VW TDI forum, Audi, Porsche, and Chevy Cruze Diesel forum banner

1 - 8 of 8 Posts

·
Registered
Joined
·
37 Posts
Discussion Starter #1
Hi to all. I stumbled onto this page in search for an overboost cure for a 2003 Jetta ALH. I have done all the preliminaries and have no vacuum leakage as per smoke test. On Vag diagnosis via MB 11, there is definate overboost resulting in limp mode. I have tried substituting N75 with no change. I suspect vane binding. I have one question concerning that, What is the full stroke of the actuator arm? This appears to only move 5mm or so measured at the actuator arm pin.
It also appears that someone has been playing with arm and stop adjustments due to the presence of broken paint at the adjuster nuts. Is ther a baseline to getting back to factory specs?
 

·
Premium Member
Joined
·
23,902 Posts
I can tell you 5mm is definitely too small but a bad actuator could be limiting it. The baseline on the actuator is set according to a vacuum pump test, there's an article in the FAQ on how to test, replace, and set the actuator. The stop is set according to the test but I ideally you should never have to adjust it.
 

·
Registered
Joined
·
37 Posts
Discussion Starter #4
Thanks for responding chitty and luke. I have pulled the actuator to take it out of the equation and I am lucky to get even 5mm of movement. I thought it looked too small at first sight. I think I will be pulling the turbo to clean the vanes. While off I can investigate the stop adjustment issue too as it appears to have been moved. I did find the document/TSB that you were referring to, but if the adjustments for both appear to have been moved then I have no valid point of reference to begin with. The adjustments, if I read correctly, are based on known actuator rod length. I can discern by paint remnants that the actuator rod measured fron the base to the center of the hole may have been about 88mm, does that sound right? Does anyone have one off that is unmolested and could make a quick check for me to confirm length?

I own an import repair shop and have inherited this problem. We replaced the engine with a new replacement due to massive destruction on the #4 cylinder caused by unknown valve train failure. The owner did state that the engine did break a timing belt at one time and he replaced it himself. I think it may have fractured a valve then because it looks like the head snapped off of one this time around. Much like the 1.8T gassers I have seen. The engine runs perfectly, just goes into overboost on a steep hill, and there are lots of those around here.

We have double checked timing and installed a set of flow matched injectors to baseline things. I have done numerous trips graphing MB 11 and 3 on Vag. The MAF if perfect, Map believable but not tested with a boost gauge, EGR functions ok, or at least no codes, substtuted N75, etc.
 

·
Premium Member
Joined
·
23,902 Posts
I think you're on the right track. There's an article in the FAQ on limp mode, sticking vanes are common issue if you've already checked the actuator, vac lines, solenoid, solenoid wiring.
 

·
Registered
Joined
·
37 Posts
Discussion Starter #6
Update:

Success. I figured it out after gleaning important facts and suspicions from all of you. I will warn you in advance that this may be a fairly long post, so I will try to break it up into the answer to my problem then the questions and theory that spawned from my discoveries. So if you just want the gory details you can skip the end.

The major problem turned out to be sticky vanes, or the shaft that goes through the housing at least. I felt that the 5mm movement that I had seen on the vane arm seemed too little to me at first too. That is one reason I was perusing this sight to find a spec on that. I was able to cheat and spray a little penetrant on the shaft, work it a little, repeat as necessary till I got full throw, which unfortunately I neglected to measure accurately too. I suspect it would have been close to 10mm though. I suspect the fact that this car sat for several months while owner was saving money to replace the engine may have been a contributing factor to the stickiness. He had no complaints of low power/boost before.

The second part of my problem was uncertain actuator arm length and stop adjustments that had that altered look. upon removal from the car I was able to fairly approximate the original starting point on actuator length by paint remnants. It measured approx 88mm from the diaphragm face to the center of the arm hole. I did similar on the stop but paint marks were not near as legible.

That got me in the ballpark on adjustments and so the trial and error began. I would lengthen the rod, drive and graph MB11 on vag, repeated as many times as necessary to get the requested and actual map to mirror each other as close as possible. I ended up at 92mm length on the actuator rod before the maps started to come together, then assuming that was close I adjusted the stop as per VW service bulletin found on this site. Overboost cured. Thanks to all that helped.

Now for my questions that aroused during the curative process, excuse me if I ask questions that have been beaten to death on this site already. (I did try the search function)
.
The sticky vanes obviously were the root of the problem but the rod length had to be altered quite a bit from it's near original length, what caused such a change? We did install matched flow injectors with a higher flow rate, could that skew the requirements? In theory, more fuel delivery and definitely more power might equate to more flow and heat to spool the turbo quicker and more than was designed into the original ecu character-map. Sound feasible? I have to believe this is not untrodden ground.

I also believe from my observations that the actuator duty cycle is hard written into the ecu as far as duty cycle response, thus the need to match the rod length to the vane actuation. I observed though as I got closer in rod length and finally settled where I did, the graphing seemed to tighten up a little and spike less the more I drove and observed. I have to wonder now if once the center of the adjustment is found and the ecu doesn't have to struggle so hard, does it start learning and adapting? Is that true or my imagination?
Kinda reminds me of the old days of K-jetronic E gas injection where I would adjust the mixture to midpoint 02 voltage RMS on my Fluke scopemeter so the ecu would then richen as often as lean the mixture and the cars ran much better and smoother. Except those were definitely not adaptive that I ever observed.

What exactly does the stop on the vane arm really do? I only tried adjusting it after getting the rod length decided and I could see absolutely no difference in mapping or performance after adjusting it, and I did have to move it a ways too.

Again thanks for all the help and apologies for the lengthy post.
 

·
Premium Member
Joined
·
23,902 Posts
No clue why the rod length had to be altered so. The actuator has a spring inside which works against vacuum supplied from hoses that go to the N75 solenoid. Vac is supplied from the vac pump on the driver's side cyl head. Are you sure there is no vac problem? I believe that the ECU flutters the solenoid back and forth to control the amount of vac going to it until it sees the correct boost. The stop on the vane arm stops the arm :D

No prob for the long posts, I hope lurkers can learn something from your experiences. And the more description, the easier it is to help people. Post moved to mk4 section as wel.
 

·
Registered
Joined
·
37 Posts
Discussion Starter #8
Thanks for moving me to my proper place, I am a newb to this site and do not know where to be exactly.

There are no vacuum or N75 issues I am sure. The first tests I perform on vacuum diagnosis are with the mity-vac and also I smoke tested the vacuum system and found no leaks with either procedure. The smoke test did however show a wisp of smoke at the valve cover end showing me an oil leak I didn't know I had yet. At most points of the diagnosis adventure I had two vacuum gauges hooked up monitoring both available and solenoid vacuum. The supplied remained steady, and the solenoid vacuum followed the duty as near as I could ascertain while trying to watch graphing, vacuum and road conditions at once. The vacuum pump is new also, and I did substite N75 early on too. Since I have dialed this one in it reads beautifully on all traces of MB11 vag. Both map traces mirror each other and the solenoid variances are reflected on the contolled map trace exactly.

You can bet I will be graphing other customers known good cars for a while to get a broader field of view. I am the owner/manager of the shop here so I really don't get the opportunity to lay hands on a car much anymore unless they are a problem child like this one. I will make it a point though to test drive a TDI now and again to further my knowlege.

I drove the car home last night. On my normal route home part of the drive includes a hill 7% grade 10 miles long. I call it my dyno, it will certainly test a car under load.

I floored the accelerator as soon as I hit the steep part, I had slowed to 30MPH and held it there till 90 MPH or so where i backed off for fear of officer intervention. The car pulled well, and did not go onto overboost/limp mode. It did get dangerously close, at one point the map was hovering near 2200mbar and the duty had locked at 94.2 which I have discovered many times is the limit, on this car at least. I backed out of the throttle before it could decide to limp home. i think I will leave it just as it is because in the real world situations I don't see anyone overboosting it like that. I could lengthen the rod a little more, but I like how it runs now, just pushing the limit at extreems.

Also, my thoughts on the stop for the vane actuation arm after thinking about all last night. I was puzzled by the fact that after changing the actuator arm length to match requested map it also threw the stop adjustment off it's adjustment as per VW TSB on that subject. I have decided it is just merely a limit stop on arm movement based on the actuator arm setting alone. I liken it very much to a gasoline car idle limit control function, where you have a idle speed determined and maintained by engine computer, but there is a lower limit mechanical stop adjustment that is set lower than the normal idle speed. If the ECU gets confused momentarily or response time is too slow, at least there is a bottom setting to fall onto so the car won't stall. I think the limit stop on the TDI turbo vane is the bottom limit of corrected boost, thus making it's adjustment logically follow the actuator change, but not being as critical of an adjustment as the actuator rod.
 
1 - 8 of 8 Posts
Top