The eternal infernal DTC 36

[fasterlaster]

For years my 94 Suburban (6.5 Turbo) has experienced an intermittent DTC 36. It comes and goes, and for a very long time seemed somewhat speed dependent, that is to say at about 55 mph it would typically trip. After trying various snake oils over the years some of which helped a bit for a short time I finally bit the bullet and picked up a rebuilt IP from SS Diesel, and installed it.

All was well for about a day, but after the first day of typical driving around it started to trip again, and now it stays on pretty much all the time. Working with SS Diesel initially we tried removing the filter harness, and did another sweep of the engine grounds. Neither of these seemed to help.

Now the potentially strange aspect. I have been monitoring the engine stats with GMTDScan program. As the Suburban sits now at idle I can watch the Injection pulse width stay at about 2ms when suddenly DTC 36 (long pulse width) triggers. Reading the shop manual it states that a 2.5ms duration is what should trip this error...so what's the deal?

Some theories:

Something amis with the ECU causing false positives.
Bad wiring somewhere which moving and installing a new IP made worse.
A bad ground somewhere I have not found (I checked and cleaned the one in the back of the engine, and the two front to the batteries)
A bad rebuilt IP (I am doubtful of this one, since this pump is throwing the exact same error as what it replaced, but more frequently.)
A gremlin with a desire to see me sell the beast.
Any other suggestions welcome.

[fasterlaster]

Two steps forward, one step back, some progress made. I spent the last few days going over every ground wire point I could find, unfortunately the performance of the truck continued to degrade, with trouble starting. I observed that it seemed to start fine when it was hot, but anything cooler, it would be a real bear to start. On a whim I unhooked the coolant temperature sensor and suddenly the problems went away. Easy to start with no SES light at all.

Now it gets interesting. I make the assumption that the sensor is bad, and replace it......but no luck. Hooking the new sensor back up brings up the same behavior as before, with SES light and poor running. Here is what I have done so far. The sensor has two wires. One ground wire which is common with multiple ground wires from other sensors. I verified that I have connectivity between these grounds by checking in between the multiple sensors wires and see less than 0.001 ohms. For the other coolant sensor wire I checked it back to the ECU with less than .001 ohms from the sensor.

The big question is does this indicate a fried ECU? It will run with the coolant sensor disconnected, but hooking it back up will cause it to run very poor, and the SES light to come on. Thoughts?

[Robyn]

What is the TDC offset ??? should be somewhere between -1.25 and
-1.94

Unhook and spray contact cleaner on the two large plugs at the rear of the engine.
These are the plugs on the harness that goes under the manifold.

Make sure the Ground on the PMD harness is connected to the IP as it was from the factory.

Unhooking the coolant temp switch tells the ECM that its very very cold outside and advances the timing waaaaaaaaaaaaay up (much Rattle)

How old is the PMD ??? A PMD with issues can cause a 36 to set.
Its not a normal occurance, BUTTTTTTTTTTTTTT it can happen.

Another area to check is battery connections.

The 94 trucks feed the ++++ to the chassis from the RH battery through a #2 cable down past the engine and then up to the junction block on the firewall.

The alternator is connected to the LH battery +++.
If there is a ground issue at the LH battery it can cause several issues.

These cables can be bad internally and never show.
What happens is that the core of the cable near the connectors corrodes and depending on the type of cable and if it was soldered can render the cable JUNK.

I have see cables that look great even at the side terminal have a totally corroded core and very high resistance.

Having ground cables bad can really make you chase your tail :eek:

Be sure that with the rig running that you have 14V at the LH and RH batteries. a difference of a couple tenths from side to side is no biggy.

If you have 14V on the LH side and 13V on the RH there is a problem.

Any major difference between the two is reason to suspect troubles.

A bad ground at either battery or the engine block can make things get real hinky.

Make sure that you have a good ground strap from the BODY to the frame and to the batteries.


The RH battery feeds the chassis and the starter and the LH one is the source for the alternator feed and it adds Amperage for crankng.

A poor crossover cable will cause low voltage at the RH battery and low voltage is a killer on these systems.

Another ground point is at the rear of the engine on the RH side (passenger) Ther are a few grounds from the main harness that connect to a bolt on the intake manifold.

Make sure these are all intact and secure with good clean contact.

Wrangle this stuff all out and if things are still hinky the issue may be the ECM.

ECM's do not fail that often.

Poor grounds and voltage issues can and do cause real grief with these systems.

Anything that has a grounding issue and is backfeeding through the system can really louse up what the computer wants to see.

Check this stuff out and see whatcha get

Missy

[fasterlaster]

Thanks for the reply Robyn. Slowly working though these, see if anyone can see anything I missed.

What is the TDC offset ??? should be somewhere between -1.25 and
-1.94

Offset is -1.5

Unhook and spray contact cleaner on the two large plugs at the rear of the engine.
These are the plugs on the harness that goes under the manifold.

Pulled these apart, and also did a connectivity test for all leads to the sensors/whatever they went to.

MMake sure the Ground on the PMD harness is connected to the IP as it was from the factory.

The harness ground is grounded to the PMD cooler, and I have added a jumper wire from this ground point to the IP

How old is the PMD ??? A PMD with issues can cause a 36 to set.
Its not a normal occurance, BUTTTTTTTTTTTTTT it can happen.

I have a known good PMD which I swapped out, with the exact same behavior

The 94 trucks feed the ++++ to the chassis from the RH battery through a #2 cable down past the engine and then up to the junction block on the firewall.

The Ignition voltage in GMTech shows 13.3V

The alternator is connected to the LH battery +++.
If there is a ground issue at the LH battery it can cause several issues.

These cables can be bad internally and never show.
What happens is that the core of the cable near the connectors corrodes and depending on the type of cable and if it was soldered can render the cable JUNK.

I unhooked the cables and checked resistance, nothing found. These cables are actually less than 4 years old as well, so I suspect they are fine.

Be sure that with the rig running that you have 14V at the LH and RH batteries. a difference of a couple tenths from side to side is no biggy.

If you have 14V on the LH side and 13V on the RH there is a problem.

Across both polls on both batteries were ~13.3

Another ground point is at the rear of the engine on the RH side (passenger) Ther are a few grounds from the main harness that connect to a bolt on the intake manifold.

Make sure these are all intact and secure with good clean contact.

I pulled these and cleaned them with contact cleaner including the studs they go down on, but no change observed.

Wrangle this stuff all out and if things are still hinky the issue may be the ECM.

ECM's do not fail that often.

Since my last post per the shop manual I ended up springing for an ECM, unfortunately the same behavior remains. DTC 36, but the GMTDScan software indicates that the injection pulse is rock solid at 1.95ms

Poor grounds and voltage issues can and do cause real grief with these systems.

Anything that has a grounding issue and is backfeeding through the system can really louse up what the computer wants to see.

This really seems to be that way, but I can't for the life of me find what the problem is. This same behavior would occasionally come/go before swapping out the IP, now it is consistent. So either the IP is bad in some way, or swapping it out has disturbed more of whatever wire is the problem.

Thanks for the help and suggestions.

[JohnC]

'94 HD (VIN F) engines had a unique injection pump and required a matching chip in the ECM. All LD (VIN S) '94's and all later engines regardless of VIN were compatible with each other and with all "upgraded" '94 HD's. Are you sure the pump and ECM are compatible?

[fasterlaster]

'94 HD (VIN F) engines had a unique injection pump and required a matching chip in the ECM. All LD (VIN S) '94's and all later engines regardless of VIN were compatible with each other and with all "upgraded" '94 HD's. Are you sure the pump and ECM are compatible?

I don't know. How could i verify this? While I did change the ECM, I kept the same chip that was in the truck.