Hey guys, I'm trying to diagnose a water temp gauge that is not quite working correctly... When the car (75 gt4) is at ambient temperature (77F/25C, hasnt been run in a couple days), and with the key turned to position II (motor off), the water temp gauge will read 90C (center of the gauges range). Using Dave Helms' suggestion from another thread, if I disconnect the temp sender unit and ground its wire to the block, the gauge will peg on the high side. With the temp sender unit disconnected and its wire not touching anything (open), the gauge will peg on the low side. In my mind, this could be a couple possible problems: -temp sender is off -gauge is off With the sender still disconnected, I get 2kOhm if I stick an ohmmeter across it. I did a search but couldnt find what the resistance of the sender should be at room temperature; does anyone happen to know this? Any other feedback on whether or not Im heading in the right direction would certainly be appreciated. Thanks, David
I found this in my files from back in 1974 You can put a VOM on your sending unit and check these resistances: Temperature: Bottom of red area 19.7 ohms Bottom of scale 198 ohms I've never tried this test. Let us know what you find.
I used a potentiometer to simulate the temp sender (so I didn't have to heat the thing up). I got around 122 ohms on the low side and around 80 on the high side (this is the parallel combination of both the potentiometer and the gauge). This makes me believe that the gauge is probably the culprit. That said, the gauge is about 10 times as expensive as the temp sender... so I'm going to try replacing the sender first. When I get it I'll report my findings. -David
David, I am also having similar problems with my GT4. I purchased a new sender and still had the problem! The guage on my car was pegged to the right "max temp" the sender was reading 2.5K at room temp. I added 2.2K inline at the engine compt harness connector and the went all the way left when running guage now reads 195 during normal operation. The fans are turning on and off around that temp. I still need to get a more accurate reading The manual states the fans turn on and off at a cooler temp. Please let me know what you find, I hope this helps. Brad
No I did not check ohms on the car I benched new and old senders. I installed new sender and still pegged the guage to full hot. I will check ohms on the car at gauge and sender this weekend
Hey Brad, It sounds like we are having the exact same problem. I too bought a new sender and found that it didn't change anything. I can see how adding a 2.2k in series with the sender would bring it back to the center of the gauge under normal operating conditions but I'm not sure if it would prevent the gauge from pegging on the high side when you actually want it to peg. I suspect with the 2.2k in series that the gauge would never read much beyond around 210F. I'm guessing here. You could test this by disconnecting the wire from the sender (making sure that your 2.2k is still in the path) and grounding the sender wire to the block. If this doesn't peg the gauge on the high side, it will never get there under an overheating condition. I tried to re-calibrate the gauge myself (with no success). So no one else has to go down the same path (or hopefully so that someone knows the correct way to do this and can help me), here's what I did: First, I mapped the resistance of the temp sender. The sender was suspended in a pot of water that I brought to a boil. I had my ohmmeter connected on one side to the threads and the other side to the pin that goes to the gauge (only the side connected to the threads was touching water). Also in the pot of water was a meat thermometer. I recorded the resistance every 10*F from 80 to 210. The meat thermometer was found to be within 5*F of correct as it registered about 206-207ish when the water was at a rolling boil. I have an excel spreadsheet of this curve if anyone is interested. Armed with the resistance curve of the sender, I set out to cal the gauge. My dad managed to gently pry open the bezel of the gauge. Once that is done, and the two screws on the back are removed, the guts of the gauge can be pulled out. It consists of some windings and a magnet that hangs between them (the magnet is connected to the needle and controls its movement). For testing purposes I used a potentiometer instead of the sender (so I could change the resistance without changing the temperature. My test was setup like this: -battery positive terminal->gauge power pin -gauge signal pin-> potentiometer->battery negative terminal -I set the potentiometer to mimic 195*F (about 210 ohms from the curve). There is a screw connected to the needle assembly portion, I thought that this might set the tension of the spring (and that the spring tension was my problem). It does not set the tension and it did not appear to be the problem. Then I found that with a little force, the needle can be moved relative to the magnet. Using this I set the needle to read 195*F when my potentiometer was set for a 195*. I thought this was it but I found that when I tried to test higher temp levels that the gauge could no longer peg. After all this, I decided that this fix was no longer something I could handle and gave up. My fix (I know, it is not correct) is to replace the gauge and sender with VDO parts (ordered, to be installed next week). If someone knows where I went wrong, I would certainly appreciate some advice (so I can get the Veglia back in there). -David
Impressive work, David. Note that pegging a d'arsonval movement (coil in magnetic field) can cause it's demise and must be avoided. That may be why you didn't achieve the results you wanted. Also when calibrating a meter, final stages should also include input voltage variations. Quite often a bridge circuit is used to reduce this type of error but not always. What I'm suggesting is to vary the voltage source from 12 to 14 volts in perhaps 1V steps allowing things to settle for 15 seconds before noting any difference in reading. When parts are cheap it is easy to throw parts at a problem. But as sensors and gauges become more expensive and harder to find, we might need a database with normal sensor values to rely on. It's good to see such work as this for water and oil devices. Voltmeters and their usage should be part of everyone's arsenal of tools. It's reassuring to see such work as this being attempted.
Thanks, Paul. I didn't know the proper name for it... d'arsonval movement was helpful for google-ing. Interesting to note is that in this particular meter, the permanent magnet is the moving piece. The coils are fixed. -David
David, I grounded the wire with the 2.2K resistor inline. The water temp. was 50F. The guage responded up to 180F. I have also had a bad feeling that I may never see a full range of the guage with this resistor inline. I now need to take into account that as the water temp. increases the resistence will drop at the sender. " Hopefully" allowing my guage to continue up to the full range. I theorize that my sender may not have the proper resistence values needed the work the full range of the guage because the guage seems to function normally otherwise. Can you please post your data you collected. With your data I will boil up some water with 2.2K inline a let you guys know if I find anything. I.E. simulated overheat condition. Thank you for all your valuable work, Brad
Brad, Here's the water temp sender curve (resistance vs temp). It's in Excel 97 format. I created it in OpenOffice2009, so hopefully it works for you. If it doesn't work, it would be no trouble to give you a text file with the same data (but no plots). Am I understanding you right that when you grounded the wire (with the 2.2k inline) that the gauge only read 180F? If I'm thinking about this correct (which may or may not be the case), your gauge (with that 2.2k in series) will never read past 180F. Good luck, hope this helps! -David Image Unavailable, Please Login
