P0299 Underboost

Hi,

This issue has made the car nearly useless to me as I commute daily over a ~2K ft grade that causes the limp mode driving in either direction. I really need to figure this out. Any suggestions are appreciated.

The vehicle is a 03 Jetta 5 Speed TDI. Getting P0299 kicking in under lighter and lighter load conditions. It used to be just on uphill in 5th gear going just above 50 mph and trying to maintain speed or accelerate. Now it will go into limp mode just on a moderate slope trying to accelerate even from 60 mph plus.
Here's what I've done to troubleshoot:
New vacuum hoses all around.
Tested MAF against a spare with no change in result.
Swapped N75 with N18, no change. Swapped back.
New fuel filter.
New Air Filter.
Tested vacuum from vacuum pump, > 25in/Hg
Tested vacuum coming from N75 to VNT at 20.5 in/Hg
Pulled vacuum on a line connecting to the VNT and it held steady at 25 in/Hg.

The following were learned with the help of MaTT, one of the local TDI Gurus, who also did my timing belt job.
Using VCDS to cycle the N75, the actuator on the VNT moved to the stop.
He noted a normal amount of play in the turbo. Not sure if that means the vane plate or the actual turbine.
After looking at it for 2-3 hours he was stumped.

So it looks like the vacuum system is working properly, and it also looks like the VNT is operating normally. Yet I am getting limp mode often, occasionally accompanied by a CEL with the P0299 code.

VCDS shows that actual boost curve sits significantly below the expected curve. Alarmingly so according to Matt. With no noises that would normally accompany a intake leak.

Couple more things to add........
The timing belt was done about 1 week before this round of limp mode started showing up. Previous to that I thought I had it beat by doing much of the above. The limp mode has appeared intermittently for the entire 7 years I have owned this vehicle. I think the timing of the timing belt job and these symptoms are coincidental, but could be relevant.

I had previously removed the intake as part of trying to figure out limp mode, back at about 90K miles. There is now 190K on the car. During the 2 plus hour session with Matt, the EGR was removed and the intake has only a slight sheen of oil on the insides, no coking to speak of at all. This leads me to thinking that the turbo and actuator are relatively clean as well.


What's next? Pull the turbo and do a complete cleaning?

Thanks,

John

Actively check the rubber boost hoses for condition. Look for cracks or clamps not in place fully. Cracks can be difficult to spot, only showing up under pressure. If possible, pressure test intake track.
How is smoke level?
Sometimes boost leaks can't be heard.

Thanks for the reply. Smoke is nonexistent. I put a mytivac on the check valve closest to the vacuum pump, in line with the brake booster and wrestled the hose going toward the brake booster, and could get leakage, or drop in pressure if I tweaked it jst the right way. I replaced the crimp clamps with hose clamps. Still got underboost.
Should the N18 allow vacuum to get by it? I tried pulling vacuum from the same spot as before, but this time with clamps on the line going to the EGR solenoid, the line coming off the back side of the N18 to the airbox, before the tee going to the N75, and the line going to 13 on the diagram, which is the EGR. These were clamped with vise-grips. I still got leakage when pulling vacuum. Is the N18 supposed to slowly pass vacuum or could it be bad?

Thanks again,

John

oldpoopie said:

Actively check the rubber boost hoses for condition. Look for cracks or clamps not in place fully. Cracks can be difficult to spot, only showing up under pressure. If possible, pressure test intake track.
How is smoke level?
Sometimes boost leaks can't be heard.

Click to expand...

Stop checking the vac system and really rule out boost leaks. Best way is to pressure test.

Built a boost leak tester:
https://plus.google.com/u/0/photos/...5905732625278690962&oid=103195593054411129757

Sprayed soapy solution everywhere I could reach. There is a minor amount of leak at the detector itself, but the only other place I get leakage is at the oil level check tube.
I don't hear any leak when I put the end of my stethoscope in the oil filler port in the valve cover. If I remove the dip stick I get massive air blowing out when pressurized with my compressor dialed down to about 40 psi.

Could this be causing my underboost or does this mean there's a broken ring or something major like that?

Good times,

John

oldpoopie said:

Stop checking the vac system and really rule out boost leaks. Best way is to pressure test.

Click to expand...

Did u block off the where it goes into the intake? Otherwise you will blow past the valves. Isolate just the plumbing bypass the engine

Thanks for chiming in here.
I disconnected the rubber hose between the EGR and the intercooler pipes, clamped a rubber (nitrile) glove onto the rubber hose end. There is still a path into the engine via the down tube connecting to the EGR, but I clamped a balloon over the dip stick spigot (?) where the dip stick tube connects to the block to close off that pathway. The dip stick tube was loose, now it's gone, to be replaced tomorrow.
Anyway, I isolated the plumbing as well as I could without removing the intake and sprayed dish soap solution at all the places where one pipe connected to another, as well as all over the fins on the intercooler while under pressure with no obvious bubbling.
At this point the VW is winning the battle. Ugh.

John

Antsrcool said:

Did u block off the where it goes into the intake? Otherwise you will blow past the valves. Isolate just the plumbing bypass the engine

Click to expand...

Disconnect the turbo inlet pipe and e rubber elbow to egr. You need to pressurize fromt he turbo to the top hose. It should hold at least 25psi with no leaks.

This is the intercooler pipes leak test right? It will just test the intercooler and related pipes and connecting pieces.
I'll try it tomorrow, have to figure out a way to make a plug for the fittings that are about 2.75" i.d.
One way I thought of is to buy an innertube for a passenger car and gut enough of the rubber around the schrader valve to be able to clamp the tube on, using the valve as an intake, then use some more of the innertube to block off the far end. Use the regulator on my compressor set at 35psi. Suppose I could light a cigar and blow a bunch of smoke into the system before pressurizing to simulate a smoke test.

More later.

John

oldpoopie said:

Disconnect the turbo inlet pipe and e rubber elbow to egr. You need to pressurize fromt he turbo to the top hose. It should hold at least 25psi with no leaks.

Click to expand...

I use pvc fittings

OK, I isolated the boost system. Made a plug out of 1" pvc and a cap, clamped it to the rubber hose removed from the output of the turbo. Made another pvc based jig with a pneumatic adapter and clamped that to the upper rubber hose disconnected from the EGR. It holds 35psi and I get no sign of bubbles anywhere when I spray dish soap solution at all the joints and on bot sides of the intercooler.

So it looks like the issue is not a leak in the boost system.

looks like you ruled out the piping, I'm guessing you already read the stuff in this link

And you are 100% certain that the actuator is adjusted properly? It should begin to move at about 5 or 6 on the mitivac

It starts to move at 4" and hits the stop by about 17". I'm 100% certain of that. I'm not 100% certain that's the proper spec.
This situation has gradually gotten worse, and I noticed an acceleration in symptoms about a week after a Guru done timing belt job. I'm assuming that's a coincidence.

Check to make sure there aren't any rags or obstructions inside the inter cooler. I've seen rags stuffed in there with the intention of keeping dirt out, left in accidentally and causing issues.

I hadn't seen that page before. It's not much help though.
It looks like the boost being produced by the turbo is low. So now the question, what controls the amount of boost being created? The N75 uses vacuum to regulate the position of vanes on the turbo via the VNT based on input from the ECU.
In my case, the N75 pulls, or allows through, about 18"Hg at idle. If I slowly raise the rpms, the vacuum smoothly drops to about 10". If I stab it at all, especially above about 2400, vacuum goes right back up to about 18".
The VNT actuator appears to move freely, meaning the VNT itself is working and the vanes on the turbo move.
With a vacuum tester, the VNT actuator starts to move at around 4" Hg and stops out at around 18". So that system seems to be working.
So to me it looks like the turbo is just not spinning fast enough, or ...?

Something slowing the throughput of air through the engine.

BobnOH said:

looks like you ruled out the piping, I'm guessing you already read the stuff in this link

Click to expand...

oldpoopie said:

Check to make sure there aren't any rags or obstructions inside the inter cooler. I've seen rags stuffed in there with the intention of keeping dirt out, left in accidentally and causing issues.

Click to expand...

That's funny.
Do you pull the intercooler out through the bottom? One bracket bolt and an intake and outlet clamp?

Just pulled off the top and bottom hoses from the intercooler and looked in using a mirror and flashlight. Also stuck a piece of wire in and twirled it around. No rags. Verified that the snowscreen has been removed as well.

Get vagcom.

OK I now have a VCDS. Here is a link to one of the group 11 runs I did today.

This was in 3rd gear on a level stretch of road. Limp mode kicked in before reaching 4K rpm.
https://plus.google.com/u/0/photos/103195593054411129757/albums/5913373204685721745?hl=en

Please comment.

clogged cat?

Bingo!!!!