2002 Golf - I'm back in a TDI!

[spifflifkin]

Happy you got it sorted! I am honestly still amazed that it did run and drive ok without this. Just an other testiment to how great these cars are.

Yes! These are great cars!
It's been pretty cool working on this. In 2003 I had a Jetta TDI, 5-speed I bought new. I had a lot of problems with it. At the time I took it to the dealer to fix since it was under warranty. Now that I've had one mostly torn apart I realize that I probably should have just fixed the Jetta myself. The only thing that I may not have bene able to fix is no electronics worked randomly. You'd start the car and nothing electric worked (windows, gauges, lights, etc). It was weird. I had to stop and restart the car until the electronics started working.
I had it for 5 years.

 

[spifflifkin]

taken a log now that it has been fixed? 011-001 or 011-001-004 ?

I took an 004 log last night. I did it in 3rd gear 1500-3800-ish until it went in to limp mode with a turbo over boost error. I'll do some logs after I put in the new turbo.

 

[burpod]

whats wrong with the current turbo?

 

[spifflifkin]

I get an over boost error. Matt at Cascade German helped me troubleshoot it. I'm getting impatient to get this thing on the road so I just bought a new one from them.

 

[burpod]

hmmm. doubt its actually the turbo, 2 min with mityvac would determine if it had any major vane issues. usually a full cleaning would solve it (sticky vanes).

 

[spifflifkin]

Check out post #170 on page 6. Is that info helpful?
I also checked the turbo actuator. It contacted the stop at 18-19 Hg.

Edit: Added info

 

[burpod]

Check out post #170 on page 6. Is that info helpful?
I also checked the turbo actuator. It contacted the stop at 18-19 Hg.

Edit: Added info

i don't see what the problem is? did you watch and check wiith mityvac if the vane lever moves freely through the whole motion? if it does, the turbo is fine. or worst case, might still be a little sticky with soot which can be cleaned. sometimes the control ring can get worn and lead to sloppy boost control, but pretty unlikely that would be the cause of such large boost issues

 

[spifflifkin]

Here's a video I took testing the turbo actuator. I misstated in the video that it contacted the stop at 17 Hg. It's really between 18-19 Hg.

 

[P2B]

Hard to be sure, but it looks to me like the vane lever may not be fully returning to the no boost position when vacuum is released. Total movement should be 3/4".

This can be caused by incorrect actuator adjustment, rust inside the actuator, or sticky vanes. You pretty much need to disconnect the actuator from the vane lever to work out which is the issue.

 

[spifflifkin]

I got the intake manifold out after lunch. After work I'll take out the turbo and have a look.

 

[J_dude]

I got the intake manifold out after lunch. After work I'll take out the turbo and have a look.

Just disconnect the actuator to diagnose, no sense taking the turbo off if you don't have to

 

[agent_jwa]

using a 14mm open end wrench pull up on the little arm that controls the vanes, i bet it will make a squeek/crunch and move up further. after that it will probably stick, so work it up and down and hopefully it will clear the carbon build up away and allow the vanes to move their full travel.

 

[spifflifkin]

I put on a new turbo today. I talked to Matt at Cascade and based on what I'm looking for he recommended sticking with a VNT-15.
There are no more over boost errors.
Now all that's left is to put on a few remaining body parts.
Here are some log files after the new turbo install.
https://drive.google.com/drive/folders/1Q9iZRycDidw8_LYbgJ0S4ZaVrGHt8m7b?usp=sharing

 

[burpod]

boost is definitely "undercontrol", although from what i can see, IMO, it's rather under-fueled for the boost. (also over-advanced). you can see how the n75 values stay low and plateau early, like its needing to "force the boost". other things can contribute/cause this of course, like a boost leak, or rod length being longer.

did you take apart the old turbo to see what was going on inside?

 

[spifflifkin]

I haven't taken the turbo apart yet. I can do that next week.
Could you point me towards somewhere I can learn how to understand the logs? I've read about the individual components (N75, G149 etc.) and understand their functions. I'm lost on the injection quantity and the timing and their relations to each other and the other logged values.

In your opinion, is it good to start commuting in?

Thanks!

 

[burpod]

the best i can do now is my "enthusiasts guide" thread

 

[spifflifkin]

Cool. I'll look for that. I just started reading your malone tune thread.

 

[spifflifkin]

the best i can do now is my "enthusiasts guide" thread

That's a great write up! I'm going to have to go over it a couple times to get a better understanding of some of the items. Thanks for writing that!

 

[Matt-98AHU]

hmmm. doubt its actually the turbo, 2 min with mityvac would determine if it had any major vane issues. usually a full cleaning would solve it (sticky vanes).

I walked him through all these checks via email. Vanes moved freely, actuator did need adjusting, but it still didn't fix it. We easily verified the control side was doing what it was supposed to with the vacuum gauge and running basic setting 11.

It's actually quite common for high mile VNT turbos to have a worn vane mechanism that causes these to overboost. Where the lever and the vane ring interface, they can develop a massive amount of slop there. So, I automatically go to "just buy a new turbo" when all the other possibilities that can cause overboost have already been verified as not being causal. No sticky vanes, Actuator adjusted right and the vacuum supply on and off from the N75 verified as working as it should, then by process of elimination, a worn vane mechanism that you cannot see and test for is often the only remaining possibility.

And, as evidence, his boost logs post replacing the turbo look great. Actual and requested track very well.

This is also why I don't trust when someone automatically likes to blame a tune as being causal of boost control oddities. It's more often than not a mechanical problem that no amount of tweaking the tune is going to ever fully resolve.

 

[burpod]

I walked him through all these checks via email. Vanes moved freely, actuator did need adjusting, but it still didn't fix it. We easily verified the control side was doing what it was supposed to with the vacuum gauge and running basic setting 11.

It's actually quite common for high mile VNT turbos to have a worn vane mechanism that causes these to overboost. Where the lever and the vane ring interface, they can develop a massive amount of slop there. So, I automatically go to "just buy a new turbo" when all the other possibilities that can cause overboost have already been verified as not being causal. No sticky vanes, Actuator adjusted right and the vacuum supply on and off from the N75 verified as working as it should, then by process of elimination, a worn vane mechanism that you cannot see and test for is often the only remaining possibility.

And, as evidence, his boost logs post replacing the turbo look great. Actual and requested track very well.

This is also why I don't trust when someone automatically likes to blame a tune as being causal of boost control oddities. It's more often than not a mechanical problem that no amount of tweaking the tune is going to ever fully resolve.

i guess we shall see, and i did mention that they can wear and lead to sloppy boost control, but i would be very surprised to see a control ring worn to such a degree it would lead to anything more than a little slop - i know i've seen a pic of one, but i would have to bet wear that bad has to be pretty rare. now i don't take turbos apart day after day, but i've done enough, and seen enough from others, to know that most all of the time there is no wear. my wifes 220k mi vnt17 i took out ~last year showed ZERO signs of wear. been on the car for a long time...so why one could wear so bad? bad metallurgy? combined with a car in a state of tune/hardware issues + the type of driving, such that the control ring is bouncing around a ton - i suppose.

most likely the differences seen are from something else. new actuator on the new turbo perhaps has a stiffer spring (old one maybe degraded?) - which would effectively "lengthen" the rod. more force per the HG to get movement required. the set screws could easily be off and that can throw off rod length a good bit too. so unless you're measuring that with good accuracy, and a feeler gauge for "rod length" and being very careful, combined with the accuracy of the cheap vac pumps, very easy to have rod length be "off" a good bit. perhaps some small exhaust/boost leaks in the mix that weren't noticed and fixed (or present) for the new turbo install

also 2 pulls in a log really isn't enough to see all that's going on. full driving log with a variety of pulls is best. there's a lot of variables to get good data based on 2 pulls. intake air temps can make a huge difference too. not only because of the physics of air density/etc, but the tune is going to be advancing timing 1.5-2.0deg at 10C intake air temp, vs no advance for 20C air temp - not much of a range, easy to get data from 10C and 20C and see very different results. timing will have a big affect on stuff.....

 

[burpod]

there are other factors too... how hot the oil was during the logging - now i don't know exactly how much the viscosity of the oil would contribute, but it stands to reason that if oil temp during one logging session hadn't been up to temp (car just warmed up to hot coolant then logged, but oil temp still at 150F - compare to full hot maybe oil @ 215F) the turbo isn't going to spin as freely - making less boost, similiar to lengthening the rod.

an old turbo could conceivable be a bit "looser" and spin a bit more freely too. not too long ago i saw someone post with an opposite problem - the shaft was quite tight, and having underboost problems and abnormally low n75 values as if the rod length were very long.

i'd be willing to hedge my bets upon turbo inspection that there is actually nothing wrong with it, but the differences seen could simply be some combination of effective rod length and environmental conditions. one can see how many variables there are at play which can compound. so it's simplistic and almost a little deceiving to draw such a conclusion that the turbo is just "bad" by putting in a new one. too much that can be going on.

but full driving log is the best - it contains a huge amount of important data about the boost - *especially* if you have the tune in front of you. you can see how much boost it makes per the n75 during coast down (say 3rd gear coasting down from 4500rpm) - this is almost like a baseline of how much boost it can make per the rpm/n75 without any fuel! very useful to know!. how boost/n75 is for cruising areas or steady mid-throttle - you can compare all these numbers (within a particular car - across cars its more complicated -- lots of variables) to the higher IQ amounts and that will tell you a whole lot more about what is going on....

 

[Matt-98AHU]

i guess we shall see, and i did mention that they can wear and lead to sloppy boost control, but i would be very surprised to see a control ring worn to such a degree it would lead to anything more than a little slop - i know i've seen a pic of one, but i would have to bet wear that bad has to be pretty rare. now i don't take turbos apart day after day, but i've done enough, and seen enough from others, to know that most all of the time there is no wear. my wifes 220k mi vnt17 i took out ~last year showed ZERO signs of wear. been on the car for a long time...so why one could wear so bad? bad metallurgy? combined with a car in a state of tune/hardware issues + the type of driving, such that the control ring is bouncing around a ton - i suppose.

I've taken apart enough myself to know it's a lot more common than you think it is. Biggest factor is type of driving. Cars that do a lot of stop and go city driving, they wear faster because the vanes are being moved through their full range of motion more frequently and are also likely to experience more soot build up that ends up meaning the vanes require more force to move than one that is under a more consistent higher load on the highway.

When I was in the Bay Area, I saw numerous cars with only 100k miles that had a worn out vane mechanism, because they were largely stop and go cars. This is also why whenever someone would ask me what kind of car they should get, I first ask them another question in response: what is the most common type of driving you do?

If the answer came back as a lot of around town and stop and go, high traffic driving on local freeways that come to a stop often, I tell them to buy a hybrid or EV instead. Diesels just end up being more problematic and expensive in that environment, especially ones equipped with a VNT turbo. But, if you do a ton of long distance highway driving, there is no better tool than a diesel, and the diesel will be a lot happier in that environment, give you far fewer issues than they do in frequent stop and go driving. This is even more true with modern diesels equipped with DPFs...

And at this point, when you're talking an ALH, it's more than a 20 year old car now. Band-aid fixing an old turbo is rarely going to give you the long-lived fix you hope for. If it's still on its original turbo, that's more than 20 years of wear on that lever and control ring.

 

[spifflifkin]

I really appreciate the input from everyone that's commenting on this thread! I've learned a lot from all of you!
It's awesome to have access to the wealth of knowledge on this site!!!

 

[spifflifkin]

I've drove it about 50 miles last week. It's not done so I'll keep it at work a bit longer. Just getting used to the different noises it makes. It sounds like something in the suspension might be loose. In the front end every suspension part is new except the springs. In the rear it's all new except the cross beam and the springs. Everything was torqued to spec. Every bolt or nut got torque seal immediately after torqueing. No TTY bolts were reused. Maybe they're just normal noises. I'll crawl under there and look things over again next week.

It runs well and is quick! I know quick is relative. I've been driving a 1-ton Duramax for the last 8 years so most everything I drive feels quick to me! My other vehicle is a 1959 Single cab VW. That's even slower than the Chevy.

The new valve cover from ID Parts is leaking on the back side of the engine. I talked to them about it. Josh asked me to check to see if the seal was folded. I pulled it off and it was fine. It sucks that the valve cover is a combo cover/seal. Not sure if I should exchange it or use RTV...

There is a bubble in the clear fuel line that won't go away. It's about 3/8" long. I read that may be caused by a leak in one of the fuel lines. The filter to firewall fuel lines are a bit rough so I ordered new ones from Cascade German this week. The filter to IP like are new ones I got from Cascade.

I did a compression test with the engine warm (90°C)
#1 420 psi
#2 420 psi
#3 422 psi
#4 425 psi
#1 420 psi

That's pretty close to what I got with the engine cold.

So far I'm loving it! I love the tractor sound!

I took a composite log for 15 minutes from startup to just before doing the compression check. 001-004-011
https://drive.google.com/drive/folders/1CDS9OovM96Pjj6mFA3bK_eLlg56d06pQ?usp=drive_link

 

[P2B]

There is a bubble in the clear fuel line that won't go away. It's about 3/8" long.

That's normal, as is a stream of champagne bubbles at idle. It's how you know there's actually fuel in the clear line.

 

[spifflifkin]

That's normal, as is a stream of champagne bubbles at idle. It's how you know there's actually fuel in the clear line.

Awesome, that's a relief! Thanks for letting me know that.

 

[burpod]

That's normal, as is a stream of champagne bubbles at idle. It's how you know there's actually fuel in the clear line.

i must disagree with the 100% - it's completely NOT normal to have any air bubbles other than a sporadic small bubble or maybe a few more if you've run the tank down low. a stream of bubbles indicates you have a large leak in the fuel system - but one you can't really see with the "naked eye". the injection pump is capable of a HUGE amount of suction - enough so that once many years ago, when my tank vent somehow clogged up on my mk1 with a metal fuel tank, it actually caved in the tank to the point it was loose and rattling on the straps!!!

so if you have any major restriction somewhere (clogging fuel sender or maybe a bad checkvalve), any weak point in the system could turn into an air leak. the most usual culprit is the stupid useless thermo-tee/-orings at the fuel filter. it's also possible the corrugated lines in the fuel sender could be cracked or one of the checkvalves back there.

one of the best things you can do is put in a BEW PD in-tank lift pump, which can be done very cleanly wiring-wise. you can add a relay into the relay panel, a fuse into the fusebox, and make a little drop-in, no splicing/no solder harness - very simple, it's only a few wires. this will turn any leaks that cause air to get into the system to actually leak fuel out that you can see, and then fix. and you'll never have air in the system.

air in the supply line can most definitely cause problems, but often it's not enough of a problem that one would notice in daily driving or by "butt dyno". sometimes it can be very obvious - like stuttering sputtering at low rpms light throttle. it probably can mess with the case pressure in the pump and likely cause little issues with timing.

you most definitely want to fix this. no TDI i own or have worked on has *any* bubbles in the supply line save but a random small one now and again.

 

[spifflifkin]

it's completely NOT normal to have any air bubbles

Thanks for the info! I was little suspicious about this. I put a lift pump in my Duramax a few years ago too as a preventative measure. Eventually I will swap out that CP4 with a CP3...

The BEW lift pump sounds like a good idea.

Thanks!

 

[spifflifkin]

Here's that darn bubble in the fuel line.

I drove the car to work today. It's a little chilly out, but not freezing. It took a few more cranks to start today and smoked quite a bit.

The idle is still a little rough for a minute. The engine has slight surges.

You can see the headlights and the dash lights get brighter and a little dimmer during the surge.

Once things smooth out the lights don't change brightness with the engine speed.

 

[P2B]

That is odd. Pretty much every ALH I've seen has that bubble after being parked for a while, but I expect it to disappear as soon as the engine is started.

However, I doubt it's causing rough idle and surging, I suspect something is going on with the IP.