Bug Eye WRX PPP - low boost
Discussion
Noticed a few weeks back that the car didn't feel as fast as it had done previously, as I'd just acquired an OBD dongle I did some logging and found that I was only getting 0.6 bar of boost rather than about the 1.1 I'd expect from the PPP map. Double checked the MAP sensor's figures by plumbing in a separate boost test kit which confirmed the readings from the ODB, thinking I might have a boost leak I gave the car to my mechanic to pressure test and it all came back good.
Next thought was to try the boost solenoid so I picked one up off ebay and fitted it, after jump starting the car (stupid battery went flat while it's been parked up
) I took it out for a run with the OBD on, initially I was only seeing 0.7 bar but after 20 mins or so of driving I did another run and saw 0.8, a further run and I was seeing 1 bar which is encouraging if a little shy of the aimed for figure. What I'm wondering is whether the flat battery means the ECU was in some kind of learning mode and slowly increasing boost as I drove or whether I've still got issues?
Other things to throw in the mix is that the last round of work the car had done was to correct a couple of exhaust leaks - most notably the up pipe was shagged and I replaced it with a decat item - I don't have any readings to confirm but I'm pretty sure that immediately after this work and for a good couple of weeks afterwards it was boosting correctly. I'm going to do some more running over the weekend and see if I'm now consistently getting this 1 bar figure but in the meantime has anyone got any suggestions?
Next thought was to try the boost solenoid so I picked one up off ebay and fitted it, after jump starting the car (stupid battery went flat while it's been parked up
) I took it out for a run with the OBD on, initially I was only seeing 0.7 bar but after 20 mins or so of driving I did another run and saw 0.8, a further run and I was seeing 1 bar which is encouraging if a little shy of the aimed for figure. What I'm wondering is whether the flat battery means the ECU was in some kind of learning mode and slowly increasing boost as I drove or whether I've still got issues?Other things to throw in the mix is that the last round of work the car had done was to correct a couple of exhaust leaks - most notably the up pipe was shagged and I replaced it with a decat item - I don't have any readings to confirm but I'm pretty sure that immediately after this work and for a good couple of weeks afterwards it was boosting correctly. I'm going to do some more running over the weekend and see if I'm now consistently getting this 1 bar figure but in the meantime has anyone got any suggestions?
Vixpy1 said:
Get someone with Delta Dash to plug in and reset the ECU, and check the MAF voltage when ign on, but not running
Don't know anyone with DeltaDash 
Also I'm guessing the flat battery will have reset the ECU?My OBD kit is a bluetooth dongle with Torque Pro, not sure if it can read MAF voltage or not - will investigate over the weekend. Should probably have mentioned that there are no fault codes logged on the ECU
You'll make half to one bar boost just on actuator spring pressure. Even if you had the wastegate wide open you'll still make some boost.
Hopefully you've checked the vacuum pipes going from the turbo to the boost control solenoid and made sure there are no leaks from perished piping and anything else like that.
Could be map sensor although they do rarely go wrong but not saying it wouldn't be that as it does happen.
MAF sensor could be completely gone or on the way out, very common, but not often i see even a completely knackered MAF sensor stopping boost from happening altogether, even if you disconnected the MAF sensor it would still make at least half bar+ of boost but it would run a bit rough and not idle so well.
It's possible your ECU has detected a bit of an event, enough so it's pulled that much timing that it gets into the realms of very low IAM, low enough to stop making any boost for safety reasons.
You could have a pretty serious boost leak somewhere but it seems you've covered that one.
If a reset doesn't sort it, again as has been said you need to get a scooby specialist on the case to plug in and take a good look at what is going on.
All this long distance diagnosis is mere speculation at best and apart from giving you a few ideas what to look at and maybe getting a lucky hit you really need to see a man that can.
As Vixpy1 says, try a reset and it may well sort it out, but if it's happened once and all you do is a reset it may well happen again if there is an underlying problem.
Hopefully you've checked the vacuum pipes going from the turbo to the boost control solenoid and made sure there are no leaks from perished piping and anything else like that.
Could be map sensor although they do rarely go wrong but not saying it wouldn't be that as it does happen.
MAF sensor could be completely gone or on the way out, very common, but not often i see even a completely knackered MAF sensor stopping boost from happening altogether, even if you disconnected the MAF sensor it would still make at least half bar+ of boost but it would run a bit rough and not idle so well.
It's possible your ECU has detected a bit of an event, enough so it's pulled that much timing that it gets into the realms of very low IAM, low enough to stop making any boost for safety reasons.
You could have a pretty serious boost leak somewhere but it seems you've covered that one.
If a reset doesn't sort it, again as has been said you need to get a scooby specialist on the case to plug in and take a good look at what is going on.
All this long distance diagnosis is mere speculation at best and apart from giving you a few ideas what to look at and maybe getting a lucky hit you really need to see a man that can.
As Vixpy1 says, try a reset and it may well sort it out, but if it's happened once and all you do is a reset it may well happen again if there is an underlying problem.
Didn't manage to get any more logs over the weekend although driving it on Sunday morning it once again felt down on power according to the butt dyno. It's booked in for a proper diagnostic session with a local specialist this afternoon - from describing the problem to him over the phone he seems to think that like ScoobieWRX suggested that the ECU has detected a problem and is pulling boost.
Probably time I updated this.. Car had a diagnostic session with a local specialist on Saturday and after plugging in and running it on the dyno he said that despite actually making quite decent power (261bhp at the fly) "it's knocking it's tits off" and that he thinks the likely cause is ringland failure
He also plumbed in his boost gauge and determined that it's not agreeing with the MAP sensor, suggesting that this might have failed as well.
At this point I really don't know what to do with the car now - if it is ringland failure then fixing it is anything from a piston swap to a full rebuild apparently, and with an initial quote for full rebuild coming in at £3k that doesn't sound feasible
He also plumbed in his boost gauge and determined that it's not agreeing with the MAP sensor, suggesting that this might have failed as well.
At this point I really don't know what to do with the car now - if it is ringland failure then fixing it is anything from a piston swap to a full rebuild apparently, and with an initial quote for full rebuild coming in at £3k that doesn't sound feasible
You should be able to get a full rebuild for less than that. Give Simon at API a call, he'll sort you out with a reasonable quote for a straight forward rebuild to factory spec. If you want forged pistons and all that of course the cost can be quite high but if it's a like for like build you should try API. Just tell him Francis pointed you there.
ETA: If you have ringland failure you ought to get a compression test done. How does he know it's knocking it's tits off to use your term. Can you hear something unusual outside of the engine or did he plug some Detcans in to listen for knock. P.S. Testing the manifold pressure is a completely different thing to doing a compression test.
Also a bit pusszzled by your earlier post in that when initially checked the ECU was giving the same info as your boost gauge/pressure check but this specialist says the MAP sensor is reading wrong.
You need to go over a few things yourself before you commit to anything.
ETA: If you have ringland failure you ought to get a compression test done. How does he know it's knocking it's tits off to use your term. Can you hear something unusual outside of the engine or did he plug some Detcans in to listen for knock. P.S. Testing the manifold pressure is a completely different thing to doing a compression test.
Also a bit pusszzled by your earlier post in that when initially checked the ECU was giving the same info as your boost gauge/pressure check but this specialist says the MAP sensor is reading wrong.
You need to go over a few things yourself before you commit to anything.
Edited by ScoobieWRX on Monday 1st July 14:44
I think the rebuild quote in question was including forged pistons to be honest - it did feel a little high even then though. My usual mechanic said he's going to have a think and give me a ballpark for how much he'd charge to do it but I'll check API out as well, even if I get my guy to do the build they might be a good source for parts! 
As for how he detected knock I think he was using detcans (he definitely had some sort of headphones on anyway!), my ears can't really hear anything unusual externally and neither could he the other week. I'm not sure why his gauge was reading different to mine - I know he plumbed in to a different place than I did when I did it which might make the difference. The boost readings he showed me from the MAP did look to be erratic in places and didn't seem to tie in with what he was seeing mechanically. Also the car was clearly making more than 0.6bar to record 261bhp!
As for how he detected knock I think he was using detcans (he definitely had some sort of headphones on anyway!), my ears can't really hear anything unusual externally and neither could he the other week. I'm not sure why his gauge was reading different to mine - I know he plumbed in to a different place than I did when I did it which might make the difference. The boost readings he showed me from the MAP did look to be erratic in places and didn't seem to tie in with what he was seeing mechanically. Also the car was clearly making more than 0.6bar to record 261bhp!
You've got to be looking at c. 1.1bar or so to make 260bhp with the standard turbo.
Best place to plumb your gauge into is the inlet manifold but it has to be said that the MAP sensor and ECU readings are much more accurate than a mechanical, analog type gauge unless it is a very expensive gauge that is very sensitive, properly calibrated and finely scaled.
It shouldn't be that far out either but it has to be said that boost is never constant. It does still waver up and down a little bit, and even when well controlled, it's never rock steady.
If it was all over the place it could just be down to the mapping not controlling the wastegate duty properly, too much boost mapped in causing compressor surge etc etc....
The ECU does compensate and works off error tables when controlling boost and because it's up and down all the time it's constantly correcting plus or minus boost all the time. Small changes are pretty normal but big changes usually means there is an issue somewhere but not neccessarily ringland failure.
You sure he's not just trying to pull your pants down.
If it's getting a lot of det maybe it just needs mapping properly. It could be down to a dodgy sensor.
Get a 2nd opinion from a subaru specialist you don't know. Tell them the symptoms you're getting and go from there.
Something doesn't sound quite with the diagnosis although i could be completely wrong.
Best place to plumb your gauge into is the inlet manifold but it has to be said that the MAP sensor and ECU readings are much more accurate than a mechanical, analog type gauge unless it is a very expensive gauge that is very sensitive, properly calibrated and finely scaled.
It shouldn't be that far out either but it has to be said that boost is never constant. It does still waver up and down a little bit, and even when well controlled, it's never rock steady.
If it was all over the place it could just be down to the mapping not controlling the wastegate duty properly, too much boost mapped in causing compressor surge etc etc....
The ECU does compensate and works off error tables when controlling boost and because it's up and down all the time it's constantly correcting plus or minus boost all the time. Small changes are pretty normal but big changes usually means there is an issue somewhere but not neccessarily ringland failure.
You sure he's not just trying to pull your pants down.
If it's getting a lot of det maybe it just needs mapping properly. It could be down to a dodgy sensor.
Get a 2nd opinion from a subaru specialist you don't know. Tell them the symptoms you're getting and go from there.
Something doesn't sound quite with the diagnosis although i could be completely wrong.
Pretty sure he plumbed his gauge into the Inlet Manifold and he was saying that he was seeing 1bar on it when the MAP sensor was showing 0.6 (apparently the MAP did occasionally show 1bar), as for the power figure I'm obviously not sure how accurate the Dyno was - the last time my car was on the same rig it made 253 IIRC which sounded about right and that was essentially the same setup but with the OEM up pipe in place, like yourself I'd be very shocked if it was making anywhere near that with 0.6bar.
As he described it to me the ECU was seeing 0.6 bar and was pushing the boost solenoid duty to 75% all the time trying to get higher boost then as it was getting det it was winding the advance back. I'm not sure how on the level the guy was - several people have said he's a decent guy (including David @ API) and given I was quite prepared to pay for him to map the car on the day if it needed it I doubt he was trying to sell me a map as I was already pre-sold as it were I'm thinking I need to trust he was being honest even if he hasn't correctly identified the issue.
The plot has thickened somewhat in that I spoke with the guys at API yesterday and they seemed to think that it would be highly unlikely to be ringland failure as this wasn't a common issue until the 2.5 engine, they do however think that it could easily be the beginnings of a crankshaft failure or HGF or similar.
At this point I'm honestly at a loss of what to do next - wondering whether it is worth getting a secondhand MAP sensor off ebay and swapping that out and maybe doing a leakdown/compression test?
As he described it to me the ECU was seeing 0.6 bar and was pushing the boost solenoid duty to 75% all the time trying to get higher boost then as it was getting det it was winding the advance back. I'm not sure how on the level the guy was - several people have said he's a decent guy (including David @ API) and given I was quite prepared to pay for him to map the car on the day if it needed it I doubt he was trying to sell me a map as I was already pre-sold as it were I'm thinking I need to trust he was being honest even if he hasn't correctly identified the issue.
The plot has thickened somewhat in that I spoke with the guys at API yesterday and they seemed to think that it would be highly unlikely to be ringland failure as this wasn't a common issue until the 2.5 engine, they do however think that it could easily be the beginnings of a crankshaft failure or HGF or similar.
At this point I'm honestly at a loss of what to do next - wondering whether it is worth getting a secondhand MAP sensor off ebay and swapping that out and maybe doing a leakdown/compression test?
I map for API
My first port of call is a quick compression test just to see what state the engine is in. You can glean a fair bit from that in terms of whether or not the engine is wearing evenly and if not which cylinder/s are causing issues.
I'd agree with API in that it's unlikely to be ringland failure on a 2ltr and they know their stuff on engines. Not so sure about crankshaft on the way out though. Not something you can really diagnose without physically being there.
If the MAP sensor is reading low then all the knock could well be coming from the fuelling being very lean as a result. If the external gauge is reading 1bar then at nearly 0.5bar difference it will make a significant contribution to lean fuelling therefore causing knock.
Personally i would run a wideband firstly to see if it's fuelling correctly. If it isn't fuelling right then you have to be looking at the usual suspects like MAF and Lambda sensors. Then monitor the MAP sensor to see what is what. Ideally if you can borrow a MAP sensor that would eliminate or indicate a problem there. MAF and Lambda sensors are pretty easy to diagnose issues with, without having to swap them out.
Equally all other parameters can be monitored to see what effect this is all having on the general running as well as sticking on a set of DET cans to listen for and see at what point/s the det is occuring.
Needs another opinion i think and a bloody good long diagnostic session on the rollers
My first port of call is a quick compression test just to see what state the engine is in. You can glean a fair bit from that in terms of whether or not the engine is wearing evenly and if not which cylinder/s are causing issues.
I'd agree with API in that it's unlikely to be ringland failure on a 2ltr and they know their stuff on engines. Not so sure about crankshaft on the way out though. Not something you can really diagnose without physically being there.
If the MAP sensor is reading low then all the knock could well be coming from the fuelling being very lean as a result. If the external gauge is reading 1bar then at nearly 0.5bar difference it will make a significant contribution to lean fuelling therefore causing knock.
Personally i would run a wideband firstly to see if it's fuelling correctly. If it isn't fuelling right then you have to be looking at the usual suspects like MAF and Lambda sensors. Then monitor the MAP sensor to see what is what. Ideally if you can borrow a MAP sensor that would eliminate or indicate a problem there. MAF and Lambda sensors are pretty easy to diagnose issues with, without having to swap them out.
Equally all other parameters can be monitored to see what effect this is all having on the general running as well as sticking on a set of DET cans to listen for and see at what point/s the det is occuring.
Needs another opinion i think and a bloody good long diagnostic session on the rollers
Thanks for all your continued help Scoobie.. I'm going to definitely owe you a beer at this rate!
What you're saying about the MAP does make sense, worth investigating at any rate - I'm guessing even if the MAF is metering the air coming in correctly it will still fuel based on the boost it sees at the MAP?
Is the MAP sensor the same between the WRX and the STI? I've seen this on ebay and wondered if it was worth picking up to try it?
Compression test does sound like a good idea as it would hopefully show up underlying mechanical issues? I'll speak to my trusted mechanic and see if he can fit me in.
What you're saying about the MAP does make sense, worth investigating at any rate - I'm guessing even if the MAF is metering the air coming in correctly it will still fuel based on the boost it sees at the MAP?
Is the MAP sensor the same between the WRX and the STI? I've seen this on ebay and wondered if it was worth picking up to try it?
Compression test does sound like a good idea as it would hopefully show up underlying mechanical issues? I'll speak to my trusted mechanic and see if he can fit me in.
Gassing Station | Subaru | Top of Page | What's New | My Stuff