I have a similar problem, and wonder if it's a failing sender,
The gauge alternates between Low and what is expected temp. A buddy with a scanner hooked it up and was able to get the proper temperature on his unit when the gauge was reading "Low" The fans were still working and came on when expected and it also reading "Low" when they both failed and it boiled over. Restarting the engine sometimes created a proper temp reading. The reading sometimes erratic. It's cold and wet here lots, The dash panel also working oddly at times, barely readable dim, but still showing data, The suspect sender likely the head mounted one as opposed to the one in front of the manifold, This one giving the signal to the fan and ECM and working correctly.
Mine was toggling between 200 & 250, but the signal was good at the connector (5 volts) and ground was good at both the connector & the ECM. Sensor on a cold engine spec'd about the same as a new one - around 3100 ohms at 70 degrees. If I jumpered the harness terminals about 5 or 6 times, it would finally read 305 on the scanner and removing the jumper, I'd get -38, all normal. Let it sit overnight and a new scan before starting would have it pegged at 200 and turning the key, it would start toggling again. Checked for a short to ground between C10 (sensor signal) and D10 (ECM ground just opposite in the connector) and got 0 volts no matter how much I wiggled the harness. I was pretty sure by then it was the ECM and a new one fixed it. Anyway, another sensor is less than 10 bucks and it comes with the harness leads which can be used to compare the old and the new. You don't need to install the new. On a cold motor - both should ohm out close to ambient, but as long as they are within a couple of hundred ohms of each other, the old is probably ok. Then check the signal (yellow for 5 volts at the connector and if it's not there and with it still disconnected, check it at the ECM ). Verify the ground by jumpering the harness connector with your test leads - should be 5 volts. Check for a short in the connecter with the hot on the ground and the other lead to ground. Should be less than .4 (you might see a little voltage here because the TPS and MAT grounds are spliced into the circuit and can create a little resistance. Check the ground at the ECM too for less than .4, PIN D2.
I think Pete K has the right idea to at least try before you start sinking some folding $$$ into parts. Clean that bulkhead connector as the unit on my 89 was a cruddy mess! Cleaning the connectors has eliminated the "fading lights" problem on my cluster. It is an easy job, but is messy and takes a couple of hours!
SAM
'89 ECM harness is on the other side, behind the right or passenger side cylinder head. Far easier to access with the head off, but there are no contacts to clean. Instrument panel (except for the ambient temp switch ground which is bundled with the ECM harness for '89), ignition and lights enters behind the battery. It's right below one of the fresh air inlet drains and doesn't have the best seal, so it's prone to corrosion. Cleaning it up, though, won't do anything for a whacked out CTS or any of the other Engine Management functions.
Also be sure to check all the ECM grounds at the left rear of the engine. There is a ground right above the oil filter and one at the transmission housing bolt just adjacent to the one above the filter. If the ECM grounds are loose you can have all sorts of problems.
Open ECM ground would have the opposite effect - Lowest temp or about -38 on the scanner or 5 volts at the signal pin with the CTS plugged in. Also, if there isn't a 21 or 23 and the TPS and MAT values are correct, that's a good sign that the ECM ground circuit is ok.
What kind of scanning equipment are you all using? I have the same hard-start problem in my '89 and bought the AKM cable and using the Datamaster software to troubleshoot, but can't link up with my ECM. I keep getting "No PCM link".
My data ports show "open"; I can get "Code 12" and read codes with terminals A&B shorted, but just can't get the scan to work. My ECM data comes out on Pin E.
Thanks in advance.
I'm using Auto X-ray and I've had intermittent difficulty establishing communication. As best as I can tell, it's either in the cable, the scanner (it acts up with some of my other cars too) or the serial data splice from the Cluster is mucking it up. My problem started as a hard start - virtually no start really - though it did eventually fire and run like crap with no fast idle, fan on, and no codes. Once I verified key on fuel pressure, I hooked up the scanner and saw that the CTS was toggling between 200 and 250 which explained the symptoms. You might try starting yours with the CTS disconnected. Depending on what's wrong, it may generate a 15, but if it's only the sensor, the engine is going to at least think it's really cold and might pulse enough fuel into it to fire it up.
Forgot to mention that you can also get a rough idea about the CTS signal with your DVM by measuring the voltage at PIN C10 (sensor connected). On a cold motor, should be above 2.2 volts. If it's below that, it probably won't pulse enough fuel to start it.
Just reviving this thread to report that my '89 with the same hard start problems and symptoms as SunCr also turned out to be my ECM. I used WinALDL with AKM's cable and my old desktop to scan my 1227165 ECM.
My before-start CT was at 113 F. even though my CTS ohmed-out for a 70 F. ambient temp. Right after start, the scan would show CT getting erratic, jumping from 175 to 205 F when my temp. gauge was showing a gradual rise from 125 F. Figured by then my ECM was hosed; replaced it with a new one (reman) from the dealer.
Now it starts on the first try, idles smooth, and does not back-fire and sputter when I drive off. Sooo nice after battling this problem for soooo long.
BTW, the 1227165 ECM, which was replaced by 16198259, has now been replaced by 88999194 with a 1 year warranty.
Corvette Store
