86 Wonder Wedge

The wiring can be bad as well at the IAC plug/harness (right there by the TB), that was my IAC problem. And it sounds like the ECM is trying to compensate/stabilize the idle.. Is/Was your A/C OR defrost on in that video? Does it make a difference if you hit the A/C or defrost? Does yours work/ do you have it still installed and hooked up? Also, check for corrosion in/around the coil terminals.. could be getting a bad tach signal to the ECM.. Also, have someone do a charging system test/alternator test to make sure the diodes are still good (little to no AC voltage ripple). "Pulsing" voltage can wreck havoc on the system, especially the MAF since it's wired right to the IGN circuit (not powered from the ECM) Or just trying unplugging the CTS, TPS and MAF sensors.. it'll still run, and should run pretty well (since it's "flying blind") or atleast more stable.

When it dies, can you hear the fuel pump cycle when you turn the ignition on (but do not try to start it)? Should hear/feel it for about 2 seconds. If not, could be an intermittent ground or the fuel pump driver in the ECM dying (fairly common). EDIT: Is you clicking sound cyclical (every 2 seconds or so)? Make sure the green and white/clear connectors by the fuel filter (in the engine bay) are unplugged. These are the ECM self-check connectors for dianostics and sorta puts the car into zombie mode. It cannot compensate for load/driving conditions and could cause a stall or weird running after it warms up...

Try unplugging the MAF and Coolant Temp sensors. This will force the car into an open-loop mode (a default, "run-blind" mode) which the car uses made-up info to make the car run. I have a feeling you have a dying IAC valve or a damaged/faulty MAF. The car will close the IAC (like you're in cruise/accel) and bump up the timing/fuel to fast-idle as you warm up. But as the car warms-up (according to the CTS (coolant temp sensor)), it will try to open the IAC and bring down the timing/fuel to normal idle. Sometimes they can get stuck or fail to open.. Also check for a large vac leak (intake gaskets, the intake duct itself, badly leaking oil pan/valve cover, pcv hoses, brake booster hose, ect. )

General Altimax RTs are 50 each on tirerack.com in the original 13 tire specs. Fantastic tire. Bought a set of them leas than a year ago and I love em. Plus, at the time they came with a 50 debit card, so the mount and balance were free.. Had them shipped to my local Goodyear where they have lifetime mount and balance. job done.

I tried couple months ago, but they said there were none left to order..

Bosal makes a nice OEM-style direct fit unit that doesn't look like hell (The Catco unit I ordered for inspection was a major disappointment for 200!) The Bosal at Advance is about 220 retail, I think, but if you order it online, you usually can shave 15-20 off the top at any given time, and free shipping (no one will have it in stock, it's a factory direct item)

Is the idle real slow to respond, like if you rev it, it'll take a few seconds to come back down? Could be a coolant temp sensor, or a vac leak of some kind... I know when my 91 Loyale is COLD (like 20F outside temp), it's a bit slow in bringing the idle back down when coming to a stop, until the blood starts flowing a little bit better..

Yep, should be the 10-pin connector, and on the wire side (not ECU/ female side) it's the top row, dead center. (RIGHT under the release tab). IIRC, GR/Y wire.. but don't quote me.. And It should be 10+V at the ECU plug position 45 (described above, unplugged, of course) KOEO. If not, go into the engine bay and unplug the IAC connector and test for 10+V at the harness side connector on the W/R wire (I believe, one will have voltage, the other won't).HERE is where I had my problem. When I unplugged the connector, it would rest and I would have 12V. But when it was bent and plugged in, it would lose connection. Moral is, flex the connectors of both the IAC and harness side and check for intermittent connections. If it is still ok, check ohms across IAC valve connector. Should be between 5-12 ohm. If it still tests ok, then unplug the smaller of the 2 round large harness connectors on the driver's side, under the coil, aft of the battery tray. Check for voltage at the appropriate pin for 10+V, KOEO. If STILL ok, then do continuity (resistance check) test from the body-side round bundle connector to the pin 45 in the ECU plug. After that, and you still have 12V at the ECU, it might be the ECU. Also, if the system checks out, plug the ECU back in, KOEO and actuate throttle with the IAC feed hose disconnected so you can see the pintle through the tube. As you bring the throttle off idle, the valve will open and shut. You could also have a mal-adjusted TPS (the car thinking it's never at idle and the valve never opens to allow air into the chamber at idle). Run the voltage/resistance check according to the 89 FSM (the SPFI system is identical). Good Luck!

I'm still running 5w40 Mobil1 Turbo Diesel Truck in my 130k EA82 because a) it covers the operating temps year round, more zinc than any other 5w40 out there, and c) can actually do the 7500 mile interval Subaru recommends in the manual. Yes, my valve cover gasket leak has increased slightly due to the detergent nature of the synthetic oil, but I get no more lifter tick, the oil still is 60% "clear" at 5k miles and when teamed with a Purolator PureONE, Blackstone said I can run a 10k mile OCI with that setup. And with the high zinc, it adds just a bit more protection for the vulnerable cam towers (and their lack of cam bearings. Really, Subaru?) Castrol is a nice oil, been running GTX in my 265 Pontiac V8 when my old Diesel spec 10w30 rotella T ran out.. Has the lowest paraffin content of any similar oil..

Unplug the ECU plug (at the ECU) and check for voltage at pin 45 (small, yellow connector 10 pin) should be above 10V with KOEO. I had an intermittent CEL for the idle air valve and turned out to bed a bad connector at the plug in the engine bay AND a missing pin-anchor at the ECU plug. Would NOT idle when cold and would dip/stall when coming down from 3K RPM.. You can also run jump leads right to the pins on the IAC valve from the battery (or 12V source and listen/look for the valve to open and close)

Actually, those plugs look pretty good! I touch on the lean side, but far from a problem... And at 233K, bearings are always a risk and considering the lack of availability of a spare motor, I'd have a back-up plan (i.e. save up enough to build/import another one, or find a shop that can bore/deck subaru blocks) But those bottom ends are pretty stout and given the stock-usage of the car and in a significantly lighter application, you should do just fine. EDIT: for safety's sake, you can have an oil analysis done on the old oil and cut open the old oil filter an look for any bad signs (or lack there of)

Best place to buy is online as most parts houses want 100+ for the CTS (coolant temp sensor). Rockauto.com is my favorite as they have good prices on SMP (standard motor products) which IMO is a good line. (SMP is also BWD, fyi) What you could do as well is unplug the CTS (green, two wire sensor just under the thermostat goose neck) and see if it changes how it starts/runs.

Try unplugging the MAF and go for a drive, see if it makes a difference. I also had a similar problem with mine, recently.. it got so bad it wouldn't rev past 5200 RPM. In the end, I didn't actually change anything except playing with the wire routing at the ECU, but I suspect it was/is a dying injector. It was noisy, sputtered at idle slightly and my fuel mileage isn't the best, one day it just "fixed" itself. I kept my foot down, it bucked a few times, then has been running like a top ever sense, but the idle problem is still there... Anyway, I'd pull a plug or two and see how she's running..

Actually, the SPFI computers will run without ay sensors connected (minus the CAS for the spark and injector pulses). It's designed to "create psudo signals" so you can still drive the car.. I've personally driven it without the MAF, TPS, CTS, Neutral switch OR O2 and it still ran pretty well (gas milage was atrocious, but it runs in "open loop" until it's fixed). For no top end power, I'd double check your timing. And what sort of fuel are you running? Some places here in Ohio have been "testing" E15 without telling people (minus the ones that come back with complaints and poor milage/problems) which is a major no-no. (Let alone it's not approved in cars before 1996 (i.e. OBDII)) And trust me, fill up with a tank of E0 (Pure gasoline, not the blended stuff) and then a tank of the standard E10 (99% of gas stations, this is standard 87, 89, 91 and 93 octanes) and you'll notice a difference. The power just isn't there anymore and since these cars cannot compensate nor adjust to use the knock-retard ability of the E10. We just get poor power and worse milage.

IIRC, the older EA82s had the 3 slotted adjustable coil seats on them where later ones did not (that's where the air-adjustable setup took over). Chances are, someone has put 2WD struts in the rear (the 2wd struts are approx 2 inches longer than the 4wd struts to account for the different mounts for 2WD cars/trailing arms) If the distro in on the driver's side by the firewall, it's an EA82. Up front by the alt, p/s and A/C, EA81. Also, you COULD have an LSD, but it would be the clutch-plate style, not viscous. Newer subies have them in the rear (WRX, some outback sports and legacy outbacks) and only the STi got a LSD up front (which IIRC, is a Torsen). You also DO NOT have a fulltime/AWD gear box. Treat that 4WD like a truck's 4WD. Only on loose surface and in a straight line. But nice find! Luckily, these are like legos. Also, I'd inspect/retorque the front spindle nuts (145 lbs-ft) and if you can, to to save and reuse the OEM axle.. unless it's an EMPI new axle, you risk vibration problems or popping CV joints.. these transaxles do not take kindly to poor quality CVs...

Yep, which is why I pulled the trigger on a new ECU... while I checked all the ECU grounds from the connectors (with ECU unplugged, of course) they all spec'd out 0.04 ohm (not K ohms this time.. ) but I didn't do a voltage test.. hmm.. BUT THE GOOD NEWS. I FOUND MY CODE 24 PROBLEM! I go to try and solve my "no grounding" issue, and I find THIS time, there's no voltage at pin 45 with KOEO. So I pop the hood, unplug the IAC conenctors, find 11.79V at the ECU side of the harness from IGN (W/B wire), 9.6 Ohm on the IAC conenctor across the terminals, >1M ohm to the IAC valve body. So I plug them back together, find 11.79V going into the valve, 11.79 out (on the IAC side) but 9.1V out on the ECU side at the round bundle conenctor. So I isolate the "engine harness", run my DMM from the IAC connector to the bundle's appropriate pin, and get 88 ohm. But then I flex the conenctor (since the wiring and conenctor were approx 90 degrees to each other) and then I lose conduction. I move it back, 10 ohm, then 1, then 0.4, back up to 80. I've got a bad wire. HALLELUJAH! Now, now next question is, when replacing it, do I use another shielded wire? Looking at the FSM, it mentions nothing about the IAC ground wire being shielded (unlike the TPS and O2 sensors), but this has honest to god shielding. It starts about 1.5 inches back from both ends, has black insulation, with fine silver strands, and the factory white insulated 18GA wire beneath it. Luckily, the bad kink is before the shieding, so I wil be soldiering in a new section, but is it that important? And the shielding doesn't ground anywhere, unlike the TPS and O2 sensors... Ideas?

Could be a lying Coolant temp sensor. How's the gas mileage? Also just have her try and hold the throttle 50% and try to start. An idle air control valve can do that too (failure to open, stuck, ect) so an open throttle can "mimic" and idle air valve allowing enough air to get in to start. Also tell her to keep an eye on the water temp. My 86 EA82T was doing the "refusal to start/idle" when warm just before the headgaskets when.. but that's the extreme end for the SPFI...

Alright. New, hair-pulling info. After angrily keeping my foot in the firewall when it was doing it normal bogging/power loss above 4K routine, it bucks several times, some relays click, and then takes off like a bat out of hell. Code 24 still exists, but now I've got full power all the way to 7K with no hesitation. The hunting/popping idle still exists, but she hasn't felt this way in months. After reading the TPS resistances at the ECU plug (with ECU unplugged), I got the same numbers as before (0.62K and 4.55K ohm) I also checked voltage (with the ECU first unplugged) at pin 45 at got 11.70V KOEO (same at BAT voltage) with code 24 set. When I plug the connector back in, and apply a test light to pin 45, engine running, the code 24 goes away, and the idle picks up, and the code 24 doesn't come back until a few key cycles later or I unplug the connector at the ECU. I also bit the bullet on a new ECU, same designation and ZERO difference. It was only 50 bucks, so I figured "hey, if the FP driver fails on the original one, I've got a back-up". I'm doing the voltage before/after 1 min test today now that I've got some time..

I assume you mean the T-belt covers, and pull off the easy-side covers (since the middle one requires you to remove the crank pulley) and 1) make sure they're still spinning, and 2) make sure the timing is still on. Sometimes, the rear covers are old/warped and when pulled together with the front cover, can bind on the cam sprocket causing it to seize and either jump the belt or break it..

The FSM dictates that I check the resistance of the TPS, isolated (connector unplugged) verifying that certain resistance are met at certain times (idle, WOT, and the idle switch opens/closes at the correct associated degree of valve opening). It also dictates I check the line resistance at the ECU (connector unplugged) to verify the resistance are withing spec (or matches the readings taken at the TPS). It also has you verify 5V at the TPS plug with KOEO (which it is). Well, today, she fired up with no code 24, fast idled and let me spin into 6K without bogging (some flat spots above 4K, WOT and feels like fuel cut at spots, but much better). But then when I pull it into neutral coming off the express way, she stalls and upon restart, code 24 is back. But now I'm starting to feel it's a ground/power distro problem cause sitting at a light before it stalled/code 24, I could hear a VERY light click from behind the dash (maybe radio/clock area?) and the shudder at idle and lower reading on the tach were synchronized like atomic clocks. I couldn't hear it now on the way home (upon several rolling restarts, the code 24 just decides to clear itself ) and sitting in the drive way. But now the RR window won't respond and I hear a cyclical clicking coming from that door. Awesome. I'll work on that today (and the rear brakes) and "hopefully" the 1 junkyard within 250 miles (25 from me) has got 2 loyale ECUs left that matches mine AND they aren't shot AND they're open on NYE. Fingers crossed.

I did and they were the same all the way to the ecu.. On the drive home tonight, I discovered that it won't rev past 5200 no matter what, even in neutral. So I unplugged the MAF forcing it into open loop, no change.. So I'm leaning toward a worn out injector, or the ecu is failing...

Yeah, that's a typo.. And I haven't checked the new O2 sensor, but I did a resistance check between the IAC harness and the bundle connector and got 2.3 ohm, and between the bundle and ecu connector, 9.6 ohm. But I could not find and corrosion and no moving of wires yielded any better... Also, I found something interesting... When i check the resistances for the TPS, on the sensor alone, it starts at 0.62 ohm at idle(0%) from the sensor pin, up to 4.55 ohm at WOT. But when I plug in the connector and backprobe the connector, I start at 0.60 ohm, max at 1.82 at 80% throttle, but back down to 1.62 at WOT. Wtf? Does that have something to do the ECU? But I figured the path of least resistance would still be the same...

Yep, getting 11.76 (BAT voltage) at the plug and a resistance of 10.6 across the IAC plug, but does not "cycle" after 1 min like the FSM says it should. I've also hooked the IAC to a 12V source and cycled it manually and it still open/closes.. I tried testing the voltage at the ECU per the FSM, but the FSM I'm using is the MY89. I've read that the MY91 and MY89 SPFI are identical, but after 2 min, I wasn't getting voltage but can't rule out the computer since the wire colors didn't match (even though the ECU plugs had the same number of wires and connectors) The only thing that makes me scratch my head is that once up to operating temp and I restart the car, the code goes away and idles/ starts up just fine until it cools down again. But the accel/misfire is still there even if the code is not.

GD is right... The double pulley setup is used when the alt is between the P/S pump and A/C compressor (Panasonic-made compressor) whereas some EA82s and EA81s had the non-panasonic compressor (name escapes me ATM) where the Alt was mounted outboard (hence the one groove pulley). I'm also finding more of the reman alts that come from Remy (Duralast/Worldwide) as being shipped with NEW double pulleys. From what I've seen, you can use the double groove pulley, just run the belt on the groove closest to the alt body so it will line up with the other belts.

More updates: Driving around now, the code 24 is constant when cold and does not clear until operating temp and a system restart. NEW SYMPTOM: It will not rev over 5250 RPM with WOT. In any gear, from any specific speed UNLESS I back off to 75-90% throttle. And it loses MAJOR power from 3K on.... Also dropped in a new CAS (OEM Hitachi. ) with zero change. The connectors are all clean and clear, the ground resistance between motor and body, body and battery and motor and battery at 0.03K Ohm. Near zero. All the bundle connectors are spotless, wiring is all in it's original loom and wires, and the alt has 0.03 mVAC output (negligible, thinking it was screwing up the ECM). Also checked the injector to check for split orings or a missing isolator (to account for the loudness) and both were like new when I put them in 5 months ago and assembled correctly according to the FSM... However, I cleared the codes approx 1-2 weeks ago and ran the D-Check mode again today and came up with codes 11, 13, 23, 24 and 31. What the what? And the only code that pops with the CEL on (when cold) is the code 24. Now I'm throughly lost...