The Authoritative Galant VR-4 Resource

Join the best E39A 1991-1992 Mitsubishi Galant VR-4 community and document your GVR4 journey.

  • Software Upgraded - Reset Your Password to Login
    In order to log in after the forum software change, you need to reset your password. If you don't have access to the email address you used to register your GVR4.org account, you won't be able to reset your password. In that case, follow the instructions here to regain access to the forum.

Fuel and temp gauges not working properly

Bruceda

Member
Joined
Jun 19, 2020
Messages
5
Location
New Zealand
Long time lurker first time poster here.
As the title says, my fuel and temp gauges are not working properly, they both max out randomly. When it happens it is often both gauges at the same time but not always.
What I have tried so far:
Remove the gauge cluster and clean the connections
Visually checked the traces on the back of the gauge cluster
Reseated all screws on the back of the gauge cluster
Checked continuity to ground of the fuel level wire in the boot - direct continuity to ground, gauge was reading off the scale at the time. Removed the gauge cluster and there is no longer continuity to ground
Checked continuity to ground of the temp gauge wire in the engine bay. Gauge was reading ok at the time and no continuity to ground.

It doesn't seem to he heat related as it comes and goes.

The car is a JDM "facelift" vr4 with the single color gauge cluster

Has anyone seen this before? Any ideas or troubleshooting tips would be greatly appreciated

Thanks!

Bruceda
 

toybreaker

iconoclast
Joined
Apr 30, 2006
Messages
3,581
Looks like you've done a pretty good job running down many of the likely causes.
It also sounds like you've put in some time into troubleshooting the issues with some logical testing
Your test of grounding the fuel sender wire was exactly the correct way to "quick test" the whole system

The gauges themselves rarely give problems (in fact i've only ever seen tach problems in these clusters)

They are of an analog design, utilizing wheatstone bridge circuitry to "measure" things by varying one leg of the bridge and comparing it to the known legs (this leg is connected to the wire running out to the sensor)
full technical explanation / theory
...before glass cockpits in new aircraft, this is how all the gauges in aviation "worked",and they were very accurate
You don't really need to know any of this, it's just background behind the reason your earlier "ground the fuel sender test" was such a valid way to show that the gauge is receiving power, and the wiring is intact enough that it';s time to investigate other possibilities

It's also worth noting that the engine temperature sender for the gauge is a single wire unit, totally seperate and not connected in any way to the two pin engine temp sender the ecu uses


Does any of the wiring behind the panel look like it's been tampered with?

Sometimes, some scabby work can be done by shops/owners installing aftermarket gauges, stereos and the like
Usually, most of that carnage is down behind the radio, but i've seen some dedicated mooks bodger things up behind the cluster to add mad "a" pillar gauges as well. You're looking for signs like extra wires not in the loom and the the loom not secured / being out of position. In addition, lot of these cars have had the dashes out, and the properharness routing may not always have occured upon reassembly. These kinds of problems can take months/years to show up, but eventually out of position harnesses will strain individual wires or cause chafing against sharp dash bracketry, and bada boom weirdness manifests.

The vr4 cluster keying into fixed plugs can (and does) create some problems with alignment when it's removed and re-installed. It's a robust design, and rarely gives problems in service, ... but impatient bogans can cause many problems
when they are wrestling with the speedocable, and things can get bent / damaged / spindled and mutilated. A few moments carefully inspecting the contacts of the harness and the circuit board contacts in the sockets of the cluster may find an anomoly. I know you said you looked at it, but take your time and carefully inspect the board traces where they contact the harness pins, they can get rub thru's that are hard to see ( a 10x magnifying glass is very helpfull here)


There is one other avenue to investigate, but it's uncharted territory for me on these cars
I only mention this because of the commanality of BOTH of your gauges showing errant readings
following info NOT verified on the vr4 platform!

The systems I'm familiar with (in aviation) use a voltage stablizerin each gauge unit to cut gauge system supply voltage down to a level well under nominal system operating voltage. This is done so sagging voltage due to an alternator/generator failure will not cause errant gauge readings untill the voltage gets waaay wonky. Aviation engineers design very robust and accurate systems that will work under adverse conditions ... it would totally bring the suck if you were showing half tank after shutting down an engine ,only to flame the other engine out over the tasmanian sea due to fuel starvation

... automotive engineers ... well, they care more about saving a nickle a unit ...

Many german cars i serviced back in the day used a seperate, stand alone voltage stabilizer on the back of the circuit board. (little three prong jobber) This was used to manage gauge supply voltage to all the gauges on the board from one common unit to save the cost of individual stablizers in every instrument casing . It was common failure point in the late 70's and 80's and some manufacturers had issue all the way to the 90's
... many was the time replacing the three prong magic pixie wrangler made everything golden.

i just peeked at a vr4 cluster, and i do not see a standalone unit on the back of the board, so it *may* not be used in this application. If i get a chance later, I'll see if i can find my service manual for this platform and look up the diagram to see if it's a potential player

Sorry i couldn't be more help, but hopefully, one of the smarter guys will be along in a minute to help you resolve your issue

Cheers!

[edit]
Can you borrow a cluster from a friend to perform a test of substitution?
it's a quick and accurate way to narrow things down

How's your charging system / battery health
Are you noticing any other anomolies when you are having this problem?
If the battery voltage is fluctuating, weird things like this can happen

And, lastly, grounds, grounds, grounds
Anytime you are having weird electrical faults, a good starting point is to clean each and every ground everywhere in the system. ... and they are in random places on these cars.
I'm not sure where the tech library is now, but Mr Fixit did an awesome job catalogging electrical diagrams, including ground locations.
 
Last edited:

Bruceda

Member
Joined
Jun 19, 2020
Messages
5
Location
New Zealand
@toybreaker Thanks for your response, some really useful insights.

I could look at borrowing a cluster but because it is intermittent this could be a big ask from the lender. Plus, the gauges are a PITA to get out and I wouldn't ask anyone to do it unless it was an absolute last resort.

The charging system and battery seem good. Datalogging shows good voltage with the engine off and on.

Grounds is an interesting one. I will check them and would really appreciate any guidance on where to find them, as you say, they are all over the place. Having said that, wouldn't a poor earth connection make the gauges read lower not higher? The temp and fuel gauges respond to resistance and the lower the resistance the higher the gauge reading. Poor or no ground would read like high or infinite resistance and would therefore make the gauge read low? The gauges in question do share a common ground according to the circuit diagram but as they can fail independently I have largely discounted this as a likely cause.

I realised overnight that the testing I have done targets the loom and gauges, not the sender or sensor in the equation. I can't rule out coincidence here and it might just be that both are intermittently broken. I do think it's unlikely that the temp sensor would fail like this as it is essentially solid state. The fuel level sensor is more understandable, especially given the rust problem I had in the tank. The direct continuity to ground could be explained by the gauge being maxed out when I was testing. Unfortunately the sender and sensor are harder to test but I will try.
 

toybreaker

iconoclast
Joined
Apr 30, 2006
Messages
3,581
Sounds like you have a legit and logical approach.

I'm sorry i don;t have my manuals here,so i can;t dig into them for the grounding locations.

We had some docs in the library section on the old format, but i don;t see them now?
Rich (Mr Fixit) did an awesome job collating information and was an invaluable resource on the jdm cars

The thing about grounds is that one bad ground can cause all kinds of wonky problems in systems that would *seem* totally unrelated. When a ground is faulty, things will backfeed other systems "looking" for ground, and it can really do crazy things.

Is there any aftermarket equipment in the car (turbo timers / stereo gear, etc) ?
I've seen the center console ground put in the wrong spot a time or two, and it caused some kinda crazy intermittent problems if it was on a bracket instead of the structure where it belonged

A visual inspection/cleaning/and sealing grounds with di-electric grease can really make a difference, so i often start there.
Is it humid where you are?

Sorry, i gotta run, I'll check back in later

Good luck!
 

Bruceda

Member
Joined
Jun 19, 2020
Messages
5
Location
New Zealand
It does have a turbo timer and I will start looking around that area to check the earths. It also has an aftermarket stereo which is not in there at the moment but the loom was hacked to fit it so worth checking in the centre console too.

I live in New Zealand so not hot and humid, more like bland and mediocre in terms of weather.

This isn't a new thing by the way so I can't correlate the problem to doing something in particular to the car. I have owned it for over 20 years and it has been off the road for about 15 of those. It has pretty much always done it as best as I can remember. I am just throwing a lot more time at it now than I have for a long time. I am trying to iron out all of the little niggles and undo my old bad decisions.

Thanks again
 

toybreaker

iconoclast
Joined
Apr 30, 2006
Messages
3,581
.. heh ...

You came to the right place
... most of us have made bad decisions,
... and some of us ( like me ) continue to do so

This problem is quite likely going to be a challenge to find if it's been there for a few years.

One of the best things you can do is tidy thing up as you go.
I know I constantly find "temporary repairs" on my own car years after the fact
"oh yeah, i remember that, it was dark/raining/had my wife with me when i did that particular roadside bodge"
I had a plastic menu from a restaurant in Winemucca Nevada all folded / wadded up isolating a chafed wire for the headlights under the fuse box for quite a few years on my tow pig. Ironically, it worked for over a decade and only caused a problem when i powerwashed the engine compartment and dislodged it. Luckily, the smoke show was in my driveway at home

I often find intermitent electrical problems in bad splices/crimps i made years ago
I try and fold the barrel over more thoroughly and completely cover them with glue lined heatshrink nowadays


Do you notice that hitting a bump in the road changes the readings?

Sometimes a little percussive maintenance can find these problems while stationary
just give it a lil fonzarelli love tap on top of the dash and see if it full ranges the gauges
you can also wiggle the harness while someone watches the gauges to see if the readings change / spike

Good luck, and keep us posted on your progress

[edit] i found a pic of the back of a usdm cluster. It does not appear that there is a common voltage stabilizer visible on the rear of the cluster, so it is likely that each gauge is on it's own circuit.
 
Last edited:

paul j

Well-known member
Joined
Apr 27, 2004
Messages
204
Location
Lone Tree, CO
This is very informative guys. Maybe the lost tech library will turn up after this discussion. I would like to know the proper way to splice in pigtails that toybreaker mentioned. I have a few to do.
 

toybreaker

iconoclast
Joined
Apr 30, 2006
Messages
3,581
... for the searchers;

I'm helping an overseas member troubleshoot some dash / instrument cluster problems before I send a cluster out. Shippin is pretty spendy to his neck of the woods, so we're working on eliminating any other potential problems before we pay to ship a known good unit out. He's got (what is probably) the only usdm vr4 in his country, so parts are hard to come by for a test by substitution.

Let's start by examining the printed circuit board on the back. I'm not sure about the JDM cluster unit that Bruceeda is workin on, but the usdm models actually have two flexible printed circuit boards on the back of the cluster. It looks like they just stacked them on top of each other, and they have been rubbing on each other since the day the car rolled off the assembly line, so there's potential for shenannigans from board to board if the traces get the mastic rubbed off. This is worth investigating, as one of the clusters showed evidence of contact where they both bend around the bottom of the cluster. ... It goes without saying that cross contact between the boards will do all kinds of weird shite ... intermittently ...

While the cluster is out, it's worth giving all the fasteners on the back a once over to make sure they are tight. Don't go all McGilla Gorilla on them, just make sure they are all snug n then some. Check all the floating wires on the back of the cluster as well for chaffing / other damage (kinda funky how's there's so many external jumpers on these clusters, almost like they use the base cluster module in other models and had to add these after the board design was finalized?) At any rate, check them at their attachment points for cleanliness and security


Now, lets talk about the ground side of things;

The instrument panel has two ground points (labeled "6" in the configuration diagram)
Both are directly behind the cluster, one above and one below the dash harness
These are the primary grounds for the cluster assembly, (and the only ground for the temp gauge)

There is also #7 above the harness on the far right side of the dash above the harness
#8 behind the glovebox ~ish (a possible player, according to the wiring diagram)
#9 down on the center console bracketry

For the fuel sending unit, it's ground ( #13 ) can be found behind the tail light carpeted trim in the trunk
#13 is also the ground for the fuelpump, so it's worth noting if you upgraded your pump and/or power feed without also upgrading the ground side of things, that wiring/ground is operating outside of it's design parameters, and willcause some anomolies like a fuel guage that won't ever read "full"

... it's worth noting there are 5 ground points in the engine compartment as well as a few more located in various locations on the chassis/body that should all get a once over. Pictures of each ground number/location *should* be in the factory service manual wiring diagrams found in the library


Finally, for Bruceeda;
A quick test for troubleshooting the senders can be done using an old school test light
Next time things act up, unplug the sender and insert the test light into the circuit,
(probe tip of the light into the harness connector, and the coiled wire of the light to a good ground)

With the key on, the test light should light and the gauge should read less then pegged (because you added the resistance of the testlights bulb to the circuit)

If the test light does not illuminate, and the gauge is still pegged, the circuit is grounded somewhere before the sender
... as mentioned before, I see a lot of issues with fuel gauges not reading correctly on these cars. Many times, it's because the float arm was bent during a fuel pump swap. Other times, it's corrosion on the wiper element of the sender. Finally, an undersized/overtaxed ground from the tank sender to the ground behind the trim panel in the trunk can cause some pretty crazy shite. The tank is kinda grounded by it's attatchment hardware and surface contact with the chassis, but once corrosion begins to occur, that ground path becomes very unreliable.

FWIW, when i upgrade the pump wiring, I always add my own 10awg ground wire under one of the sender nuts and run it over to the factory ground point behind the trunk trim ... end of problems ...


Finally, here's the specs on the sending units;

The engine temperature sender only lists one resistance spec; 104 ohms @158*f (70 celsius for those who are unfortunate enough to have to use that silly system of measurement ;) )

The fuel level sender lists an empty and a full resistance;
empty; 3ohms (with a tolerance of +/- 2 ohms)
full; 110ohms (with a tolerance of +/- 7 ohms)


... hope that helps
 

Bruceda

Member
Joined
Jun 19, 2020
Messages
5
Location
New Zealand
@toybreaker thanks for the awesome write up! This is a really useful post for me. I haven't done any more testing for a while as I have been sidetracked rebuilding the brake calipers. The gauges pi$$ed me off enough to look for something else to do before I took the "percussive maintenance" you described too far :ROFLMAO:. The cluster is still out of the car so I have attached a pic of the rear showing the flexible PCB. JDM's (well mine at least) have a single layer of circuit board and two interfaces to the vehicle loom. As you suggested, I have already cleaned and re-seated all of the screws and I used contact cleaner on the copper strips in the loom interfaces. No change sadly but peace of mind that it is one less possible cause.

The brakes are very nearly ready to go so once that is done (and work stops getting in the way of my fun) I'll take another look at the senders. Great idea about running a new earth for the fuel sender and the test light idea too. The boot lining is all out still so it will be easy to get to everything.

I will report back once tested

Thanks again

Bruceda

20220201_112416.jpg
 

toybreaker

iconoclast
Joined
Apr 30, 2006
Messages
3,581
Thanks for the picture!
While the basic design is similiar, our USDM clusters are considerably different in both layout and configuration

I think your best bet is to just check all the basics we've covered while everything is apart, and then troubleshoot after everything is back together. I don't like powering stuff up with the car apart and a lot of things unplugged, as there's waay too many ways for things to go sideways

I would love to see a thread on someone rebuilding brake calipers!
Do you have any pictures from the process?

Thanks!
 

Bruceda

Member
Joined
Jun 19, 2020
Messages
5
Location
New Zealand
I would love to see a thread on someone rebuilding brake calipers!
Do you have any pictures from the process?

I followed another guide from a DSM site. I am not sure if we are allowed to post links to other forums so if you want the links DM me. I will post pics when I am done but right now I am having a bastard of a time finding decent rotors that don't cost my first born child.
 

yubh8tn

Well-known member
Joined
Aug 21, 2019
Messages
166
Location
coeur d alene, idaho
I followed another guide from a DSM site. I am not sure if we are allowed to post links to other forums so if you want the links DM me. I will post pics when I am done but right now I am having a bastard of a time finding decent rotors that don't cost my first born child.
we're allowed to. i'd be ashamed if we werent. information is information.
 

Ludachris

Administrator
Joined
Feb 12, 2017
Messages
1,094
Location
California
I followed another guide from a DSM site. I am not sure if we are allowed to post links to other forums so if you want the links DM me. I will post pics when I am done but right now I am having a bastard of a time finding decent rotors that don't cost my first born child.
Depends, which DSM site? :) Just kidding.
 

iceman69510

Turn Right Racing
Staff member
Joined
Mar 5, 2001
Messages
10,907
Location
Michigan
Yeah, we don't want it to be any of that Tooners nonsense. :eek::ROFLMAO:
 
Support Vendors who Support the GVR-4 Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications RixRacing RockAuto RTM Racing STM Tuned
Top