I'm looking for what the resistance value, in ohms, should be for the temperature sending unit on an XJS V12 (1984) at normal operating temperature (180-190 F) and/or what it should be for a "hot" (220F) temperature.

If anyone has any information on ohms @ degrees, this would really help me in diagnosing a "false hot" indication I am getting.

An infrared beam thermometer proves the engine is at normal operating temperature, but the needle is going up halfway between the N and H positions. The symptom has become progressively worse.

The circuit is simple: +12V to one side of gauge, and then the other side reads resistance to ground through the temp sender.

Thanks for any help!

 

Blackjack, check out the table at http://www.2carpros.com/forum/1989-j...-vt103213.html. That is for an 89 XJS, but it should be very similar to yours. The easier way to prove what is good and what isn't is to get two or three known resistances and then plug them into the computer side of the plug. That way the computer will display what it thinks temperature is and you can judge from there. If the reading appears off, then you know it is something with the car. If it is reading dead on, then you know it is the temp sensor. Having a variable resistor will be handy in this case and you simply use the multimeter to adjust it to the values that you are after.

 

Thank you so much, Chris. That information was exactly what I was looking for!

It pretty much proves my temp sensor is a whack job. Even after sitting for over an hour, my sender reads 178 ohms--which would, by the table, mean that it's temp was over boiling--over 212F. And clearly, the motor was cool enough to rest a hand on so it's giving an erroneous reading.

Next step, tomorrow I will wire a 1K ohm potentiometer in series with the sensor lead and engine block, to pick various values and see what the car temp gauge believes for the values in the table.

 

Such a temp. sensor is almost certain to be a "thermistor" of the negative temp. coefficient type, i.e. it's resistance falls with rising heat. With your handy multimeter on the low ohms range, connect it across the terminals to measure it's natural resistance. Then pour boiling water over it. There should be an immediate 'dive' in the resistance value. If you don't get this, it will be knackered. Thermistors were used in degaussing color tv screens, and they had a habit of falling to bits inside, so when you extracted them, shaking the thing produced rattles from inside. Other types of thermistor had problems where metallizing was put onto the carboniferous element for the connecting wire -- it just fell off, prob. due to temperature cycling.
Leedsman.

 

OK, something is amiss.

My sensor indicates only 54 ohms at operating temp (185F). This is with the lead disconnected, reading from the spade lug to engine block for ground.

I got a variable resistor and attached it between the disconnected sensor lead, and engine block, to be my "simulated" temperature sending unit.

At 190 ohms, the needle in the car only goes to about 1/8" above the C indicator.

To get the needle to move to the bottom of the "N" indicator, an applied resistance of only 93 ohms is necessary.

My car being a 1984 model...I checked under the hood of a 1989 model and the sender is the same.

So, either the table in that blog is incorrect, or my sensor is all messed up, or my gauge itself is faulty...I'm kind of back to square one!

Also, if there were a faulty ground....ground leads that are faulty go UP in resistance, not DOWN...so that would produce a "false cold" reading, not a "false hot", since any additive resistance on the ground side will cause a net effect of increased resistance, producing a colder reading.

 

Blackjack, first, with the temperature sensor out of the equation (ie, disconnected), it sounds like your dash is reading erroneously. Not having the repair manual for your car, there is only a few things that it can be. One thing that I would tell you to try is setting a resistance of say 250 ohms (this will be the relative temperature that you car is going to run out assuming you are using a 180F thermostat) and then accessing the gauges in the dash, lift the needle off of your temp gauge and reinsert the needle where you want (ie, middle of the 'N'). That should get your gauge to read a proper level at normal conditions. When you turn off the car, it should drop so the needle is pointing at the low end. If not, then you have an issue with the instrument cluster and/or the temp circuit (not sure if it is built into the cluster or if it has some sort of 'computer' associated with it. Older cars simply used 2 resistors in series (the temp sensor and a fixed resistor in the instrument cluster). The gauge then reads off of the fixed resistor. So, as the temperature is cold, most of the voltage is dropped across the temp sensor and very little over the fixed resistor. This would cause very little current to flow through the fixed resistor and therefore can only push the gauge a little ways. But, as the temps rise, the resistance of the temp sensor drops causing it to drop less voltage and more voltage to be dropped across the fixed resistor, therefore causing more current to flow and driving the gauge higher. It is very possible that due to the age of the car, that fixed resistor has altered in value, therefore disrupting the ratio of voltage between the two resistors at a given temp. Hopefully this will help you understand how the circuit works.

So, you know the temp sensor is fubared because it is reading wrong values. So, you need a new one of those at a minimum. IT may be one of those things that since the temp sensor is just an indicator, you have to "learn" what is normal for the car until you can get the gauge fixed.

Wish you luck.

 

Thank you Chris,

No computer...all old-school analog stuff.

I am really fortunate to have the original owner's manual folder for this car, including a big fold-out schematic.

For the temp gauge circuit, it shows +12 volts going into one side (and this same rail feeds the fuel gauge) and the other side of the gauge just reads through the temp sender to ground. The resistor you speak of must be inside the instrument cluster and wired to the gauge; it does not appear on the schematic. Nor have I seen it when I've had the occasion to take out the cluster, which has been repeatedly. I completely re-worked the cluster cleaning every connection and checking every bulb, the week after Christmas, so it's in fine shape.

Yes; that sender reading so LOW has got to be a problem. That's job one at this point, I think. Although it's new, it's an aftermarket one. I ordered a real Jag OEM one that should be here in a couple of days.

Or, I guess I could add a temp offset resistor of 40 ohms in series with the sender....Nah....Last resort.

Thanks to anyone else with ideas or experience in this area.

 

The values in the link provided are for the coolant temp sensor for the fuel injection system, not the sending unit for the dashboard gauge. Besides the values themselves, another clue is that the text mentions measuring "between" terminals of the sensor. The sending unit for the dashboard gauge is a single terminal.

I've never yet seen (nor looked for) published values for the temp sender for the gauge...but I'll see what I can find !

Cheers
DD

 

Thanks Doug.

Would someone mind please measuring the resistance of their XJS temp gauge sender when the car is warmed up and the gauge needle at the center of the N? The sender is located on the RH side of the engine, at the front, just behind the thermostat housing. It can be seen between the air filter box and the intake manifold. It has one wire with a spade lug connector at the top and it isn't too hard to get to.

Much appreciated,

Lou

 

I really wish someone replied to this as I need the info too!

If you do have it please post or message me.

Thanks!