Imperial Homepage -> Repair -> Fuel -> 1981-1983 -> Rough Running
Question from Rolland (1981):
I have been chasing a problem of dying during warm-up and I did a little lab work on coolant temperature sensors. I am listing the data in hopes that it may help someone.
I have three coolant temperature sensors. Two I purchased new from NAPA within the past two years and one I purchased new from Chrysler when they were still available. Perhaps 5 to 6 years ago.
I ran a test of resistance Vs temperature. This was not terribly scientific. I used a pan of water on the stove, a candy thermometer, and a decent digital ohmmeter. I did not let the temperatures stabilize but took them on the fly. It took about ten minutes for the water to go from room temperature to 200F. Perhaps this can be added to the archives and will help someone.
Temperature Sensor (readings in ohms) Chrysler NAPA # 1 NAPA # 2
40 F 919 905 866
70F 984 964 924
100F 1049 1034 994
125F 1108 1094 1043
150F 1186 1182 1119
175F 1260 1261 1189
200F 1336 1348 1261 Both NAPA unit numbers are TS 5008
Although I did not change each of these and then take a test ride immediately after I have had all three installed for a period of time and I cannot say that I detected any difference in performance or starting. I have tried different sensors to try to solve a warm up or idle problem. It would seem that an 80 or 90 ohm difference would be noticeable but I could not detect it.
Experience of others and comments would be appreciated.
Replies:
From Bob:
I have some comments on your Temp Sensor studies. The specs on this device are: COLD: 500 to 1000 ohms; HOT: more than 1200, but not to exceed 1500 ohms Your cold readings are on the high side of the allowables, but not excessive; your warm water numbers, are low except for the readings @200 degrees.
During Start-Up, (hot or cold), most defaults are ignored, (the Memory is in control), then the water temp sensor becomes important and the Memory parameters control the quality of engine as to idle speed, fuel mix and ignition system until it is warmed-up. I too have had problems with this sensor and they were solved when I replaced it. My experiences were with more severe resistance readings. But I also, as a result of this, looked at some other items and looked at all of the devices using vacuum to operate and found two aged hoses at the solenoid valve and one on the Canister which were leaking - this fix resulted in a dramatic improvement. Hard starts followed by stalling are usually a fuel problem, (heat soak fuel evaporation), and cold starts with stalling are sometimes caused by the Closed Throttle Switch being dirty, hence insulated from the system, the Heated Air System not functioning correctly, (vacuum diaphragm, stuck damper door, leaking vacuum hose or the Standby Memory Circuit, (check the red wire to the memory - should have at least 10 volts; this wire has a 620 ohm resistor in it).
If your engine, while idling, has extreme highs and low idle speeds, (severe Bracketing), this also may be caused by a leaking vacuum hose since the air volume for combustion is measured by the Air Flow Sensor, any additional air leaking in will cause a severe correction by the Computer and then the opposite occurs because it becomes too rich followed by too lean. Maybe this will help
From Phil:
This is just a general grease monkey opinion, but stalling on warm up is usually due to a lean condition, similar to a choke opening too soon. I wonder if you may have a vacuum leak somewhere? Are any of your plugs white? If changing your sensors makes no difference, it makes me wonder if the problem is elsewhere. Maybe something that is leaning out the fuel mixture more than the computer on the car can control, or possibly, sense. If these sensors control fuel mixture, and changing them makes no difference, I suspect something else is affecting the fuel mixture, beyond the control of the sensors.
From Leo:
I recently changed my original Mopar temp sensor in my 83 EFI to a Delco Remy and have had nothing but problems with cold engine stalling, doesn't want to keep running when first started, etc. Two days of the last 21 it has run decently. Haven't got a clue why. The rest of the time it runs when it feels like it. I think I am going to put my original sensor back in. All I had was a slight warm engine stumble then. The resistance readings of both were very close.
Follow-up from Rolland:
Thanks for the input on the cold temperature stalling. I have checked all hoses for vacuum leaks and also the air cleaner seals etc. I don't believe I have a vacuum leak.
I may have stumbled upon the problem this afternoon. I put my lexan air cleaner cover on so I could observe the fuel flow while idling during warm up. I also had a small light bulb connected to the fuel tank pump resistor to try to determine if the electrical system was shutting off before the fuel flow stopped. This is a bit tricky to observe but I think the power to the EFI stayed on until the engine died.
I did, however, observe rather severe leaking from the "O" rings on the fuel nozzle assembly. It was leaking quite badly. I replaced the "O" rings and the leak stopped. I tried the car out and it did not die during the test run. However, this has happened before. I will try it again tomorrow. I haven't quite decided whether the engine was flooding or starving with the leaky "O" rings. Since even the fuel that was leaking found its way through the throttle valves the amount of fuel should not have been any different with or without the leak. However, since it was not passing through the nozzles I would expect the mixture to be less consistent that normal. The control pump, since it is a positive displacement pump, delivers fuel proportional the DC voltage applied as directed by the computer. I don't think it cares whether the fuel passes through the nozzle or leaks into the throttle body.
In any case I will continue to try it and see if it is corrected. My current assumption is that the mixture, since it was not all delivered by the nozzles, was inconsistent enough to cause the engine to die.
This page last updated October 21, 2002. Send us your feedback, and come join the Imperial Mailing List - Online Car Club
