Hey guys. New to the forum here. I have a 1998.5 Cummins that has been in my wife's family since day one. My wife grew up getting to drive the truck to high school when her Dad wasn't using it for work. He also used it to tow his mega truck "Crawldad II" to events in the southeast. Anyways he passed away 2 years ago and my wife wanted me to make the truck my daily driver a remembrance of her dad. When I got back from deployment this past spring we flew to TN and proceed to drive the truck from TN to WA with no issues. I only flushed/changed the fluids/filter and replaced the brakes. Outside of that the truck drove great. About a month the truck started surging at idle. When it surges the ECT will immediately drop to zero. First I replaced the ECT sensor and the issues continued. Then the VP-44 and lift pump decided to quit. I thought this may of been the problem. Replaced the VP-44 (from thoroughbred diesel) and installed a 165 FASS pump. When I did the pump I also did a timbo apps sensor. The surging problem continued. I thought it may of been due to old batteries. I replaced those. Still there. Replaced alternator as it had excessive AC noise before and after I did the ground mod. I have replaced the Tstat with a cummins Tstat and burped the system for about 30 minutes. The surging continues and the ECT drops with it or even fails to respond with a change in coolant temp. The tan and brown wire for the ECT sensor has 5v at it. The sensor itself shows a change in resistance for coolant temp (decreasing as the temp goes up) I found the vacuum line from the power steering pump to the manifold to be missing (this explains the blow by/not going into 4wd/only having defrost on the AC) replaced that line with a flexible rubber hose with clamps. Solved that problem but the surging continues. It happens in park and drive. I have videos of the surging if needed. I'm at loss. It's throwing no engine codes. The WTS light always comes on immediately. With a scanner it reads everything at the right value expect the ECT will flash from NOT to -40, or will stay at -40 for long periods of time. I am assuming the ECT is causing the surging due to the changes in fuel delivery and timing.

 

Any advice would be greatly appreciated. I am trying to get this fixed before I deploy again later this week.

 

There are no mods done to the truck besides a larger horn for the intake. It had a banks engine calibration module but nothing was plugged in and I don't know anything about it so I took it all out. 

Another symptom when I unplug the sensor it reads 293F. I'm assuming the ECM is intact and there's a short in the wires. 

I wish I could help! Hopefully someone knowledgeable on this subject will be along soon.

did you happen to clean all the grounds?  a bad ground on these trucks will give all sorts of gremlins.

I will go through and clean them all again. Can you give me a list of all of them? I want to make sure there's none that I'm missing. 

All the wiring maps are here. 3 pages of grounds for these trucks.

 

I think you have a break some where in the wiring. With no sensor hooked up it should read -40°F because its an open circuit (I think). 

 

Are there any codes being set? p0117 or p0118 are coolant temp sensor high and low respectively. If its staying at -40 for a long time it should set a code for coolant temp too low..

 

I don't think that that will fix your surging though. How many miles are on the truck? 

 

Post the video to YouTube then post the link here. Might help us determine what actually is happening.

Edited October 14, 20213 yr by Silverwolf2691

This guy had a surging issue with a bad coolant temp sensor. 
https://www.cumminsforum.com/threads/probably-beating-a-dead-horse-at-this-point-but-heres-a-dead-coolant-temp-sense.2553531/
Your issue does sound like a bad ground. 
 

First off welcome to the family...

 

As for ECT sensor it supplied the +5V and sense on the same lead. So when the sensor is unplugged the ground reference is gone so the temp should max out like it did this is normal. For the ECM to report -40*F means the sense lead is shorting to ground. That is the only way to drop to low volts.

 

I didn't see any mention of reading codes. Could you grab a OBDII reader and list all the P codes here it would help. 

 

Don't thing its unusual to see a internal short inside the ECM. I've seen cases where the +5V reference is gone or it seems like every one of the sensor shorted to ground. Worse yet to have the ECM either all hi volt or lo volt codes for everything. 

 

Take notice that the ECT sensor is sharing the sensor ground with other sensors. This change through the years but not by much. 

Take notice that 

1 hour ago, Silverwolf2691 said:

I don't think that that will fix your surging though.

If the S165 ground is bad, being that it is shared by the ECT and Cam/crank sensor (if applicable to 98.5 models), then I would think fixing the ground would fix both issues.

Yes it's not reading any codes. I've hooked up both my cheap scan tool and a friend's snap on scan tool. Nothing whatsoever. You could see on the snap on it would flatline the ECT and then spike back of to NOT. 

 

I'm assuming I should find s165 and inspect it's physical integrity? 

 

The day the VP44 died it threw several codes. P118 being one of them. However when I installed the new equipment and cleared codes nothing came back. The surging and ECT gauge bouncing stayed. 

 

Here's videos: 

 

 

 

 

Edited October 14, 20213 yr by sethreesh

Ok my suggestion is to pull the ECM plug and test with a ohm meter between the ECM plug and the ECT plug it should show 0 ohms. No more than 5 ohm is allowed. Now you do one more test leaving both plugs open. Test between a pin and ground to verify there is no wire shorted to ground this should show infinite ohms or open circuit. If any ohms show then there is a short to ground. If these two test are OK. Then you need to send the ECM in for rebuilding being the problem is most likely internal.

 

Another way is to find another ECM from a friend or shop that is like model truck and swap temporarily and see if the problem goes away. If the problem goes away then the ECM is damaged internally. 

 

ECT sensor controls idle speed. As the ECT drops lower it will raise the idle speed up for warm up. I found this out during the development of my high idle kit and using a rheostat on the ECT leads as I dropped the temperature the idle speed rises.  Even running one of the foolers 3 or 6 cylinder will drop it to 123*F but no real effect on performance or idle speed. It once it down in the double digits that you see idle rise.

Are there any concerns/precautions I should have prior to pulling the ECM plug? I will be testing that resistance once I get off of work today. I'm really hoping it's the wire and not the ECM. 

@LorenS, would seem I'm wrong lol. I'm fine eating some crow. I forgot about how everything interacts with the vp44 timing/fueling tables. With the tach working however, I don't know if its the ground. I think we will find out from the ohm test Mike mentioned.

 

What are the odds the computer is trying to go high idle? Listening to the videos it sounds like its hitting the accelerator pedal on its own and then backing off.. Don't know if there's a delay for the factory high idle after it shuts off (up to temp). Bit of extra torque from the rpm increase moves the motor to connect the circuit and drops out of high idle, then sees -40  because the motor moved back, and wants to bump the rpms again... repeat ad nauseum.

Edited October 14, 20213 yr by Silverwolf2691

12 minutes ago, Silverwolf2691 said:

With the tach working

Yes, that would likely blow my theory.  Did 98.5 trucks read their tach signal from another sensor?  Thankfully I haven't had to deal with a 98.5 since 2005 when my girlfriend of the era bought one so I bought a 1999 FSM.  I guess I could look up the FSM on here to answer that, but really should get back to work :D

Did the wire test between the ECT/ECM. I had 0.16 ohms for both the sensing signal wire (tan/black/#14) and the ground wire (blue/black#13). However there is some bare open wire on the tan/black wire right as it goes into the plug. Any advice on how to change out that wire. I've never done that before. 

 

Replaced the wire. Now the truck reads 293 on the scan to with the key in the on position. I'm assuming the ECM is bad at this point.

Something isn't right. Yeah if the signal line is open it will show max temp and trip the CEL.

What's your recommendation for my next step forward? 

 

Im going to put everything back together and run it and see what it does later today. 

 

Weird thing is if I unplug the sensor the p0118 comes on. But if it's plugged in it reads 293 now. I'm getting 5v on the sensing wire. 0.2 volts on the ground wire. And the resistance on the sensor is about 19k cold. It's in the low 50s. 

Edited October 16, 20213 yr by sethreesh

Might be time to send the ECM in for testing. If it test out OK that means the wiring is having issues. If the ECM is found with a problem then that it.

I deploy tomorrow. I'm gonna pull the ECM out and have the wife send it for testing and coordinating that. Who do you recommend I send it too. 

Thank you for your service.