legacycentral bbs

I've noticed over the past few weeks that my car has been missing throughout the rpm range. So I changed the spark plugs (one range colder and gapped to .43) and wires and added some fuel injector cleaner. The car runs a LOT smoother but it will still miss/hesitate on full boost in the lower rpm range (3000-4500 rpm). It's not a hesitation like I used to have but almost like a fuel cut. Just to check I tried morgie's one-way check valve, but the valve I have doesn't vent well without also ingesting some vaccum from the engine . . . no go there.

Here's the situation. Whenever the engine misses a/f will goes way lean and then way rich for a quick millisecond and the car will buck. After that, it runs smooth and strong. Boost gauge will read a little over 14.7-15.5psi and then drop and then go back to ~14.7 psi. That's why I think this might be a fuel cut issue. However, when full boost hits at above 4500rpm, the engine runs strong, even though boost hits higher than 15psi!!

Any ideas?

Hmm... You have a gauge that indicates injector duty cycle, right? That should tell you for sure whether or not the ECU is cutting fuel.

Some experiences people have had indicate that our ECUs have a complicated fuel cut system. If for one reason or another the ECU decides that a certain set of situations is bad (for example, engine speed=3500rpm, manifold pressure greater than 14psi), and bad enough that the only response it has is to cut fuel (rather than just retarding timing or changing injector pulse width), it'll remember to cut fuel again when that situation comes up again.

I haven't read any official Subaru documentation of this, though... But it would be pretty easy to check. If your duty cycle is indeed dropping at those times, try resetting your ECU and see if it fixes the problem at least temporarily.

YOu know, I've been so engrossed in checking a/f ratios that I didn't even check the IDCs!! I'm trying to remember, now. At any rate, I'll check for the IDCs during my "lunchtime drive" and let you know. If it does cut fuel, I'll see about resetting the ecu.

Any other ideas should it not be a fuel-cut issue?

When it's missing/hesitating at lower RPMs and full boost, do you have the throttle open all the way or is it partly closed?

It could be inappropriate air/fuel ratios due to some sensor problem or something, I guess... But you say the problem has only appeared in the past few weeks.

The weird thing is that it's doing fine at full boost at high engine speeds, because that's supposed to be the most difficult environment for the fuel burn.

Let us know what the lunchtime drive reveals.

Okay, I was in the process of resetting the ecu dy hooking up the connectors under the dash. I started it up and drove around and for a good 10-15 minutes the ecu read the "all clear" five continuous flashes . . . but then it started reading a number 44 code (I'm trying to look up that code as we speak). Pulled into a parking lot and as I was coming to a stop the car starting idling REALLY low. It eventually stalled. When I started the engine up again it still showed the code. I disconnected the connectors and started it up again. NO CODE SHOWING!! I haven't been able to flog the thing yet, as I was afraid I would get stuck and not be able to return to work.

As I pulled into my garage, the car started idling low just for a second and then the problem disappeared. This seems similar to when my TPS sensor was bugging out a couple months ago.

I'm suspecting that the 44 code is a idle switch.

Number 44 is a wastegate solenoid, if that helps ...

Hmmmm. So . . . is that the boost control solenoid (the cylindrical looking thing). As far as I know it's still attached to the ecu, but I don't have any hoses attached since I have an MBC. I'll check it again to see if it's still plugged in.

It's not entirely clear that the "clear code" procedure (hooking up the read and test mode connectors) actually clears the ECU's learned memory. I'd suggest disconnecting your battery and waiting a few minutes. Some people suppose it helps to press the brake pedal while the battery is disconnected.

subyluvr2212 is right -- 44 is the wastegate control solenoid. It's not unlikely that your wastegate control solenoid is dead; they do often eventually go after 10 years or so. Or is the solenoid disconnected electrically? That would also cause the code.

In any case, though, there's no way as far as I can see that a code 44 can have anything to do with the problems you're having.

So the idle speed fell until the engine stalled out, but then you were able to start it up again and it ran okay? Bizarre. I suppose it could be the IAC valve being iffy or something.

I wonder if the two problems are related. I'm stumped for the moment.

Ack! You posted while I was posting!

Yes, the boost control solenoid is cylindrical in shape. It's mounted on the passenger side strut tower. The cylinder is mounted vertically below the pressure sensor and pressure exchange solenoid. It has two wires coming out of it (black and pinkish-red, I believe) and at least on my car there is an asbestos sheath around those wires.

I'm continually stumped by this as well. Now I haven't been able to check up on the "fail-safe" measures enacted by the ecu when the boost control solenoid code is thrown, but would that possibly do anything to the running condition of the engine?

Well here's the scoop on the TPS switch. A couple months ago I had a bunch of codes thrown by the ecu. One of them was the idle swtich. Going by Legacy777's instructions, the check of the resistance values on the switch were COMPLETELY backwards! And I could NEVER set the switch correctly. I bought a new TPS switch (attached to a throttle body) that I never installed but the code seemed to disappear without the replacement.

I drove it around fine, and then these problems started occuring recently.

I'm just as stumped as to why the car would misfire under low rpms and not high rpms as well. It just runs STRONG at the higher rpms!! Now I've got to drive the car like a honda . . . no real power until the higher rpms . . . V-turbo or something like that. lol!

Well, the solenoid has two basic modes of failure. One is to be stuck closed, which results in only wastegate boost. The ECU doesn't have to do anything in this situation. When it's stuck open, though, boost is not controlled at all. My understanding was that the ECU responds to this with a fuel cut.

Huh.

With a fuel cut.

Maybe your ECU actually is confused in a weird way making it cut fuel at lower engine speeds when boost is high.

I still don't see what that would have to do with the idle, though. I remember your troubles with the TPS. Could be two unrelated problems.

My suggestions: When you get a chance, cajole a buddy into following you in his car while you go for a drive so you can push it and see what the IDCs are while still having a way home if somehow the car were to stop working. And, try resetting the ECU by disconnecting the battery, pressing the brake pedal for a few seconds, waiting 10 minutes or so, and reconnecting the battery. See if anything changes after the reset.

Well, I just let the car idle for 10 minutes to see whether it still exhibited the same symptoms. Nothing new at idle. Checked the connection to the boost solenoid, and it's still connected. Is there a diagnostic check for the boost solenoid.

I'll flog it later tonight if the weather cooperates.

Okay, I ran the PISS outta the car this afternoon on the way home from work, but I couldn't get it to exhibit the same symptoms!! Even the idling problem mysteriously dissappeared? Now I'm confused.

don't worry, be happy. hope it never pops up again. and if it does you'll have more info to take a snag at it.

Wow... Your car has a mystical ability to heal itself.

I'm thinking that your trouble has got to be intermittent wiring problems.

It would be a royal pain in the behind, but you might one day want to test the entire engine management harness. You'd set up an ohmmeter with really long leads and check that all the pins from the ECU have good continuity to all the pins on the sensor connectors, and then try to jiggle all the wiring and the car in general to see if you could make the connection less than solid.

I think my IAC valve's harness connector is in bad shape. Every once in a while the idle goes all wacky and the ECU throws a CEL for the IAC valve, and then it heals in a few minutes. The valve is only a year old.

ya know i was just bout to call it a nite when i decided to check the forums and read this thread and its wayyyyy to similar to what is going on with my car. and if i remeber right boostie some of ur probs started after a washn and so did mine..then it ran fine for a while and then the low idle / high idle then ok idle, then acting up until i got to higher rpms and then ran like a bat.o.o.h. then check engine light soz i hooked up the connectors to read the codes...#3injector and coolant sensor. did the battery thingi and cranked it up again check engine light again soz did the connectors again read all the flashes and two more codes.. #1injector and cannister valve...and still the weird idle. now everything that has been thrown is on the right bank so i am getting the idea that there is a short in my wiriing, after all this car as many others on this board is near 200k miles and wiring does get brittle so i am thinking about replacing wires..i had a spare coolant sensor that checked out and replaced the oem..no difference. so after more thought i realized i had not changed the fuel filter in a yr plus so off with the old and in with the new, now i just need to try to remember how to clear all the codes and reset the ecu but i still have the nagging feeling that wiring is at the root of this

I'm almost certain it's a wiring problem now. It's been raining today (relative humidity 100%) and I've noticed that my a/f ratios are not as ideal as I would like. At WOT it dips quicky down into the STOICH area, where I shut down. I'm will to bet as soon as it dries out around here that I'll see them go back to normal.

*waiting for the dry weather*

ha!just when do ya think that will be..it has been pouring here offf and on all day...wait that was yesterday no the day before and looks like the rest of the week!our feet are begining to web and mold after reading ur probs and others it has to be the wiring, i think i yanked on the wires too much everytime i was changing things out or moving things out of the way to get at something else. after comparing the wires under the hood with the wires in the protective confines of the dash and under seats its obvious what heat and exposure to the elements has done to the wiring harness. looks like a system by system wiring project under the hood is in order

Well, the gremlins in the legacy decided to take an afternoon off . . . It actually wasn't raining this afternoon and ambient temps were ALOT cooler than normal this time of year.

Boosted back up to 16psi at 3000rpm in fifth gear. No problems whatsoever!! I'm guessing part of the problem might be heatsoak of the intercooler . . .

I'm still confused why it's fine at higher rpms when the weather's hot. Is it because the air actually flows causing less heat buildup in the intake charge?

Electrical gremlins can be weird like that. It could just be at those middle RPM ranges, the ECU relies more on certain sensors whose signals get muddled, or the engine's vibrations resonate with stuff around wiring harnesses, or, yeah, something about the airflow keeping stuff fresh...

i'm in the gang ! yeah, got 22-23-24-35-44-45-49 codes today, and a max boost of 5psi *despite my EBC* (this 5psi limit is now gone) , anddd, when i start my engine, CEL blinks 2 times slow, 2 times quick, and 2 times REAL Quick ! ... never seen that before :\

I'll make some switchs to enable / disable Green and black connectors so see ecu codes and do diagnostic ... i'm a bit lasy and tired of getting under the dash to hook / unhook them.

morgie,

Definitely clear the codes, and recheck because it's very unlikely you have problems with all those sensors.

As i always say, 'Another job well done' !


Bring the engine codes now, i can handle them ! haha

(I alway knew there would be a use for those old computer 5"1/4 bay covers)

HiRes : http://images2.fotki.com/v21/free/309ce ... img-or.jpg


yellow : Abs switch
Green 1 : Diagnostic (green connector)
Green 2 : Engine Codes (black connector)
black round thing (dont know how to call that in english) : boost adjustment.

Heh, nice, Morgie. I have two switches on my dash for the read and test mode too. I stuck 'em where the air suspension control switch would otherwise go.

(It's called a knob, by the way. :)

Morgie, I just know someone (most likely a girl) will get in your car and start playing with that knob!

Et, est-ce que vous aimez manger les petites chiens. C'est bon!!