Dashboard Gauges Bouncing & moving Crazy

[SwerveMonkey]

So recently have been having issues diagnosing a problem as described in the title... but it seems to be intermittent.
Or at least every time i touch something or try one of the Three big fixes for this issue the problem resurfaces.

1. REPLACE ECM/TCM fuse
2. clean wiring harness going to the ECM
3. clean/replace Oil Pressure sensor

The only other solution to the problem I seem to find is to correct any grounding issues.... either making sure it isn't grounding where it shouldn't OR making sure that the engine is grounded correctly where it should be grounded... sigh.

Wondering if anyone else in these forums has had the same problem, it seems common enough in the cobalt forums/community.

 

[Markm]

Are your wires at the ipc plug squeezed against the bar? Most gauge issues in the cobalts were caused by harness chafing by the oil filter.

 

[SwerveMonkey]

Are your wires at the ipc plug squeezed against the bar? Most gauge issues in the cobalts were caused by harness chafing by the oil filter.

will have to check that IPC plug... maybe it is time to add more spacers behind the cluster.
I dont think the oil filter scratching is an issue in this build as most of my wires at that area are ziptied away, will snap a pic later...

 

[jirwin]

Does the engine die with it?

 

[Gtstorey]

Let's start with a better description of the problem since you seem to have tried things that don't really match "gauges bouncing and moving crazy".
Is this speedo, fuel gauge and tach? Does this include warning lights? Coolant temp and other DIC messages?

You don't have really have an oil pressure sensor just a low pressure switch and that isn't tied into anything but the warning light, unless I'm misremembering the wiring diagrams.

 

[SwerveMonkey]

So on the trip home today the issue dissolved sporadically on its own as i pulled into my street.
not sure if it has anything to do with the gauge cluster plug clearance yet as it didnt seems to have any significant "pinching" going on and the wires were not pinned down or difficult to adjust.

The engine doesn't die as a symptom. but it is restricted and things are set to "limp mode"... i.e loss of power, no heavy acceleration, power steering disabled and that's only what i directly noticed.

the speedo, tach, & feul gauge all move sporadically and will rise and fall randomly while driving
the tach and fuel will rise and fall randomly while at a standstill and idling

sorry for misnaming the low oil pressure switch, but that was something that was mentioned in the cobalt forums as causing a short for someone at some point. likely due to oil leaking out and infiltrating the plug that goes into the switch.

 

[Ross]

My Instrument Panel would sometimes die, then restart, causing the gauges to do their normal startup, which is one full sweep, then back to the starting point. I eventually found the electrical problem. For me it was the power steering unit, and the electrical connector to it. When the electricity was intermittent, the IP could do this startup cycle many times in a row.

 

[Gtstorey]

Probably the most likely reason is something going on with the high speed network but also might be a power issue with ECM. Have you checked for any codes.

 

[Robinjo]

Are your wires at the ipc plug squeezed against the bar?

Mine did exactly this. I used nuts and washers with longer bolts to get the gauges off the bar a little bit more, maybe by 1/2". It solved my lost signal or bouncing gauges.

 

[Gtstorey]

Mine did exactly this. I used nuts and washers with longer bolts to get the gauges off the bar a little bit more, maybe by 1/2". It solved my lost signal or bouncing gauges.

Did this cause it to go into limp mode?

 

[Robinjo]

No, mine didn't go into limp mode when the cluster lost communication. I would just have gauges stop working or work sporadically.

 

[Gtstorey]

No, mine didn't go into limp mode when the cluster lost communication. I would just have gauges stop working or work sporadically.

This is why I would lean toward an issue with the High Speed Network or ECM wiring. I doubt that an issue with the Instrument Cluster would cause a limp mode (reduced power). But a real reduced power mode should have a DTC.

 

[Joebob]

Limp mode and wild gage needles is a communication problem usually at the BCM behind the footwell. I found that I had several pins with loose capture in one of the BCM multiplug. I ended up getting a junkyard BCM pigtail and soldering in new wires where loose and solved problem.

You can see what youtube says to look for as well.

 

[SwerveMonkey]

the problem is exactly this... funny enough this is also one of those videos I watched trying to get to the bottom of this, ha ha ha.... I'm going to check if the BCM plug is problematic as well as testing several grounding connections with the ECM, low oil switch, and IP

EDIT/ follow-up....
went to start diagnostics today and the problem is not present... I did however remove the kick plate for ease of access in the future.
while there, checked other 10A fuses to see if they were any worse for wear and found a few were browned... must be corrosion or rust somehow.
since the problem was not present I went ahead and drove it to the parts store & grabbed a fuse kit to replace as many as I can that are "rusted"

 

[Markm]

Try wiggling wiring harnesses with someone watching your cluster and see if it starts acting up

 

[Dsteinhorst]

Most gauge issues in the cobalts were caused by harness chafing by the oil filter.

This was my problem. In my case, it was the crank position sensor wiring. I ran new wires to the sensor from the ECM and my problem was solved.

 

[SwerveMonkey]

Yessir! that was one of my first tests that i performed.
had a coworker watch the guages as i moved things around on the rear fuse box then the BCM fuse box under the kickplate.
neither test was conclusive, as the guages held steady. I did find that there was a LOT of corrosion onf the fuses on the BCM, and replaced them with new ones.

http://imgur.com/3IQkTdH

all my fuses there were in bad shape...

http://imgur.com/vRHtcqQ

 

[SwerveMonkey]

I'm still waiting a few days of driving it to see if the problem resurfaces after having changed the fuses... .
If the corrosion on the fuses really ends up being the "silver bullet" then I will go back to the first post with an edit to point to the solution being the fuses.

Thanks everyone ^_^

 

[Jake Falcon]

My Instrument Panel would sometimes die, then restart, causing the gauges to do their normal startup, which is one full sweep, then back to the starting point. I eventually found the electrical problem. For me it was the power steering unit, and the electrical connector to it. When the electricity was intermittent, the IP could do this startup cycle many times in a row.

This was the same thing mine would do back when it was still a Cobalt. The shops and dealership could never diagnose the issue properly and I ended up changing the power steering motor which fixed the problem.