After I had my engine re-built, I noticed that the car seemed to run about 15 degrees hotter. I just had an overdrive transmission installed and during the process my mechanic, using an infrared device, told me that the temp. gauge was reading 15-20 degrees too hot, which came as a great relief. Question: Is there an easy way to determine whether I need a new gauge or a new temp. sensor?

Thanks in advance because I know someone has already been down this road and has just the answer I need, which is what I love about this forum!

 

welcome Greygost. I to have been down this road before. What is happening is the sensor is giving a different electrical restiance at at specficed heat tempature. the new sensor is "calaberated" ( i use the term loosley) to the gauage. I suggest to use your orignal sensor OR Grab a hole boat load from your builders "lying around" stash.
what I had to do is boil about 8 of these puppys in my kitchen at 200 degreees. then take each one (while still hott) and measure the electrical restiance with a multimeter. write down the "scores" on paper.
next put one sensor in the engine and warm it up. compare the gauge reading to the ACTUAL temp reading (use this as an excuse to buy a infred temp reading tool our use your wifes kitchen thermoter..but DONT TELL HER. They tend to get pissed). if the gauge is reading "higher" than the real engine temp, then go with one with a lower electrical restiance vaule.
Once you get really close you can add 10ohm resistors to help fine tune.
In addation these sesors are not "lineaor" with the restiance vaule they give. as they get hotter the sensot will drop 1/2 a point in valule while others will drop 1/3, or a hole point. My reason for saying this last point is that once you fine tune your average engine tempature, the gauge will become more incaaure as the temps rise.
I have learned that when my gauge says 165 the engine is at 165
gauge:175= engine 178
gauge:200= engine 210
gauge:210= engine 230
Anything above 175 I dont care cuz, that means somethings wrong with my cooling sys and im pulling over to fix it.

cheers
chad


[Modified by 71'AirStrike, 1:13 PM 8/18/2003]

 

New GM sensors are a different resistance than our guages, nice of them to change isnt it. anyhow, solder a 15 ohm resistor in line with the wire coming from the sending unit and your guage will be very very close to correct. I made a short extension so that i would not have to butcher the stock harness, and my guage now reads correctly.

 

I see these Infared sensors that you guys are talking and had a question.
Where do you shoot it at to get a reading of your engine temp?
Do you shoot it at the radiator or do you shoot it at your thermostat housing to get the best reading or where?
Do they actually give you a correct reading?

 

OR - you could buy the correct sender from Lectric Limited and just plug the wire in.

 

Where did he "shoot" your motor? 15 degrees is really not a big deal IMHO. Had his "shooter" been calibrated on a boiling glass of water, which can vary a bit in degrees just by atmospheric pressure? Way too many variables here to get worried, I would think........

 

my .02
I worked in the R&D lab of a major temperature sensor company ( Minco Products)

Store bought IR sensors are OK but you have to take into account the type of material you are shooting toget an accurate temperature. Shooting a 212' piece of glass will show up as a different temp that a 212' piece of cold rolled steel...

Also remember that you need to measure the temp of the water in the head, not the head itself, shooting the head is the wrong place. The water is cooler than the head by definition. It's cooling water! The thermostat housing is where the water is the hotest since it's after the heads, no more heat to adsorb.

The absolute beast way to calibrate ( ok verify) these sensors is with the gauge you are using. The easiest way to do this is to hook the gauge to a sensor, and dip the sensor into a of water at full rolling boil, use distilled water not tap water! This will be within a degree or so unless you live on a mountian...


I used a camping stove to boil my water.
You can put the sensor into a plastic bag to keep you connection dry.

Um, Yea, Connections... Hook the sensor to the gauge using the gauge wire and ground the sensor to the engine by wrapping a wirearound the body of the sensor and then around an engine part.

Whe you check the gauge ( don't forget to turn the ignition to 'ON' no need to start it tho) te needle should be pointing to about 212. If it a little off just remember, or mark the gauge to indicate where 212 is.


[Modified by 427V8, 10:07 PM 8/18/2003]

 

That's some really interesting info guys... i've heard of the factory tachs being as much as 2-300 rpm but never gave it a thought to the temp sender being off..does that hold true for the oil press guage as well?

 

anything closer than 10% would be dreaming.

 

Fellas,

You can correct the problem by trying a buttload of different sensors.....or resistors (white things attached to your printed circuit (for 77 and later). Trying different resistors will also work. Unfortunately, I only have 2 or 3 different ones and none of them are accurate. My engine reads 220 when I'm at 200. It reads 240 when Im at about 210.

So I always read hotter than I am, but I've had it shot w/ infared and my engine is running fine.

 

OK, I have to say this.

Putting a resister in series or parallel with a tempweature sensor of any kind will destroy it's response curve.

This of course does not apply when the factory uses them in circuit to calibrate the sensor aka, the '77s

Temp sensors typically use a thermistor, which has a negative and nonlinear resistance vs temperature slope. Putting a fixed resister ( which actually has a small positive slope) in series or parallel with a thermistor effectivly make the sensor usless. Yes it will be calibrated at one temperature, but you might as well just use the fixed resister. The temperature reading will be off by a lot just a few degrees above and below that temp.

Don't pad your temp sensor. Get the right sensor and calibrate it per the above procedure.


[Modified by 427V8, 4:51 PM 8/20/2003]