Multiple Injection pump failures

[Neil]

Greetings all,

We are a non-profit operating in rural Haiti and have 3 chevy 6.5 deisels, a 1997 and two 1998's. Fuel for all three comes from the same tank. Service and usage is about the same on all three

In the last year we have had to replace the inejction pump on one of the 1998 models three times. Each time the pump lasts about a month before failing, when it fails the pedal becomes very sensitive. Small presses of the pedal result in huge rpms swings.

At each failed pump we have drained and cleaned the fuel tank, changed the filters etc. We check the timing gears and timing chain, all look good.

Hooked up a scanner and it shows no error codes, the only thing that seems different than the other trucks is that the fuel rate shows as 0 m3.

Can anyone give us some ideas? No chevy dealers here, we do all the work ourselves, but we're pretty fair at it. Been driving chevys here on some of the worst roads on earth for 18 years and never had anything like this happen.

Neil
Haiti Outreach
1997 K2500, 1998 K2500, 1998K3500, 2005K2500

[gmctd]

Try swapping PCMs between the two '98s, see if the symptom follows the PCM or stays with the engine.

What is the source for the replacement Inj Pumps?

Orig should have been 5288, replacement would be same, or latest 5521

[Neil]

Thanks for taking the time to answer.

We swapped out the computers between the two and the engine runs the same, as the pump is bad. I guess we could just leave them swapped and see if it trashes another pump. That would at least help narrow it down. As we're putting in a new pump tomorrow, we'll switch the PCM's and see how it does, but it might be a month before problems show up...

All the pumps came from a Chevy dealer in FL. I don't have the number here but I can get them in the morning.

Thanks again

[gmctd]

No problem, Neil, and welcome to the forum.

Assuming you've also swapped the APP module, you might check the main PCM harness for fraying, where it crosses the a\c pipe out of the lower firewall.

What about the main harness grounds at the back of the engine, around #8 cylinder area - head bolt, intake stud?

Are the replacement IP's rebuilt? New? Furnished with a new IP\PMD harness and PMD installed?

What are the actual DS4 spec numbers - 5288, 5521, ???? ?

Is the Cal Resistor installed in the PMD?

[Warren96]

Changing the PCM is a a good idea,you dont even have to buy it .Have you checked for bubbles in the return line?

[DA BIG ONE]

There is a problem w/harness inside pass compartment at the frt/passenger side where the harness comes through the firewall then goes to computer behind glovebox and can cause all kinds of problems, as follows;

When someone sitting in frt/pass seat places their feet forward then stretches out this stresses the harness to the point of pulling pins out of computer socket plug (this is the 3 sockets that plug into computer) the pins may or may not pull completely from the socket plug. The socket plug w/most problems is the one closest to the glove box door. A non conductive probe can be used to seat all the pins that have been pulled out, if you use anything metal to push pins back be extra careful.

Good luck........

[Neil]

Thanks to all for the suggestions. We will look at them this morning.

Can any of you explain to me how the wiring/computer can ruin the injection pump? I mean these pumps run fine for a month or two then quit completely, we can only get it to run again by changing the injection pump. Once it stops, changing the computer doesn't help.

We have tried changing the "black box" on the side of the injection pump that controls the throttle, no difference.

Thanks

Neil

[Neil]

Ok the pump we are swapping out this morning ends in #5521

I assume all the pumps are new they come from The Chevy dealer in West Palm Beach, FL

Then you have to expain the acromyns to me:

APP module? Is this the "black box" on the side of the pump? If so a new one comes with every pump, we have tried changing it too, after we noticed the problems, no chage.

PMD? Cal Resistor?

Thanks

Neil

[gmctd]

Good - the 5521 is the supreme acme pinnacle of current Stanadyne-Roosamaster IP gloriousness.

The PCM-swapperoo was to see if the fuel rate error followed the PCM or stayed with the Inj Pump - PCM can create the appearance of IP failure, but probably not cause the failure.

In that case, the PCM from the running truck would have resulted in a most miraculous 'healing' of the failing truck.

So - the real problem is in the vehicle wiring or the fuel tank - is the IP furnished with a new harness, and if so, do you install it, or use the original?

How does the Fuel Injector pulse width look?

Should be 1.84ms - compare all parameters to the good truck , immediately upon installation of new IP - Des\Act timing, TDCO, FI pw, fuel rate, idle rpm, etc

Idle rpm will be different between manual trans and auto-trans vehicles - around 600 for auto, 650-675 for manual in the "F" series.

[JohnC]

3 failed pumps with no pump codes? What's the common denominator? Must be the truck...

Try unplugging everything you'd unplug to change the pump, Then put it all back together. Look for miracle cure...

Also, check the grounds at the rear of the passenger's head.

[gmctd]

Yes - always make sure your passenger's heads are fully grounded - reduces static from them, along with back-seat driving.:eek:

However, as we all know is JohnC's intent, check the major PCM harness grounds on the rear of the cylinder-head to your left, as standing at the front and facing the engine bay - in the case of left-hand passenger vehicles vs our familiar right-hand passenger units.

From the files of the lazy-wrench inventor............;)

[DA BIG ONE]

Thanks to all for the suggestions. We will look at them this morning.

Can any of you explain to me how the wiring/computer can ruin the injection pump? I mean these pumps run fine for a month or two then quit completely, we can only get it to run again by changing the injection pump. Once it stops, changing the computer doesn't help.

We have tried changing the "black box" on the side of the injection pump that controls the throttle, no difference.

Thanks

Neil

Again, I would seriously suspect some loose pins within one or more of the 3 plugs you have to remove and replace when changing out computer. When a pin or pins are loose they will only touch the socket it is supposed to fully seat into, as the vehicle is driven the loose pin/pins will vibrate away from the socket it did not seat in causing all kinds of issues. If perhaps the pin/pins that are not seated all the way don't back out or away from socket they can vibrate causing arcing, intermittent issues, or even a no start condition.

Consider, your pumps may not be the problem.

[gmctd]

My apologies - overlooked this

APP - Accelerator Position Potentiometer, the module on the accelerator pedal

FSD\PMD = Fuel Solenoid Driver \ Pump Mounted Driver, the module on the side of the Inj Pump

Cal Resistor - inside the PMD connector with a new Inj Pump, specific to that IP and must be installed in any replacement FSD

[Neil]

The PCM-swapperoo was to see if the fuel rate error followed the PCM or stayed with the Inj Pump - PCM can create the appearance of IP failure, but probably not cause the failure.

In that case, the PCM from the running truck would have resulted in a most miraculous 'healing' of the failing truck.

So - the real problem is in the vehicle wiring or the fuel tank - is the IP furnished with a new harness, and if so, do you install it, or use the original?

How does the Fuel Injector pulse width look?

Should be 1.84ms - compare all parameters to the good truck , immediately upon installation of new IP - Des\Act timing, TDCO, FI pw, fuel rate, idle rpm, etc

Idle rpm will be different between manual trans and auto-trans vehicles - around 600 for auto, 650-675 for manual in the "F" series.

No miracle cure by swapping out the PCM...we tried that but the pump was already shot.
We have never received a new wiring harness always used the old one.
Just got everything back together new pump and new computer behind the glove box:

Fuel injector pulse width:
Des/Act Timing: 12.3 deg / 12.2 deg
TDCO: 0.0 deg
FI pw: I don't see this on our Matco scanner...Pro Scan Plus
Fuel Rate: 8.75 mm3
Idle RPM: 692 rpm (manual Trans)

Engine is currently running fine, but then it always does for a month or so...We will continue to monitor it.

Thanks again for everyones input,

Neil

[Neil]

GMCTD,

Thanks for clearing that up

we did try and change the APP and the FSD to no avail. Niether made any difference in the way the truck ran, The only way to get it going again is to put a new Injection pump in it.

My concern is that we will have yet another pump fail and have no idea why...I just don't understand how a wiring problem or computer problem could do physical damage to the pump. I understand how it can destroy the FSD.

In any case we'll continue to watch it and I'll report back

Thanks

Neil

[DA BIG ONE]

Someone streching out the diesel supply using kerosene or worse? Until you can be sure its not a diesel fuel mixed w/who knows what I'd be using lots of lubricating/c-tane additives.

Someone correct me here if I'm wrong, when I researched different fuels that can be used in the 6.5 only the military IP pumps and the later w/ceramic rollers DS4 pumps were up to the task of running kerosene, and other mixtures for short periods, or?

Anyone open the pumps to determine the failure?

[gmctd]

Loose or bad connections can result in same symptoms as a failed device, so wiring is as important to the system as the PCM and the connected device - you've eliminated the PCM by substitution.

I was refering to the short harness on the Inj Pump, connected to the PMD and the Fuel Solenoid, with the black ground wire on the top of the IP, as being furnished with each new IP

TDCO should be -0.5deg, but iirc that Matco unit is same as the SnapOn, so no telling what it will read.

Are you resetting TDCO after each replacement?

How about FS closure-time - is that listed?

And, all readings should be taken at 190deg F engine coolant temperature, so PCM will not be calling for cold advance and fuel.

The readings look normal except for TDCO, and maybe Des\Act timing, depending on ECT

We need the FI pulse-width or FS closure-time number