Fairtax4me Posted May 25, 2012 Share Posted May 25, 2012 This has been an ongoing saga on my 96. Pretty much ever since I bought it it has had this code and I've tried several times to cure it, only to have it snicker and slap me in the face with that CEL. Now I have a somewhat serious running issue and I can't help but think it may be related. It feels like the car is running out of fuel. I give it some throttle and the engine just falls flat on its face, barely has enough power to get out of its own way. Then after a few seconds it picks up and carries on as if nothing were wrong, shift gears and the fun starts all over again. It does this more when cold, once it's warmed up it kinda goes away, but not completely. Never stalls, doesn't misfire, it just doesn't want to accelerate at low Rpm. When it gets wound up to about 4000 rpm it seems to go away. I've swapped MAF and Knock sensors with spares that I have tested and know are good. No change. Checked fuel pressure and get steady 28psi at idle, picks up to almost 40 when driving and never fluctuates, even when the engine is struggling. Checking various readings with my scanner, the only thing that stands out is the MAP, which seems happy to sit at 0.5 inhg constantly. I have two spare MAP sensors which I have swapped at other times in the past to try to cure the MAP code (never fixed it) and will swap one of them in today and see what the reading says with another sensor. I've also tested and swapped the PSS solenoid, which again made no difference. I will also hook up a vacuum gauge later today and see what, if anything, that tells me. Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted May 27, 2012 Author Share Posted May 27, 2012 Not much progress on this so far, although I may have found the source of the running problem. I tried a different MAP, still the same reading. I just had the idea to bypass the PSS solenoid and see what it says, will try that tomorrow. Vacuum gauge read ~18in of vacuum at idle. Vacuum drops periodically when the throttle is opened but climbs quickly afterward and doesn't change enough to raise any eyebrows when the engine acts up. Anyway, I did get somewhat of a break today, tried to start the car after swapping the steering rack and it wouldn't run. CEL came on immediately, so I whipped out the scanner, got a P0100 MAF sensor code. Seemed strange for it to die, but I swapped it with the one I took off two days ago and still no change. Found a broken wire in the harness, one that I had soldered together several years ago while troubleshooting, of all things, a MAF sensor problem. Repaired the solder joint and the problem went away. Haven't driven the car yet because of an issue with the extended steering shaft from the lift not quite fitting properly. It seems the input shaft on the "new" rack is about 1/8" longer than the old one, and it was causing the u joints to bind. So when I get that dremeled out tomorrow I can drive it and see if the power loss issue has been solved. Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted May 28, 2012 Author Share Posted May 28, 2012 Kinda hard to say, I got the car driveable today and I think the power loss issue is gone. I couldn't give it too much, the driveshaft was wobbling around like a wet noodle under heavy throttle. Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted June 20, 2012 Author Share Posted June 20, 2012 (edited) More MAP sensor fun today. I hooked up the Scanner and had a constant steady reading of 0"hg. Drove all around looking for SOME minuscule change and got nothing. Connected the vacuum line straight to the sensor (bypassed the PSSS). Same 0 reading. Swapped MAP sensor with another one, then another one, same thing. So then I went chasing after broken wires poor grounds and low voltage from the ECU, none of which I found. It was hot so I gave up. Might have a go at checking voltage and signal while I wiggle the wire harness on the ECU tomorrow morning before it hits 90. Posed to have 100+ heat index tomorrow so if it doesn't happen before 10 it wont happen at all. :Flame: I have 5.1V reference at the sensor. Approx. 0.9 V on pin 2 at idle, but it doesn't change with RPM increase. I looked for a while but couldn't find any info on sensor output compared to vacuum. Edited June 20, 2012 by Fairtax4me Quote Link to comment Share on other sites More sharing options...
markgcarter Posted June 20, 2012 Share Posted June 20, 2012 Using a scan tool, watch the MAP sensor value with the key on, engine off. Compare the BARO reading with the MAP reading. They should be roughly equal. The voltage for the MAP sensor should read approx. 4.5 volts. Now start the engine and look for a significant drop in the MAP sensor voltage indicating the MAP sensor is working. Quote Link to comment Share on other sites More sharing options...
idosubaru Posted June 20, 2012 Share Posted June 20, 2012 chased broken wires - you checked for continuity from the MAP to the ECU? what a mess. Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted June 20, 2012 Author Share Posted June 20, 2012 Using a scan tool, watch the MAP sensor value with the key on, engine off. Compare the BARO reading with the MAP reading. They should be roughly equal. The voltage for the MAP sensor should read approx. 4.5 volts. Now start the engine and look for a significant drop in the MAP sensor voltage indicating the MAP sensor is working. The MAP sensor on these cars serves double duty as both Baro and MAP sensor. A solenoid inline with vacuum before the sensor switches the pressure source when the ECU tells it to. Only have a MAP value on the scanner. I hadn't though about checking voltage output from the sensor with the engine off. That should tell me that the sensor is responding to vacuum. Thanks for that idea! Yes checked continuity between the sensor and ECU. ECU isn't hard to get to though, just pull the carpet back and 4 screws to remove the cover plate. Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted June 20, 2012 Author Share Posted June 20, 2012 I get 0.914 V on pin 2 with no engine vacuum, when reconnected to engine vacuum it drops to 0.903. Vacuum Gauge shows close to 20" at idle. Still get no reading from the scanner. Tried wiggling wires, unplugged the ECU multiple times to try to clean the contacts, sprayed out the connect with electronics cleaner, and got no change. It's down to two things. Either the ECU is bad, or I have three bad MAP sensors. Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted October 23, 2012 Author Share Posted October 23, 2012 Update time! Basically I left this alone until just a few days ago when we had time to play around with some stuff in Electrical class. So I grabbed my two spare MAP sensors and put them on the test trainer in class. 5v and ground hooked up, I got 3.8v from both sensors on the signal pin with no vacuum. Hooked up a vacuum pump and at 19" I got 1.6 v from both sensors. Finally found some info in the FSM about the voltage to expect from the sensor and this falls not exactly in spec but close enough for me. Put a meter the one in the car and tested 5v on pin 3, no back voltage on ground (good ground), and 1.2v on signal pin w/no vacuum. Hooked up vacuum pump got 1.2v. Now is where this gets interesting... Tested the same sensor out of the car, 5v and ground hooked up... 3.8v on the signal pin. 1.6v at 19" Hg. Back to the car, pull the ECU connector off, recheck wires, continuity good, no shorts. Decide to pull the sockets out of the harness connector at the sensor. Hook up power and ground pins, but NOT the signal pin... Key on, 3.8v on the signal pin. Key off, plug in the signal pin, key on again, now have 1.2v signal. Pull the ECU connector off, pull the signal wire out, key on 3.8v at the sensor, 3.8v at the ECU end with the signal wire UNHOOKED. ???????!!!1 Pretty much what I had assumed before but this finally confirmed it. The ECU is bad. Quote Link to comment Share on other sites More sharing options...
idosubaru Posted October 31, 2012 Share Posted October 31, 2012 Pretty much what I had assumed before but this finally confirmed it. The ECU is bad.wow, what a crazy mess, would have never guessed that. i have a 1996 Legacy Lsi EJ25 ECU if you want it. so - is it probably the shorted wire that fried the ECU? Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted November 1, 2012 Author Share Posted November 1, 2012 See that's the problem. I can't find any shorts. I've checked as much of the wire harness as I can get to and I see no chewed or chafed or crimped wires. The ground for the MAP is (according to the FSM ECU I/O) shared by several other sensors and since none of them are acting strange I fail to see how that can be the problem. Got a solid 5V at the sensor, reference is good. The only thing I can think, and this is based on schematics for a Toyota ECU so it may not be the same for the Subaru, is that there is a resistor that bridges the reference and signal traces in the ECU (a "pull-up" resistor if I understand correctly). The resistor would have to be open internally, providing no pull-up reference to the signal trace, which would tell the ECU that the signal voltage is out of spec. Or, what I think is more likely, there is a poor solder joint or connection, or corrosion (some form of high resistance) inside the ECU on the MAP signal trace. Unless there's something I'm missing. I'd like to open it up to see if I can see anything obvious, but don't want to take the ECU apart without having a known working spare on hand. Quote Link to comment Share on other sites More sharing options...
Fairtax4me Posted December 8, 2012 Author Share Posted December 8, 2012 Gonna put the P0106 epic to bed. Finally got to the junkyard and picked up an ECU from a 97 Legacy L. First improvement: I get a MAP reading on the scanner! It seems low at 10" at idle, but its there! Second improvement: Made two stops on the way home and no CEL, no DTCs pending. Two ignition cycles is all it took before to trip the CEL. Power loss issue may be resolved, a few more drives will confirm but I won't be too disappointed if it isn't, as long as the CEL stays away. Fuel mileage has been kinds low lately, hoping that now the ECU has a proper input from the MAP it'll lean out a bit. Again here, not gonna be too disappointed if it doesn't. Kinda figure this thing needs an O2 sensor by now anyway. It's only been like 4 years? since this issue first popped up. I should organize some of my experiments throughout this story into a write-up. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.