Lift Pump Or What???

[WileyCoyote]

My 2001 ETH/DEE with 36000 miles all of a sudden became hard to start--sometimes. Never when cold, but usually after sitting for 10 to 30 minutes after being fully warmed up.



Occasionally, the engine would need to crank for 5-20 seconds and then slowly catch. It would then run fine--no stumbling or any hint of a problem.



My review of topics on the site indicated that my lift pump might be bad, so I bought a new 3990105 lift pump from Cummins Southwest. It installed fine, and the truck was back to normal for a few days. I also changed the fuel filter--nothing bad was found--looked great.



Now the new pump is doing the same thing that was happening with the old pump. Maybe I changed the wrong thing. The questions now seem to be:



1. If the fuel pressure is low, will it hurt to run the truck? Several people have told me that the VP44 has its own low pressure pump and noharm will occur as long as the engine is running. I am scared to use the truck at the present time.



2. The new 3990105 came in a Cummins box loose with no packing or plugs on the inlet or outlet. It may have been a repack of a returned pump. Is this packaging normal?



3. The new pump will sometimes run for 20 seconds after the key is turned on/starter bumped like it is struggling to pump up to pressure. This is definitely not normal.



4. What controls the pump run? Does it have an internal pressure switch or is the switch that commands the lift pump somewhere else in the fuel system?



5. Where is the best test port for fuel pressure? If it is the back of the filter housing what is the fitting?



6. What is the operating plan for the lift pump. It definitely is needed for starting, but what happens after that? Is it supposed to cycle on and off or run continuously?



Thanks for the immediatle help.



Harry

 

[LSMITH]

I can answer a few of your questions. The lift pump is supposed to run continuously once the engine is started. It is normal for the pump to run for 15-20 seconds when the starter is bumped, it is running at half speed however. The pump is controlled by the ECM, it tells it when and how much. There is no pressure switch in the system at all, excess fuel is returned to the tank. Continuous operation with a dead lift pump is harmful to the injection pump, it cannot pull all the fuel it needs for cooling and lubrication the way the fuel feed is set up on our trucks. The best place for you to test the lift pump pressure is going to be on the injection pump itself. There is a schraeder fitting on the rear of the pump, that is where you want to check it. There is another schraeder on the filter housing, and it will tell you what the pressure is like pre-filter, but what is getting to the injection pump is what you need to be concerned with. What you describe could very well be caused by a dead lift pump. Could also be a leaking return line or a bypass valve in the injection pump that is stuck open bleeding the fuel supply back to tank. How to check the bypass valve I do not know. Check the pressure at the inj pump and look for leaks first. Hope this helps.

 

[WileyCoyote]

I am now a little smarter than I was. I used my air conditioner service lines to hook up a pressure gage to the rear of the filter housing.



When the key is turned on, the pump will initially run for about 3-5 seconds and the pressure will go to about 13-15 psi. The pump will then shut off and the pressure will bleed back to zero over a time of about 5 seconds. It will not come on again unless the key is shut off and turned on again (rebooted).



If the engine is started, the pressure at idle will stay about 13-15 psi. My next plan is to plumb a gauge that can be used while driving and monitor fuel pressure at the inlet to the VP44.



This almost seems like there is an intermittant problem with the control module deciding to turn on the lift pump before the engine is started. Also, if sufficient time elapses between initial key on and start, the pressure in the filter housing goes to zero. ???



Also, it looks like the back pressure contoller might be on the banjo fitting where the return line comes out of the VP44. Is this the right place?? Could this be cleaned with compressed air or must a new one be installed??



Thanks for your help so far.



Harry

 

[Evan A. Beck]

Henry,



I have replaced dozens of lift pumps, and built fuel systems of my own, so I think I may be able to help you out here...



First, the fuel pump has two different cycles. Full load, and cranking load. Depending if you are starting the truck, or if it is running, the ECM gives it a different duty cycle, so it can vary the pressure. High pressure while cranking can cause hard starting. The pump operates at full load all the time, except when cranking. If you crank the engine and it doesn't start (bumping the starter), the ECM sees this as a signal for going into "priming mode," which means your lift pump will run for 25 sec be tween cranking cycles. Make sense so far?



The fuel pressure is regulated using an overflow valve, which is the large banjo bolt on the VP44 where the return line is connected. If this is faulty, it can cause loss of prime, and/or low pressure. To check it, remove it from the truck, and stuff the end of it in a hose that is connected to regulated air pressure. I use a small section of 1/2" fuel hose and a gas-type air regulator. The little check ball in the valve should start to open around 11 psi, and be fully open no later than 13 psi. I have seen only one of these go bad on an ISB--low pressure is usually caused by a bad pump and/or filter.



The best way to measure fuel pressure, is to do a driving fuel pressure test. See what it sits at when just cruising along at 60, and do some WOT runs too. I have seen pumps be ok at idle, but give very poor performance under load. Connect your fuel pressure gauge at the test port on the supply line at the VP44.



I like to see a minimum of 4 psi at WOT... As long as it doesn't go below that, your pump should be ok.



As for there being an intermittent failure in the ECM causing power to be interrupted to the pump, try connecting a test light to the pump wiring, and see if it blinks or goes out. Check this one while driving too. Also, be sure the connection between the factory wiring and the lift pump is in good shape... A bad connection can cause the problem you are describing.



Check out this link http://www.dodgeram.org/tech/tsb/2000/Sa_00-11.htm for a TSB regarding hard starting.



Keep us posted.

 

[RJOL]

Excellent summary Evan... This is perfect example why this site is a great place to learn more about your CTD...

 

[Evan A. Beck]

Just glad I can be of some help.

 

[WileyCoyote]

Evan--



Thanks a bunch. I will try all the things you suggest. The current question is: Did the problem I described (long cranking before starting) sound like a bad lift pump to you or something else?



Also, did the packing I described with the new pump sound like a factory pack or a repack at the dealer? I was surprised that the new pump did not have any plugs or other devices installed to keep it cleaner before installation. It was simply loose in the Cummins box with some packing paper around it.



Thanks,





Harry

 

[Evan A. Beck]

Harry,



The packaging sounds normal... Was the paper kind of a waxy, sticky material? If so, that is how Cummins usually packs them. There's usually no caps on the openings...



The problem you described definately sounds like a fuel pump that is on the fritz. If everything in the fuel system checks out ok, then it is time to dig a little deeper, and look at sensors and wiring.

 

[WileyCoyote]

This is stage two of getting smarter.



I removed the return line banjo bolt and hooked it up to a regulated air supply with a low pressure gage.



This banjo bolt has a very small bleed hole with tiny jet installed just upstream of the relief ball and spring. This jet will permit a small flow at all pressures.



I cranked up the regulator. The bleed jet ran all the time, and the ball/spring relief started relieving at about 10 psi. It was very consistent over several cycles.



It seems to be operating normally. However, what the bleed seems to mean is that the VP44 will always bleed to zero inlet pressure after the lift pump shuts down. I suppose Cummins/Bosch have a reason for this.



My plan now is to permanently install a pressure gage to monitor pressures at the VP44. If I see more symptoms, I'll be back.



Evan, BTW, I notice that your signature has the Peterbilt oval. I have a 1995 379 in our business. Great piece of machinery.



Harry

 

[FHawkins]

lift pump

I too had the same problem that lasted for about 8 weeks (60,000 miles). Took it to the dealer under warranty several times without getting the problem solved (they said "some trucks just start slower then others") to make a long story short the truck finally wouldn't start and had to be towed to the dealer. They replaced the lift pump (for the 2nd time) still wouldn't start, turns out the injection pump went out. They replaced it and all is well again.



It sounds like the injection pump may be going bad to me I would check it out closely