Stinkin temp gauge

[chicofuentes0224]

Hey I know this is an old post but I have a Cherokee from Canada with the metric system installed. I'm begining to think these senders are designed for those vehicles. I had a Comanche which just about drove us nuts with this same issue.

 

And what did you do to "fix" the problem.

 

Back to trying to get the gauge to read right. Anybody ever heard of a resistor that you physically attach to the wire to get the electrical reading to change? Someone here at home brought it up saying it might cure the wrong reading.

[brigarpeon]

There's 3 electronic devices on an 88 4.0;

Coolant Sending Unit - cylinder head

Coolant Temperature Sensor - Left side of the block under intake

Radiator Temperature Switch - Radiator Drivers side

All must be connected and in viable working condition

[chicofuentes0224]

There's 3 electronic devices on an 88 4.0;

Coolant Sending Unit - cylinder head

Coolant Temperature Sensor - Left side of the block under intake

Radiator Temperature Switch - Radiator Drivers side

All must be connected and in viable working condition

 

All three are connected and brand new. And actually the sending unit for the gauge has been replaced 3 times now to see if I can get a decent reading on the gauge.

The truck is not running hot. (200-205* at the thermostat housing)

3 differnt clusters have been used.

2 differnt gauges have been tried. (One checked/calibrated by Hornbrod)

So the problem seems to excist somewhere in the wiring between the sending unit on the back of the head by the valve cover and the actual gauge. Hence a friend said that he believes you can actually buy something so whatever voltage is going thru that wire to the gauge can be changed to be a differnt voltage so my gauge will read correctly.

Does my gauge need to read correctly? Not really. But it's driving me nuts that something so simple does not work correctly and it's become somewhat of an obsession to corect the false readout of that gauge.

[brigarpeon]

OK then, I would start looking for a melted wire or places where it could of gotten stripped abraided, a loose wire from one pin connector etc... That's making the resistance go down by providing a better path or bigger wire, thereby increasing the guage reading.

Remember size of the wire determines the resistance .

[HOrnbrod]

So the problem seems to excist somewhere in the wiring between the sending unit on the back of the head by the valve cover and the actual gauge. Hence a friend said that he believes you can actually buy something so whatever voltage is going thru that wire to the gauge can be changed to be a differnt voltage so my gauge will read correctly. Does my gauge need to read correctly? Not really. But it's driving me nuts that something so simple does not work correctly and it's become somewhat of an obsession to corect the false readout of that gauge.

 

Chico-man, below is a pic of the back of the temp gauge with an arrow pointing toward the meter movement calibration resistor. Voltage drop across this resistor is what determines meter deflection. Unsolder one side of this resistor and accurately measure it's resistance. Should be about 80 ohms IIRC. Then substitute other resistors in about 5 ohms higher increments until your gauge reads just a bit higher than your laser thermometer reading at the sensor. Of course this is just a bandaid; your real problem is the wiring between the sensor and the gauge. Can you run a clean new wire from the sensor directly to the gauge input (pin A1 on connector A)? I'd try this first.

 

Image Not Found

[Eagle]

The ohms readings at the various temps should decrease as the temp gauge goes up, so I'm interpreting your 180* reading as 232 ohms; I think you're reading the scale wrong on your ohmeter. :eek: That sensor is going in the right direction, but it's range seems wrong. Factory cal specs are: 100* = 1365 ohms; 220* = 93.5 ohms; 260* = 55.1 ohms.

Are you sure on those factory calibrations? My understanding was that the 84-90 vehicles had the temp sender for the gauge rated at 0 to 88 ohms, and the one for 91 and newer was rated at 104 - 0 ohms. Are you sure those numbers you have aren't the ratings for the temp sender on the side of the block that inputs to the ECU?

[brigarpeon]

So the problem seems to excist somewhere in the wiring between the sending unit on the back of the head by the valve cover and the actual gauge. Hence a friend said that he believes you can actually buy something so whatever voltage is going thru that wire to the gauge can be changed to be a differnt voltage so my gauge will read correctly. Does my gauge need to read correctly? Not really. But it's driving me nuts that something so simple does not work correctly and it's become somewhat of an obsession to corect the false readout of that gauge.

 

Chico-man, below is a pic of the back of the temp gauge with an arrow pointing toward the meter movement calibration resistor. Voltage drop across this resistor is what determines meter deflection. Unsolder one side of this resistor and accurately measure it's resistance. Should be about 80 ohms IIRC. Then substitute other resistors in about 5 ohms higher increments until your gauge reads just a bit higher than your laser thermometer reading at the sensor. Of course this is just a bandaid; your real problem is the wiring between the sensor and the gauge. Can you run a clean new wire from the sensor directly to the gauge input (pin A1 on connector A)? I'd try this first.

 

Image Not Found

 

Ya wouldn't suppose that the slotted, white cover post is also the guage needle adjustment? I don't know but I have seen gas guages that work the same way to in effect either zero it at empty or full depending how it works up or down. I find it alittle hard to believe that 3 temp guages all had bad resistors but hey crap happens.

[HOrnbrod]

Are you sure on those factory calibrations? My understanding was that the 84-90 vehicles had the temp sender for the gauge rated at 0 to 88 ohms, and the one for 91 and newer was rated at 104 - 0 ohms. Are you sure those numbers you have aren't the ratings for the temp sender on the side of the block that inputs to the ECU?

 

The temp gauge sending unit (back left side of the cylinder head), not the sender in the stat housing, carries the same part number for 87-92 MJ six cylinder gas engines, 53005309.

 

The temp gauges also are identical; I have used 87-90 temp gauges in the HO clusters and vice versa. The gauges that are not interchangible between the two are the speedo, fuel, and tach gauges. The calibration resistances are straight from the FSM.

[HOrnbrod]

Ya wouldn't suppose that the slotted, white cover post is also the guage needle adjustment? I don't know but I have seen gas guages that work the same way to in effect either zero it at empty or full depending how it works up or down. I find it alittle hard to believe that 3 temp guages all had bad resistors but hey crap happens.

 

That's not an adjustment, just a plastic post, part of the housing that the meter windings are wrapped around. In Chico's case I do not think there is anything wrong with his temp gauges since they are all reading approximately the same.

 

I think he should try the new wire trick first: Can you run a clean new wire from the sensor directly to the gauge input (pin A1 on connector A)?

[chicofuentes0224]

Ya wouldn't suppose that the slotted, white cover post is also the guage needle adjustment? I don't know but I have seen gas guages that work the same way to in effect either zero it at empty or full depending how it works up or down. I find it alittle hard to believe that 3 temp guages all had bad resistors but hey crap happens.

 

That's not an adjustment, just a plastic post, part of the housing that the meter windings are wrapped around. In Chico's case I do not think there is anything wrong with his temp gauges since they are all reading approximately the same.

 

I think he should try the new wire trick first: Can you run a clean new wire from the sensor directly to the gauge input (pin A1 on connector A)?

 

Donnie baby, I can run anything you want me too, but can you tell me which one connecter A is. Is it the one where the wire is running from the temp sender to the firewall?

[HOrnbrod]

Connector A where the harness connects to the cluster in the middle. But thinking about it some more, it would be easier to run a wire directly from the sender tip on the head to the temp gauge. As you know, the temp gauge has three screws that attach it to the cluster chassis and printed circuit foil. One is 12V, one is ground, and one is the variable signal from the sender. That's the one to connect your test wire to. You can figure it out by following the three foil traces attached to the meter screws, then verify it with your continuity checker. Be sure to have your original sender wire disconnected. This way you eliminate all the wiring and connectors from the sensor to the temp gauge. Stick the cluster back in with the new test wire, and see if the gauge indicates correctly. If it does, then there's a wiring problem between the sender and the gauge. Make sense?

[chicofuentes0224]

You got it Captain!

I'll let you know how it works out then :D :cheers: