CTT boost leak analysis

[Zakowsky]

I have been doing some data analysis on the boost profile of my 2004 CTT which is having some performance issues (almost 8 second 0-60, yikes). Lewis has pointed out that I have a boost leak, but I thought I would start a thread in case this is useful for others, or if anyone can provide more feedback from experience.

 

So when I first got the vehicle it was notably faster, then after a few weeks I started hearing a wheezing sound when I let off the throttle and the turbos slowed down. Logging the data during light acceleration looks like this:

 

Green is speed in kph – so this is cruising at ~90 then slowly up to ~120 kph. The blue is RPM/10, so you can see it downshift then come up from 2000 to 2600. The red is boost in mbars, so it is climbing to about 550 fast, then tapers off, but when I let off the pedal it goes negative and that is when I hear the sound.

 

In this case I am doing a 0-60 mph, so the green goes up to ~ 100 kph, and the engine is shifting at a little over 6000 RPM. It is remapped, so the boost goes to almost 1.1 bar, but rapidly shuts down, then when I let off, a really big negative spike.

 

 

I don’t think the 1.1 bar is too high and triggering something as it will do this even when the initial spike is only 0.6 – 0.8 bar.

 

So I am going to do a smoke test, but does this negative spike and quick loss of turbo pressure mean anything specific to anyone? I will post results as I work on it. Cheers.

 

 

[GSpence2]

great post thanks

[Zakowsky]

Thanks - hopefully this will be useful for others too.

 

So an update; a big clue came from looking at the measuring block reading for Idle, Part Throttle etc...the car basically never shows Full throttle, it goes from Part Throttle right to Overrun! And when Overrun hits, the boost pressure drops like a stone.  The only time I can avoid this is by not pressing the throttle all the way, during which time I can get the boost pressure to hold at around 0.7 - 0.8 bar. This would explain a lot in terms of the performance; it often seems stronger when I don’t put my foot all the way down.

 

Now the hardware side of this; Lewis suggested the N75 value which makes sense in terms of its tie in to shutting down the boost when an overboost is detected. But I am not so sure it is the valve itself. I know its been talked about, but just to make sure, how should the air flow go when the vehicle is off? In this picture I have labeled the two input hoses (top and side) and the two ports they attach to. The one out the bottom is to the wastgates I assume.

So when I blow into the side hose, there is no air flow (although it feels like you are blowing against a diaphragm). The top hose vents to atmosphere. This are normal I believe. The top port is shut (can’t blow air in), but when I blow in the side port I can hear air leaking out the passenger side down by the turbo. Is that normal? My first guess is that I have something wrong with the wastegate on that side.

 

Anyway I think this is narrowing things down...

 

[lewisweller]

5 hours ago, Zakowsky said:

Thanks - hopefully this will be useful for others too.

 

So an update; a big clue came from looking at the measuring block reading for Idle, Part Throttle etc...the car basically never shows Full throttle, it goes from Part Throttle right to Overrun! And when Overrun hits, the boost pressure drops like a stone.  The only time I can avoid this is by not pressing the throttle all the way, during which time I can get the boost pressure to hold at around 0.7 - 0.8 bar. This would explain a lot in terms of the performance; it often seems stronger when I don’t put my foot all the way down.

 

Now the hardware side of this; Lewis suggested the N75 value which makes sense in terms of its tie in to shutting down the boost when an overboost is detected. But I am not so sure it is the valve itself. I know its been talked about, but just to make sure, how should the air flow go when the vehicle is off? In this picture I have labeled the two input hoses (top and side) and the two ports they attach to. The one out the bottom is to the wastgates I assume.

So when I blow into the side hose, there is no air flow (although it feels like you are blowing against a diaphragm). The top hose vents to atmosphere. This are normal I believe. The top port is shut (can’t blow air in), but when I blow in the side port I can hear air leaking out the passenger side down by the turbo. Is that normal? My first guess is that I have something wrong with the wastegate on that side.

 

Anyway I think this is narrowing things down...

 

As per my last email the side hose should be closed, this means you have a hose leak or a leaking wastegate actuator diaphragm. 

I replaced both side as they were cracked and leaking a bit. 

If you check my post you will find it. Thomas has also replaced all his with blue silicon hoses and pipes.

Basically they are brittle but also hard to pull off the spigot fitting on the turbo and I just sliced down the side or the hose with a Stanley blade (feeling around as you can't see really). My new pipes fitted well enough I didn't have to use any securing clip or clamp. 

Edited January 16, 2017 by lewisweller

[Zakowsky]

So it does seem my N75 valve is not normal, and as per your email Lewis, great idea to disconnect the T to the diverters and simulate a changeover valve failure. I'm starting work now but will try it tonight. I keep thinking there is something else going on here - this is a graph of just normal driving...is it normal for the turbo to only come on when accelerating a bit hard? You can see how mine only comes on 10-20% of the time; and sometimes for no real reason, although if I hit an incline it will, which wouldn't show on the graph. In this one purple is boost, blue is RPM/10 this time, red is speed in kph and green is advance BTDC.

Good luck with the other fuel pump.

[lewisweller]

17 hours ago, Zakowsky said:

So it does seem my N75 valve is not normal, and as per your email Lewis, great idea to disconnect the T to the diverters and simulate a changeover valve failure. I'm starting work now but will try it tonight. I keep thinking there is something else going on here - this is a graph of just normal driving...is it normal for the turbo to only come on when accelerating a bit hard? You can see how mine only comes on 10-20% of the time; and sometimes for no real reason, although if I hit an incline it will, which wouldn't show on the graph. In this one purple is boost, blue is RPM/10 this time, red is speed in kph and green is advance BTDC.

Good luck with the other fuel pump.

Mike we need to find the requested boost and graph against actual, this will tell if the boost system is doing as it's told or not. The graphs of each of these should follow each other quite closely or it spells an issue. 

 

 

Fuel pump was a bust! I'm holding back the tears........ Lol

Edited January 18, 2017 by lewisweller

[Zakowsky]

Well the diverter valve test was a great idea Lewis – I disconnected the line from the changeover valve at the T, hit record and went to the beer store. No difference in the negative pressure, wheezing sound when I let off the throttle, and no stumbling or anything else new. So it look like my diverters are actually doing nothing. Since they seem alright there is a good chance the valve is bad, or there is a vacuum leak to the valve. And for some reason disconnecting the diverters made the other problem, the quick turbo shutoff a little bit better.  Red is boost, blue RPM/10 and green speed in kph.

 

 

You can see the boost stays up a bit longer than before, although this was not winding it out in any way – I didn’t want to blow anything out with the diverters disconnected. Also the idle was smooth and it just ran really well. So I am ordering a changeover valve and an N75 valve and as I wait weeks for them to come in I will check for leaks at the back of the engine.

 

That sucks about the fuel pump. I’ll keep logging my injector timing and maybe some logic will come from that. I am still getting Overruns all over the place.

 

[Zakowsky]

Edit - post on hold, I might have found something....

 

Edited January 22, 2017 by Zakowsky

[ekstroemtj]

Did you continue to look for your Problem? Found it? Or maybe you posted the result in onother threat!?

[Zakowsky]

Hi Thomas,

 

I kind of dropped off line due to some unfortunate life related events, but ready to get back into it, and this would be a good way to start! Since making the original graphs I have optimized my recordings, and I now have what should be final boost pressure versus rpm and speed. Here are two current runs I did.

 

So speed in kms is in blue, rpm in red (divided by 10), and boost in green (eg 700 = 0.7 bar). The time scale on the bottom is in sample units, and works out to a somewhat disappointing 6.4 second 0-60mph time. Couple of things I am seeing, the boost levels are not as high as I saw when I first got the vehicle, just over 0.7 seems the max. Boost starts at about 2800 rpm, which I guess is normal, although a little earlier would be nicer, since it hasn’t even peaked when you shift out of first. It is also dropping after the peak a bit fast, down to 0.45 - 0.5 by redline (6200 rpm). Could shift higher but the tach is way off the actual it seems, as I thought I was well into the 7000 rpm range.

 

So since there are no major or obvious leaks or malfunctions (and the boost holds well on long uphill climbs), I am thinking there might be some leaks at higher pressure at the places I have taken things apart and not used new gaskets on, like the throttle body to Y pipe, and the intercooler line where you drain the oil out of. Going to try sealing things and testing again, and hoping to get the boost back up closer to +0.8, dropping to 0.75 on shifts, and see if that can pull it closer to the 5 secs I’m hoping for. Btw it is a remapped ECU, although I have no info on the specifics.

 

[ekstroemtj]

Sorry to hear. Hope you could leave the unhappy things behind.  

Thank you for your detailled instructions.  I am still hunting for leaks too. Preasure is too high with me that i get constant overboost codes. Somewhere the engine is reading a low boost and thats why its building more and more pressure. Have to doublecheck all places again

[Zakowsky]

Hoping for some feedback from any turbo gurus who might see this, I think I don’t know enough about how the system works here...

 

So I went over the intake side looking for leaks, specifically I was concerned the Y-pipe to throttle body o-ring was loose; there was noticeable play between the two. A bit of teflon tape and it was nice and snug, and here are some results:

 

 

Turbo boost is in green (700 = 0.7 bar); RPM/10 in red. So basically no improvement was made by sealing up the connections better. But if you look at what is going on in detail it seems strange. Observations and questions: boost seems to start at around 2500 rpm – this is normal, correct? But look what happens in first gear when RPM gets to about 4500, boost stops at 0.55 bar and starts heading back down, until the transmission shifts at 6200, at which point the boost starts climbing again, all the way up to 0.7 bar, with the RPM still dropping because of the shift. This can’t be normal(?). If there was some function that shut the turbos down at 4500, that would explain why they shut down while the engine is still building RPM, but a second later boost builds all the way to 0.7 bar with the engine revving much higher. But it is always backwards; except for low RPMs, the boost builds as the engine slows down, and drops as the RPMs rise. Except once on the first graph, at the end of third gear where there is a quick peak and it holds boost through the gear.

 

So here is a test of its ability to hold boost, going not too fast up a steep hill.

 

 

Seems fine here, part throttle, and it holds a reasonable boost level as RPM slowly climbs. Which I know it does from towing uphill, it can keep high boost on for minutes under certain conditions. Btw, don’t trust you gauge too much, mine was reading 0.8 -> 0.75 when it was actually 0.7 -> 0.65.

 

So the questions are, why does the boost go dramatically in the opposite direction to RPM in first and second over 4500? And why did it stop building in first, when it couldn’t be overboost, because it can hold at much higher levels under different conditions? Plus I never get overboost codes.

 

The only thing I can think of is there are different pressure sensors that are being used by the ECU to make these decisions. If you look at the first recordings I did, where I wasn’t completely sure of the pressure sensor being tracked, boost pressure is shown to spike above 1.0 right at the start. The new recordings which show final, actual boost don’t have the spike, but the initial boost cut in first sort of happens at the same time. So is there a second sensor somewhere that doesn’t cause an overboost code, but dumps pressure, and you only get a code if the actual pressure goes over max?

 

It would be interesting to hear from some turbo experts. And I really wish I could program the ECU curves myself!

 

[lewisweller]

14 hours ago, Zakowsky said:

Hoping for some feedback from any turbo gurus who might see this, I think I don’t know enough about how the system works here...

 

So I went over the intake side looking for leaks, specifically I was concerned the Y-pipe to throttle body o-ring was loose; there was noticeable play between the two. A bit of teflon tape and it was nice and snug, and here are some results:

 

 

Turbo boost is in green (700 = 0.7 bar); RPM/10 in red. So basically no improvement was made by sealing up the connections better. But if you look at what is going on in detail it seems strange. Observations and questions: boost seems to start at around 2500 rpm – this is normal, correct? But look what happens in first gear when RPM gets to about 4500, boost stops at 0.55 bar and starts heading back down, until the transmission shifts at 6200, at which point the boost starts climbing again, all the way up to 0.7 bar, with the RPM still dropping because of the shift. This can’t be normal(?). If there was some function that shut the turbos down at 4500, that would explain why they shut down while the engine is still building RPM, but a second later boost builds all the way to 0.7 bar with the engine revving much higher. But it is always backwards; except for low RPMs, the boost builds as the engine slows down, and drops as the RPMs rise. Except once on the first graph, at the end of third gear where there is a quick peak and it holds boost through the gear.

 

So here is a test of its ability to hold boost, going not too fast up a steep hill.

 

 

Seems fine here, part throttle, and it holds a reasonable boost level as RPM slowly climbs. Which I know it does from towing uphill, it can keep high boost on for minutes under certain conditions. Btw, don’t trust you gauge too much, mine was reading 0.8 -> 0.75 when it was actually 0.7 -> 0.65.

 

So the questions are, why does the boost go dramatically in the opposite direction to RPM in first and second over 4500? And why did it stop building in first, when it couldn’t be overboost, because it can hold at much higher levels under different conditions? Plus I never get overboost codes.

 

The only thing I can think of is there are different pressure sensors that are being used by the ECU to make these decisions. If you look at the first recordings I did, where I wasn’t completely sure of the pressure sensor being tracked, boost pressure is shown to spike above 1.0 right at the start. The new recordings which show final, actual boost don’t have the spike, but the initial boost cut in first sort of happens at the same time. So is there a second sensor somewhere that doesn’t cause an overboost code, but dumps pressure, and you only get a code if the actual pressure goes over max?

 

It would be interesting to hear from some turbo experts. And I really wish I could program the ECU curves myself!

 

Hi Mike. Not a turbo guru but I will add my pennies worth. 

1. The turbo comes in at 2500 this is same as mine. Don't forget I have the E81 kit for 500bhp and 0.8bar + boost if the gauge is to be believed. 

2. Boost is lower in first gear yes, mine is the same, it's done to protect the engine and drive train I think. I notice the redline is hit quickly in first gear when rpm flies from 4000 to redline, it's almost so quick you bounce the limiter if in manual with TC off (it won't auto shift).

3. Second gear will and should deliver peak boost from 3000k to gear change rpm. Next gear should also go to peak boost assuming full throttle.

4. If you short shift I can hold mine pegged at max boost it won't drop and build up like if you hit the shift point. Try that on a run be interesting to see if you system will hold the peak boost??

5. Your first graph looks quite good boost peaks and drops at gear change as expected in relation to rpm. Second graph looks wrong where has the boost gone after the gear change ? This is electronic issue (boost map requested this) or a leak or the wastegate is stuck partially open after the shift? Maybe why you don't build boost as high as one would expect from a remapped ctt ? 0.65-0.7 is standard boost pressure not modified. 

6. Have you cleaned the map sensor ? Have you tested the wastegate actuators? They should hold pressure and open in Tandem with 10psi or slightly less. If this isn't spot on with no leaks and they don't open properly and they stick then your boost will never be correct. 

 

Edited April 23, 2017 by lewisweller

[Zakowsky]

Thanks Lewis,

 

That explains the cut off in first...really wondered about that. The drop off during the run up in second and third is what I would really like to change, as you say it should be climbing not falling off. I will try short shifting and see if I can hold the boost that way. Combined with the speed recordings, I wonder which will produce the faster 0-60, boost or rpm? I still have to get used to the transmission though – this is literally the first automatic I’ve owned in my life – I still instinctively go for the clutch, then get all messed up and over rev. I also didn’t know at first that if you over rev in manual with the PCM off it lets you go really high...I hit like 8000. But I will play around with it. I was also trying to slip the torque converter, holding the brakes until at least 3000 but it doesn’t seem to work too well. I have some recordings of me blowing shifts...they are kind of amusing.

 

I’m feeling this is a map issue rather than a leak these days. Wg actuators are OK, one squeaks a bit but works fine, the same as the other side anyways. Although maybe they are taking too long to close, but if that is the case I’m boned, because they are basically inaccessible. The MAP sensors are clean, as with the boost sensor etc. It’s like the map they put in is as crappy as the other work this tuning shop did to the car. But I don’t want to spend +1000$ on a new map if it is the same as what I have. It would be interesting to see your E81 curves – using the VAG-COM you just record from the instrument controller, channels 1 and 51.

 

I wish I could get anywhere near the wastegates; I could at least douse them in WD40. And I am still tempted to spray something inside the actuators...and I want to add a resistor to the boost sensor and ramp down what is reported to the ECU. I'm just full of bad ideas!

 

[Zakowsky]

First, apologies for forgetting you can’t run any 0-60’s due to the torque converter leak Lewis. All I have to complain about is the endless snow here (it’s almost May, come on already).

 

So it might be of interest to absolutely no one, but a few things are dawning on me here. First of all, Porsche published 0-60’s are around 5 secs, but other car mags of the day were reporting closer to 6, so mine is not off by that much. But two factors I didn’t consider. First, transmission shift times. Looking at the last graphs, 1-2 could take almost a second; now I am not used to automatic transmissions and just assumed they were horribly slow, but that seems awfully long. I tried a bit of manual shifting (hard to wind it out with the snow), and as I have noticed in the past, it seems almost worse. If you are not full throttle you get a relaxing 1.5 maybe even 2 second shift! PCM on or off doesn’t seem to make a difference. But WOT it seems you can get the same (1 sec) shift, but it is really hard to do, because if you leave PCM off and shift for 1-2 at 6000, it hits close to 8000 before it actually shifts.

 

So is my transmission failing? It had a new valve body put in several years ago, and the endless threads on tranny problems all focus on hard shifting, which mine only does after long highway runs, going from 5-4 with PCM on. May I should flush it, see if that helps.

 

But then the second thing I realized is I now have my summer tires on now, and I checked old shift times versus the new recordings, and could see shift times as low as 0.5 seconds. The computer could well be adjusting to the increase torque needed with the higher gear ratio caused by the tires size. And for the 0-60, my “small” rims are 20” for the winter, and I now have 22”, which over stock rims of as small as 18”, that could explain a lot. Could turn a 5.9 sec 0-60 into 7.2 secs. That old math thing again…

 

Anyway the bright side of all this is while sealing up the Y-pipe to throttle body junction didn’t do anything for performance, the RPM dip at idle seems better.

 

[Zakowsky]

To possibly close out this thread, just wanted to post a last 0-60 log. After flushing the transmission the shifting improved, and so this is how it looks with the Y-pipe and intercooler hoses well sealed, engine running well and trany shifting properly. Speed is in blue (kms), boost in green (700 = 0.7 bar) and RPM red (600 = 6000 rpm).

The time samples correspond to about a 6.2 second 0-60 mph time, boost is well into the 0.7’s now, the 1-2 shift is cleaner and given the tires size that time is OK I guess. There is still a small vacuum leak somewhere, because I still get a slight dip at idle, but the air blows hot theses days when on Lo, so maybe the vacuum line to the heater control unit is leaking and I haven’t checked into that yet.

 

[Zakowsky]

I was thinking of posting this in your over-boost thread Thomas, but it is a pretty specific question for boost issues in general. I happened to meet someone with a lot of experience with such problems on the Cayenne yesterday; he checked out how mine ran and strongly suggested new diverter valves. I know this was the suggestion from Lewis way back at the start of all this, and it really makes sense now. My rapid loss of boost and the weird sound from the wastegates opening a bit too early just says the springs in the diverters are shot. And in your case maybe they aren’t opening fast enough (?).

 

Any I was wondering which diverters people here were using? Many people seem to say just get the stock ones rather than aftermarket, which sometimes need to be oiled and offer no real gain. The stock Porsche part is 95511071001, but that seems to just be a Bosch 710P, which some people use, or some people use a 710N which has a stiffer spring. Is there a VW or Audi part number that can substitute? I don’t know if I can get a Bosch number here in town. Thanks.

[ekstroemtj]

Hi Mike. Thank you very much for sharing your info. I replaced both DV valves for new OEM Porsche and afterwards I tested the old ones and both where working. Didnt solve my problem. 

The wastegates , actually Lewis pointing to this,  both working. At least I can hear them when I am making this pressure test.

[ekstroemtj]

Mike ..... You mentioned you sealed the intercooler pipes. You are talking about the pipes down coming from the charger or you checked them even up when they meet the y - pipe?

 

[lewisweller]

3 hours ago, Zakowsky said:

I was thinking of posting this in your over-boost thread Thomas, but it is a pretty specific question for boost issues in general. I happened to meet someone with a lot of experience with such problems on the Cayenne yesterday; he checked out how mine ran and strongly suggested new diverter valves. I know this was the suggestion from Lewis way back at the start of all this, and it really makes sense now. My rapid loss of boost and the weird sound from the wastegates opening a bit too early just says the springs in the diverters are shot. And in your case maybe they aren’t opening fast enough (?).

 

Any I was wondering which diverters people here were using? Many people seem to say just get the stock ones rather than aftermarket, which sometimes need to be oiled and offer no real gain. The stock Porsche part is 95511071001, but that seems to just be a Bosch 710P, which some people use, or some people use a 710N which has a stiffer spring. Is there a VW or Audi part number that can substitute? I don’t know if I can get a Bosch number here in town. Thanks.

Use the 710P it's the revised part and works well. 

[Zakowsky]

3 hours ago, ekstroemtj said:

Mike ..... You mentioned you sealed the intercooler pipes. You are talking about the pipes down coming from the charger or you checked them even up when they meet the y - pipe?

 

 

So the main one that was leaking on mine was the one you undo to drain the oil out from. It was easy to tell that it was leaking because there was oil sprayed around that area. I sealed that up with Teflon tape and the car ran better. I have since ordered the proper replacement gasket so no more tape.

 

I was worried about where they meet the Y-pipe as well, and did my Teflon tape test there as well. But that didn't make any difference so I took it out. I think the steel clamps around rubber onto plastic holds well enough.

[Zakowsky]

5 minutes ago, lewisweller said:

Use the 710P it's the revised part and works well. 

 

Perfect, thanks.

[ekstroemtj]

52 minutes ago, Zakowsky said:

 

So the main one that was leaking on mine was the one you undo to drain the oil out from. It was easy to tell that it was leaking because there was oil sprayed around that area. I sealed that up with Teflon tape and the car ran better. I have since ordered the proper replacement gasket so no more tape.

 

I was worried about where they meet the Y-pipe as well, and did my Teflon tape test there as well. But that didn't make any difference so I took it out. I think the steel clamps around rubber onto plastic holds well enough.

 

About which gasket you are talking Mike?

[Zakowsky]

Number 39 on mine (2004), I think 25 is an alternate version on newer models.

 

[Zakowsky]

Although I started this in the overboost thread, figured I’d end it here. Not a great ending really, but for others who might be doing the same. I finally replaced the diverter valves, and just a few notes on the job. Passenger side is obviously easy, but (in hindsight) the driver’s side can be even easier with the right tools, in that taking the wheel tubs out on the passenger side in a PITA on mine for some reason...well out is easy, getting them back in is a pain.

 

For the drivers side I messed with the top clamp for an hour with a variety of needle nose vise grips and pliers and such, then said screw it and went and bought one of these for 30$.

 

 

So not the standard straight ones which most places sell, got this in a set of two at Canadian Tire. Using the one with the angle in them it took me literally 30 seconds to take the clamp off and the valve out. Other tip is when you put the new valve in, put the pliers on the clamp before you install the valve. It ratchets, so lock it open, slide it over the hose, install the valve then slide it down. 30 minutes wasted before figuring that out too.

 

But before the results, does anyone know wtf this is? I took out the driver’s side valve and this fell out of it! A short steel tube, doesn’t look like part of the valve. Why would this be in there? I thought maybe it fell down the hose from what ever is above?

 

 

 

When I saw that, and that the wrong diverter valves were in there to start with (0 280 142 108, which is for an Audi) I thought the new ones (710P) would be the answer. Not so however. The wheezing noise was a bit louder and the 0-60 times a bit worse if anything. Here are the graphs:

 

 

So with new diverters I’ve learned for sure that the wheezing noise is just my wastegates opening. It happens mainly when I accelerate lightly and let up quickly, like it was getting ready to built up fast but gets stopped, and dumps out the air. Doesn’t happen with WOT. Putting one new DV in did drop the turbo start RPM from ~2500 down to ~2200, and putting the second dropped it further down to ~2010. But I paid for that on the other end, as now they don’t break 0.7 bar. But over all it holds boost better in medium acceleration, they kick in faster so I guess in typical driving they will be better. Just sucks to lose top end.

 

But what really is a bummer is when I am not WOT I can actually get better performance...here is a slow acceleration before going to about 3/4 throttle. Almost 0.8 bar! And holds better too.

 

 

There must be a reason for this. Any ideas, and does anyone know what that steel ring is for? I didn’t put it back in btw...thanks.