Page 1 of 2
10-11psi stock?
Posted: Wed Nov 12, 2003 3:20 pm
by Dizzy
So, just for giggles, I installed a boost guage into my turbo wagon. If I drive like a sane human being, the car holds between 5-7 psi. If I put my foot in it, the car will hold 10psi, with occasional spikes to 11-12psi. I can reliably repeat this behavior.
I'm going to tee in another guage to see if this one is off. However, I used this guage in my WRX and have no reason to doubt it's accuracy (within reason). Also, the car pulls pretty hard at 10psi. My only mod is intake silencer removed with K&N panel & tubing from the fog light.
What are you guys seeing out of your stock setups? TIA.
PS - Altitude here is ~5400ft. I'm not worried about running out of fuel. However, I am concerned with running the VF11 outside of it's efficiency range. Does anyone have the compressor map?
Posted: Wed Nov 12, 2003 4:59 pm
by boostjunkie
In my conversations with vrg3, I've been slowly learning about how our ecu works. Don't quote me on this, but the MAP does tie into a pressure exchange solenoid, which switches between manifold pressure and atmospheric pressure. Now, I'm curious as to why the ecu would need atmospheric pressure readings . . . but, I would speculate that it would have to do with changes in altitude and corresponding atm pressure changes.
Does the ecu compensate for higher elevations by increasing boost? That might be one possibility . . . when getting stock boost numbers from people it would be good to also have them include their elevation (or at least an estimate of elevation).
Of course, I could be way off base.
Posted: Wed Nov 12, 2003 5:35 pm
by Dizzy
Some simple math coupled with previous observation seems to bear out your theory.
Atmospheric pressure here is ~85% of sea level. To get the ratio of Colorado:sea level pressure, invert & multiply: 0.85^(-1) ~= 1.1765. Solving for sea level turbo legacy boost (b):
b = 10/1.1765
This yields ~8.5 psi, which seems to be what folks at lower elevation are seeing. This seems to be the OPPOSITE of what I saw with my WRX. :shrug:
Posted: Wed Nov 12, 2003 5:38 pm
by boostjunkie
Wow! You mean I might actually be right?

Posted: Wed Nov 12, 2003 5:50 pm
by Dizzy
Wow! You mean I might actually be right?
Hey, your title is "Know it All", so I would hope so.

Posted: Wed Nov 12, 2003 7:48 pm
by Legacy777
yes the ECU needs to know barametric pressure readings for altitude a/f ratio corrections
Posted: Wed Nov 12, 2003 8:14 pm
by vrg3
Yeah, I think you guys are right.
The factory service manual says that the ECU attempts to maintain an absolute manifold pressure of 1210 mmHg. At standard pressure (750 mmHg), that means 460mmHg, or 8.89psi, gauge pressure. At Colorado pressure (~640mmHg), that means ~570mmHg gauge, or ~11psi.
The ECU does check atmospheric pressure for a split second every time it sees manifold pressure rise above what it believes to be atmospheric. In simple terms, every time you start to boost it checks atmospheric pressure. It makes perfect sense that it would do this to know how to fine-tune its boost control signal every time.
It does also use the atmospheric pressure reading to trim fueling to compensate for thinner air at high altitudes.
1990-1992 non-turbo models also send the atmospheric pressure signal over to the TCU, which uses it to change shift points at high altitude.
Posted: Wed Nov 12, 2003 8:41 pm
by boostjunkie
I actually notice this switchover occuring since my boost gauge is tied into the same line as the MAP sensor. I'll see readings go from .65 to .35 and then back to .65 in about a second.
As for your wrx, if you have opposite readings, I'd hazard a guess that the ecu does not have this function.
Posted: Wed Nov 12, 2003 9:03 pm
by mikec
Hmm... Anybody think that a malfunctioning pressure sensor might be the cause of low boost then? I'm still only getting 6-7 psi max, but I don't think its the gauge, as my idle vacuum readings are correct.
I'm still waiting for my MBC to show up, which I'm hoping will allow me to better troubleshoot this.
Posted: Wed Nov 12, 2003 9:33 pm
by Brat4by4
I think an MBC will help break whatever the ECU was trying to protect by sticking to wastegate. You need to find what codes the car is throwing that makes it go into limp mode. Mine did the same thing when I had a coolant hose break on top of the engine... I don't know what it got wet, but whenever the CEL came on I only had wastegate boost.
Posted: Wed Nov 12, 2003 9:34 pm
by Dizzy
Mike -
That sounds possible. If you do have a bum sensor, I wonder what else that is affecting (effecting? I can never remember which one to use).
As for the WRX, we were guessing that the ECU was using the value of atmospheric pressure to determine how much boost should be run. Instead of an absolute value, it used a PR value. Hence, those of us at altitude got really screwed.

Posted: Wed Nov 12, 2003 10:11 pm
by mikec
Brat4by4 - No codes being thrown for a while. :S That's why I'm so confused by this problem. Unless my boost gauge is reading wrong, but only when the pressure is positive.
My plan with the MBC is to set it to run at wastegate pressure, and see what my gauge says. And then I guess take if from there.
Ah well.... As long as nothing blows up, I guess I'm happy.
Posted: Wed Nov 12, 2003 11:26 pm
by vrg3
WRXs don't compensate for altitude? That's weird. That's one of the big advantages of having a turbocharger.
mikec - You don't need a boost controller to run wastegate boost. Just connect the compressor outlet nipple and the wastegate actuator nipple together directly.
I would guess, though, that your boost control solenoid is just dead. They do seem to go kaput after enough thousands of miles, and when they do, typically you just get wastegate boost levels. An MBC is probably the simplest and cheapest repair, but you can replace the solenoid (or sometimes clean it out).
Posted: Thu Nov 13, 2003 3:19 am
by free5ty1e
I see... so perhaps a '94 Legacy with say, 200,000+ km on it, could possibly have a dead boost control solenoid and be running 5-6 psi regulated only by the internal wastegate? And my MBC will fix things? That... that is some good news. I was wondering why WOT didn't quite feel much faster than halfway open...
Posted: Thu Nov 13, 2003 3:37 am
by vrg3
Uh, WOT should still feel much faster than 50% throttle. Wastegate boost is regulated before the throttle and intake manifold, so you should be getting much more air at 100% throttle than at 50%.
Are you getting "full" boost at half throttle?
Posted: Thu Nov 13, 2003 1:11 pm
by MCO-Ryan
I seem to be having the same problem. At times, the car will only boost to 5-6psi, and at other times it'll push almost 10psi (everything's stock except for a K&N panel and the muffler removed). Its even done it in the same drive. I also get the same feel that the car isn't really any faster at full throttle than at 50%, though I'm not showing full boost at part throttle.
Posted: Thu Nov 13, 2003 4:02 pm
by vrg3
Ryan, I'd have to guess that your boost control solenoid is also failing... I don't see why you guys wouldn't feel acceleration increase much between 50% and 100% throttle, though... do you guys have really clogged air filters or something?
Posted: Fri Nov 14, 2003 3:35 pm
by free5ty1e
well, I thought the air filter could be the culprit but a new K&N drop-in didn't fix it. I mean the car pulls nice up to about half throttle but above that doesnt seem to increase much. Its very strange, do these cars need throttle linkage adjusted every so often? I wonder if my TB is opening all the way.
Posted: Fri Nov 14, 2003 4:18 pm
by vrg3
Hmm, I do recall someone once mentioning a problem where the cable was adjusted wrong so that when the pedal was pressed all the way, the throttle plate moved more than the usual 90 degrees, so it started closing back up again.
You could check by removing the plastic tubing going from the compressor outlet to the throttle body and having a buddy press the pedal while you watch the throttle plate. You may need a mirror to get a good look.
Posted: Fri Nov 14, 2003 10:01 pm
by Matt Monson
Dizzy,
I found it...
Ok, so here's a question for you all since I am at the same altitude as Dizzy. Assume when I get my gauge on here next week, and I see the same numbers. When I put on a MBC, the max the car is going to allow with a stock ECU is 13psi, correct? So without overriding the ECU, Dizzy and I can only go 2psi over stock with just an MBC, right? While you guys at sea level can go like 5?
Also, on the WRX, I think you guys are correct, it isn't compensating. But a Cobb or a Vishnu reflash can remedy that to my understanding.
Posted: Fri Nov 14, 2003 11:03 pm
by vrg3
Matt, I believe that you are correct; your car is already increasing boost to try to compensate for the altitude, so you can't go as much higher as those of us at sea level.
Posted: Sat Nov 15, 2003 3:34 pm
by free5ty1e
yeah, I'm going to have to check the butterfly plate and see if it's opening properly. There are adjustments for the throttle cable, right?
Posted: Mon Nov 17, 2003 3:13 pm
by Dizzy
Matt -
Keep in mind that we may not want to go much higher than 12psi on the stock turbo. I don't have the compressor map, but I imagine that increasing the boost is not going to yield much power due to the turbo operating at an inefficient level.
Posted: Mon Nov 17, 2003 4:46 pm
by Matt Monson
Dizzy wrote:Matt -
Keep in mind that we may not want to go much higher than 12psi on the stock turbo. I don't have the compressor map, but I imagine that increasing the boost is not going to yield much power due to the turbo operating at an inefficient level.
Yeah,
I realize that. For the short term I was planning on running 12-13psi through the winter. I've decided to not tackle the WRX TMIC project until the spring and just deal with the MBC, snorkelectomy and turbo back for now. I figure it is cold enough here for the next 4-5months that I can get away with it. Plus it isn't like this car will see any track duty or anything.
Come spring I'll be looking at replacing my leaky head and an upgraded turbo and then will go for more boost...
Posted: Mon Nov 17, 2003 5:59 pm
by Dizzy
Cool. Which turbo-back are you going to use?