Harness number three on my Spec E30 has gone bad. During the last Club Race at Roebling this past December my temperature readout on the IQ3 in car went nuts. I have backup oil pressure and coolant temp gauges, so I wasn’t concerned and finished out the race. This past weekend I worked on that issue, figuring it was a bad sensor, bad connection, or bad analog input channel. But no, it can’t be that simple, easy, and cheap to fix.
The IQ3 reads the resistance of the OE gauge sensor using a 5v source and current limit resistor. I pick that signal up at the cluster connector. While diagnosing the problem I found that the input to the IQ3 was sitting at 10v when it should have been less than 5v. Disconnecting the wire from the IQ3’s input channel proved that the cause wasn’t a blown input module. Disconnecting C101 took the voltage to zero and disconnecting C191 didn’t make any difference. Therefore the problem is between C191 and C101, i.e., a short in the harness.
This is the third harness that has gone bad. The first harness in the original car worked for about a year before it developed an intermittent cut out problem. The car might behave all weekend, it might act up occasionally, or it might act up all weekend. The cut outs only occurred on track and above 5k. In the quest for that over the course of a year every part of the engine management system was changed at least once without ever solving the problem. That car was wrecked at Road Atlanta in 09 and I built another Spec E30 using what I could from the wrecked car.
The PO of the second car had cut out C191 and spliced the wires which meant that you couldn’t take the harness off the engine (which had a blown head gasket). So I used the harness from the old car. Once I got the car put together the cut out problem was still there randomly (new engine, new car, etc). I finally found the cause with a battery operated storage scope. As the rpm climbed, the CPS input signal to the DME distorted. It didn’t matter what DME or CPS was in the car, so that left the harness as a probable cause. For grins, since I had a way to find the cause I tried th first harness and saw the same thing. My guess is that time and heat had taken it toll on the coax in the harness that routes the CPS data to the DME.
I grabbed a damaged harness from a 525i, cut off the half of C191 I needed and spliced it into this car’s original hacked harness. I put that harness into the car and life was good. Well up to now. Late this summer the car occasionally would loose fuel pressure. A new pump, relays, cleaning the socket contacts, etc., all seemed to help for a while and it seemed that the last contact cleaning work had fixed the problem. Then the problem with the temp sensor showed up. Now I’m thinking that the gauge and fuel pump relay problem are related and caused by damage inside the harness. My fiddling with the relay sockets may have just disturbed things enough to fix the fuel problem. I’ve checked the splices I made at C191 (solder & heat shrink) and the problem isn’t there. It is some place within the harness. It could be chaffed insulation or insulation damage from other causes.
Given my experience with harnesses I’m thinking that, as expensive as it is, a brand new harness is the only thing that makes sense. Poking around it looks like I can get the engine harness for ~$490 and the injector harness for ~$87. That sounds expensive, but if I get another used harness and have it cause problems during a race weekend the lost weekend will cost about the same as the new harnesses.