Truck started stuttering again

[IBMobile]

Thank you.   Those on signals have me a bit concerned.  

[wil440]

16 minutes ago, IBMobile said:

Thank you.   Those on signals have me a bit concerned.  

I can't see why a FP shutoff would fluctuate like that

[wil440]

I hooked up my Modis to my truck, there was at first start when  cold a spike every say 75 % across the screen, after a short time just flat line no spikes, got the pics the wrong way round, 1st one is running a minute or 2, 2nd is straight after start up, sorry for the rubbish pictures I just couldn't stop the glare from the sun, I did do a record and if you want it I can get it off the modis and post it but the first picture should confirm it should be flat line with no spikes, my second picture has me thinking I have a problem as I cannot see why it should spike without it is a ecm test on first startup to check it is working, ECM could see a rpm drop to coincide with the solenoid cycling and if it does the test is stopped,  if you look at the frequency on mine 2nd pic, it is quite far apart and it was uniform, yours is quite random ish

[Haggar]

Interesting....

@Dripley has my Solus.  Maybe we can get him to look at the same parameter real time on his truck if he has it with him.

 

The only thing I caution, we are watching the output of interpretative software....  It may not always be the real information.  It is bits and bytes (pun intended) of the real info filtered through another software package.  

 

If you suspect it is real (boy it makes ZERO sense) do a measurement of the control wire to the fuel pump relay, does the command signal really fluctuate?  In my small mind, I envision it does not.  (that wire is going to be hard to back probe....  The ECM commands it.   So one end is waaaay down there the other is buried under the PDC.  you could use one of the relay jumper kits.  or you can just hotwire 30 and 87 and test 86 to ground with high speed sampling. )

 

Good luck!   This is very interesting. 

Hag

Edited June 17, 2020 by Haggar
tagging Fail.... still not sure how to get it right...

[Mopar1973Man]

6 minutes ago, Haggar said:

If you suspect it is real (boy it makes ZERO sense) do a measurement of the control wire to the fuel pump relay, does the command signal really fluctuate?  In my small mind, I envision it does not.

 

Fuel pump relay drop is will cut the power to the PSG and the the P1689 code is tripped. 

 

As for @wil440 I thought it was the Fuel shutdown lead. That showed the pulses? 

[wil440]

2 hours ago, Mopar1973Man said:

 

Fuel pump relay drop is will cut the power to the PSG and the the P1689 code is tripped. 

 

As for @wil440 I thought it was the Fuel shutdown lead. That showed the pulses? 

In my 2nd picture I have the wrong one highlighted on the Modis screen. The correct one is the 2nd line down with the 2 distinct spikes

I'm going to check again tonight and watch what happens very closely

[Haggar]

Moparman,

 

We are seeing the theoretical command to the fuel pump relay.   Agreed, if the command goes to off the truck will not run since that is the power to the VP PSG.  Since this is the command to the relay and it is a very short spike the coil could keep the contacts closed (so truck does not shut off) and re-energize before the truck shut down....

 

BUT this has some huge and interesting implications.....  If the coil is de-energized momentarily, there is a back EMF wave pushed towards the ECM.....  This voltage spike can be quite tremendous.

 

 

 

This picture is of an injector.  But the concept is the same.  When the command to the coil is interrupted, a spike of voltage is produced from the collapsing magnetic field.....  (this is voltage, not necessarily amperage) so if we are worried about stray voltages damaging our ECM and if our programming actually turns this on and off (for no apparent reason) this stray voltage has to go somewhere, and back to the ECM it goes....  This is an ECM output, how does it handle a stray input...  Probably not well.    This shows an injector, the coil is being returned by fuel pressure and a spring, so the amplitude is higher than we might expect.  But the similarity is there. 

 

It will be awesome to see if the signal is really there, or if it is a software interpretation.

 

 

Hag

[wil440]

When I checked mine last night the spikes were uniform as can be seen in the 2nd picture it was like that from cold  start and did not change, I looked away for second to turn the interior light off to stop the glare on the screen for a second better picture, looked back and flat line no spikes and at no point did it spike again. This is why  I felt it was a ecm test to check engine stop. Will look close tonight if this storm  stops

[wil440]

I checked it again tonight and it is different from yesterday   Truck had not been run since last night when I checked the first time,  on startup there were random spikes that got less and less the longer the truck ran, also seemed to be able to set a spike with a stab of the throttle especially early on while still cold, it is quite warm and humid here today, Fuel temp was 27 deg C.

As the truck ran it did go mostly flat line but still did have the odd random spike, blew my theory out of the water.... I'm going to check it again tomorrow also.

 

I doubt a multimeter would see it backprobing the correct wire as it is so quick but the scope will on the modis or similar

 

[Haggar]

Yeah a regular voltmeter won't see it, too much averaging and balancing.

 

you would have to use the oscilloscope function to watch directly at the relay coil for the ECM command. 

 

I suspect (truely hoping!!!) it is a false signal.  Something dropping out in the interpretative software of the snap on that no one caught....

 

Very interesting none the less.

 

Hag

[wil440]

I'll check it at the relay but it won't be tonight as it is raining heavily here for the 2nd whole day and truck is outside

[Haggar]

I just had a brain fart. (gas gas gas) I think I feel better, and pretty sure I didn't get any on me....

 

I think I have a possible explanation, but your testing will confirm it.

 

TL:DR  go to bottom, otherwise we are heading into the rabbit hole....

 

We are displaying("seeing") the COMMAND from the ECM to the relay.  What if the relay is internally latched through programming to stay energized until another condition is met.  If this were a PLC program in ladder logic, my line would have a string of conditions that had to be met for the first firing of the relay coil.  BUT i can latch that output with another set of conditions before it will turn off . 

If we were watching the logic run we would never worry that the set of conditions to make that output high (coil energized) happened again.   But IF we were watching this logic program run by looking at the output that commanded the rung to go high, WITHOUT respect that it was already high,  we would see a "new" command to go high any time the conditions were right.

 

TL:DR  we are seeing the command.  The command may be valid at multiple times during running...   SPST switch for starting vs 3 wire start.

 

It is a shot in the dark, but I can see that reverse engineering my PLC programming might work like that.

 

HTH
Hag

Edited June 18, 2020 by Haggar

[wil440]

It took me 2 reads to get that... you say maybe we are seeing a command on top of a command thats already happened ?

 

What does this relay actually do..... bear with me here cos I haven't had the chance to take a look at manuals and wiring diagrams (work hours)  I'm asking because on the fastest key off/on I can manage nothing on the scope changes, is it engine key off or collision off, I checked key off/on as my ign switch isn't the best ?

[Haggar]

It (the fuel pump relay control you are scanning with the snap on) is what the ECM commands to turn on the DSG in the VP.    It is what powers up our VP.

We are seeing the software command for the output from the ECM terminal 36.   The relay coil that it energizes  completes the circuit for fuse 3 in the PDC (which is a constant hot, not interrupted by the ignition switch.) to the VP44 control circuit board.

 

 

 

Thanks for reading it a couple times.....  That is as simple as I could make it, but it is still a weird combination of words to describe.  what is worse is my programming kung-fu is weak, so I may not use the correct words or current words for proper programming.

 

Hag

[IBMobile]

I don't think the injection pump shut off signal is controlled by the injection pump relay but by an off/on signal generated by the ECM.  Look at page 8W-30-31 in the 01 FSM,  there you find pin #33 on the ECM labeled "fuel shut-off signal" with a lb/rd wire going to the fuel injection pump pin #5 labeled " fuel shut-off signal".  

This could be a ground signal generated by the ECM   I tested the signal with key on and engine stopped and got the same random off/on pattern. 

 

One way to test if the signal is generated by the fuel injector pump relay or ECM is remove the fuel pump relay and jumper connectors  30 and 87,  turn key to on and see if the spikes are still there.   If no then it's the relay.  If yes the it's the ECM.  i

Edited June 18, 2020 by IBMobile

[Haggar]

Thanks IBM,  you are right.   I misread or just didn't catch you were watching the fuel shutdown solenoid, I envisioned it as the fuel pump solenoid (or relay)

 

hmmmm,   That is interesting...

 

I realized why I don't like your  graphs....  notice they are all  low while running, and spike high.   That is exactly opposite to how either of those circuits should work.  (fuel cutoff solenoids are always held open by power, so they close on loss of power.  (same with the fuel pump solenoid.  power holds it energized.) 

 

I think the same sampling error could be the problem...  we are seeing the software command, but the output is latched.  (though that doesn't explain the inversion) 

 

I want to think of the FSO to be digital like it is in a Ppump (and most other diesels).  It is commanded high to (open and) allow run, removing power allows it to close and stop the fuel flow and therefor the run.   Could Cummins/Bosche have decided to modulate this command?  (I don't think this would be a Chrysler thing, the Engine, ECM and Inj pump were supplied by Cummins as a functioning unit.

 

 

 

My brain hurts.

 

Hag

[wil440]

3 hours ago, IBMobile said:

I don't think the injection pump shut off signal is controlled by the injection pump relay but by an off/on signal generated by the ECM.  Look at page 8W-30-31 in the 01 FSM,  there you find pin #33 on the ECM labeled "fuel shut-off signal" with a lb/rd wire going to the fuel injection pump pin #5 labeled " fuel shut-off signal".  

This could be a ground signal generated by the ECM   I tested the signal with key on and engine stopped and got the same random off/on pattern. 

This is what was confusing me as I turned key off/on quickly so as not to stop the Modis but nothing changed on fuel shut off signal I was watching, Is this a collision fuel off ?  In the body module that controls the locking seatbelts in my truck there is a collision switch which is a ball in a track IIRC when I took it to bits to try and fix it 

Edited June 18, 2020 by wil440

[wil440]

I had no time yesterday to test as it rained again all day

 

I checked with the modis again tonight and the spikes are still there but a lot less, I swapped the ASR and the inj pump relay with others but still the same 

 

Page 8I -3 of my 2001 manual refers to the asr  but it doesn't specifically mention diesel, hopefully I'll finish work at a better time tomorrow to be able to test

[IBMobile]

The ASR powers up the PCM and has nothing to do with the ECM or injection pump.  It is used in gas models to power the fuel pump, injectors and O2 sensor. 

 

One of the injection pump conditions that is monitored with the Modis is the injection pump voltage.  I've seen voltage readings start at 14V and go down to 12.2V but never lower.   If the voltage from the ECM is greater tan 14.5V or less than 9V then a p0232 fuel shut-off single to high.  One might think this code would be set if the ECM hasn't commanded the shout off.

Edited June 23, 2020 by IBMobile

[wil440]

I managed to jumper terminal 30 to 87 tonight, set code P1285  injection pump control always on   same flat line on the Modis  truck started and stopped as normal, no spikes

 

replaced relay, and cleared code, started truck and spiking but quite wide apart and not uniform

 

removed relay, truck no start as expected BUT same flat line as jumpered above no spikes, forgot to write the code down but it was no comms between ecm and inj pump as you would expect

Replaced relay and cleared the code, started the truck and flat line with spikes spaced quite a way apart and what appears to be random