PDA

View Full Version : Wierd Electrical Problem, out of ideas, need help!



KASilviaAR
09-09-2012, 02:43 PM
OK, so I have had my friend's S13 with a redtop SR over at my house for the last month or so, trying to figure out what inna he!! is wrong with it. It's a stock longblock with a T25 and some basic bolt-ons (FMIC, BOV, etc.), and it's running stock boost. Engine management is a stock #62 ECU, with a very good stock lengthened harness, stock SR or KA24E MAFS. He has an SAFC2 hooked up to it in "Monitor" mode only - i.e., he doesn't have the MAFS wires hooked to the SAFC2.

It starts and runs perfectly at idle and while it's cool. Once it begins to warm up, it begins showing knock on the SAFC2 - levels are anywhere from 4 to 20 when not on the gas, and higher when you accelerate. Somehow the ECU is actually seeing this, because it does pull timing, and it won't hit full boost once it's hot.

Here's what I've done:

1. I've set the base timing with a light. One thing the car does that's weird is that it won't act right when setting timing as per the FSM procedure. When you unplug the TPS, the base timing drops way down. So, then I adjust it to 15* BTDC with the CAS. Then when I plug the TPS back in, the dam thing pulls the base timing back down to near 0*. I tried doing this procedure several times, and it just keeps doing the same thing. So, my "work-around" was to leave the TPS plugged in and just adjust the CAS to 15* BTDC while it's running (although it's hard to make it get to 15*). Then I unhooked the negative batt. cable and hit the brakes a few times to reset everything. When I recheck timing after doing this, it seems to hold pretty close at 15* with the TPS plugged in.

2. He's replaced the knock sensor, but not the short extension harness for it. The engine still acts the same as before he changed the sensor. So, I unplugged the harness at the intake manifold, and I soldered a 1-MegOHM resistor across the 2 wires, on the engine harness side of the plug, thus eliminating the knock sensor and extension harness completely. Whaddya know - the car acts exactly the same as it did before!!! No knock shown on startup, but then it shows knock as the engine warms up, pulls timing, won't get full boost, etc. - same exact story! But now there's no knock sensor hooked up at all - it's been bypassed.

He's replaced every part in the engine control system - he's even tried a brand new uncut engine harness, different ECU, different MAFS, different TPS - everything! And it still does the same exact thing. The engine has ground wires like spaghetti all over it, going to the body, frame, and everywhere. The alternator shows correct voltages when it's running. The long block is physically very healthy - no knocks, smoke, blow-by, oil usage, etc..

I'm totally stumped by this one! I'm looking for ideas!

Thanks, guys!

Mike_TX
09-09-2012, 03:34 PM
That IS strange. Timing is slipping when it warms up.

Have you done a MAF reground? That seems to have been an issue with a lot of SR20's and could cause it.

Also, there is an Engine Elec (B) fuse in the fusebox in the dash that is on the O2 sensor circuit. If bad it can affect idle, and maybe only when warm.

PVC and related hoses good and no leaks?

Just some shots in the dark after some internet surfing. :thinkingsmiley: :shrug:

.
.

KASilviaAR
09-10-2012, 02:39 PM
On the MAFS reground, it "shouldn't" be that, because he's tried replacing the entire engine harness with a brand new OEM one, and it did the same thing.

I'll look at the fuse tonight and report back.

Thanks for the ideas! Keep 'em coming!

Kingtal0n
09-11-2012, 09:43 AM
I would start trying parts on another vehicle. Take the ECU/TPS/CTS/etc... and try it on a good running vehicle with the same engine.

If the ECU is actually the culprit- that is, if the ECU is responsible for the change in timing, then either the ECU itself is causing the problem, or something in the sensor/wiring is telling the ECU.

Could be the SAFC wiring is messed up somehow with the harness causing a problem. When the new harness was installed, a new SAFC was also used? If not, the safc is still suspect. I would remove the safc all together first, then set the timing based on the indicator mark on the CAS, like the FSM proceedure. That will get it very close to 15* BTDC without guessing or using a timing light... because your timing light could also be wrong, make sure you use a high quality light and check the polarity of the pickup.