MEP-004 run down to idle after 25 minutes, then dies - fuel system clean

[bimota]

Ok, here is another tough one. We have a mission critical MEP-004 that starts and runs fine, so we power up the loads and it runs perfect for roughly 20-25 minutes then slowly drops to idle then eventually dies.

Work done:
complete fuel system cleaned (like three times), pulled float switch from lift tank - changed with another float switch (interesting, one unit had two floats and one had only one float on the float switch) - reset speed switch, checked that fuel was at injection pump, ... all several times.

Our theory was the lift tank was not getting filled via float switch but that changes does not seem to have an effect plus the unit will start right back up without priming via the battle switch. No change in behavior.

Since the injection pump starts and runs the unit fully loaded - how can that fade after 20 minutes? This is a nut cracker problem - need skilled advisory!

 

[Ray70]

I'm not that familiar with the -004 and up but It sounds like you may have a fuel supply problem going to the IP. I've seen other types of equipment with similar problems caused by clogged or closed vents on fuel tanks and/or restrictions in the fuel lines/filters.
Would it be possible to install a pressure gauge close to the input to the IP and see if you loose fuel pressure just before the unit starts to die?
Also, have you verified the issue is not related to the throttle cable or IP linkage? When it starts back up are you back up at 60Hz?

 

[DieselAddict]

Have you checked that excess fuel is being returned to the tank?

 

[glassk]

Sounds like a fuel line collapsing, ??

 

[Stonepicker1]

Any red lights showing on the Test indicator on the control panel when it shuts down ?

 

[s011883]

Have you pushed the reset button on the speed sensor?

It sound like you are just running on the day tank.

 

[R Racing]

I'm thinking its the day tank and or its float in the day tank not telling the pump to fill it.

 

[s011883]

I had the same thing and pushed the reset button on the speed sensor. It fixed my problem. Good luck

 

[s011883]

I had the same thing and pushed the reset button on the speed sensor. It fixed my problem. Good luck

 

[Isaac-1]

I think the MEP-004 uses a Roosa Master / Stanadyne injection pump, these pumps are known to do this exact thing if they build excessive back pressure inside the pump. Pull the timing cover plate off the side of the injection pump and look for "coffee grounds" inside the pump as a sign the rubber flyball holder is disintegrating and likely clogging the glass ball in the check valve on the return line.

Ike

 

[goldneagle]

There is a check valve on the return line coming out of the IP that intersects with the return line coming from the injectors. On my gen set that check valve was clogged and would build up pressure on the IP and shut down. I drilled out the glass marble inside the check valve and now the generator runs great now. This fix was suggested t o me by a SteelSoldiers member.

 

[bimota]

Thanks all. We do have a spare unit so can swap parts but sure would be nice to diagnose prior to swap. The speed switch reset seemed to solve it but then the problem continued. We've cleaned the check valve and i-pump several times. Frustrating because we have to run it for 20 plus minutes just to see if the change did anything.

Looks like we'll need to wire up test lights to float pump and speed switch. Anyone know why the speed switch reset might fix this issue?

 

[1800 Diesel]

Thanks all. We do have a spare unit so can swap parts but sure would be nice to diagnose prior to swap. The speed switch reset seemed to solve it but then the problem continued. We've cleaned the check valve and i-pump several times. Frustrating because we have to run it for 20 plus minutes just to see if the change did anything.

Looks like we'll need to wire up test lights to float pump and speed switch. Anyone know why the speed switch reset might fix this issue?

I don't believe the speed switch has anything to do with a slow rundown, but instead the shutdown time relieves the back pressure on the IP, thus allowing a successful, but temporary restart. Listen to Ike & goldneagle & either clean or drill out the check valve. I had the same symptoms you described & the removal of the ball along with a brief start-up (with the check valve/return line disconnected from the pump) which flushed much cruddy fuel out of the pump body--this resulted in a good running & dependable unit with at least a hundred hours of loaded run time.... That being said I did see others posting valid concerns about the possibility of unfiltered fuel getting to the IP when the check valve is removed. I haven't studied this scenario myself but I do intend to replace the original check valve with another one with a larger passage on the current unit on my bench. I'll do at least 4 hours of load testing to validate the aftermarket part.

Edit-I missed where you cleaned the check valve...try flushing like I did & I forgot to mention that I did run into a similar fuel delivery problem but it was a clogged feed line between the SS day tank & the IP. Since the fuel is gravity fed to the IP, even small pieces of trash can clog this line right at the fitting where the hose connects to the day tank.

 

[smartguy74]

I'm with Ike. My unit did had the same problem. It'll frustrate you to know end until you get those grounds out.

 

[bimota]

So we gave up and switched out a complete IP and check valve from another unit and got this customer on his way. That fixed the unit that was down but now the problem is moved to the other 004 we have. So let's talk more about removal of the check valve, that's interesting - I try to avoid reengineering these things when possible but if it works, it works. Does anyone know the main reason for the check valve? Can we verify a replacement aftermarket check valve? Since the IP runs these under load no problem, does that isolate the run down to IP back pressure from the check valve not functioning properly?

 

[bimota]

Ok, a little manual review is helpful - maybe:
The -24 manual shows the parts and calls the items we call "check valve" as a "tee"

See Figure 64 - Engine Assembly and Fuel Injection System - page 256 on my -24 parts manual TM 9-6115-464-24

So what's up with a check valve being in the fuel system when the manual calls it a "tee"?

 

[Isaac-1]

Removing the check valve is only a temporary fix, the injection pump needs to be rebuilt because the rubber components are deteriorating. Note Stanadyne / Roosa master injection pumps built (or rebuilt with parts) from after 1994 (I think, may be 1986) use a rubber compound that is less prone to deteriorating than the older pumps, they also hold up better to biodiesel.