MAP logging and diagnosis

[Artmasterx]

I have a 2000 Seat Leon 1.9 TDI (110hp, AHF), about 120,000km.

I have the typical limp mode issues and have logged the behavior. Basically I can get it to limp when full throttle in 4th gear at about 130 kph. It seems difficult to get it to do it in 3rd gear. Fault is standard: "Charge Pressure Control: Positive Deviation"

Is there any way to diagnose more precisely what may the issue based on how the profiles are? Essentially the boost spike seems to be too high initially (I have 2.1 bar is typical) and then it cannot regulate the boost quickly enough (pressure goes up with increasing RPM after the initial attempt at regulation). The two primary symptoms seem to be:
1. Too slow to regulate the initial peak to 2.1 bar (getting 2.5 bar instead)
2. Cannot regulate boost to 1.9 bar under full throttle
3. May not be be related, but my fuel consumption is about 6.5L/100km (44 mpg), which seems a bit low.
(note, my MAF readings look fine)

The turbo was supposedly cleaned around 5000 miles ago (when I complained to the place where I bought it), so I am leaning towards another issue... but can't really be sure. I posted on the previous (same) issue here:
http://www.seatcupra.net/forums/showthread.php?t=280772&page=2

My next three thoughts are that either the N75 valve or the VNT actuator on the turbo are not completely functional. My theory would be that:
1. The N75 valve is malfunctioning and is not causing the appropriate vacuum to be pulled on the VNT actuator, so that the vanes done move as much as they are supposed to.
2. The VNT actuator has malfunctioned or the diaphragm has a leak so that even when the system tries to pull full vacuum on it, it does not have the proper travel to fully actuate the vanes in the turbo, causing overboost... (the only problem with this theory is that I would assume atmospheric pressure means low boost and high vacuum means high boost... is that right?)
3. Leak in a vacuum line or a constricted atmospheric pressure lines so that the N75 doesn't deliver the proper pressure to the VNT actuator?
4. Faulty MAP sensor so that the ECU is getting incorrect pressure readings?
5. Maybe they did a shitty job cleaning the turbo the last time...

Thanks for your help!

My log:

Typical log (per Ross-tech website): Note the efficient regulation of pressure...

 

[Seatmann]

Have a look at this bud.

http://www.myturbodiesel.com/1000q/multi/limp-mode-TDI-fix.htm

 

[Artmasterx]

Thanks for the link. I have looked at that and it seems like there are a number of possible causes.

Does anyone know if the profile of the MAP reading can give an indication of the specific problem. It clearly shows that I have a problem with overboost, but does it give any other indication?

For example, does the fact that the boost regulates itself partially mean that the vanes are able to move somewhat, but maybe just can't open (or is it close?) the vanes all the way to regulates the boost enough under high load?

Maybe the actuator arm is not adjusted properly and can't move the vanes over their full range of movement?

Anyone have an idea?

 

[Seatmann]

You might want ot find a user on here called MJ, he might be able to help on that one.

 

[Artmasterx]

Thanks, I sent MJ a PM...

 

[Artmasterx]

Here are some additional plots showing N75 duty cycle and the RPM readings... maybe they would be helpful. I can try to post any other comparisons that may be helpful. Thanks.

From what I have read several places, the duty cycle of my N75 seems to be too high (meaning that it is really trying to limit boost, but can't, as it hits its limit at 94% or whatever). This means that it is trying to get atmospheric pressure (correct?) to the VNT actuator to get it to relax into low boost mode. Since it is trying to apply atmospheric pressure, it makes me think it is not a leak in a vacuum line (maybe the line to the airbox is blocked, but I check that earlier and it didn't seem to be). I have read that the N75 duty should be somewhere between 60-80% at 4000 rpm... and I am significantly above that.

This makes me think that I have oneof this problems:
- The VNT actuator arm is too short and can't return the vanes to minimum boost position (causing the N75 to work its hardest)
- The VNT actuator is messed up/corroded and can't return to its fully-extended position
- The turbo is sooted up making the vanes unable to return to their minimum boost position (I am thinking/hoping this is not the problem because I had this problem earlier, had the turbo cleaned, and still had the problem)

Any sounds more plausible that the other? I am planning to test the VNT actuator once I can get my hands on a vacuum tester. I have also ordered a used N75 valve that should be here in a few days so that I can test that and see if the logs look the same. The resistance measured on my current N75 was about 16 Ohms.

Here are some good discussions about the VNT/N75 issues:
http://uk-mkivs.net/forums/p/353504/2407857.aspx
http://forums.tdiclub.com/showthread.php?t=277746

Limp mode happens on the far right of the plot, and you can see the N75 duty flattens out around 92% once limp mode kicks in.

 

[MJ]

if i were you mate i'd remove the under tray, get a vacuum tester and check the turbo actuator and vanes open/close correctly. when some people "clean" vnt turbos they take them off, look at them then spray the impellars with carb cleaner - this will do nothing unless the turbo is completely stripped, split in half and completely de-choked.

 

[Artmasterx]

Thanks, I will see what I can find. I tried to find a vacuum tester today, but it seems like I may need to hit the internet... the 2 Belgian auto parts stores I tried today didn't have them.

I will update when I get a chance to do some of these things...

 

[MJ]

ok buddy.

its a really simple test:

-plug the tester on
-vacuum the actuator (see if the actuator moves freely)
-release the pressure (if the arm moves back in one smooth, fast motion then the vanes are ok, if it steps back in stages then the vanes are sticking)

 

[Artmasterx]

Thanks again... I just need to get my hands on a vacuum tester.

 

[MJ]

ebay might be your best bet.

 

[Artmasterx]

Well, I tried out the replacement N75 and it basically performed the same. I haven't had a chance to do a log with it yet... but qualitatively the same and was able to induce limp mode.

I also did a crude vacuum test using my biological vacuum pump (lungs), and was able to get the actuator to almost touch the stop screw. Upon releasing the vacuum the arm appears to snap back in a smooth motion without any jerking.

Then I turned on the car and used the vacuum system to actuate the turbo. The arm seemed to move over the full range of motion, and did so smoothly. I even removed the vacuum slowly and it still returned to the default position smoothly. I did this from above the engine, so I didn't have a perfect view.

I also took the vacuum hose from the EGR valve and plugged it, but it still went into limp mode... I didn't think it would help, but it was easy to test.

I am running out of idea...
- Again, maybe the VNT actuator stroke is too short and needed to be lengthened?
- Could intake manifold build up cause an overboost situation?
- Could the MAP sensor be messed up and reading too high?
- anything else.?....

 

[Artmasterx]

anyone other ideas?

 

[Artmasterx]

No new ideas? Here is some more data...

I have been having the same problem for a while now, but haven't been able to get it worked out. The car is at the shop again right now seeing if they can figure it out. Basically it is an independent used car place that has been trying to fix the problem since I noticed it the day I bought the car. Up to now, they have been trying fixes themselves (cleaned turbo, new vacuum lines, fixed a minor leaking oil return pipe from turbo). They have finally broken down and taken it to an actual Seat/VW dealer to have a full diagnostic to try to get it fixed once and for all... hopefully. We'll see.

You can see some of the details in these two threads:

http://www.seatcupra.net/forums/showthread.php?t=300976
http://www.seatcupra.net/forums/showthread.php?t=280772

The typically turbo cleaning didn't seem to fix the problem with mine, and the vanes and actuator seems to move smoothly. I tried a different N75 valve but that didn't help either.
You can see my current theories in the other threads, but I will certainly update once it's fixed for good.

The Innotec thing requires you can get access to your turbo by taking the downpipe (?) off so that you can fill the exhaust side of the turbo with a solvent to dissolve the buildup. It's not a simple DIY at home for most people, but better than having to take the turbo off completely.

It could be a split vacuum hose, but doubtful. Typically this only occurs under heavy load. You could also try logging data and seeing what it looks like, you can see my examples in the other thread.

Hopefully you can get it sorted quickly and easily... definitely let us know what fixes your case.

Well, apparently the Seat/VW dealership recommended replacement of the MAF... which I could have told them would do no good based on the data I have logged and various rules of thumb heard here. And it didn't fix the problem... F-in A man.

So, I decided to see if I could determine if the intake or EGR was clogged and causing too high a back pressure. I did not take the EGR off, but took off the hose leading to it from the intercooler so that I could at least get a few pictures (below).

Maybe you guys know better what to look for, but certainly on the EGR side things look pretty clean. I couldn't get a good view to see much with the intake, but it doesn't appear to be heavily clogged from the pictures (you can see one with a pretty far focal point, presumably somewhere in the intake past the EGR).

What do you, does this probably rule out a clogged intake as a potential source?

This one has a pretty far focal point that makes me thing the intake is fairly clean...

 

[Artmasterx]

Supposedly, the shop did some analysis on the turbo and apparently the previous owner had modified it in some way (not sure what), and messed things up. It's probably that it had a tune on it at one point, since the modified turbo doesn't make any sense otherwise.

Anyway, a new turbo did the trick in this case. I wasn't able to find out what the previous owner did to it, but it is not longer overboosting, which is excellent.

Now it just looks like I need (want) to adjust the VNT arm a bit to improve the responsiveness and the get the N75 duty optimal (to between 70-80% at about 4000 rpm, correct?)

Thanks for all your insight, but I guess if someone FUBAR'ed the turbo, what can you expect other than a new turbo...