legacycentral bbs

as of recently my car has only been boosting 5lbs. there are no active codes and the only stored ones were maf and 41( a/f mix). maf replaced the day after i got the cel for for codes 23 & 41 .
is this something the ecu is doing or is it the boost control selenoid? curroius if i just run a aftermarket mbc if it will fix the problem or is it gonna be more involved

CEL just came on again tonight maf & a/f (which i think are stored) the only thing i have changed this month is the rotation of the maf (have an intake) when i put a new support on it when i replaced my inlet boot. possible i got another bad maf too. gonna try cleaning it and rotating it tomorrow see if that helps. so my guess is that every time the maf goes bad it forces the car to run 5 lbs?


** and just to add. the car runs perfect with no ill effects that normally comes with a bad maf. only having the boost issue

**another edit**
just ran the car in Dcheck mode. only am getting code 41. i had a torn inlet boot for awhile, but now have a new non leaking one

whats your vac at idle?

vac is normal. with this stupid code 41 aparenlty is the bigest pos thing to deal with. acroding to the FSM the order of operatiosn to fix it is checking...drum roll please

operation of injectors-> air flow sensor->water temp sensor->throttle sensor->o2-> fuel pressure -> injectors-> then replace the ecu

the thing is that none of those are throwing codes!

I would grab a DMM and test your water temp sensor and go from there. The likelyhood of it being the injectors or PCM is not very high.

Have you reset the ECU? I'd suggest clearing the codes and then see if code 41 comes right back.

well tried clearing the ecu last night. drove today and the cel poped back up. just replaced the intake with a factory box and clean filter. gonna take it for a drive now and pick up some maf cleaner too. still running perfect but only 5lbs boost still

drove today no cel, still 5lbs. stopped by SIR and used their subaru select monitor. no active codes but it was unable to clear the memmory . so replacing the ecu seemed to be the next step. left the shop got on the freeway and voila back to boosting normal again! still gonna source out a replacement ecu in case this happens again

I am actually experiencing this same exact problem. i tried putting an intake on, and now (even after i put the stock box back on) i am only boosting 5 lbs... i haven't tried clearing the ecu but i am definitely wanting to find out how to get this fixed. i will keep you guys posted if i find out anything on it. i am going to do a boost leak test today, and then an ecu reset. (Cross your fingers)

What about your O2 sensor, last time it was replaced?

comes to find out that with my setup. i have a leak somewhere when i did my boost leak test. but i can't find the major leak. along with that i have a knock sensor code and a throttle position sensor code... i searched for the leak all over and couldn't find it any tips on locating it? also i am ordering a new knock sensor, so hopefully that will fix the problem.

If you can find a "fogging" or something that will smoke, you may be able to try and put it around the engine to see if you can find out where the leak is coming from.

any tips on that fogging method? w/out spending alot of money? like a smoke can or something?? ha ha

If it was around the 4th of july I'd suggest smoke bombs from the fireworks store. Dry ice in a pan of water might also work. Some of the hunting/camping stores may have something like a smoke signal type thing if you're stuck in the wilderness.

i have a smoke bomb, would that cause any trouble with the engine? thanks for the advice. it is greatly appreciated.

No, it should be fine.

thanks i'll try that out and see what happens.

I am having a similar problem. No codes are stored and no check engine light. Sometimes it will boost 8+psi but about 80% of the time it maxes out at 5psi. I've replaced a lot of the rubber lines (pcv and vacuum) and I don't believe there are any leaks. I used the diagram here http://www.surrealmirage.com/vrg3/vacuu ... vacuum.png to verify the correct connections. I've also tried swapping out the map sensor, the pressure switch and the boost control solenoid with no change.

Do we know all of the scenarios when the ECU will limit peak boost? Does it do that when knock is detected or something?

Edit: Found out that line J in that diagram has a restriction pill inside of it. I put the old hose back on and went for a few pulls. the fourth out of ten was at 8+psi and all the rest were at 5 After thinking about it I realized the pill would only prevent overboosting not underboosting.

Did you verify or check the pressure exchange solenoid? If it is bad or old, the MAP sensor may not see proper boost pressure.

just an update.

no bad sensors and the problem/CEL never came back. just the ecu being a little bitch

Legacy777 I wasn't sure if you were referring to me but I did try swapping out all three of the items on the strut tower with used ones I got from the junk yard a while back (map sensor, switching solenoid and bcs)

For my personal situation switching to 91 octane versus 89 seemed to make it run in high boost mode more often. I think there is some threshold in the ECU which disables high boost but doesn't throw a code. I am wondering if detected knock or maybe some type of intake leak causing a rich/lean condition and it goes into low boost mode until 30 seconds of under load driving with no problems has passed? One of the old service records noted that the intake manifold gaskets were leaking (which if true could be the cause of my rough idle). I also wonder if the 260k mile injectors might be worn out. I did have one partially fail in my NA with 165k miles that sometimes didn't work at idle causing it to run on three cylinders but was fine at higher rpm. It didn't throw a code either.

It makes me think something like the IAM that is monitored on newer cars that should always be at 16 or whatever. Maybe the ECU runs waste gate spring pressure until the ECU is back into a happy fueling or knock feeling?

These ECU's really aren't that smart....they are quite basic, so I have my doubts they have that kind of logic. Any chance you have an older laptop and can run Vikash's scan tool to see what the ECU parameters are doing?

The ECU might be smarter than you realize. I drove for three hours today after fixing my intake manifold gaskets and after the initial ten minutes of driving (still in limp mode?) I hit full boost EVERY single time. My orange gas light confirms many full throttle pulls
http://bbs.legacycentral.org/viewtopic.php?f=5&t=49091

My very strong feeling is that strong knock or fuel ratio errors can put the ECU into safety mode and it takes a few minutes of driving over a certain load to restore the "good mood" just like in the newer ECU. In newer cars there is a method to fake a full strength IAM by driving with the brakes lightly applied at 3500 rpm for 1 minute iirc. I did something similar before my boost was restored to normal levels.

Hopefully tomorrow it will stay at full boost and I wont have to return with sad news.

Yeah, ciper, I'm with you. I'm pretty sure the ECU does limit the turbo to wastegate boost when it detects knock. That's one of the mechanisms it uses to cope with someone using the wrong fuel. I don't know about air/fuel ratios coming into play though. I suppose it's possible.

When I bought one of my old Sport Sedans, the previous owner said he'd been filling it with 89 octane because he thought 91 was a waste. On the drive home I watched with my scantool and the boost control duty cycle was constant at 1.56% -- effectively zero. There were no trouble codes, stored or active. The tank was about a quarter full so I filled it the rest of the way with 94 octane and reset the ECU, and boost behavior returned to normal and stayed that way.

When you go to aftermarket boost control you lose this safety feature.

Maybe we should ask the robtune guy. I know he never finished working on it but he did get the US Legacy Turbo rom partially reverse engineered.

This safety issue is not designed properly in my mind. I bet there are LOTS of Legacy turbos running around with no check engine light and only running 5psi of boost. Without a gauge you'd have no way of knowing.