I have a customer unit, brand new under warranty, that is failing to start after it sits a week or so. We suspect possibly a suction leak. I am thinking of taking the ends off the suction line and pressurizing with a slight bit of air and looking for diesel leaks and/or soap bubble checking the whole thing. Any wise tips and tricks would be appreciated. This unit is some distance from the shop.
I have used about 5 psi of air to test fuel system suction side integrity. Cap the line to the injection pump and pressurize the pickup tube end. Just had one this week with a fuel water separator bowl leak. It would lose prime, but did not leak fuel. Pressurizing the circuit revealed the culprit. I have also used clear lines as well. Place clear plastic lines on the suction and return sides with vertical loops. Run unit to fill the lines. While running see which has air in it, if any. Let sit then see if one of them is empty.
I can't add anything........MG covered this well!
Agree all the way, who's warranty is it under? why not let the major corporation pay for the repair?
I'm the one performing the warranty under flat rate. Which is why I ask here. I have already been out on it once. It seemed at the time that it was just never primed, brand new. But then a week went by and this happened.
It this a single unit or one of multiple units pulling from the same fuel source?
I always scrutinize the water separator and priming pump. So many issues arise from a rolled o-ring or flat used o-ring. Have also had bad crimps on brand new outta the box filters. The pressurised low side has always revealed itself. The only difference between my way and MG2361, I only cap the return to tank. That way I get all the emissions and possible flame start solenoid or any other consumer that might introduce air. Good luck.
Do we get extra points if we guess what you are working on? I use an old gear transfer pump driven with a drill and a liquid eye to check for air . Hook it up at the transfer pump intake on your motor. Bob
It is a Mitsubishi 4 cylinder generator motor pulling from a tank right under it. Bosch style A pump. I have been meaning to get some clear hose on the truck, will pick up some tomorrow to help with the job.
Clear hose is usually the easiest but some times i cheat and put shaving foam on connections to see if it gets sucked in with the air.
Got out on the job today. There was an updated flash so installed that first, it starts with a much more aggressive fuel position. At least it started but it started terribly again, seemingly one cylinder at a time with shaking and smoke. I am sure the customer will not be real happy. After it got started, it was very clean each time after that. Installed clear line on the return, from a distance you would say it was 100% solid fuel but up close with a flashlight you could see very fine bubbles whizzing by. The thing has a check valve at the top of the dip tube, not leaking at all, zero fuel drops back to tank. Pulled dip tube, same deal, it's clear plastic and does not lose any fuel back to tank over any amount of time. Pressurized the entire system from dip tube to return with air including transfer pump and main pump, soap bubble check, zero leaks. The only part of the system that is under vacuum is the dip tube, first check valve and a rubber hose between the check valve and the transfer pump. After the transfer pump is pretty significant pressure on these A pumps IIRC and even checked all of that with air pressure, nothing. Pulled a fuel sample, looks perfectly clean. Somebody will go look at it next week and see how it starts after sitting 1 week.
I don’t know for sure, but I don’t think they came with the check valve.?? I’m thinking someone put that inline to stop a drain back problem.?? MAKE SURE IT’S INSTALLED CORRECTLY.. (1 way check valve) Good luck
Yes, the check valve is just a pipe fitting type valve screwed on right above the dip tube. It works perfectly when I was testing it, never drained a drop back into the tank, let it sit a couple minutes and watched. And I was able to re-prime the dip tube just by yanking the pipe up and down until fuel came out so I know it opens and closes without much resistance. Here is a picture of the pump numbers, looks a little different than what I have seen in the past.
That pump should have an overflow valve on the outlet side of the pump.. banjo bolt.. if u have one in your arsenal, replace it with yours and then you’ll have to wait to see if it works..
You mean the bolt with the ball/spring/washer stack relief valve in it, might be leaking away? I may or may not be able to scrounge one of those from an old Isuzu or Mitsu in my yard, are they all the same setting? This one has the injector leak off rail teed into it, so I wonder if that might make it a different size or length.
Post a pic of whatcha got..
Isuzu 6 cylinder. This one looks promising but it is old, mid 90s I think.
Mitsy S3L2 but this one seems to be dead headed with no overflow valve so probably no use.
Isuzu C240. Overflow valve is on top of the filter head with the rubber hose connected to it. Injector return might interfere.
The installer has been back after a week, and reports it just as bad as ever. Smokes and sputters and shakes like it is only running on 1 or 2 cylinders, until it picks up and takes off eventually after several crank attempts. It makes a lot of smoke too because it is a turbo engine with no anaeroid so it is going full fuel on 1 or 2 cylinders for quite some time until it finally catches them all. Once they all catch, it runs fine the rest of the day.
Any fuel knock while cranking. It almost sounds as if you've got an injector leaking off.
I had thought that also. However the original flash only gave it about 10% rack when cranking to make it not smoke. And then was absolutely no smoke or fuel visible during extended cranking. Only by giving it rack by hand did it start up with the same sputter and smoke originally. We have since put the current flash in it. Now it cranks with a healthy rack position and this is what we get. Were you thinking a cylinder leaked full of fuel? Of course it sits a week between exercise so any leaked fuel would have plenty of time to make its way past the brand new rings into the crankcase. It does seem rather like injectors because of long cranking, lackadaisical firing most cylinders at first, and then it just takes off when it is time.
Yes, I was thinking a faulty nozzle leaking and pudding in the pistons combustion bowl. That would be accompanied by a fuel knock once she starts to fire and a smoking hole for some time after running. I think i would have them tested. It's really the only unseen variable. You've investigated the low side , the return valve and the anti-drainback. There's not much else.
The builder has no idea what is going on and they have no individual parts for the pump. They are going to have us install clear hose on the suction and the return and see what happens. Every time I call in I get a different person with a different idea. There is no fuel knock on startup, just like starting an engine that has run clean out of fuel, lots of cranking before any smoke or fire. It has glow plugs that are working. TPG, what can you tell me about this pump arrangement? Something fishy here. The turbo looks like it may have been an add on by the OEM, and the breather separator looks like it came out of a Summit catalog. There is a rubber lined bracket for the slobber tube that is no longer being used. But if that was the case they would have had to change the injection scheme somehow. Either adjust the pump or replace it completely. The transfer pump has "Zexel" and the numbers clearly on the tag. The tag on the IP looks a lot different than I expect on this style pump. The governor housing has a blank plate where the tag would go. Both the IP and gov have the standard -Z- cast into them though. The unit has a grand total of 3 hours on it.
Unfortunately the cure is gonna be, either change the overflow valve to a Bosch 2.5bar opening pressure. Or install a low pressure electric “helper” supply pump.. BETWEEN the tank and the oem supply pump.. To test my theory, BEFORE STARTING, after it sits a week .. Use the hand primer on the supply pump.. pump it up until u hear fuel lifting the overflow valve and fuel returning to the tank.. SCREW IT BACK DOWN.!! “NOW”.. start the engine.. Lemme know what happens..
For all we know, the hand primer isn’t screwed down..??
I use a fuel primer for when I change a fuel filter or run out of fuel. I just use a electric fuel pump for a carb and put one line in through the filler cap in the tank and the other line in the fuel filter port with a air quick coupler and a valve. Run the pump till you hear it coming out of the return and either close the valve or leave it open and on. Fire up the engine. Should be able to rule out a fuel line issue this way.
The hand primer has been screwed down the whole time except when priming. It has another light check valve as it leaves the tank, so I am not sure an electric pump could do much. Because the only place for it to pump is into the transfer pump, fuel filter, and overflow valve and we know the electric pump cannot overcome the overflow valve. Tell me if I'm wrong. Of course it's under warranty and the mfr did not suggest that so we will not get paid for that. One funny thing is that it cranks quite a while before injecting any fuel at all. In most cases of suction leaks I have seen, it will start right up and run on what was already in the pump and then die when the air gets to it. This is the opposite. Since the fittings are high on the pump, where can the fuel go? Is it leaking into the engine oil? Need to take a sample. TPG, what are you saying about the existing overflow? It has a pretty stiff spring and ball in it now, do you think it is a different pressure than what you suggest? Also, I remember you saying that you can decipher any numbers on any pump - do these numbers make any sense? I suspect this is kind of a gray market product, nothing like it in the US and they seem to have added the turbo which means they must have also changed the injection somehow.
Lol.. ur calling me out on the # info.?? Lol.. those aren’t Zexel #’s, sorry..(I haven’t tried) The problem “sounds like” a bleed back problem to me..& the only way I know of solving it is, either find the problem (overflow valve or supply pump, loose connections) or install a low pressure feed pump. AND LIKE U SAID, they ain’t paying for the supply pump. Have u tried using the hand primer BEFORE STARTING.?? THAT WILL get the fuel up into the fuel gallery and allow the pumping plungers to get a full shot of fuel from the get-go, instead of air.. Just make sure to pump it until you HEAR the overflow opening and pushing fuel back to the tank..
On MORE THAN 1 occasion, while doing warranty work, I’ve installed electric supply pumps and filed paperwork for loose connectors..(turn in copper gaskets and Bill for diagnostic time) The electric pumps can b had for less than 20-30.00.. and the customer doesn’t have to know.. All they need to know is YOU fixed the problem.. It’s obviously a manufacturing design problem ..and getting one of those “egg-heads” to admit it is a novel by itself.!! One more thing that’s come to mind.. The supply pump that’s on the injection pump might b having a real hard time overcoming the check valve that’s in the line.?? I understand your frustration, seeing it only does it after sitting a week unused.. so u only get 1 whack at it..THEN have to pack up and go home.. wait a week and go back to see if u fixed anything.. BEEN THERE, DONE THAT..(sux)
Is it possible or common for fuel to leak internally from this style pump down into the oil pan?
No, not with that few hrs..
Is it possible this pump is a clone and/or put together defective? The strange numbers on the side make me wonder where it came from.
The # is a Mitsubishi.. But doesn’t come up in Google.. The common rail pump comes up, 32K65-00010..