I made it to school. About 20 miles of mixed driving. I had to use the AC to turn the rad fan on again. I took the 710 cap off and reved the engine. I could see oil splattering around. That's a good thing. I pulled the codes again. Got Code 34 and Code 45. I unplugged the MAP harness while it was idling. It died immediately. It has a new O2 sensor.
Now that I think about it, I may have caused the Code 34 when I disconnected the MAP.
[This message has been edited by Boostdreamer (edited 10-17-2013).]
So I'm thinking that if the ECM's coolant sensor is good and the fan switch is good and all the wiring is good, maybe the gauge sender is bad.
I'm guessing that the fan switch is NOT connected in any way to the gauge sender. If that is true, then maybe the reason the fan isn't coming on is because the engine really isn't as hot as the gauge shows. Here's the problem with that. The gauge reads fine for a while then starts climbing. When it gets to the red, I can hit the AC and that triggers the fan. Then the needle starts walking back down to a vertical position. If the sender was bad, why would it react with the rad fan?
Confusing. Here are my thoughts on temperature control:
1. Thermostat has to be working to allow coolant flow. Both of the side pipes get hot so it must be working. 2. The ECM coolant sensor must work to help regulate the fuel/air ratio. If not, it dumps fuel which causes the engine to HEAT up. (is this correct?) ** 3. The coolant must not have air bubbles trapped. They can prevent coolant flow and give false high temp readings if caught at a sensor. 4. The coolant system must be closed tightly to allow a build-up of pressure which lowers the boiling point of the coolant.
** I'm leaning toward this and the MAP sensor right now.
Here's my plan of attack. I'm going to change the gauge sender. I'm going to park the car, nose down, on a steep slope. I'm going to remove the thermostat and cap and let it run to burp. Going to re-affirm that the fan switch turns the fan on when grounded.
Am I missing anything else that would cause the coolant temp to rise out of control?
[This message has been edited by Boostdreamer (edited 10-18-2013).]
You're correct that the fan switch, the Coolant Temperature Sensor and the Gauge Temperature Sensor are totally independent of one another. Either can be incorrect without affecting the other. Any sensor that has a single wire will need a good ground contact through it's base, so no teflon tape on those. In the case of the gauge sending unit, that's a dual wire sensor, but it also is a dual purpose with one lead being for the temperature light and other other for the gauge. Again, no teflon tape on that one. The CTS is a different story, but I still wouldn't use tape on that one, either.
There's a Harbor Freight in Johnson City. They have an infrared/laser temperature sensor that I've found to be very handy in checking the gauge accuracy. I point the laser dot at the brass base of the gauge sending unit, get my reading then compare it to the gauge reading. Right now it's on sale for $25 if you print out the coupon. http://www.harborfreight.co...0AIjoiODkwNSJ9%0D%0A
Your temperature gauge should sit on the 100* mark when you turn the ignition on without starting. That would be at a normal ambient temperature and the engine hasn't been running. If it is either side of 100, then you'd have to compensate for the difference to determine what the true temperature is.
The gauge sending unit is reacting when you turn the fan on because the temp goes down. Doesn't mean it's giving you an accurate reading, though. My 86 was showing that I was running hot during RFTH scouting runs and I was doing the same thing with the AC. When I started checking the temp with the laser, I found out my sending unit and gauge were wrong. The gauge sits way below 100* but when it warmed up, it was quickly shooting up beyond the 235 mark.
Thanks for the confirmation. I may have to go get that laser thing, too.
I did check the fan switch. It does engage the fan when grounded. I didn't use tape but I did use some thread sealer. I only used it on the upper threads, above where I could see that the threads were clean from being backed out of the block. I expect it to be fine and it looked as if they may have had something similar when new from the factory. They prolly have a dry putty-like ring around the upper threads when new. If I continue to have trouble, I'll take it out and clean all of that off.
I pulled the thermostat out and changed the gauge sending unit. I drove to get gas with the thermostat out. Naturally it never got past a quarter of the gauge range. This tells me that it is either a gauge problem, a thermostat problem, or a fan switch problem. The cooling system is obviously capable of keeping the engine cool if controlled properly.
My hesitation and stumbling is greatly improved but not completely gone. I'm still getting Code 45. I've got another MAP sensor to try. Don't really know what to do beyond that for the code.
The ECM will set code 45 when the ECM detects a high voltage from the oxygen sensor, throttle is applied and the system is operating in Closed Loop (conditions must exist for longer than 50 seconds and engine must be running for at least 1 minute).
Check the oxygen sensor and replace if necessary.
Check fuel pressure. The system will go rich if the fuel pressure is too high.
Check for rich injectors.
Check for leaking injectors.
Check for fuel contamination (specifically for contamination with oil).
Check for proper ignition module shielding. If the ignition module is not properly shielded, the ECM might mistake the electro-magnetic interference for reference pulses, causing too much fuel to be delivered to the system.
Check the canister purge for fuel. If full, check canister control and hoses.
Check the MAP sensor. If the ECM detects a lower than normal vacuum then this will cause the system to go rich. Disconnect the MAP sensor. If the rich condition goes away, then a problem exists with the MAP sensor.
Check for leaking fuel pressure regulator by checking the vacuum line to the regulator for fuel.
Check TPS. An irregular TPS output will cause the system to go rich due to a false indication of accelerating.
[This message has been edited by Boostdreamer (edited 10-18-2013).]
I put the 86 gauge sender back on. Went for a test drive and the needle was as steady as a rock. In fact it was reading cold, directly between 100 and the half-way mark which is 220.
I tried three different MAP sensors and none of them cured the Code 45. Since I think the problem came with the 87 engine, the next thing I'll change will be the cold start injector. It was fine on the 86 so I'll swap them.
It's OBD1. I doubt that anyone is building an app for that.
Of course, I could be wrong.
I figured there was an app for everything! I did a search here and didn't find what I had in mind.
I went out and bought a 15/16 deep well socket to change the cold start injector switch. It worked great on the stripped engine but not so well on a fully dressed one. That didn't work so I just left it unplugged for a test drive. No change. Still getting the Code 45. With that eliminated, there's no need to swap it now.
Still having a bad hesitation off idle. Maybe it's a TPS thing. Gonna swap that next.
I just found out that the timing procedure for manuals and automatics is different. This engine was timed while mated to a manual. I suppose I should check it now. Could be the cause of the hesitation. It would be nice if it cured the code, too.
I swapped TPS and made a little adjustment to the timing. There was some improvement but there is still some hesitation. I went for a long test drive and when I was sure the Code 45 issue was solved, I decided to go home and that's when the engine light came on again.
Next I'll swap coils. Then get new spark plugs. Then try new plug wires.
Been busy swapping parts. I changed the ICM, the rear bank of plugs and wires. I cleaned the distributor cap and rotor contacts. None of that has helped the hesitation or eliminated the codes. In fact, I got a new code! Code 32. I swapped the EGR solinoid and it seems like that was the solution for that one.
Next on the list will be the fuel filter. Doing that first thing in the morning before driving to school.
I'm running out of things to try. There's still the front bank of plugs and wires, the fuel sock, and the distributor.
Try adjusting the TPS. There is a specific voltage value that it needs to be set to (which I don't really remember) but if you don't have a scan tool, you can make a close guess. (You won't hurt anything, in any event.) Watch which way the TPS moves when you open the throttle. You can bend the tab just a bit, to move the TPS in that same direction, with the throttle closed. Make small adjustments, and see how they work. This will make it look like the throttle is open a little more, and should make the ECM react more quickly to throttle opening. (This is for the hesitation. Don't think it will help the code.)
The TPS is the one thing that reacts as quickly as the throttle opening. The MAP sees it too, but there's a delay.
[This message has been edited by Raydar (edited 10-22-2013).]