Fuel sending units read WRONG time after time!

-
I am stumped. I should get in line.
I have had the modification working on the bench and temporary halfway installed in the dash. But as soon as I put the cluster into the dash, the darn thing stops functioning. I am not going to let this be, I am going to get the modification working installed in the car.
 
I am stumped. I should get in line.
I have had the modification working on the bench and temporary halfway installed in the dash. But as soon as I put the cluster into the dash, the darn thing stops functioning. I am not going to let this be, I am going to get the modification working installed in the car.

What modification ?
 
Im wondering if anyone has tried removing the bi-metal in the back and did the transistor, cap, etc. swap that I read up on from another website.

Unfortunately, the link is on my other computer that crashed and I cant remember the site I read that on. I was going to try it but obviously cant right now.
 
What modification ?

If and when I get it working, I will let every one know. So far, I have spent less than $5 in electrical parts. Plus my modification is user friendly if I need to tweak it in the future.
 
I think I am making progress. But the last hour the multimeter started being nuts. So I realized the battery had not been changed in at least 10 years.
So between my embarrassment about the battery and the garage is 113 degrees, I decided to take a break.


I am finding an abnormally in the voltages on the Instrument cluster. Incoming voltages are over 12.5 vdc. But the input voltage of the regulator is a full 2 volts less. Plus the output voltage on the regulator is not regulated and not 5 volts
Being a semi conductor I understand that it may lose a few tenth of a volt turning on. But not 2 full volts. I need to go further into the electrical.
 
4 different times I have gotten the dash cluster working just fine until I put it back into the dash. then nothing works properly, even the lights are haywire
 
That is very strange! Do you have a nick in the wire and its grounding out when you slide the dash back in? Just a thought?
 
Nothing that would cause the lights and gauge to error like this. It must be the way the cluster is grounded, as that is the only thing common to all the errors.

I just had a thought. I have the light switch unhooked from the cluster. I am on my way to look into this

BTW the temp sensor that measures the temp of the coolant. It is a variable resistor. Does anyone know the range. I measure 1.4k on the ohm meter.
 
4 different times I have gotten the dash cluster working just fine until I put it back into the dash. then nothing works properly, even the lights are haywire

I asked "what mod ?" before. Doesn't matter really. You have to assume that's where the problem is. Here's a tip , everyone else's panel is just the opposite of yours in that they wont work properly while dangling from its harness connectors.
 
The mod is using a variable resistor (POT) either in series or parallel with the gauge to get the meter reading correctly. I am using a 100 ohm and 24 turn pot.
I did not want to mention that until I knew for sure if it would work. I do not know if I need to put a cap or coil in with it.
Believe me, when I know I get it to work. I would have told you all.

I don't know about everyone else's panel I usually have mine on the bench and use my power supply and electronic parts to get it to work. I tried hanging it once, but that did not work properly.

I think the temp sending unit is screwy. The fuel sending unit uses about a 100 ohm variable resistor (reostat) and the two gauges seem to work about the same. So 1.4k ohm temp sender is way too high for the temp gauge in the cluster
 
All the senders operate on the same range of 10 to 80 ohms except for the warning lamp senders/switches that open /infinite or closed/zero ohms. 1.4 ohms suggests a bad or wrong type sender.
What instrument panel is this ?
What IVR is supplying on 3 volts ?
 
All the senders operate on the same range of 10 to 80 ohms except for the warning lamp senders/switches that open /infinite or closed/zero ohms. 1.4 ohms suggests a bad or wrong type sender.
What instrument panel is this ?
What IVR is supplying on 3 volts ?


Yes these the temp and fuel gauges are basically operating as an ammeter. They are configured to measure the amount of current running thru them.

When I got the car there were at lease 15-20 wires running somewhere that were not connected in the correct place or not connected at all. The coolant temp sensor was connected to a wire that ran directly thru the firewall and then not connected to anything. The coolant sending unit has 1.4k (1,400) ohms thru it. The oil pressure sensor what not connected to anything. I found a wire about a foot away that was covered in grease. When I traced it I found it went to the oil press sender.
The instrument panel is what I call the instrument cluster.

you mention the IVR, I am not familiar with that abbreviation.

I still need to find the voltage drop I measure on the cluster PCB.
The car is drivable and the engine runs great, I still need to finish checking the wiring to the a/c blower motor switch so I can charge the ac
 
Yes these the temp and fuel gauges are basically operating as an ammeter. They are configured to measure the amount of current running thru them.

When I got the car there were at lease 15-20 wires running somewhere that were not connected in the correct place or not connected at all. The coolant temp sensor was connected to a wire that ran directly thru the firewall and then not connected to anything. The coolant sending unit has 1.4k (1,400) ohms thru it. The oil pressure sensor what not connected to anything. I found a wire about a foot away that was covered in grease. When I traced it I found it went to the oil press sender.
The instrument panel is what I call the instrument cluster.

you mention the IVR, I am not familiar with that abbreviation.

I still need to find the voltage drop I measure on the cluster PCB.
The car is drivable and the engine runs great, I still need to finish checking the wiring to the a/c blower motor switch so I can charge the ac

I still don't know what car you have or what instrument panel you are working on. I can tell you the original fuel or and temp gauges are nothing like an amp gauge. They don't measure current. A strip of bi-metal inside bows due to heat. You could make their needle move with a candle.

IVR = instrument voltage regulator. If you built your own and it puts out only 3 volts, my guess is your solder iron overheated it.
 
71 duster v318

My voltage regulator puts out 4.975 volts. I did not overheat it. I use heat sinks when I solder semiconductors.

The gas gauge uses a float that moves up and down a wrapped wire resistor. if you look at the schematics the gauge is in series in the circuit. That is how a ammeter is hooked up, the Temp gauge is also hooked up in series like a ammeter.
The bi-metal is just how it moves the needle.

if it was a volt meter it would be hooked up in parallel to the circuit.

I went to school a very long time, long time ago. I am sure the laws of electricity have not changed since 1971.
 
I don't argue, whether topic is who has the most edumacation, how the gauges operate, or anything else for that matter.
I'm out.
Good luck
 
I was not arguing. but clearing up things that you inferred about me
 
You get the readings you want with the instrument cluster out of the dash, but when you put the cluster back in the readings go nuts, right??? Well, OK then. Normally your gauge cluster is grounded to the car body by contact with the metal in the dash and by the screws that hold it there. With the cluster hanging by its wiring you have NO GROUND, and whatever readings you're getting are bogus! Hook up a ground wire from a metal part of the dash to the ground trace of the instrument cluster and start over. When you have no dedicated ground path (either with the cluster installed or with a dedicated ground wire referenced previously) the voltage supplied to the cluster will find some kind of path to ground through the other circuitry, producing the bogus readings you're getting. GOTTA have a good solid ground path!!

BC
 
Thanx 64ragtop.

A couple posts ago I wrote
"Nothing that would cause the lights and gauge to error like this. It must be the way the cluster is grounded, as that is the only thing common to all the errors.
I just had a thought. I have the light switch unhooked from the cluster. I am on my way to look into this"

That is the path I am following now. As I sit here I am thinking that on the bench it was hooked to the wall ground thru the power supplies I was using, not the car's ground.
I need to follow up on this.
 
I think I found the voltage drop on the PCB. The wire that feeds the fuse blocks switched fuses is severely burnt. It is still in place, but has probably developed a resistance change from the heat. It is 11 pm here so I will wait till tomorrow to verify my theory. This may account for all the abnomolies I have seen in the instrument cluster.
I found that once I rebuilt the heater blower switch and got the fan running, it seemed to be running too slow. I then powered the switch from the fuse box with jumpers. The fan jumped the speed at least twice the RPMs. So tomorrow I will replace that burnt wire and see if I am correct.
If I am correct, I will start to button her up and charge the a/c and be on my way. At that time I will be able to tell if my modification to the fuel gauge will work.

Ed
 
Gauges do go bad...I have a new sending unit and good grounds, but My gauge is goofy, have not replaced it. Full is full, and when first hits empty is half a tank. It drove the guy I work with on my car nuts, he is OCD and and was ready to explode. He adjusted the sending unit to do what it is currently doing, which will kill the gauge faster, oh well. I just keep track of when I fill the tank go 200 mile ans refill.

As long as the sending unit is good and grounds are good. Check Gauge they are old...and do fail.
 
Gauges do go bad...I have a new sending unit and good grounds, but My gauge is goofy, have not replaced it. Full is full, and when first hits empty is half a tank. It drove the guy I work with on my car nuts, he is OCD and and was ready to explode. He adjusted the sending unit to do what it is currently doing, which will kill the gauge faster, oh well. I just keep track of when I fill the tank go 200 mile ans refill.

As long as the sending unit is good and grounds are good. Check Gauge they are old...and do fail.


I know that these work, but I do not know if they work to spec. I have added a 24 turn 100ohm pot in series with the gauge to get more resistance. Or I may need it to go in parallel with the meter. But I need to finish a few more things before I know for sure.
 
-
Back
Top