joz8968

iTrader: (13)

Cool.

simonds1

Guys, its started happening again!! I don't ****in believe it

simonds1

So I'm guessing these are the options......

1. The fitting of my new (second hand) boost solenoid off ebay didn't actually make any difference at all because the boost solenoid isn't actually the problem.......and whatever is the problem just happened to work for a bit and it was purely coincidence that the car seemed to improve after fitting the boost solenoid.

2. The fitting of the new solenoid did initially solve the problem, but something else is wrong on the car and is causing the boost solenoid to fail after a period of time? Oil getting into it or something? I have no idea.

3. The new boost solenoid isn't actually fully working. I mean, it was from a MY98 car, which is 2 years older than mine anyway. So if my problem is that my boost solenoid is on it's way out, then I'm pretty sure it's reasonable to say that one that's 2 years older could be on it's way out aswell?

4. The boost solenoid has got nothing to do with the problem at all!


Ahhh well, I dunno. I give up!

TimH

iTrader: (11)

Classics are definitely prone to oil contamination of the boost control solenoid. You have probably been asked, and answered, this already (too lazy to check back!) but is the oil overfilled? The advice when I had my classic was to only ever fill to half way between min and max otherwise you run the risk of contaminating the solenoid.

Check that, and also (re)clean the pipework to the solenoid with carburettor cleaner.

simonds1

Hi Tim, no I had not been asked that. Not that I can remember anyway! The oil is always just below full when I get the car back from being serviced, and I have never ever had to top it up in between. I'll have to give the pipes etc another clean and see if that helps at all.

Also, another thing I'm going to try.....



The boost solenoid I got off ebay came still attached to the bracket you can see in the picture, and it had part 'C' on it aswell. Someone thought that was a 'pressure exchange solenoid' when I asked before? But I'm not really sure. Anyway, I didn't change that, I only changed the boost solenoid. So I'll try changing that aswell to see if it makes any difference

joz8968

iTrader: (13)

Yeah,

A. 2-port BCS
B. MAP sensor
C. Press exch sol

simonds1

Just to update, changing the pressure exchange solenoid hasn't made any difference

simonds1

Hey guys, I'm guessing everyone's bored of this thread now but I just wanted to update you again......

I can absolutely 100% confirm that the problem is NOT the boost solenoid like we first thought. I thought because the replacement one I was trying was a 2nd hand one off ebay, there's a chance that it might not be a fully working one. So I was looking for another one to try the other night on ebay, got carried away with my bidding and accidentally ended up winning two lol. So now I have tried THREE different ones on the car......and they all make bugger all difference and I've still got the car to suffer from the non-accelerating problem with all of them on. So yeah, if anyone ever needs a boost control solenoid .....



So my next question is of course....does anyone have any ideas as to what I might look at next? One of the first things I thought of was throttle position sensor, because that's what the problem actually feels like when ur drving the car. Just feels like the car isn't responding to your throttle press like. But according to the logs, the TPS is working fine. Splitpin got me to do a logfile recording only the TPS voltage, and to gently press and release etc to check I got smooth voltage changes and yes I did. So yeah that seems okay. The only other thing I can think of is the wastegate actuator sticking? Another thing that was mentioned previously was the posibility of an air leak? But it can't be an air leak surely, because the problem is intermittent? Like you wouldn't have an air leak one minute, and then not the next surely? If you've got an air leak, you've got an air leak.

So yeah any ideas guys, please throw them my way cuz I'm getting desperate now! Gonna have to actually think about getting rid of the car if I can't solve this, and I really don't want to do that

TimH

iTrader: (11)

Feel for you

I guess the clue here is that changing the BCS - at first - seemed to fix the problem. So I'm wracking my brains trying to think what would have been sorted - albeit temporarily - by changing the BCS if the issue isn't the BCS...aaarrrggghhh!!!!

simonds1

Hi Tim, that's exactly what I was thinking, and that's why I bought another one (well, two lol) to see if when I initially fitted it it would be okay......and then eventually fail......but this was not the case. Car faultered with both of them pretty much straight away after fitting them. I think before when I thought I'd fixed it must've just been a complete fluke that it was working that day, and a coincidence that I had just happened to change the BCS. Doh!

The mystery continues!.....

TimH

iTrader: (11)

Does it idle OK? Does it ever stall? Clutching at straws here...

simonds1

It does seem to have different 'moods' every morning in regards to idling. Some mornings it seems to idle high and the revs will kinda rise between gear changes rather than drop when I'm first pulling away from my house like. Always smooth idling tho whether it's high or normal, it never hunts around for idle. It's never idled low enough to stall either. When I say it's idling high, I mean about 2kish rpm like, not stupidly high or anything. Also, if it is idling high, a little tiny blip of the throttle will get it to settle down lower. Dunno if that means anything?

TimH

iTrader: (11)

Reason I mention it is that I've had an idling issue which - I think - I've just fixed. It would often try and stall, and would sometimes idle high and I found that the throttle cable had no slack in it at all and was sticking slightly. So, when taking foot off throttle, the throttle would not always close properly.

I *think* that was confusing the ECU and resulting in strange values being sent to the ICSV.

As I said, clutching at straws, but it might be worth checking your throttle cable, and/or logging the ICSV percentage - mine was all over the place...although, that said, maybe that's normal.

Other than that, have you got any friends or acquaintances with the same model? If so, you could swap sensors (MAF being the obvious one).

TimH

iTrader: (11)

Reason I mention it is that I've had an idling issue which - I think - I've just fixed. It would often try and stall, and would sometimes idle high and I found that the throttle cable had no slack in it at all and was sticking slightly. So, when taking foot off throttle, the throttle would not always close properly.

I *think* that was confusing the ECU and resulting in strange values being sent to the ICSV.

As I said, clutching at straws, but it might be worth checking your throttle cable, and/or logging the ICSV percentage - mine was all over the place...although, that said, maybe that's normal.

Other than that, have you got any friends or acquaintances with the same model? If so, you could swap sensors (MAF being the obvious one).

simonds1

No the throttle cable is fine. It was one of the first things I looked at back when the problem started because like I said earlier, the problem when you're driving does feel like it's throttle pedal related......just because of the fact it doesn't seem to respond to you pressing it. Unfortunately no, I don't know anyone else with an Impreza that I can try swopping sensors with That would be ideal tho!

simonds1

Another thing I meant to ask......I've just been reading around the forum about other people with similar problems etc, and in one thread someone got asked to check if their 'restrictor pill' was still present? And also that it was at it's correct position in the pipe - approx 30mm away from the 'tee'? Does anybody have any idea what this 'restrictor pill' is? And could it possibly be contributing to the problem I'm having?

Splitpin

Blimey, the monster's back, thought this one was done and dusted!

Without wanting to drag the thread off Simonds' topic, if your throttle won't close properly (or the throttle position sensor isn't adjusted correctly), it definitely will, as you were suspecting, confuse the ECU's idle control strategy.

Right, now the awkward one. Deep breath...

Splitpin

2000rpm is unusually high on these - unless (and Bob would tell you) he significantly altered the cold start behaviour when he mapped it. Not sure why he would though, not even sure Ecutek allows you to make those sorts of detailed, low level changes.

To give you an idea of what normal is, a typical cold start on a 99-00, on AE801 base code, in current ambient air temps (so a coolant temp of around 15-20 celsius) should see a peak rpm of around 1650 and then immediately dropping back down below 1400 and continuing to drop as the car starts to warm - should be below 1200 within 30 seconds of start. Once it's up to temperature, it should idle at 750-800 consistently and settle back to that within a couple of seconds of you pulling up.

So, given the above, would you say yours is "normal", or not?

Hmmmm. It certainly means something. Difficult to work out what by remote control tho. And, going back to before, "emissions mode" would drop the boost control solenoid duty - so this can be pretty definatively ruled out as your prior logs didn't show such a drop.

You did that log before of the sensor's behaviour with the engine off - and it came up fine. It's always possible that there's some sort of intermittent wear with either the pot itself or cabling back to the ECU that only exhibits with engine running, but your earlier log data didn't demonstrate anything obvious. In your logs though the WOT voltage has always been consistent, hasn't it? That early log you did when the sensor voltage wasn't maxing out, have you definitively put that down to not fully nailing the throttle?

That sounds like one of my posts. The restrictor is (should be) in the piece of boost control pipe between the compressor bleed outlet and the tee-piece. If you feel along the pipe you should feel a hard, pretty much pea-sized object inside it - and it should ideally, as above, be around 30mm away from the tee. There will probably be a dab of paint visible on the pipe in the correct location - assuming (and my memory's going to need a refresh!) the pipework is still the standard Subaru stuff.

From what you've said before it's all but certain that the restrictor is present - if it wasn't you wouldn't be getting much more than actuator boost. As Tim has said it's also possible that it's being contaminated by vent oil, but this would normally give over- rather than under-boost. I did mention cleaning the boost control pipework out a while back. Did you try this, and did you see a lot of oil coming out when you did?

Also if you were to roll a bit of tissue paper up and poke it into the ports on one of your solenoids, would it come out soaked in oil, or just with a light covering?

Further, what oil level are you running? Does it go above the full (top hole) mark when hot?

It would also be well worth trying the previous suggestion to link the compressor bleed outlet direct to the wastegate actuator, and drive the car around like that for long enough to provoke the problem (or become convinced that it isn't occurring anymore). You'll be stuck with about half a bar of boost, so you won't be breaking any speed records, but doing this will help you a hell of a lot in your diagnostic process. If the car behaves totally consistently (albeit with reduced power) while the direct connection is made then the overwhelming likelihood is that the problem is somewhere in the boost control area of the car.

If, however, you still get inconsistent acceleration running on actuator pressure, then you can almost certainly rule out the boost control system as the cause - and if that's ruled out, you can focus more specifically on other potential influencing factors.

Either way, you will come out of this process with a far better idea where to look. It's simple to do and you don't need anything exotic to do it (a few inches of 4mm fuel pipe plus a pair of suitable hose clips).

Also - does your throttle cable have a tiny little bit of slack in it when fully closed? Shouldn't have much, but definitely shouldn't be taut.

And, to close, can you try plugging your laptop into the car for a couple of morning starts - until you get some data on one of those days when the idle is "high"? It would be interesting to see exactly how high it goes, and also whether there's any obvious causation in the other data parameters. You might look, for example, for the throttle sensor giving inconsistent readings when it's cold, or something like that. If it normally registers, for example, 0.50 volts at idle, watch for it suddenly giving different values. Idle voltage with the engine on should basically never vary.

simonds1

No Bob didn't say anything about changing the cold start behaviour, and like you said I'm not sure why he would've done that. I tried to log a start thismorning but my laptop was saying something about an SSM1 diagnostic port not existing everytime I plugged the USB lead in?? Anyway, I'll sort that when I've got time! I was already running late for work lol. Speaking of work, when I came to leave work I reminded myself of idling behaviour that quite often happens. I park on a very slight downward hill at work, so everynight when I come to leave I have to reverse back out of my space like......and when I do that, it's then that the car tends to start revving highly. After I've reversed briefly and am getting ready to drive off forwards.I don't really know if that's relevant to this problem or not, and I spoze nobody else does either at this stage!

Yes it has yeah.

Yes, I've put that down to me not nailing the throttle. I really should get my facts straight before coming on here sorry!

I haven't got time to try this tonight sorry, but I'll give it a go tomorrow night for sure, and then report back.


Have never checked the oil when the car's hot. When it's cold tho, it sits just underneath the top hole mark. Again, I'll do this tomorrow night and report back.

I will definitely have to try this at some point aswell yeah. Not sure if I'll have time tomorrow night but will give it a go - might aswell get it all sorted at once!

Tiny bit of slack yeah, not fully tort like. But not too loose either I'd say.

I didn't actually try the pipework no, mainly because I couldn't get most of it off! It ain't half stuck on there I tell you! The actual boost solenoid itself was the only thing I cleaned out and there was a fair bit of oily colour coming through that before it went clean looking. This reminds me of something I meant to ask back when I first tried this. This is the stuff I used to clean out the solenoid, and I'm wondering if it's the right sorta stuff? (bit late now I know!!).....



I'll try cleaning out the boost pipework tomorrow night aswell. How dry and free of the carb cleaner do the pipes have to be before I put them back on the car? I don't want to damage anything by letting that stuff get onto any sensors etc if it can knacker them.

So I think that's it for now. Looks like tomorrow night is gonna be a busy one for me!! Cheers again for all your help, speak soon

simonds1

Sorry, forgot another thing I meant to ask.....

Do you remember me mentioning this little fella?....



I was searching this forum lastnight for other people with similar problems etc, and there was one thread with people referring to this little bit as a 'one way valve' and telling the person to check it's operation as this could confuse the MAF readings if it wasn't working? Could a boost problem similar to mine be caused by this little thing not working? I don't really know how I check it's operation either, but just thought I'd mention it and ask if anyone thinks it's worth even looking at? Cheers

simonds1

Splitpin, I was just re-reading your post in preparation for what I need to do to try out your suggested test, and I just wanted to check something with you first. You said that when I do this I'll be stuck with about half a bar of boost? Well I am quite mathematically challenged most of the time lol, but I have just worked that out to be around 7.3psi? So isn't that around what people were referring to as 'actuator pressure'? And this is what I was seeing in my data logs every time the car suffered from the problem? Sorry if I'm missing something quite obvious here, but I don't understand what I'm meant to be looking for in the car's behaviour when I'm doing this test? If I will be limited to half a bar of boost while I'm doing it.....and that is what I'm getting when the car is suffering from the problem.......then doing this will only make the car behave like it's permanently suffering from the problem? Or am I just being really stupid here? (Be honest, I can take it )

Splitpin

That valve you've pointed out under the inlet pipe is the canister purge, and shouldn't be a contributor to an issue like this, as all it does is allow (or disallow) the vapour in the carbon can to discharge into the inlet tract.

And, yes, linking the compressor and wastegate will give you actuator pressure all the time, that's the point. From what I recall in your earlier logs (don't have them in front of me at the mo), you appeared, and especially in the way the turbo spooled, to be getting less than what actuator pressure would be expected to give.

As such the rationale for doing this would be, as above, to see whether you're still getting inconsistent response even with the direct compressor-actuator connection.

simonds1

Apologies that this post is so all-over-the-place, but I'm just trying to cover everything that's been asked as best I can

Yes you re-called that correctly, in one of the logs I did the other night after trying another new boost solenoid, the boost was only going up to around 3ish psi when it was suffering from the problem.

As you can see from the pic below, there is hardly any oil on the tissue after poking it into the boost solenoid ports....



Infact, that pic is actually of the same bit of tissue that I put in and out of both ends of all three of my boost solenoids lol. So yeah there's hardly any trace of oil between the three of 'em, which is good I'm guessing!

In regards to linking the compressor bleed outlet directly to the wastegate actuator, I just wanna check I'm planning to do the right thing here....



Okay so I've colour coded two pipes in the above pic - The light blue section is the bit I thiiink you are meaning me to link directly together? If that's the case, then I'm wondering what to do with the yellow pipe coming out of the T-piece, which goes to the boost solenoid. Do I need to block the end off, or can I just leave it hanging? Can somebody also recommend where I can buy new sections of this sort of piping? I remember earlier on in the thred I mentioned some samco pipes I was thinking of buying to replace all of my boost pipework just so I could rule out damaged pipes/air leaks as a problem, and I was told that the samco pipes are **** and to not get them. Also I think I was looking at the wrong type of thing - I was looking at vaccuum hoses, which am I right in saying cannot cope with high pressure situations like turbos etc? Splitpin, you mentioned 4mm fuel pipe, so I'm guessing that's what I need?

Also, while we're on the subject of pipework, I had a look for the 'restrictor pill' we were talking about a few posts back, and if I have correctly understood what the 'compressor bleed outlet' is (in my pic above where the furthest away end of the highlighted light blue pipe connects to) then I don't think my restrictor pill is present in the location it was described to be in. There is no obstruction of any kind in any of those pipes coming from the T-piece, and also no blob of paint on any of them. I think I mentioned this before, and I'm sorry if I didn't, but Bob did modify this pipework in some way when he did my remap. I always remember asking him about changing to a 3-port boost solenoid when I had the remap done but he said it just wasn't worth doing it for the mods I had sorta thing, and that he could do a little pipe modification which does a similar sort of thing to a 3-port boost solenoid? So yeah anyway, am not sure what pipes are standard or not standard anymore because I don't actually know what he did.

joz8968

iTrader: (13)

News just in... Fingermouse in hunger strike shocker!

TimH

iTrader: (11)

LMFAO

joz8968

iTrader: (13)

simonds, you just need a new bit of pipe (temporarily) going from compressor bleed nipple to actuator - with NO T-piece off to the BCS.

The whole idea of the test is to take the BCS/PES/MAP assembly - i.e. the ECU's electronic control of BDC - out of the equation, so as to see how just the w/g and turbo are behaving (albeit at reduced pressure i.e. the w/g spring's crack pressure of c.0.5bar or so)...

TimH

iTrader: (11)

And it won't matter if the boost control solenoid is open to the winds

simonds1

Aye I know this, but I wasn't entirely sure where the 'compressor bleed nipple' was, that's all (because I am stupid lol) - hence my earlier post with the picture asking if I had got the right places to connect together. So in my pic my intention is to connect ONE piece of pipe along the light blue route, and ignore the yellow pipe & T-piece. And going from what you just said I think I did have the right idea

joz8968

iTrader: (13)

That's it!

Splitpin

Yep, this is the important thing - and under these circumstances I'd expect running a direct link for a few days to help your diagnostic process.

Yes, that pretty much rules out oil contamination as a significant contributor or cause for the issues you're seeing.

That's it. You should ideally block the "yellow" pipe leading to boost solenoid with something, as otherwise you will get a small amount of unmetered air running through it. Shouldn't be enough to cause any significant issues, but bearing in mind this is a troubleshooting process, you don't want to introduce any issues if you can avoid them.

Some vac hose can cope with pressure, but a lot of the cheap silicone stuff sold as vacuum hose can't, no. It balloons. Braided fuel hose is almost overkill but it will do what you want properly.

That's where it *should* be, yes. When you take the pipework off to make the direct link, examine it all closely to see where the restrictor is. Maybe Bob has moved it, possibly to the other piece of blue pipe on the actuator side of the tee or something like that, or maybe to one extreme end or other of the pipework so you can't feel it distinctly under the pipe next to the compressor outlet or something.

Question the man himself may need to answer - although once you have taken the pipework off you'll probably find it somewhere.