jbates346

Hopefully we have some electricians in here. I post this in another section but i want to get a lil deeper in it. The problem was my ECM showed me 248 degrees after about 2mins of running. The gauge in the dash said about 120. Obviously that one is more accurate. From what I understand the gauge works off the water temp sensor and the ECM goes off the coolant temp sensor. I just changed the sensor a lil bit ago and i'll be hooking it up to the scanner tomorrow to see what happens. Heres what throws me off. When it first read 248, I unplugged it. After I unplugged it read 113. So how in the hell can that happen if the sensor is unplugged? The only thing i could figure is a partial short to ground. I never hear of a PARTIAL short tho. Is this possible? How can you test this circuit to see if theres a partial short? Any ideas or hints will greatly be appreciated.

BigBadLou

iTrader: (3)

You could unplug the sensor and measure its resistance. Then plug it back in. Unplug the ECM connector and measure the resistance on the appropriate pins that bring the sensor reading to the ECM. The resistance should match. If it doesn't, you have a wiring problem.

Hope this helps.
Lou

jbates346

I believe its shorted to power somewhere. Somewhere.....thats the question isn't it. If the ECM does not get a signal from the ECT does it look at another Sensor (i.e. IAT) for the information? I'm thinking that it might be partially shorted to the power side of the TPS, just cause the pin is right next to the ECM. When I did unplug the sensor it told me that it was 113 degrees. My question is where is it getting that voltage from, and how in the hell could the ECM even get a input signal with the sensor unplugged (open curcuit)?

jbates346

bump

BigBadLou

iTrader: (3)

Variable resistance sensors are routed to the ECM where they are connected into a voltage divider (simple resistor bridge). If the sensor's resistance increases, the voltage goes up and the ECM translates it into a lower temperature (for example). If the sensor's resistance decreases, the voltage in the divider drops and the ECM translates it into a hight temperature reading. That's how the ECM "sees" real world values from a resistance sensor.
Now if you unplug the CTS, for example, the voltage spikes, is translated as -40°F and the computer throws a code for sensor out of range.
If the CTS (or its wire) is shorted to ground, the voltage drops to 0V and is translated by the ECM as 285°F and another code is thrown.
So that's how it roughly works.

Please measure the voltage on the two CTS pins on the ECM with the CTS connected and disconnected. That might tell us if there's a ground problem.

Also, when you disconnect the CTS and the ECM, measure the resistance of the wires between each other and also each against good ground.

Let us know what you find!
Lou

jbates346

Those are some good tests. I appreciate the ideas. Tomorrow we're gonna dig it to it and do that. I was thinking about doing something similar too. Today we had the scanner up to it and it said 121 degrees, good right? Well......we started looking at some other stuff and then about 15mins later i said 241. How the hell does that happen? Nothing changed, the sensor was plugged in all the time. When we did unplug it and it said 110. What the hell is going on!! So tomorrow we gonna use a test light and disconnect the ECM if it still has a ground then it has to be a pinched wire. If the light goes off then the ECM is grounding it for some reason meaning the ECM is bad. I mean its just making up s***. Another note i want to point out is that my A/C request said no but my A/C clutch said yes. So why is it turning on the A/C clutch when it knows i'm not requesting it? Also it doesn't like to start when it sits for awhile. i'm thinking it might be the cold start injector. What you think? Does it sound the ECM needs replacing?

BigBadLou

iTrader: (3)

Before you start spending money on new parts (ECM, cold start injectors), it's good to make sure that they actually do need to be replaced. That's why you'll be doing all the tests, to determine whether it's the wiring's fault or the ECM's fault.

The A/C request comes from you as the driver, not the ECM. The ECM only "sees" the request from you (when you move the hvac control lever to A/C) and it adds a little more gas to keep the idle steady. You might have a bad connection somewhere if the ECM doesn't detect the A/C request.
(diagram here)

Hope this helps.
Lou

jbates346

Ya i totally agree. I was supposed to get in it today but something came up of course. I get what your saying, but what i don't understand is how can the ECM say A/C request no, but fan clutch yes. So it knows i don't want the A/C on but its grounding the A/C clutch? What is it doing that. Furthermore, the A/C doesn't even go on. I'll check the voltage at the conn. and see if its even getting anything there. If not we obviously have an open circuit.

In answer to your other question my plan was to disconnect the engine fan and the coolant temp sensor straight from the ECM. If i see ANYTHING on the scanner's pin data i know the ECM is bad because at this point it has to be just making stuff up. What's your take on this?

BigBadLou

iTrader: (3)

The A/C request does NOT come from the ECM. It's a common misconception. If you look at the diagram, the DRIVER controls the A/C circuit by the HVAC switches and the wire going to the ECM is an input wire, signaling to the ECM, NOT controlling from the ECM.
The A/C can work normally even without the computer. Like for example on carbed engines. The A/C is a completely independent circuit.

Hope this helps.
Lou

jbates346

I understand that part. But my question is how come the ECM is turning on the A/C Clutch if it sees that I'm NOT requesting it?

BigBadLou

iTrader: (3)

Once again the ECM has NOTHING TO DO with the A/C circuit. YOU control it 100%. You engage the compressor clutch by flipping the switch.

The ECM equipped cars merely READ the status of the clutch and if you turned the A/C on, the ECM adds a little fuel to keep the idle steady.

Lou

texasautobike

Sorry to bring up an old thread, but I had an issue just like this on a 91 tpi. Cold start temp was at 109* when it was 50* ouside. Hard to start without opening the throttle a little. Then at the 120* ish mark would jump to 245* and the fan would kick on, idle is perfect and ran perfect.

Voltage at the C16 pin (yellow wire) was 3.65v (unplugged) instead of the 5v it should have been being disconnected. Because this voltage was wrong at the ecu I pulled the chip and found a modification in the file to adjust the top voltage. Like someone wanted the fans to kick on early which could have been done differently. Changing the voltage output told the computer to never run cold mode startup, it thought it was already warm. Without modifying the ecu (prom) you would have to just deal with the cold start and horrible fuel mileage. Or adjust the throttle screw to make starting easier, this inturn won't allow the IAC to do its job and would hunt idle around 850-1000rpms and still have terrible mileage.