Since my last post I have been going around in circles with my instrument cluster. I installed the new tachometer in the cluster, modified my wiring, and then reinstalled the cluster in the dash. I started the engine and the tachometer wasn’t working. I triple checked my wiring and couldn’t find anything wrong. The Tachman had sent me pictures of him testing the tachometer, so it led me to suspect that I had done something wrong in the installation. So out the cluster came and I returned it to my workbench. This tach is designed to not return to zero until it receives power, so I would have expected the needle to move when power was applied. I applied power to it using a 12 volt power supply and there was no movement of the tach needle. So I removed the tach from the cluster to check my wiring. Everything looked fine, and when I applied power the needle moved to zero. After reinstalling the tach in the cluster housing I repeated the same test and it failed. I loosened up the screws holding the housing in place and the tach tested as working. Tightening down the screws caused the tach to fail. Now mind you, I wasn’t using gorilla strength to tighten the screws. I was just making sure they were snug. So my solution was to find the point where the tach was secure but not too tight to work.

After all of this trial and error I installed the instrument cluster once again. This time the tach worked fine, the dash lights worked, and the gauges appeared to be working. So I went for a test drive. One thing i noticed was that the temperature gauge was reading very low. Almost like the thermostat was stuck open and the engine was not warming up fully. I grounded the sending unit wire for the temperature gauge and it read full scale. The smart thing to do at this point was to check the thermostat. Instead I ordered a replacement. After all, the temperature gauge and sending unit had been working fine, and a stuck thermostat wouldn’t be a big surprise on a car that had sat with the same coolant for 15 years or so. Plus, I wasn’t looking forward to pulling the instrument cluster out again.

The new thermostat arrived and I first tested it out in a pan of water on the stove using a thermometer. It worked just as it was expected to. After installing it in the car I started the engine and warmed it up. The thermostat remained closed until the engine warmed up and then opened. I looked at the temperature gauge and saw that it was still misbehaving. Obviously I had replaced the thermostat in error. Time to do the testing and diagnosing I should have done from the start.

The best place to start would be to verify the gauge and sending unit. The sending unit is designed to vary its resistance based on the water temperature. To test the gauge I used resistors of several values to simulate the sending unit at various temperatures. I was able to get different readings from the gauge, but I didn’t really know what the characteristics of the sending unit are. A search on the Internet turned up a chart showing the expected resistance values for given water temperatures. Here is that chart.

I selected resistors I had on hand that matched the values in the chart as closely as possible. When I tested the gauge using these resistors I found that even though it read full scale when grounded it didn’t even come close to matching the chart. I had an old instrument cluster stored away in my garage so I tested the temperature gauge in that cluster and it matched up pretty close to the chart. So I swapped gauges and reinstalled the cluster for testing in the car. This time the temperature gauge worked as expected. I buttoned everything back up and called it a day. Too bad that I just remembered I forgot to reconnect the speedometer cable. Guess I get to take everything apart one more time. Oh well, maybe the 4th time will be the charm.

In case anybody is interested, here are the readings for my old and replacement gauges compared to the chart above:

And here is the tester I made using spare parts and cardboard: