense inspection on the return side. Thank you dlochner and capta

3) I didn't know there was such a thing as "dye kits for tracing leaks". Thank you, MikeHoncho, I will look at that option if the 'free' inspection options dont turn up anything.

4) I now appreciate the many subtle and specific parts that make up the entire fuel line and, therefore, all fittings and parts that can and should go wonky and need to be replaced. It makes sense to change all such items as a matter of routine maintenance as I change out adjacent parts, etc. Thank you, jssailem

A bit belated, but here are the pics.