Archived 02 stranded, appears to have electrical issues

[ERuhl]

Driving home from work last night the truck developed a random studder that got worse as time went by. Within minutes I was limping into a fast food parking lot with the coolant sensor buried in the hot position. It was ~22°F last night so my first thought was frozen water in the fuel clogging a filter, despite the indicated 15psi fuel pressure (measured after the filter), but when I saw the coolant sensor pegged I knew something else was up.

I popped the hood and felt the coolant hoses and they were just warm, not hot, consistent with the fact that I had just started the truck a few minutes earlier. I even removed the cap (minimal spillage) and felt the fluid, just warm not hot. I checked the fuel bowl/filter and that looked perfectly fine, and when I turned the key on it took a second for the pressure to build back to 15 psi so I know the gauge is working. I also noticed that the coolant sensor started out at the cold position, and then very erratically made its way to the full hot position, all this while the engine was off (behaves identically with the engine running). I started the truck and it idles fine, but pressing the throttle is like stepping on a sponge, the rpms very slowly rise and there's obviously no power there. One time in the past I started the truck with the MAP sensor disconnected and it feels just like that. By now the CEL and the water in fuel light were illuminated as well, but the other gauges appear to be functioning normally (voltage, rpm, oil pressure). I checked the codes (see below) then I unplugged the Edge EZ from both the data port and the map sensor, reconnecting the map sensor to the stock harness. No change. I even tried unplugging the coolant sensor (by the thermostat) and the coolant gauge was unaffected, it still moved erratically from full cold to full hot over about 10 seconds (and then it stays full hot).

Here's the lengthy list of codes, all of which came from the ECU (PCU only had the 1693 code):

• P0112 (M) Intake Air Temp Sensor Voltage Low Intake air (charge) temperature sensor input below the
  minimum acceptable voltage.
• P0177 Water In Fuel Excess water found in fuel by water-in-fuel sensor.
• P0178 Fuel Sensor Voltage Too Low, Loss of water-in-fuel circuit or sensor.
• P0217 Decreased Engine Performance Due
  To Engine Overheat Condition
  Engine overheating. ECM will derate engine performance.
• P0237 (M) Map Sensor Voltage Too Low MAP sensor voltage input below the minimum acceptable voltage.
• P0342 ??? not listed in the 24v code list
• P0562 Charging System Voltage Too Low Supply voltage sensed at ECM too low.
• P0575 ??? not listed in the 24v code list

I had no codes whatsoever before this, and I can't help but notice the repeated "voltage too low" references, especially P0562. Again the voltage meter appears to be reading correctly, it's around 12v with the engine off and moves up to 14v with the engine running. I looked at the fuses under the hood and on the side of the dash and I didn't see any that were blown, but it was also friggin' cold and I didn't pull each one to inspect it. I just looked closely at the fuse element with my flashlight. I remembered a post from ToolmanTim that mentioned a connection under the fusebox cover under the hood and that appears to be connected fine (he had different symptoms as well). Again the coolant sensor appears to be the only gauge malfunctioning, and all of the lights on the truck are nice and bright. Even the exhaust brake still works when I'm idling.

I got a ride home but the truck is currently stranded at the same fast food parking lot. I checked all of the ground points that I could see, and I'm at a loss for where to look for this electrical gremlin. Has anyone encountered this before???

 

[ERuhl]

I looked at the Fuse 3 circuit (attached), and it supplies both the ECM and the fuel pump relay. Visually Fuse 3 looked fine, and I know the fuel pump relay is working because my fuel pump is working. So it seems the problem is either in Joint Connector 2, the ECM connector (pin 48 or 50), or the ECM itself, unless there's a fuse somewhere in between JC2 and the ECM? Or is the P0562 code referring to a different voltage source for the ECM (ie. not pins 48 or 50)?



Edit: Ugh, the attachment processes hoses the image resolution. You can view the Fuse 3 circuit here.

 

[ERuhl]

I looked at the Fuse 9 circuit, and this also services the PCM which is not currently complaining about any voltage issues.

Fuse 9 circuit

It looked fine last night, but I'll pull it and take a closer look at it. I'm not near the truck at the moment, so I'd like to have a list of things to try before driving to it.

 

[fox]

Mine has gone beserk twice, both times was dirty battery cables.

Clean them real good.

 

[rweis]

Since the coolent temperature is displaying erratic electrical behavior, have you tried disconnecting the sensor? (up on the front top of the engine). I am not sure if the ECM will ignore a disconnected coolent sensor though, might?



An erratic coolent sensor could be effecting the ECM by sending a "dangerously high temperature" and the ECM taking the appropriate "degrade" actions. P0217



Bob Weis



Take some contact cleaner, brake cleaner spray, dielectric grease, with you (tools of course).

 

[ERuhl]

Thanks guys, I'll try cleaning the battery cable connections. Bob, I did try unplugging the coolant sensor and the gauge behaved exactly the same, so it must have something to do with the ECM voltage. It's lunch time here, so I'm off to go check on it and I'll report back in a bit.

 

[Gary - K7GLD]

Sure sounds like a poor ground connection somewhere - probably affecting one of the computers... Cold temps can freeze random water/moisture at grounding points, causing a poor or lost voltage return path. That becomes more evident if proper engine operation returns as things warm up and moisture/ice is eliminated.



Hope you get it located and corrected...

 

[ERuhl]

No luck with the battery terminals. I took them all off, cleaned them, coated with dielectric grease, and reinstalled them. No change. It seemed like it was better when cold, because I could bring the rpms up and you'd hear it drop out every second or so. As time went on the drop outs got longer and longer until finally it's was mostly limping with an occasional kick in. Along the way I noticed that the coolant gauge would head towards the full hot position when it was "out" and return to a normal reading when it was "in". I also noticed the cruise control indicator was turning on and off like someone was pushing the button. Actually pushing the cruise button did nothing. This time it looked like it took longer than usual to get to 14+ volts, so I hooked jumper cables to the car to give it an alternate source of power. No change whatsoever. The voltage gauge is steady, it's doesn't move as the engine kicks in / drops out. I checked the fuses I mentioned above, they're fine.

At this point I have to make plans to tow it back home where I can start troubleshooting it more effectively. I've examined all the ground connections I know of, but once I have it home I can take a closer look and break out the multimeter. Oh joy.

Edit: Fwiw, I checked the codes and they're the same.

 

[fox]

Check all the computer connections too, including grounds.

 

[rweis]

If it looks like computer grounds are part of the problem, you could make up some grounding alligator patch cords and clip one end to a known good ground and the other to the ecm ground. Just an idea.



Bob Weis

 

[Gary - K7GLD]

Check all the computer connections too, including grounds.

YUP - that's what I was referring to - since the alternator is obviously functioning and applying charge to the batteries - and there's enough battery current to operate the starter, the problem likely is in a signal or or ground return in a sensor or computer...

 

[ERuhl]

Thanks guys, I'll try and sort this out this weekend. Last night I was able to haul it back home so I'm not as concerned about it being stuck in a parking lot. Now it's just the inconvenience of being down to one vehicle. Usually my truck is pulling the trailer, and this is the first time it went for a ride on it

Someone suggested either a sensor or wire shorting to ground thus pulling down the supply voltage. Something like is described in this thread. Is the A/C wire issue described in that thread applicable to our diesels? I'll be looking for this sort of thing this weekend. First place I plan to start is measuring the supply voltage to the ECM. A friend loaned me a spare ECM so I'm prepared to swap that out if needed.

Something that may be worth mentioning, my steering column has had the infamous "pop" for a while now while turning the wheel. I thought it was just an annoyance from an internal spring and a sticky bearing, which is on the to-do list to replace, but very recently it magically stopped popping and is back to smooth turning again. Coincidence? Or is there a chance that a wire inside the column was damaged and is grounding out?

The shorted wire idea seems reasonable, since the ECM supplies the voltage to the sensors. So if the voltage is low to the ECM, that sets P0562, and because of that the voltage to the sensors is low which sets the P0112, P0178, and P0237 codes. P0177 is because of P0178, P0217 is due to the high coolant sensor reading (low voltage is interpreted as high resistance which means very hot coolant). Seems to make sense? The cruise indicator behavior may just be due to the low supply voltage to the ECM. Hopefully I'll squash the gremlin this weekend.

 

[Gary - K7GLD]

I sure hope it isn't a failed internal ECM voltage regulator...



Good luck, and keep us posted!

 

[rkressg]

Thanks guys, I'll try cleaning the battery cable connections. Bob, I did try unplugging the coolant sensor and the gauge behaved exactly the same, so it must have something to do with the ECM voltage. It's lunch time here, so I'm off to go check on it and I'll report back in a bit.

I had my truck do this about 5 years ago. It turned out to be the actual connector on the Wire Harness where it connects to the Temp Sensor. I replaced the connector and this resolved the floating gauge issue.



The ECM measures temperature by measuring the variation in resistance in the circuit. If there is an issue with resistance/impedance in the wire, the ECM will read erratic engine temps.



I will see if I can find the P# for the connector. I bought 2 when I had the issue (so I had a spare in the event it happened again).

 

[ERuhl]

The gremlin is dead

We had a nice brisk morning here in Colorado, -2°F. Perfect weather to be working on the truck outside I was using a heat gun to warm up each connector just so I wouldn't risk breaking off the retaining tabs. Eventually it warmed up to be a nice day. Anyway, I spent most of the morning looking over all of the wire routing and checking things with my multimeter. I checked the voltage at each of the sensors that the ECM was complaining about (IAT, CTS, WIF) and they all checked out with a healthy 5V. I checked the signal return leads for each connector and those appeared fine as well. So I pulled the ECM connector to check things at the ECM connection (disconnected the batteries first, removed connector, reconnected batteries). 12V supply to the ECM was fine (pins 48 and 50), as were the two grounds (pins 30 and 49). I checked the continuity and isolation of each sensor at the connector and again, everything checked out fine. So I swapped out the ECM with the spare (once again disconnecting the batteries while plugging in the ECM), turned the key, and everything was back to normal! Oo. Oo.

Just to confirm I had found the gremlin, I shut it down and swapped the old ECM back in. All of the problems returned. I swapped the spare ECM back in, and all of the problems were gone. Bingo.

So on the bright side I found the problem, unfortunately it's an expensive one With 108,000 miles on the clock I'm just outside of the Cummins warranty. Lovely.

Btw, the ECM diagram mentioned here was extremely helpful. I only wish I had the rest of the manual available, but I was thankful for this much.

PS
My TDR membership expires tomorrow, so I won't have access to this 911 forum anymore. Since my truck is running again, I'll post any followups in the 24v forum.