"Tuning" stock gauges

[dyeager535]

I’ve suspected for quite some time that GM tuned the gauges, but I never took the time to investigate, until now. Finally last fill up made me sick of seeing the thing slightly over full.


I’ll try to keep the non-technical talk to a minimum. Some pictures, some of my findings. Not nearly an exhaustive study, but I didn’t feel like disassembling any more clusters. I have plenty.


This picture shows what my gauge always read after a fill up, and this time after driving 20 miles with a trailer. IMO the gauge should never indicate above “F”. I cannot recall if this gauge cluster was installed complete, or if I changed any of them. I expect the gauge is original to the cluster, however the entire circuit from tank sending unit to the gauge cluster is not original to the cluster. The resistor color is yellow:

Here is what the same gauge reads when a known-bad (infinite resistance) resistor was used:

Here is another yellow resistor, this one off of an 80PSI oil pressure gauge. Note that with the same color resistors, but off completely different gauges, results are identical:

This is an orange resistor. Slightly less resistance than yellow. Also while hard to tell, SLIGHTLY higher gauge reading:

Here is yet another yellow resistor, a spare I just had lying around:

Here is where the difference becomes obvious. This is a light blue resistor (also a spare) with a 30% higher resistance value than yellow.

Here are the numbers for the tested resistors:

Yellow resistor value: .092
Green: infinite (broken)
Another Yellow: .091
Orange: .087
FinalYellow: .095 (poor resistor contact surface)
Light Blue: .126

Here are some facts: Resistance measured between the two terminals the resistor sits on was constant on the two fuel gauges I had, .091 WITH THE RESISTOR REMOVED. Here are pictures of the back and side of the gauge showing the resistor in place.

You can’t measure the actual resistance value with the resistor in place. If the resistance between the terminals (.091 in the above case) is the same with or without the resistor in place, the resistor is bad. The number on the gauge written with a grease pencil corresponds to the resistor installed. The number will either be written on the gauge “motor” under the resistor, or on the side of the “motor”. Here is the same gauge with the resistor removed:

Now for my diatribe. I cringe when people talk about “inaccurate” factory gauges, and/or how you must go mechanical to get accurate readings. It is obvious they ARE capable of good accuracy, or GM would not have spent the time to fine tune them in this way. Think about it. We all know that GM parts tend to be MUCH more consistent than aftermarket. But if you use a parts store sending unit on your truck, can you expect that piece to work correctly with the gauge? Of course not. No more so than you can trust a parts store ignition control module or a parts store fuel pump to last. Obviously with age these things don't have everything going for them, but it's obvious they are very robust.


If you still question their potential accuracy, then you are questioning repeatability between like gauges.

Here is the same blue resistor on the only other gauge I had laying around to double check my findings:

And here is the same gauge with the yellow resistor installed:

Here is the back of that gauge:

That gauge came from this cluster:

Which I pulled years ago from a ’91 crew cab that had the roof smashed down to the dash pad, broken leaf springs, a broken 4L80E output shaft, and a broken 4L80E-205 adapter, which was obviously dumped into a large river. The entire truck had been underwater, and I did NO work to that gauge other than swap the bad resistor.


And since these pictures are cluster related:

Note no provision for clock in the above, and thus the small fuel gauge used with tachometer equipped trucks won't work here without cutting, plus no mounting points. This one (a gas truck) has a battery charge lamp in this place. The diesel one shows “low coolant” in this position. This cluster is ’90-91, electronic speedometer.


I believe the tach mounting provisions are still here in the cluster.

Finally, the back of the gauge cluster black plastic piece, painted chrome for better light reflection.

 

[cdoggwsu]

Great writeup, Dorian! I guess this begs the question...did you ever find a resistor that gave you the "correct" fuel reading?

When people said resistors in reference to the gauges I always pictured the small breadboard style, I didn't realize they were actually plates that bolted in between the terminals. It makes much more sense now.

 

[4X4HIGH]

I know you're talking about the different resistors making the gauge read differently but you're also comparing "GM senders" to auto parts store senders and you need to keep in mind that GM doesn't make the senders so a "GM sender" versus an auto parts store sender doesn't mean that the auto parts store sender is inferior to a "GM sender" and any other part for that matter.

 

[dyeager535]

Honestly, I'm not sure yet. I would LIKE to find one that a full tank exactly equals "F" on the gauge, but the light blue one ended up reading I suspect a bit low, but closer IMO than the yellow. Since the tank is down about 1.5 gallons from "full" I'm not going to say it's dead on yet. I don't expect that it will ever be dead on, but closer is better. Yellow was way out there, even being down 1.5 gallons.

I'd rather have too much fuel than too little, according to the gauge. I think I need one in between yellow and light blue, at the next fillup I'll see how "bad" this one is off, if at all. If it's off, I'll dig through my other clusters, or some wrecking yard donors, for other values.

I didn't add this previous:

The number penciled in on the gauge...the higher the number, the more resistance, at least according to the two gauges I compared. Since the one resistor I tested was dead, and the light blue was not on a gauge, I wasn't able to get a larger sample of numbers to make certain that was a rule.

 

[dyeager535]

GM's vendors made the parts to GM's specifications. Aftermarket makes the senders the cheapest possible.

When common (read: low-mid range parts store pieces) aftermarket parts start to last as long as GM without failure, I'll eat my words. Far too many aftermarket ignition control modules, gauges and fuel pumps dying an early death for me to believe the aftermarket can hold a candle to the 20+ year old factory parts (at least electrical) that are still on many of these trucks. You probably only need look at the price difference to see that GM isn't making them the cheapest possible way...er, GM isn't using cost as the sole factor in who provides their stuff.

 

[4X4HIGH]

Maybe you and others have just had a string of bad luck with aftermarket parts but to be honest the only thing i've had issues with are starters and alternators (both seem to only last 3-5 years for me).

As for the gauge reading beyond the full mark i can tell you that i like to know how much fuel i have. I have completely drained a tank (two different blazers each with a 31 gallon tank) and then removed the sending unit and tested the OHM reading 0-90 and the sender was correct at each end of travel so then i put the sender back into the tank and the tank back into the blazer and took a 5 gallon gas can to the gas station and filled it with exactly 5 gallons of fuel then poured that into the tank and then drove the 2 blocks to the gas station and filled the rest of the tank and you can actually get 31.4 gallons into the 31 gallon tank when completely empty.

Whew that was a long sentence.

With that being said the gauge reads about 1/4" beyond the full mark when full. I have driven until the gauge hit the "E" and filled the tank and at the "E" mark there is still 4 gallons of fuel left in the tank. Now it's been so long ago that i did this that i can't remember where the needle was when i ran it out of fuel but IIRC the needle was a good ways below the "E".

 

[dyeager535]

Maybe you and others have just had a string of bad luck with aftermarket parts but to be honest the only thing i've had issues with are starters and alternators (both seem to only last 3-5 years for me).

I've actually not had any problems. I use GM stuff. It's not a string of bad luck, it's a pattern of poor quality control. Just search on here for ignition module failure, it's not just a couple of people. Do you implicitly trust engine parts and assemble without measuring anything? Of course not, and that's probably dealing with even "good" companies parts.

In any case, my point was simply that the aftermarket senders *might* be fine, but the tolerances for resistance could very well be off enough to throw the gauge off. I would just expect MORE variation out of a non-AC Delco boxed part, whether or not it actually pans out is another story.

As you indicate there are more variables to the equation than simply the gauge/resistor. I had already pulled my fuel tank sending unit, tried measuring resistance (with no success mind you), cleaned the rheostat, reassembled, swapped senders, etc., all read the same on the gauge. I rarely let this thing get under half a tank, I may need to take it on a long trip and carry a five gallon can of gas with me just in case.

 

[4X4HIGH]

I've actually not had any problems. I use GM stuff. It's not a string of bad luck, it's a pattern of poor quality control. Just search on here for ignition module failure, it's not just a couple of people. Do you implicitly trust engine parts and assemble without measuring anything? Of course not, and that's probably dealing with even "good" companies parts.

In any case, my point was simply that the aftermarket senders *might* be fine, but the tolerances for resistance could very well be off enough to throw the gauge off. I would just expect MORE variation out of a non-AC Delco boxed part, whether or not it actually pans out is another story.

As you indicate there are more variables to the equation than simply the gauge/resistor. I had already pulled my fuel tank sending unit, tried measuring resistance (with no success mind you), cleaned the rheostat, reassembled, swapped senders, etc., all read the same on the gauge. I rarely let this thing get under half a tank, I may need to take it on a long trip and carry a five gallon can of gas with me just in case.

Why didn't you have success? Just set your meter to read OHMs then hook one clip to the sender body (ground) and the other to the sender terminal and then move the float up and down to see the resistance. I can't remember if full is 90 OHMs and if empty is 0 OHMs but i think that's how it reads (it is a resistance to ground so i think that's correct).

 

[dyeager535]

I never could get it to read as it should. I cleaned it up a bunch, finally got it down to 1-100 ohms, but that is obviously going to screw with my readings a bit. I forgot that later on I "fixed" the problem, had to search out the post. It was intially real flakey readings, but turned out it was corroded pretty bad.

 

[Blue85]

I don't understand your approach. If you just want the "F" to be accurate, fill the tank and then install the needle. Don't you want to look at both the "E" and the "F"? Changing the resistor should change both. Like Scott says, it's easy to capture the resistances, but it seems like voltage would work as well. Then you can apply those voltage again and again while tweaking resistor values. You might have to alter the circuit to get the results you want - like a series and a parallel resistor, for example.

When you're listing resistor values, those are kOhms, right?

 

[79rustyk10]

FWIW, I put a coolant temp sensor in my f 150 last winter, was very hard starting when warmed up. The new sensor corrected that for a while... I am currently having the same issues as before, again. guess where I got that sensor and where I will be getting the replacement for it.

 

[dyeager535]

I don't understand your approach. If you just want the "F" to be accurate, fill the tank and then install the needle. Don't you want to look at both the "E" and the "F"? Changing the resistor should change both. Like Scott says, it's easy to capture the resistances, but it seems like voltage would work as well. Then you can apply those voltage again and again while tweaking resistor values. You might have to alter the circuit to get the results you want - like a series and a parallel resistor, for example.

When you're listing resistor values, those are kOhms, right?

Pulling and reinstalling the needle isn't a very good approach...fairly good chance of breaking the needle. Besides, they didn't come from the factory wrong, and GM did the fine tuning with the resistors, so the needle position on its post isn't "wrong" per se. I would assume whoever initially tested the gauge on the bench received the gauges assembled, minus the resistor, and was the one marking the number on the gauge "motors" after they knew which value would get it on target.

Gauge "sweep" isn't going to change if your sender is within spec, it will simply shift both F and E to the right or to the left, which is fine. If the fuel gauge reads too full, then "E" is going to be incorrect as well, UNLESS your sender is shot, which mine is not.

I don't know if I have a pic of my meter or not. What are the odds? Here's a picture of the same one I have:

So yes, kOhms, with the upper dial at the "3" position

I'm confused why it needs to be any more complex than I've outlined? GM used resistors to correct the gauges, why shouldn't we do the same, assuming we know the gauge, sender and wiring are all good? I simply started to document/catalog what GM was doing with the gauges when new, and how it can be applied now. Not being combative, just trying to figure out if I'm missing something?

I suppose with a bit more thought however, that this is a potential band aid to another problem. If the gauge is "stock", it is already calibrated to be right based on the sender resistance. The resistor on the back of the gauge is used to ensure that sweep falls between high/low. Therefore, any variation in gauge reading (without gauge or resistor failure) HAS to be in the wiring or sending unit.

In my case I know the tank sender is a bit off, but short of replacing it, there is nothing I can do about that problem. But if the gauge resistor change brings the resistance back down to where it needs to be to get the gauge right, why not? On the plus side, since the number on the gauge equates to a certain value resistor, you can always go back to the original when the problem is resolved.

 

[dyeager535]

FWIW, I put a coolant temp sensor in my f 150 last winter, was very hard starting when warmed up. The new sensor corrected that for a while... I am currently having the same issues as before, again. guess where I got that sensor and where I will be getting the replacement for it.

But have you actually tested the sender to make sure it is the problem?

Not saying you are wrong, but it's possible you've got a bad wire or something, and changing the sender "fixed" the poor connection for awhile.

No doubt I bad mouth much that is aftermarket, however I'll give the parts a fair shake before I condemn them (or let others find the failures for me lol). Heck, I've got a Bosch O2 sensor on my truck, and those are supposed to be absolute garbage according to some testing and results by others.

 

[4X4HIGH]

My original post was basically just saying that it sounds like you're trying to correct for something that doesn't have a problem. The fuel gauge from the factory read just past full with a full tank and just below empty with an empty tank.

 

[dyeager535]

I see what you are saying now. Just trying to get mine a bit closer on F than it was. IMO what my gauge shows in the pictures is excessive, especially down at least 1.5 gallons or so.

As I know the sender has some issues I have no doubt that is part of the problem. But the same would apply if I bought/used another sender that wasn't quite right.

 

[Big Ray]

I've noticed that on my '86 burb that the needle drops faster once it gets under 1/2 a tank.

I don't worry @ being full, I pay more attention to "how low can I go"...

pretty cool research you're doing. I wonder if you could install a trim pot in there, adjust to taste, and wire in a equal value resistor. Custom tuned fuel gauge....cool

 

[Blue85]

Where does the needle point when the tank is empty? If it's below the E then your problem will just get worse.

You might not get to adjust how many gallons are burned before the gauge moves. It could be limited by the float and sender.

 

[dstrbdxp-AK]

Dorian, do you have a source for these resistors? I'd like to get mine closer to accurate. As it stands with the new sending unit, it stays well past full until I've burned atleast 5-6 gallons, then clibs down, and I'm empty at what is shown on the gauge to be just over 1/4 of a tank, this was the same with the old sender until I found a hole in the tank, so I replaced the tank and sender at that time.

 

[dyeager535]

Only source I've got is from other vehicles in the wrecking yard. Don't want stuff from the 70's, I found at least one that was actual wire wrapped around the ceramic, while newer are all the "sealed" type pictured.

I believe most all GM gauges of this era (car or truck) were the same, so if you have a wrecking yard, up until the mid-90s there should be donors.

As to where the needle sits on E, I have never trusted the gauge enough to let it get close. Don't like running that low anyways with the electric fuel pump, however I will throw a 5 gallon can in the truck and test it out here soon.

At least now I trust the "F" side, and while I could be totally wrong assuming where it will end up on "E", I can certainly test it with the two resistors, the one in use now, and the original one for the gauge. I can also hook a sender up remotely and see if the replacement sender reads the same when the float is "full", and if so, have a (hopefully) good idea where E is based on the replacement senders indication of E.

Ancient dredge-edit: I did finally test where "E" was when draining the tank. Pump went dry at what's about the 1/16 point before E. Yes, I'm anal, but I want E to be E. I need to know when I'm going to run out of fuel, that is arguably the only thing that is important as long as you have somewhat of an idea of how fast the gauge will go from X-Y reading as fuel is consumed.

Also confirmed that the sender sweeps from 8-98 ohms (thus should need an 8 ohm resistor to be dead on) instead of 0-90. There is no adjustment to the float arm in this case, that is with the "wiper" as far as is physically possible to move it either full or empty. This was measured right at the sending unit out of the tank, so no measurement error.

 

[Big Ray]

I've been driving chevy trucks and cars most of my driving life.

My experience is each one has it's own "personality".

Having an ex that loved to leave them on "E", I kinda learned the hard way where the real empty was.

If you drive your particular vehicle long enough, you get a "feel" for it.