110 TDI Sticking Vane Problem Cured By Car Wash?

[Ganda]

Recently my 1998 Octavia 110 TDI has exhibited all of the classic signs of the variable vane turbo problems, a loss of power on occasions and a general lack of power through the rev range at other times. Listening to the turbo, it's sometimes slow to get up to speed (it's hard to explain the noise it makes, but it doesn't sound right), and I was psyching myself up to get the turbo cleaned.

However, I put the car through a car wash on Monday night, and the sticking problem was cured! The car went back to driving beautifully, with power all the way through the rev range, and that has lasted until tonight. When I set off from work earlier, it was back to how it was before, but did get a little better just before I got home. This has me a bit stumped.

I guess it's suggesting that the problem I've got is external dirt, rather than internal, which would make sense as I don't often keep the car at low revs, I try to change gear at around 3,000 rpm as I've read that helps with this problem. I'm going to give the turbo a good external clean over the weekend to see if that cures it, but does anyone have any other ideas as to what is causing the problem?

TIA.

[rwbaldwin]

Did you turn the ignition off at the carwash? That resets the limp mode.

Most early TDIs have an undertray, so there is no way that water could get anywhere near the turbo.

It's also very unlikely that external dirt caused the problem. The vacuum capsule pulls a lever which rotates a shaft that moves a butterfly on the inside. If the inside is moving freely, it would be very difficult to stop the external part of the actuator from moving. The cause is likely to be sticky vanes/actuator ring on the inside.

[Ganda]

Thanks, but I left the engine running in the car wash, so it isn't the limp mode being reset. The car wash has definitely done something as it's behaved itself for nearly three days in a row, which it hasn't done for some time now. Mine appears to be missing it's undertray, looking down through the engine bay I can see the ground, so it could be that water has got onto the turbo and cleaned it a bit.

I'll give the turbo a thorough external clean over the weekend to see if it makes a difference and will report back.

[Liverpool-Lad]

Water has got into a sealed turbo system? Interesting. Sounds like pure coincidence!

[Lew247]

water mist got in the air intake and blew through the turbo cleaning it?

The hot turbo would instantly vaporise the water into steam so it would be the same as steam cleaning it I guess?

[mbames]

water mist got in the air intake and blew through the turbo cleaning it?

The hot turbo would instantly vaporise the water into steam so it would be the same as steam cleaning it I guess?

That would seem unlikely unless there was a leak in the air intake or if there was no filter in the airbox.... I would simply class this as a strange coincidence. I am sure the limp mode will be back with soon :'(

[rwbaldwin]

water mist got in the air intake and blew through the turbo cleaning it?

The hot turbo would instantly vaporise the water into steam so it would be the same as steam cleaning it I guess?

So you just need to drive in the rain to solve the sticky turbo problem. If only it were that easy

[trundlenut]

I think this is just a coincidence. When I had the sticky vane problem the car would sometimes be fine for a week and then drop into limp mode several times in one day. I doubt any mist getting through the intake would make a difference as the sticky bit is on the exhaust side of the turbo not the intake side.

It would be worth getting the car scanned for error codes, clearing any that are there and see if they come back.

Edited 17 February, 2012 by trundlenut

[Ganda]

Sorry, duplicate.

Edited 17 February, 2012 by Ganda

[Ganda]

Thanks everyone. I guess it could just be coincidence, but it was noticeably better after the car wash - it looks like I'm not the only one to experience this:

http://www.vwaudiforum.co.uk/forum/showthread.php?97036-Audi-A4-1.9-TDI-%281999%29-power-loss...limp-mode&p=614624#post614624

As an experiment, tomorrow morning I'm going to chuck some water over the turbo to see if it happens again (are there any technical reasons why I shouldn't do that? I can't think of any). I'll report back with the conclusions. Sorry, I don't really believe in coincidence

[Lew247]

People on a veg forum actively use water mist in the air intake to help the engine out, and although they inject the water after the turbo, a couple have done a very fine mist before the turbo and swear it helps a lot

A new thing they are trying now and it seems to be successful is to use steam rather than cold water.

[trundlenut]

Injecting water after the turbo I think helps to reduce temperatures at the inlet and hence help with the power and prevent heat soak. You can also spray water on the intercooler to help with the cooling - the standard SMIC on the octy doesn't have exactly great air flow over it.

You just have to remember that any water vapour going in the intake has to go through the engine before it reaches the exhaust side of the turbo where the sticking is occurring. Also the problem with the vanes could be caused by rust rather than carbon build up.

[Ganda]

The plot thickens! Chucking the water over the turbo doesn't seem to have had an effect, as others suspected it wouldn't, but I have operated the vane system manually and I don't think it's the cause of the problem. When I pushed on the linkage, it moved down smoothly through its full travel, against the spring pressure of the vanes, and returned smoothly as well, which would seem to indicate that it's OK. I did, however, try disconnecting the MAF on a quick test run and it didn't seem to make much, or any difference, to the performance.

I'm guessing that the noticeable difference in performance could be down to water from the car wash getting into the filter, and the vapour improving the performance as noted above. I'm going to try taking the MAF out and giving it a good clean in Isopropyl alcohol to see if that makes a difference. Edit - thinking about it, it would make sense - if the filter got a bit wet in the car wash, it could take a couple of days to dry out completely, which is why the car went brilliantly for a couple of days, before it dried out.

I guess the only other causes of the problem, now that the vane mechanism seems OK, are either the vacuum actuator on the turbo, or the valve that controls it. Are there any simple tests that would indicate that either of these components are at fault?

Edited 21 February, 2012 by Ganda

[trundlenut]

If you do have sticky vanes then you should be getting an overboost code logged. Look in the diagnostics forum and see if there is anyone near you with VCDS who could scan the car for faults.

If the performance doesn't change with the MAF unplugged it would suggest that all is not well with it.

[mbames]

Were you able to operate the actuator on the turbo without disconnecting the vacuum pipe to the n75?  I am pretty sure that I can't do that on mine, so that might suggest a leaking vacuum pipe, a failed n75 valve or a split in the diaphragm on the turbo.

Disconnect the vacuum hose at the n75 end and suck on it - can you get the actuator to move on the turbo then?

However if you say driving the car with the MAF disconnected has no affect on performance I would start my investigations there.

[pauldazzle]

If you disconnect the MAF & it makes no difference then usualy it means the MAF is ok. It's only when you disconnect & the car runs better that the MAF is suspect. The reason for this is that if the ECU doesn't see the MAF it reverts to default values, which is why it'll run better if a dodgy MAF is disconnected.

[Ganda]

Thanks for all of the suggestions so far, everyone. Having borrowed VAG-COM from a mate, I can now rule out the turbo completely, none of the vane related fault codes has shown up when a scan was done. It has thrown up a few codes, one is MAF related:

17552 - MAF Short to ground, intermittent

The next one, which is slightly worrying, is this one:

17656 - Start of injection timing regulation : control deviation.

Looking at the Ross-Tech wiki, it suggests that this may appear after the car is stalled (which it was on the last time it was used - my other half stalled it). But it may also be related to the cambelt, which was done about 4,000 miles ago.

The last one is 17978, which is to do with an intermittent immobiliser problem, which I know about.

I think I might have a dodgy connection somewhere, which was cured by the water from the car wash - does anyone have any suggestions of where to start, please?

[Ganda]

Were you able to operate the actuator on the turbo without disconnecting the vacuum pipe to the n75? I am pretty sure that I can't do that on mine, so that might suggest a leaking vacuum pipe, a failed n75 valve or a split in the diaphragm on the turbo.

Yes, it is quite stiff, but it was moveable. It should move when it's all connected, as all you're doing is compressing any air in the vac system.