The little Yanmar has been a little unhappy lately.

After cruising at 2700rpm for half an hour, engine stalls after throttling back to idle. No problems when underway, only when returning back to idle, as if the idle screw has been backed off too far. Will restart reliably only after a bleed (at the injectors). But the fault will eventually happen again.

Air is entering the fuel system, but so far can't see visible external fuel leaks. I've replaced the hoses to and from the lift pump, put a new fuel filter housing (on the engine) with new banjo washers.

Is it possible the air is entering from the crank case on either the mechanical lift pump or the high pressure pump?

Any help and ideas much appreciated.

Many thanks,
Evan

You could also install a return line from your filter back to the tank. Any air that is drawn into the system will return to the tank before entering the injector pump. A small orifice insert in that return line will ensure some positive pressure to the injector pump.
Putting a return in the system also makes priming/ bleeding filters easy as all you are required to do is pump.
You often see a return line fitted at the filter from where the injector leak/bypass lines( top of the injectors) return to the filter.

Hi Evan
A frustrating problem
But to answer your question
Is it possible the air is entering from the crank case on either the mechanical lift pump or the high pressure pump?

It would not seem so.

The parts list doesn't show a replaceable diaphragm , but the rubber diaphragm would be the only thing in the pump that would give out
The result of a leak in the crankcase side of the mechanical lift pump (fuel feed pump in the parts list) would be fuel in the crankcase as well as air leaking into the line. Do you have noticeable fuel in the crankcase? I have no idea how you would determine this, Smell?
j30.us/files/Yanmar-2GM20-Parts-List.pdf (page 37)

The fuel pump (page 36) has some seals (parts 6 and 13) which would allow a leak back into the crank case but as with the fuel pump such a leak would result in fuel in the crank case as well as air in the line.

So sit down with a pencil and paper and draw a schematic of your complete fuel system.
Note every joint and connection.
Go back to the boat and with the drawing, check everything again,
As you said Air is entering the fuel system,
It will turn out to be simple ,(I hope)

All the best

gary

Done a little bit more looking around but not there yet.

What do you gents think of the following?

-I've cracked an injector line open at the injector at idle, one at a time. Engine loses about 300 rpm, but still runs. So I'm thinking the problem is not after the high pressure pump. I'm not sure about what you mean Ramona by the paint comment, but I don't think the problem is here.

- I've got a rubber primer bulb out of the tank. It took 2 squeezes to charge the system. I thought this was odd. Your thoughts? Kept pumping to pressurise. Looked and couldn't see or feel any leaks up to the mechanical lift pump.

-Can't tell if fuel is mixed with oil. Nothing obvious. It seemed a touch over the full mark on the stick but that could be me. Will keep an eye on oil level rising, indicating fuel in the oil.

-Tank is full way above the pickup so, couldn't be picking up air from a pin hole etc in the pickup tube.

Not winning.

Ps: Thanks for the drawings and manual, very handy.

One of the Facet type electric pumps or even an outboard motor fuel squeeze bulb pump added into the system next time you have the pipes apart. just use a jumper lead to power the pump while you bleed, no need to run it full time. Pumps are about $14.

Sometimes small particles of engine paint get lodged under seals etc and allow the fuel to leak down and just a tiny bit of air to enter the system. Cleanliness is everything.

GusTee said..
Done a little bit more looking around but not there yet.

What do you gents think of the following?

-I've cracked an injector line open at the injector at idle, one at a time. Engine loses about 300 rpm, but still runs. So I'm thinking the problem is not after the high pressure pump. I'm not sure about what you mean Ramona by the paint comment, but I don't think the problem is here.

- I've got a rubber primer bulb out of the tank. It took 2 squeezes to charge the system. I thought this was odd. Your thoughts? Kept pumping to pressurise. Looked and couldn't see or feel any leaks up to the mechanical lift pump.

-Can't tell if fuel is mixed with oil. Nothing obvious. It seemed a touch over the full mark on the stick but that could be me. Will keep an eye on oil level rising, indicating fuel in the oil.

-Tank is full way above the pickup so, couldn't be picking up air from a pin hole etc in the pickup tube.

Not winning.

Ps: Thanks for the drawings and manual, very handy.

Have you thoroughly inspected the rubber prime bulb, both the connections and the bulb itself?

Good Morning Evan
RE: Interesting find along the way: Banjo bolt out of the lift pump has a tiny centre hole, max 2mm diameter, where the fuel is fed to the filter on the engine then to the high pressure pump. Looks like a restrictor.

This is a standard Yanmar part. (page 37 No 4 . BOLT,PIPE JOINT).
No restriction there.
I know that 2 mm seems small, but these engines are not thirsty at all.
When trouble shooting an engine it is a good thing remember that the engine worked well before and to look for things that have altered.

Gary

Check this out ,, It explains leak by

I might give the bypass mod a miss for now then. My bypass line was quite dry. It lead me to think that there was no constant leak like the video, but was used for another purpose, perhaps a bypass back if the pintle/nozzle is faulty to relieve the pressure on the main pump and lines.

I've had another run with the 2GM, and she's been starting extremely well and hasn't missed a beat. Looks like the war might be over...