Hi guys,
I recently installed a used TPI system on my 86 IROC. It has been running great except for extended cranking time when it is cold. After some research i found out that the Coolant sensor is usually the culprit. Today, i found that the ground wire from the sensor was cut. I resoldiered the wire to ground "fixing" the problem. Now the car starts readily but sets an error 15 "CST low temp indicated". I measured the voltage of the sensor wire unhooked from the sensor and have 5V. The sensor measures 450 ohms which is perfect. The 5V drops to 2.5V with the sensor connected. I pulled the ECM fuse for 30 seconds to clear it with no luck. HELP!!!!

 

Are you certain you are looking at the the correct sensor? There are three temperature sensors on the '86 TPI. You need to check the one with the yellow/black wires in the connector at the front of the intake crossover.

The error code is generated when the ECM cannot read sufficiently low resistance from the sensor. It is currently seeing something over 400K ohms, which is essentially an open circuit as far as the ECM understands.

 

Thanks man, I actually figured out that the ground wire was cut on the sensor So, after repairing the ground, my car is now running much better and starting very easy. Now all i have to do is figure out why it wont go into closed loop and why the SES light is on even though there are no error codes being stored...

 

Here is my two remaining problems in more detail... I shorted ALDL A&B while the engine was running to check on the Closed/Open loop status. After the engine reaches 180 degrees and i am driving for about 15 minutes, i would expect the engine to go into closed loop but it seems to want to stay in open loop almost all the time. THe only way i can get closed is to rev the engine, accelerating, and decelerating. When i rev the engine it will go into closed loop for about 5 seconds and return to open loop. I will also go into closed loop as long as i am accelerating or decelerating. Also, after i fixed my CST sensor, my SES light comes on and stays on all the time where it was off before i fixed the sensor. I check for error codes and there arnt any being stored... Unfortunatly, Chilton doesn't include the SES light on his wiring diagrams, so i am at a loss as to how the wiring is supposed to go.
THanks in advance!

 

This is a typical system:

 

You need to check the CTS resistance cold, warm, and hot and compare them to the specifications (450 ohms = 160° to 170°F.) Resistance should go lower than 450 ohms for 180°F.
Closed loop shouldn't take 15 minutes, just idling mine takes less than 5 minutes.

 

CST resistance looks good rising and falling as expected... I have managed to get a code 13 out of the ECM relating to the O2 sensor. I checked the voltage on the ecm side and got ~2.8V which dropped to .53 when i plugged in the connector. this seems about right to me on a cold engine... Also the O2 sensor is a brand new AC Delco sensor.
Thanks again!

 

¿Any progress?

 

Eh, i still have a SES light with no error codes. Engine still wont stay in closed loop. I can only get closed loop for about 5 seconds when i am accelerating then it goes back to open. WHat would you do? I really need closed loop and i think that is why im only getting 10MPG :P Now, considering i have ran the car for about 3 weeks with the CST disconnected and a bad O2 connection... Do you think that would be enough to kill the O2 sensor? SHould i swap it out? Do you think it is the ECM? ONe more thing: I did managed to get a code 44 out of it. I cleared the code and am watching it to see if it comes back again. Can you think of a reason the ECM would set a hard error and not store a code for it? Thanks again for all the help!

 

Before you condemn the ECM, make sure the connectors are clean and tight, and verify that the CALPAK/MEMCAL in the ECM is fully seated. Disconnect the battery, drop the ECM, and check it out.

You can test the O² instead of just replacing it. A voltmeter and propane torch are all that's necessary.

Does this vehicle make nothing but short trips, or does it get some road mileage mixed in?

 

It gets driven about 8 miles a day durring the week. and maybe 25 or more on the weekends. No regular long distance trips... So i take it that you check for fluctuation in the resistance of the sensor while running propane over it and while in fresh air?

 

If you have a DMM you can test the sensor fairly easily. Remove the sensor and connect the voltmeter on a 2.0VDC scale. Hold the sensing element end in the flame of a propane torch while measuring voltage. The sensor should produce a signal of at least 0.75VDC within 60 seconds. Removing the flame should drop the sensor output to less than 0.10VDC. This does not test the quality of the sensor, but will tell you if it is completely "dead" or not. A propane torch flame is nearly devoid of all oxygen and should produce a fairly high signal as a result, just like a very rich mixture. If you don't get the voltage, your sensor may be toast, or contaminated with coolant, carbon, incompatible RTV curing agents, or oil.

If you are only driving four miles one way, the engine and oxygen sensor may never be getting hot enough to develop a good signal. Moreover, it there is any contamination, it may not be burning off. Further, four miles of driving is not adequate for the ECM to write integrator data to the BLMs. That means the table data in the ECM may never have been accurately created for the specific installation, and the vehicle is constantly operating on basically backup parameters. You really need to take about a 20 minute drive at varying speeds and loads AFTER the engine has achieved full operating temperature (not including the time to get to temperature) so that the ECM can at least begin to write data. That's a minimum. Eight miles daily isn't going to get it. 25 miles total on the weekends isn't going to get it either. That's why I asked the question. I suspected that might be a problem. Until the ECM can "learn" the sensor data for the specific installation and vehicle, things may never change, even if every piece of hardware is operating perfectly.

Removing power from the ECM will reset everything, so you'll get to do it all over again.

Also, I neglected to ask earlier, but you mentioned installing a "used TPI system" on your '86 IROC." From exactly what model, year, and engine was that TPI system liberated?

 

Wow, thanks alot Vader!!! I guess ill have to use that as an excuse to take a 200 mile trip to Twin Falls and back It is kinda strange now that you mention it... I took my friend home last night and the SES light went out :| It stayed out the whole 5 miles home too. I took the freeway (Idaho roads are very rough) to see if i could get it to come back on with the vibration- stayed off... I did some jack rabbit starts and quick stops- still off... Still in open loop but off. I guess these symptoms must be normal until the ECM calibrates. Ill take my long trip and see what happens this weekend and let you know. My TPI system is off a 305 89 Camaro. I Have the original 19# injectors which seemed to still be ok- but due to hard starts (CST disconnected ) I replaced them with a set of 22# dyno'd Bosch injectors. THe MAF is used off an 88 IROC (The guy swore up and down it was good before he threw a rod) I bought new sensors direct from GM (O2, Knock, MAT, CST, and Oil sender). However, the TPS, and IAC motor are the origionals. I replaced the fuel tank with one out of a TPI camaro (year unknown) I modified my carb fuel sender by extending the return line with a rubber hose down to the fuel reservoir. I mounted a new MSD fuel pump outside the tank and used high pressure fuel injection fuel hose to patch it up. The origional carb fuel lines went over the top of the tranny to the passanger side, i cut them there and used the hose all the way up to the fuel rails. I used felpro gaskets and "black" RTV sealant only on the very front and back of the block. i had to modify the manifold a bit to make the four center bolts fit on my 86 heads. I reused my large-cap distributer. Also, i had a custom chip made by TBIchips with no VATs or emmissions. Other than timing and firing order issues it took right off with no major problems. that is pretty much the extent of my "little" project. Thanks again for all the help!

 

As long as the RTV was labelled as "Sensor Safe", it should be fine. If the RTV used was not "sensor safe", the curing agents in the RTV may have "poisoned" the zirconia sensing element in the O² sensor, rendering it useless. You may want to search for that old tube of RTV to be sure.

FWIW, a Bosch O² sensor designed specifically for the engien would work fine as well. Bosch used to be the OEM supplier for oxygen sensors.

 

Good news! I found the problem with the O2 sensor. I looked around and found a post by Trickster on troubleshooting a code 13. I remember that the voltage on the O2 sensor unplugged was ~2.8V which ment that the O2 ground was probably open.When i assembled the system i somehow got the 85 and 89 wiring schematics mixed up. IN 85 the MAT wire was tan and in 89 it was green. Also, in 89 the tan wire is the O2 ground. To fix it i simply cut the tan wire D6 and tied it to ground. Bingo! Closed-loop!! Now all i have to do is locate the open MAT wire and patch that together, reset the ECM, and go for another long drive. Thank you very much for the help!