alpop

Don't know what my chances are for beating out three great mechanics! If you do a web search its amazing the number of people with this era subie that have this problem. Usually no indication if the person has got it figured out, occasionally they get it solved with knock sensor or ignition wires or something relatively simple which I've already checked. One person did solve it with a different (used) ECU, but apparently had to try a second one from an earlier year to do it. I plan on doing a lot more checking with the scanner, hopefully there will be a clue. The only thing I know for sure about my problem is if you resetthe ECU (disconnect the battery for 15 minutes) it runs like a champ for twenty minutes and then starts to malfunction again, done this numerous timeslike clockwork (starts malfunctioning at the same place on the same route). I assume that for the first 20 minutes it is running in open loop on somedefault settings, and the problems start when it goes to closed loop, I will check this with my scanner. With the car like this I can granny around this small town ok, but sure wouldn't want to hazard big city traffic with potentially little or no acceleration. I've managed not to throw too much money at the problem yet, I would feel a lot worse if I had spent a lot of money and still hadn't fixed it. I'm real close to dropping another ECU in there. Trouble is looks like its a lot harder to find a ECU for the 2,2, most are for the 2.5 I bought this in the winter for my daughter to drive in the snow, we had almost no snow, winter's now gone and its still not running right, might justsell it fixed or not.

ok I get it, you get more instantaneous readings if its just looking at one thing. Will do-

So that all makes sense, few things still stick out: OL-drive, wouldn't stay in closed loop spark adv. 50 this is what the ecu is requesting, it went up to this when I increased RPM to 3000, never went back down MAF, why are people suspecting this? and what more can I do to test it Would it be useful to monitor while driving? (someone else driving)

I think its been a month or two since I disconnected the battery. OL - what does it mean?

I've got data! First of all I had trouble even getting (TDT)started, had to bang on the starter a few times, there has been a brief delay of 1-2 sec after turning the ignition key lately, guess I need to change out the solenoid contacts or whatever it is when you have to bang on the starter. No codes until I started the vehicle, then it threw a P0507 (IAC rpm higher than expected), first one it threw on its own. Not a surprise,I knew the idle wasn't quite right and figured the IAC needs cleaning or replacement - wouldn't cause the acceleration problem. Here are the numbers in the order they were displayed, first is at idle (engine warm), second is when holding at 2500-3000 RPM. Fuel Syst 1: alternating between CL and OL-Drive Fuel Syst 2: NA Calc Load: 3.5; 7.8 ECT: 196-200, cooling fan activates at 200 STFT B1: 0 mostly until took it to 3000 RPM a few times then -7 LTFT B1: 0 then as above -2.3 MAP 7.1; 6.5 RPM: around 1000 Spark Advance: 0 initially then 15, then 50 MAF: 0 initially, then 0.5; 1.3 TPS: 0; 4.7 with smooth transition, but stayed 0.4-0.8 after goosing O2SB1S1: 0.875 but then 0.12 to 0.68 after goosing STFTB1S1: 0 but then -2 to -10 after goosing O2SB1S2: 0.855 but then 0.165 to 0.76 after goosing STFT B1S2 0, alway 0 and never changed Red Flags anyone? Don't know what the fuel system codes mean Fuel trims: seems funny that they were mostly 0 but then after goosing varied but always negative, are't they supposed to cycle from lean to rich? MAP: funny I measured 18-21 at idle, is absolute pressure different than what the gauge reads? Spark advance: don't know but 50 seems like an awful lot? MAF: this seems ok? TPS: this seems ok Don't know what these last four signify, is this the change in fuel trim (short term) in response to each of the O2 sensor readings?

Hmm, well I guess its good news that the engine has no major problems, and as you say no blockage in the exhaust system. I also noticed that if you drive like a real granny and only very slowly/gradually depress the gas pedal it does fine, acceleration keeps up with the pedal depressing, it bogs down only when you try to accelerate more briskly (normally). Its like the increase in fuel can't keep up with the increase in air unless its very gradual? Something's limiting the rate of fuel increase? I ordered an Innova Equus 3040 scan tool,($60)supposed to do live data and have company support too. Sounds like the Autel wasn't supported by the company because the ones at the great price on Amazon were chinese knock-offs of the authentic(chinese)product. Probably won't get to it 'till next weekend.

one more vacuum test. At idle in park at steady 21, slowly increase rpm to 2500 and hold, steady at 20.

Manifold vacuum test result: test elevation = 2560 ft. Idle in Park: engine cold, high idle steady at 18, slowly climbs to 20-21 as engine warms and goes to low idle. Blip throttle: shoots down to 2 then quickly rebounds to 22-23 (not 25) then settles back at 20-21 Drive test: when accelerating at lower RPM and its dogging, vacuum at 0-1, when rpm finally passes 2500 and acceleration becomes normal, climbs to 2-3 The gauge I got also can test fuel pressure I guess, but gasoline is a whole other thing and although the Haynes manual details the procedure it seems like they have left out a few steps.

well now I'm thinking I can buy a vaccuum/fuel pressure gauge for $25 and a OBDII scanner all for less than the probable $100 diagnostic charge. With 4 cars they should come in handy again (right, my daily transport is a bike - go figure). I see the "Autel MS509" OBDII scanner on Amazon for $44, and it seems to just about do it all, livestream ECU info, freeze frame data (what's this?) etc., in addition to reading and clearing codes. Seems like an amazing price, several reviews from mechanics are very favorable, though it sounds like its older and no longer supported by the company - thus the reduced price. Maybe you wouldn't be able to update it but hey my newest car is 2006 anyways. Anybody know anything about it?

All good info, thanks Fairtax-

1 lucky tex, thanks for the info - I will do that on the wkend probably- I did find the TCU right above the brake pedal as supposed to be, all plugged in and such - no obvious problems As I mentioned, there is absolutely no readout for gearshift position anywhere on the dash, and the area beneath the car/door open icon doesn't appear to be an active part of the multimeter. Owners manual has no reference to a shift position readout. The shifter console is not lighted (couldn't find a burned out bulb under there) so there is currently no way to know the gearshift position in the dark! I really can't believe they designed the car this way! Did they? Also noticed two non-mated dissimilar plugs under the dash to the right of the steering wheel, no where to plug them in, probably for some options the car doesn't have? Although they each have about a dozen connections, so probably not just for dash switches. These are near the smaller green check male and female plugs.

Hmm, well maybe it is time to get a vacuum gauge.

Turns out this 4EAT transmission is a 4 speed, no electronic overdrive, 4th is just referred to as the OD by virtue of its gear ratio. Here is a great write-up on the transmission: http://www.subaru-svx.net/photos/files/ZMan/13060.pdf

I'm on to thinking about the transmission now, there are a few quirks I've been ignoring When it seems to be acting pretty normal (no major heasitation below 2500 rpm) it still "sometimes" does the following: 1. Doesn't want to shift from 2nd to 3rd (or maybe its 3rd to 4th) 2. At higher speeds when you take your foot off the accelerator its like an engine brake is applied, although I don't notice a corresponding increase in rpms like it has shifted down. 3. No way to tell what gear its in except when its light, looking down at the shifter (a problem in the dark). I saw a pic of my dash board and it seemed to show there is a gear indicator located vertically beneath the vehicle/open door icon (which works) on the dash. No sign of the gear indicator on my dash. I 've done a little digging around, and there should be a Transmission Control Unit usually above the brake pedal, independent of the ECU. The TCU reacts to a number of sensors, some different than the ones I've tested. I need to find this and check it out - wonder if it is completely independent of the ECU or if it works throught the ECU?

it would be very nice to see live data. Or, if you have or can get scanner, create a CEL and get freezeframe data. Beyond me, thus the professional diagnosis idea did all 4 sparkplugs look the same? Yes does the car get reasonable mileage? No, haven't checked exactly but probably 15-20 mixed driving, not suprising the way it runs could you drive with a vacuum gauge attached and take readings when the car exhibits its failure 'mode'? Again, professional diagnosis

Thanks for the input guys- Rear O2 sensor: This is the second time I crawled under the car and couldnt find it, must be between the transmission and floor boards, enough of that, doesn't seem likely it would cause the symptoms anyways. Overdrive: Don't know much about it, is it a three speed automatic with electronic overdrive? when its dogging it and I push the accelerator down further, it might start around 1500 RPM (30 mph) and then sometimes seem to first shift down just a little at first (maybe to 2000 RPM) but seemingly not a full gear - maybe its shifting out of overdrive? Is the engine really supposed to be at 1500 RPM at 30 mph? If its electronic overdrive there should be a wire or plug I could pull to see if the problem goes away Evap system: No code but I did remove the gas cap and there was NO pressure release, come to think about it I think that has always been the case when I put gas into it, I seem to have thought something was different but couldn't put my finger on it. Cap gasket looks fine. Would this really cause these symptoms? I've been telling myself its not a transmission problem but having exhausted most other possibilities maybe I should re-consider? Not typical transmission symptoms. I Still need to reconcile with the "drives perfect for 20 minutes after resetting the ECU", Is there a separate Transmission Control Unit?

Now there's a completely new idea - let me give that some thought! I am also thinking of putting in the FWD fuse and seeing if that makes a difference. You probably could simulate the problem by artful manipulation of the parking brake -

This thread is long and unwieldy, here is a summary of my lack of success - any ideas are appreciated! THE CAR: 1999 Legacy wagon 2.2L automatic, very nice inside and out. THE HISTORY: PO removed the oil drain plug and purposely burned up the engine. A replacement engine (same vintage 2.2 phase II) was dropped in with about 140K miles on it. THE PROBLEM: very sluggish to (sometimes) non-existent acceleration below 2500 RPM under load (fine above 2500 RPM - this switch spot can vary, it surges as it switches from poor to normal acceleration), intermittent problem, but for example has been doing it constantly the last two days (and it has been rainy). WHAT WE KNOW/DIAGNOSIS: If one disconnects the battery and resets the ECU it drives beautifully for 15-20 minutes then starts to malfunction (did at least 3 times) - I assume that for these 15-20 min. it is on default settings and then the ECU changes things and it runs poorly. Thus it "may" be either a bad sensor feeding bad info to ECU or a faulty ECU. IS THIS A VALID CONCLUSION? Throws no codes unless I mess with the sensor (unplug, check voltage etc). THINGS CHANGED REGARDLESS OF ABOVE DIAGNOSIS: -Spark plugs checked, new and gapped correctly -new ignition wires installed -new PCV valve installed -new air cleaner element installed -checked for vacuum leaks (fine water spray, engine running) -checked coil pack for leaks (fine water spray in dark, engine running -checked all engine harness plugs and other - secure THINGS CHECKED ASSUMING ABOVE DIAGNOSIS: sensors checked either by, 1. replacement, 2. voltage input and output. or 3. disconnecting and see if runs better. MAF sensor (cleaned, unplugged didn't run), TPS (voltage ok and replaced temporarily-no change), Knock sensor (replaced twice-no change), Pressure sensor (voltage ok and unplugged ran the same), coolant Temp sensor (replaced - no change), front O2 sensor (unplugged it ran the same) Rear O2 sensor, cam sensor and crank sensor not checked, assume it wouldn't run w/out the latter two. ANY MORE IDEAS? Any more sensors to check? Are there (non-engine) things that one might expect burning up the engine would damage? The end of the tail pipe is very black, but it doesn't use oil w/new engine. Should I take it somewhere for a professional diagnosis or try a $50 salvage yard ECU, or a $250 rebuilt ECU?

I'm starting this new thread on a continuing problem since the old one has been going for many months w/out resolution and the thread is getting a bit unwieldy. THE CAR: 1999 Legacy wagon 2.2L automatic, very nice inside and out. THE HISTORY: PO removed the oil drain plug and purposely burned up the engine. A replacement engine (same vintage 2.2 phase II) was dropped in with about 140K miles on it. THE PROBLEM: very sluggish to (sometimes) non-existent acceleration below 2500 RPM under load (fine above 2500 RPM - this switch spot can vary), intermittent problem, but for example has been doing it constantly the last two days. WHAT WE KNOW/DIAGNOSIS: If one disconnects the battery and resets the ECU it drives beautifully for 15-20 minutes then starts to malfunction (did at least 3 times) - I assume that for these 15-20 min. it is on default settings and then the ECU changes things and it runs poorly. Thus it "may" be either a bad sensor feeding bad info to ECU or a faulty ECU. IS THIS A VALID CONCLUSION? Throws no codes unless I mess with the sensor (unplug, check voltage etc). THINGS CHANGED REGARDLESS OF ABOVE DIAGNOSIS: -Spark plugs checked, new and gapped correctly -new ignition wires installed -new PCV valve installed -new air cleaner element installed -checked for vacuum leaks (fine water spray, engine running) -checked coil pack for leaks (fine water spray in dark, engine running) -checked all engine harness plugs and other secure THINGS CHECKED ASSUMING ABOVE DIAGNOSIS: sensors checked either by, 1. replacement, 2. voltage input and output. or 3. disconnecting and see if runs better. MAF sensor (cleaned, unplugged didn't run), TPS (voltage ok and replaced temporarily), Knock sensor (replaced twice), Pressure sensor (voltage ok and unplugged ran the same), coolant Temp sensor (replaced), front O2 sensor (unplugged it ran the same) Rear O2 sensor, cam sensor and crank sensor not checked, assume it wouldn't run w/out the latter two. ANY MORE IDEAS? Any more sensors to check? Are there (non-engine) things that one might expect burning up the engine would damage? The end of the tail pipe is very black, but doesn't use oil w/new engine. Should I take it somewhere for a professional diagnosis or try a $50 salvage yard ECU, or a $250 rebuilt ECU?

"Should look the same as the plug at the sensor end of the harness." Don't know what you mean by this - There is the O2 sensor with wires coming out of it, no plug, then the wires go back and dissappear above the transmission. Is the plug which I need to disconnect visible from the (driver side) engine compartment or from underneath the car?

So in my continuing effort to test sensors I looked for the "rear" O2 sensor today, found it easy on the back part of the cat convertor but a long wire left the sensor and instead of taking the easy route of up to the firewall and the wiring harnesses there, it went back and disappeared above the back part of the transmission. I could find no sign of a plug or where the wire went after that- Anyone know where the rear O2 sensor plugs in?

Hmm, only other thing is I had a real hard time both tries with getting the ground probe to work. Both times initially I got no reading until I grounded to the chassis then after a minute or two of fooling around the ground probe started working. I left it unplugged and it just drove fine for 10 minutes, 'course that means nothing since its done that before.

well that sort of popped by bubble, 1.4v at idle and 3.4v with no vacuum. Why would Haynes be so wrong?

Well I checked out the FPR, vacuum hoses and such look fine. Then I followed one tiny vacuum hose over to the pass side strut tower and said what are these darn things? The MAP (manifold absolute pressure?) and pressure switching solenoid valve are found here. The Haynes manual has a procedure for checking both of these. The resistance on the switching solenoid came out fine, and the 5v reference voltage for the MAP sensor was fine. However, when I backprobed the MAP sensor with the engine running, I found 1.37v at idle, and this went to around 2.0v at 1/2 throttle. 1.37 - 2.0 Haynes says it should be about 3v at idle, and should decrease to around 2v at WOT, 3.0 - 2.0 Not only are my values way off, but they go in the wrong direction with opening the throttle! Bingo?? Anybody know if Haynes is right? I have found some mistakes in it.

Yes, I cleaned those ground points a while back. Occasionally the idle seems "a little" rough but not very often. It does move around a little, it can vary from ca. 600 - 900 RPM when the engine is warm. When started cold it acts perfectly, starting maybe at 1400 rpm and gradually working its way down over a period of ten minutes or so to ca. 600 rpm. I haven't yet cleaned the IAC valve and figure this needs to be done. Never hard to start warm or cold.