Overboost but low power. NOT from limp mode

Bizarre symptom. Boost gauge shows 22psi+, yet power is very low.

Note, after repeated "overboost exposure" it does throw code and go to limp mode cutting boost entirely. However, this low power issue occurs while overboosting. Every time I've ever previously experienced overboost, the car hauled ASS (obviously).

I know the turbo is shot and needs replacing, but if low power on high boost, would this also point to something else?

My first guess is fuel side. Injectors? They likely haven't been replaced since factory, but VCDS showed IQ quite good. Timing is retarded I need to fix.

Other potentially related/unrelated symptoms: Oil burn (smell), massive startup smoke and hard starting, intermittent oil smelling smoke on WOT. Thoughts?

How does the intake look? If it's clogged the high pressure and no power makes sense.

Thanks Eddie. I just cleaned intake in May so it should be good. Cleaned it 1yr before that and it had minimal buildup (after the 1yr. Prior it was filthy, prob never cleaned). I also cleaned turbo vanes in May.

Now that you mention it though, I always noticed that the top intake hose (from intercooler to EGR) was very oily. More as of late, seems like cause for concern. I could never figure out why, or where the oil was coming from.

Both times I rebuilt the turbo, the compressor side was clean as a whistle (no oil), so the oil must be entering intake hoses somewhere between the turbo and the EGR/intake.

I've drilled the intercooler and minimal oil came out there.

I think I read something about the CCV "Crank Case Vent", but could never find where it was or a good explanation of what it does. Does it enter intake line between these 2 points? Or do you guys know anything about it?

Still unsure if this would cause the power issue. I guess vaporized oil would increase manifold pressure, but again wouldn't it increase power during that symptom as well? (IE, like in diesel runaway). Hmmm

The hockey puck looking thing on the valve cover. PCV is necessary, however, can be filtered. Not so much of a worry really. Do you have an aftermarket boost gauge to compare to the vehicle? EGR sticking open, or ASV sticking closed?
ASV can stop the pressurized air from reaching the cylinders.
Sent from my Armor_2 using Tapatalk

Oooo Eddie you might be on to something with EGR/ASV. When I pulled it out, it was pretty crusty and I didn't give it as much care and attention as I should have. I'm going to have a look at it.

I don't have aftermarket boost gauge, just getting boost readings from car's existing boost sensor via OBDII scanner.

Where is the factory boost sensor on these things so I know what's upstream/downstream that could be causing issues?

Thanks so much for your help. Very very astute.

The MAP sensor is below the windshield washer reservoir in the intake piping.

It's on top of the intercooler. Easier to pull the bumper cover for access.

If you have a spare, I would switch it out. MAP can cause weird issues, however, they are uncommon. I think sometimes we miss the easy stuff.

AceTdi said:

(snip)I don't have aftermarket boost gauge, just getting boost readings from car's existing boost sensor via OBDII scanner.(snip)

Click to expand...

The ECM displays boost in absolute pressure. So 22 - 14.7 = 7.3 psi.
Limp mode territory maybe?

eddieleephd said:

ASV can stop the pressurized air from reaching the cylinders.

Click to expand...

I would start by checking this.

Also the camshaft/lifters could be toasted.

Sorry guys have been swamped. I think you're onto something with the MAP. Been watching the boost gauge via my OBDII and torque app. Sometimes the boost will stick at 22.9psi, thought it was the app causing issues so restarted the app and was still stuck.

Then I noticed something weird.

You know how when not boosting, the boost gauge will show somewhere between 0 and 1 in/hg of vacuum? (Not the vacuum system, the boost system while non boosting).

Well last night it was dropping to like 8-11in/hg. I've never seen that, seems highly bizarre to me...

Would this confirm a faulty MAP sensor? I do know lots of oil has been getting into intake, which could certainly tar up a sensor on that line.

I would assume a MAP incorrectly reading too much boost would tell ECU to reduce turbo actuator movement and thus boost. IE engine only pushing 2psi of boost because ECU thinks it's pushing 23psi. Is this logical thinking? Missing anything?

I always confirm diagnosis before throwing parts at something. Headed to check ASV now too.

(EDIT: Don't have a ratcheting screwdriver to fit in and snag the MAP, R+R delayed)

Are you subtracting the atmospheric pressure from the Torque boost reading? The gauge I use on Torque is the intake manifold pressure and with that reading you have to subtract whatever your current atmospheric pressure is.

When your car looses power scan for codes before you shut it off.
If you get an over boost and limp mode it resets when you shut it off.
When you loose power you could also shut the car for 5 seconds (for the ASV to close & open) then start it to see if the power is back.

Last, with the engine warm, not running but key on check what the boost level is in VCDS.
I'm thinking about the vacuum you mentioned. Bad sensor?

A log from VCDS recording boost, MAF, and n75% might help.

Check the actuator on your turbo, and check the vac lines. If the actuator has never been replaced, put a vac pump on it to make sure it still holds vacuum. After I replaced mine, and got the rod travel adjusted, I was getting a negative pressure code coming up with low power and no boost. it turned out the vacuum line had a split in it where it attached to the actuator. I cut that end off and spliced in a new end, and it's been working just fine since then.

Thanks for the great replies. Haven't dug that MAP off yet, cold up here and busy. But I have more reliably replicated the symptom.

The boost gauge confirmed sticks. So something is wrong with the signal, either at the MAP or I've heard ECU issues? Going to read up on that, unless you have experience?

Replicating the symptom: On initial hard acceleration after startup, car has typical power levels for about 2sec. Boost on Torque app spikes to 22.9psi on hard accel, then after 2sec or so power begins to drop off about 50% (not limp). After about 5sec, limp kicks in and you can feel all boost lost.

However, the boost guage still shows 22.9psi the whole time!

Even if I idle the car down and put it in park, still 22.9psi. Even if I disconnect the app from ECU, then reconnect it, the reading shows 22.9psi even when sitting idling (which would be physically impossible boost levels).

The odd time I can unstuck the boost by driving it. Then instead of the gauge dropping to it's normal 0.4-1in/hg of vacuum under non-boost conditions when working properly, it goes crazy up to 10-25in/hg with erratic fluctuations. Note this is intake pressure, NOT the vacuum system.

So I think what's happening is on hard accel ECU is requesting boost. It gets the boost requested as shown by spiking 22.9psi sensor reading. Then because sensed boost is too high, ECU tells actuator to reduce boost with vanes. Because sensor still shows 22.9psi, ECU tells vanes to reduce even more, reducing actual power levels but not reducing the stuck sensor reading. Finally because sensed boost (not actual) is too high for too long, ECU triggers limp mode to protect against perceived damage.

This is the only plausible diagnosis I can come up with. Would be nice to put aftermarket boost gauge on to verify this, how do people usually do this? Drill IC pipe upstream and tap one in?

Considering how much oil is coming up the intake, it makes sense the sensor was dirty. Will update and document when I get the chance to pull it.

Mr Fiat: I haven't subtracted atmospheric pressure. When the car was working right I would get consistent boost and vacuum levels under varying atmospheric levels. Can you elaborate on what I may be missing?

Wonneber When I shut the car off it eliminates limp mode until hard accel reproduces limp again. Unfortunately I don't have VCDS. Not having it makes diagnosis a huge pain with lots of creative methods. Should get it, but am also leary with commitment lol

lovemybug Some context, I've had this entire turbo system apart multiple times in past 1.5yrs for other symptoms which have been fixed. The actuator was good, rebuilt turbo twice (cleaned vanes) etc, full vacuum line tracing replacing and confirming. I haven't re-checked vane movement with mighty-vac recently, so will add that to my list.

Thanks for all the help everyone, will update when I actually do something with it. Hope this documentation helps someone else in future though.

AceTdi said:

Wonneber When I shut the car off it eliminates limp mode until hard accel reproduces limp again. Unfortunately I don't have VCDS. Not having it makes diagnosis a huge pain with lots of creative methods. Should get it, but am also leary with commitment lol

Click to expand...

There are a few scenarios that can trip limp mode.

If the boost goes over a certain value it trips.

If the requested boost and actual boost are more then 4 pounds (iIrc) apart it will trip it.
This can be from the vanes in the turbo sticking, are worn, a problem in the vacuum control or the actuator itself.
In VCDS you have to log a full power run in 3rd gear from 30mph to red line then upload it to Malone tuning to get a graph which the wizards here can analyze.
https://malonetuning.com/

Log boost actual & requested, MAF, and the N75%

UPDATE: Finally ripped the MAP out and drilled IC pipe to install manual boost gauge.

MAP screws were rusted right in, snapped off, had to fabricate to reinstall. Cleaned MAP sensor.

Symptoms persist.

Now with reliable manual boost gauge I can monitor trustworthy boost readings through acceleration. Seeing high numbers already with spikes up to 30psi, and lots of hanging in mid-20's.

The electronic gauge (via OBDII) does still stick at max boost, even when actual boost is low or 0. So something is clearly wrong there on the MAP reading it, or the ECU analyzing it.

When I reinstalled turbo this spring, I kinked the oil return line, and the exhaust blew white smoke during the test ride. Immediately I fixed the line properly, but oil obviously pushed past a seal. I wonder if it got into the vanes, and when mixed with soot, gummed them up prematurely.

The reason I've discarded the typical overboost diagnosis is because I've had overboost before when I first bought the car. The car took off like an absolute rocket (as one would expect in overboost).

Now when I get the overboost (prior to limp kicking in), power is lower than normal. There's still boost, you can feel it prior to limp triggering. But power is 50-75% of what it should be.

If this was vane sticking overboost my read end should be sucked to the seat, that's what baffles me.

This is why the back of my mind wonders if there's a fuel issue causing the overboost condition to result in less power. Or an exhaust restriction from the oil blowby on last turbo fix.

Nearing my wit's end on this thing.

At $300 for VCDS just to diagnose. Plus potentially a $1k turbo, exhaust, fuel fix, MAP, or combination of the above.... Honda's are looking quite good right about now... Just too many dang mechanical systems on these TDI's to go wrong. Argh.

If you think it's a fueling issue, what is the IQ number? If it's not where it should be, you might be triggering boost issues.

Sounds like you need a new map sensor and possible cleaning a turbo.

Sent from my Armor_2 using Tapatalk

Agree Eddie. I'm going to pull a MAP from the wrecker when I get a chance to swap-nostic it. Seeing as how this would be the 3rd vane cleaning in 18mths--and the effort to pull the thing out--I'd likely just replace turbo with a new one.

Thanks for all your help, will continue to update developments.

Hope you have a heated garage to work in.

Before pulling the turbo off maybe take the vacuum hose going to the turbo off and go for a ride.
See if you still have high boost.

You could take the actuator off to check it and if the vanes are sticking.
It's 2 nuts, 1 clip and the vacuum hose.

Move the arm going into the turbo full range several times to feel if it's free or sticking.
I've read post that if you cycle it for 5 minutes (iIrc) it may break up the carbon
build up.

Shake the actuator to see if rust falls out.
VW bulletin about this.
Test the actuator holds vacuum and moves freely.

When putting the clip back on the actuator tie a piece of thread or dental floss to it so if it flies you can find it easier.

Back to your earlier post about VCDS, there is a lite version that is free to use.
all the functions work for what you need for this.
You would have to buy a OBD cable that works with it.

If you are going to the bone yard for a used MAP sensor get one from a TDI.