Overboost then underboost: Need help understanding this boost issue

[pyromancer]

General car info:
2001 ALH Jetta
1019 injectors
Garrett VNT-17
3 bar MAP
Malone stage 4 tune

Vacuum at idle:
@n75 - 28.5
@n75 to turbo -19
no leaks I can find

IP:
6.0 - 6.2, just retested in VCDs 32768.
Hammer modded to between 5 and 7 per instructions with 1019 injectors from DBW LLC.

I am having an issue with the car slightly over boosting and then under boosting. To me it feels like there is a bit of a flat spot at low RPMs as well which is my major issue. I have had this issue since getting the tune in 2019 and it seems that Owain at Malone agrees, this should not be how the car behaves.

Original graph.


Mike at Malone suggested shortening the actuator rod to fix the lag I was feeling. Working with Owain I one turn at a time shortened the rod quite a bit. I would guess 6ish turns. Yellow on the rod is where the Lock Nut was when we shortened it, and where the Lock Nut is in this photo is it set to coming off the stop screw at around 19ish of vacuum after I set it back.

With the rod shortened to the yellow my graph looked like this.

Shortened rod

I developed a boost leak in the line to my boost gauge at that point and had to replace my 30psi boost gauge, as it wont read lower than 2 now =(
After fixing those issue, I couldn't find any other boost leaks. On seeing the above log Owain told me I am flat lining the MAF and getting the turbo to respond quicker will limit boost spikes, without giving me an explanation of what that means, or what if anything else I should try. I assume flat lining means I was spiking above the MAFs ability to read? If anyone has a better explanation would love to know.

I've been trying to troubleshoot on my own as well and reading what I can, this in particular seems to have good explanations on turbo control for an ALH, https://www.uk-mkivs.net/threads/vnt-tdi-turbo-how-to-adjust-actuator-cure-boost-problems.34834/ From what I can generally understand from reading up on stuff, the rational of making the rod shorter to the point we did has not made any sense to me at all. I keep reading things telling me shorter rod = bigger spike on a MK4. Owain keeps telling me getting the turbo to respond quicker will limit the boost spikes, but I don't understand... Like that should have happened if I was a small bit off and should have happened when we shortened the first couple turns, not taken like 6 turns right? Every time I have raised a flag, or asked about the adjustment, if I should check anything else, he has told me to go shorter on the rod, or not given me anything better than a confusing one sentence answer, but shortening that rod has only had the spike get worse and has not made the dip go away. What am I missing? I feel like I am just not understanding something or comunication has broken down, so am desperately searching for answers. We have been corresponding about this about once a weekish, or when he gets back to me, for a month now. I feel like shortening the rod is not the answer here, will only cause more problems, maybe cause more/worse damage, and there has to be something else wrong, but I don't know what else to try.

I reset the rod to having the actuator start to move at 19ish, graphed looked like first one in this post again. Then to make sure we weren't having bad communication on directions to move the rod to begin with I went longer instead to see what happened.

Longer rod

From my understanding of how the car works and the turbo control system, the above graphs would lead me to believe I have a boost leak. I always fall to about 2400mbar regardless of rod position. I have pressure tested from the turbo inlet to the EGR a few times now and can NOT find a leak at all though. Vacuum all seems good and I have no leaks there as well. N75 has been swapped for a new unit, behavior remains the same. Arm on the turbo seems to move freely when I removed it from the actuator arm. Turbo actuator holds vacuum and seems fine. I need help or ideas and am at my wits end trying to figure this out. I am terrified that if I shorten the rod again I am going to blow the turbo. The only things I can think of to try next would be: to clean the turbo(sticking vanes?), replace exhaust gaskets and nuts between the engine and the turbo (on the off chance I have an exhaust leak there?) and/or to pull the intake and replace all my gaskets for it as well (leaking boost from there maybe?). If anyone has other suggestions or some insight on what could be causing this I would love some help and input. Or even confirmation the above cleaning turbo and replacing gaskets would be my next best troubleshooting steps. I am at a lost for understanding this.

Any ideas? Again would love some help, please and thanks!

 

[pyromancer]

Turns out this was some fueling issue and was partially corrected with tune adjustment after a lot of back and forth. I was told it was likely due to 10mm pump with 1019 injectors. I did end up completely replacing my intercooler and swapping in a 1.8T pancake pipe before we determined this. Although I was fairly confident I didn't have a boost leak the behavior was similar. Car will still not reach the 2655mbar requested boost but will hit 2600.

 

[burpod]

was anything done with your hardware to correct this "fueling issue"?

 

[03TDICommuter]

Per your photo, you are lengthening the actuator, not shortening it. Where your yellow mark is though, you don't have a lot of room to shorten it from the original spot.
edit: Nevermind - I read part of your email, then looked at the photo not realizing you later lengthened it. Ignore me.

 

[pyromancer]

was anything done with your hardware to correct this "fueling issue"?

No, no hardware adjustments. I tried adjusting my IQ lower via hammer mod, but it didn't realy have any affect on the boost, other than how much I was spiking. Something was adjusted with a tune update, prior to that I was having trouble getting much higher than 2400mbar. Here are two graphs after that update with different IQ

IQ 6

Log #534610

log.tunezilla.com

IQ 5

Log #538209

log.tunezilla.com

 

[Nevada_TDI]

Although it seems counter intuitive, if the rod is too long it will cause spikes; the ECU is requesting boost, and vacuum stays on the actuator until it overshoots, and then drops the vacuum/boost.
Most tuners want the actuator set to where the boost is not over 100 millbar over the requested boost.
IIRC, 100 mb of "excess" boost won't usually trigger the n75% until 3 seconds have passed.
And conversely if the rod is too short, depending on your tuner, it will spike because it builds boost too quickly. And then sometimes you chase your tail for a while. I had kind of the same problems you are having, and the problem was the camshaft somehow got too advanced, so the valves were closing right when boost was being requested.

 

[pyromancer]

I had kind of the same problems you are having, and the problem was the camshaft somehow got too advanced, so the valves were closing right when boost was being requested.

How did you figure that out and how did you end up fixing it?

Rod length obviously wasn't my issue as it wasn't affecting anything other than how much I was spiking or under boosting. I would just always level out at around 2400mbar. As is it sits my IQ is set at 5 and I will probably leave it there and lengthen the rod slightly. I think the car starts a bit better with the lower IQ.

 

[burpod]

maybe someone can point out to where i'm wrong... but the actuator length seems to be all too often and unfairly blamed on issues, as is the "iq value", two very misunderstood things. i myself misunderstood them for a long time. but as long as the actuator is set that it hits both the set screw stop (should be around 20-24hg) and the full-open internal stop (should be around 6hg max), the exact length should not make any difference at all. on on a stock alh, the max amount the vanes can be closed (max possible boost given rpm and fuel) is 25%. this number is set in the ecu "tune" files. for alh the n75 duty cycle numbers correspond to the % the n75 is "open". so at idle, stock ecu vnt15 it's at 25% open. or 75% closed. if the actuator is hitting the stop at idle, and if is hitting the internal full-open stop with vaccum less than 5hg (car off), your actuator is fine. if boost is off, you either have a fueling issue or some other problem like sticking vanes, boost leaks, vacuum leaks etc. if you watch the n75 values on an 011 log you can deduce many of these problems.... putting a vac gauge on a tee to the vnt actuator on the turbo would also give a good picture.

 

[burpod]

i'd still be curious to see some good 10-20 minute log of 011 (commute to work or wherever you're going, just hit log and go) of how it is running now. would be much more helpful, but based on those single pulls, there could be several things wrong - there is a good bit of lag there building up boost for a little vnt17, even though the initial rpms are very low:
-you may possibly have a low fueling issue for sure - either in the tune or hardware, impossible to say based on this log only
-boost leak - would have to be a fairly decent sized one...
-vacuum leak
-n75 map values not right for your fueling

it's very hard to make good guesses as to boost problems with only single-pull graphs....

 

[Nevada_TDI]

How did you figure that out and how did you end up fixing it?

Rod length obviously wasn't my issue as it wasn't affecting anything other than how much I was spiking or under boosting. I would just always level out at around 2400mbar. As is it sits my IQ is set at 5 and I will probably leave it there and lengthen the rod slightly. I think the car starts a bit better with the lower IQ.

Prior to me doing my own timing belt, I will guess whoever did the timing belt used the "mark and pray" method. How I found out, I was using the camshaft locking tool at the back of the camshaft, the slot was a bit too advanced to go in directly.

 

[pyromancer]

i'd still be curious to see some good 10-20 minute log of 011 (commute to work or wherever you're going, just hit log and go) of how it is running now. would be much more helpful, but based on those single pulls, there could be several things wrong - there is a good bit of lag there building up boost for a little vnt17, even though the initial rpms are very low:
-you may possibly have a low fueling issue for sure - either in the tune or hardware, impossible to say based on this log only
-boost leak - would have to be a fairly decent sized one...
-vacuum leak
-n75 map values not right for your fueling

it's very hard to make good guesses as to boost problems with only single-pull graphs....

After 5 months of trouble shooting, including boost leak testing and vacuum leak testing multiple times, seriously so many damn times. Which included isolating multiple parts of the system and looking for leak down, I can assure you a boost leak or a vacuum leak were not my issue. I also chemically cleaned the turbo at one point since I was trying anything I could come up with that maybe an issue. Hell I even replaced all my intake manifold and exhaust manifold gaskets. Brand new all aluminum IC, swapped in a 1.8T pancake pipe. I at one point determined with a leak down test that the turbo itself would slowly leak down at 25-28 PSI by bumbling air up the oil feed line. Which I think would be a non-issue when you consider oil pressure from engine running would likely prevent that, but that was a fun couple days trying to decide if my turbo was shot. I was sending Malone anything they asked for or trying what ever they asked me to try mechanically, including the new IC... Since I posted in November and no one replied, which was after trying to figure out issues on my own for two months first, I had kind of forgot about this post until you DMed me asking about the issue and if I could send you logs TBH. I am here updating this now in the hopes it helps someone else.

All I can say is the conclusion was Malone/Tunezilla sent an adjustment to the tune with this note "The low boost at mid-high RPM could maybe be because of relatively small fuel hardware (10mm + 1019 nozzles) and a tune update could help the turbo hold boost better. Part of me does wonder if there's an issue with your turbo." that was right before Christmas, all I know is they did something to the tune, I would guess to adjust fueling and it improved things. After which I sent them the linked graphs with IQ at 6 and 5 above (I wanted to experiment with the IQ to see if it helped further, maybe because I don't understand it correctly) and they told me "boost looks fairly reasonable" and I was good to ship the rental tool/computer back to them, I dropped it in the mail Tuesday of this week actually. At this point honestly I am pretty over pulling my hair out on trying to figure out what was going on. 5 months of all my free time trying to read up on anything mechanically that could be causing my issue, trying everything I could think up, 59 back and forth messages with Malone/Tunezilla... I'm pretty sure the only things I could try on my end that I haven't would be replacing the turbo and/or upgrading to an 11mm pump. If you got extras sitting around you want to mail to Seattle, WA, I'm happy to give it a go haha. I'm not sure I have an issue with the turbo and if I upgrade the pump I will do it with the timing belt when it needs a new one. Also if your in the area your welcome to look at the car. Otherwise this is all I can tell you. I don't own a tune loader, was told the tune was good and to ship the rental loader back, so really not much else I can do now anyway. I would think if there was something else in the tune they would have caught it, they are a respected vendor.

Prior to me doing my own timing belt, I will guess whoever did the timing belt used the "mark and pray" method. How I found out, I was using the camshaft locking tool at the back of the camshaft, the slot was a bit too advanced to go in directly.

I'm about 30K I think from needing a timing belt, so maybe I will find some stretch or something similar when I get to doing one. I haven't done my own yet. Last one was done by a guru in Portland and I would think he did everything correctly. I seem to remember advancing the injector timing when I installed the injectors I could check that again in VCDs. I have been a bit intimidated thinking about doing the TB and looking at buying the tools TBA. I'm about to get into replacing all the motor mounts as I think they are getting tired and the thought of supporting the engine by the oil pan kind of freaks me out too, car has been really good to me and I just don't want to screw things up.

 

[burpod]

I'll have to read your post more carefully again. I wasn't meaning to imply that I thought you *did* have hardware issues or hadn't been thorough, as it seems like you were competent and covering all your bases. My remarks were more just general, as if I had just seen that blip of a log and were to comment on it without knowing full history

 

[pyromancer]

I'll have to read your post more carefully again. I wasn't meaning to imply that I thought you *did* have hardware issues or hadn't been thorough, as it seems like you were competent and covering all your bases. My remarks were more just general, as if I had just seen that blip of a log and were to comment on it without knowing full history

It's just frustration on my part. Hell I was about to rip my hair out because my graphs looked like boost leak or exhaust leak to me. I have more graphs but all are 3rd gear pulls because that's what I was asked to log. I have lots of logs for various rod lengths, chasing boost leaks that really weren't issues, or leaks that happened because of over boosting due to shortening rod, attempts at swapping N75, 5 months... I literally mailed the flashzilla tuner back this week. If there is still anything wrong that isn't mechanical what could I even do? Not that I think that is the case either, but ruffly 6 months and tons of troubleshooting mostly on my own was very frustrating...

 

[burpod]

Definitely very frustrating, I feel your pain I do know it can be very difficult to know when it's hardware or software sometimes.. or a combination.. triple checking everything, swapping things, redoing things, did I not check something stupid and obvious? Etc. Or is it purely tuning issues? Ugh!