Having said a few days ago how reliable my Octavia Mk1 (AQY engine) has been over the past nine years (http://www.briskoda.net/forums/topic/381918-yet-another-from-sheffield/), yesterday I thought I had spoken (written?) too soon. I gave my eldest son his first driving lesson in the car and he had a lot of trouble holding the revs constant to pull away. He carefully held them at around 1000-1200 rpm, and then suddenly they were at 2000 rpm. As that's the sort of thing that beginners do, I initially put it down to him, not the car. However, when I took over to demonstrate, I found that it was doing it for me too.

 

I used one of those cheap OBDII bluetooth devices, my mobile phone, and the Android Torque app, to check for stored error codes, and found that there were none. I then carefully raised the revs from idle to 1000-1200 rpm and held them there. The device reported a throttle position of 4.3-5.5%. After between 5 and 15 seconds, without moving the pedal at all, the revs then jumped to 2000 rpm. The device then reported around 11+% throttle. This happened over and over exactly the same: too regular to be caused by the throttle position sensor (or fuel pressure etc). It had, I decided, to be something to do with the software and / or calibration.

 

Looking on the internet (Golf Mk4, New Beetle etc), I found that a number of people have had idling and light throttle problems with this engine after disconnecting the battery. I disconnected the battery a week ago to remove the airbag (http://www.briskoda.net/forums/topic/381783-steering-lock-puzzle/), so I wondered whether that was what caused the problem,

 

A suggested fix - known as "the chicken dance" - involves switching from off to ignition and waiting between 40 seconds to 4 minutes before starting without pressing the throttle. (It also involves opening the driver's door and the bonnet / hood, but that's just to ensure that the throttle pedal doesn't get pressed). I tried this procedure but it didn't solve the problem. Some people think that the procedure triggers the throttle body adaptation (or alignment or calibration) process, Others insist that it doesn't. What it does to is cause a mosquito-like noise to be emitted from the throttle body for around 20 seconds, followed by a clunk (the mosquito biting?). (Some insist that the noise is from the fuel system rather than the throttle body, but it's not. You can easily prove this by pressing a wooden stick against the throttle body and your ear. The noise clearly comes from the throttle body).

 

Next I disconnected the battery all night. The following morning I reconnected it and then did the chicken dance i.e. ignition on for 4 minutes without starting the engine. This caused the famous mosquito-like noise, but - crucially for the arguments surrounding this - did not cause the full throttle adaptation process to happen. It might have done some sort of calibration near the closed throttle position, but it did not fully open the throttle as I understand happens with the full adaptation process. There was also no noise at around 3 minutes - no thunk etc - as some people describe.

 

After that, the light throttle problem was completely sorted. I checked it repeatedly from cold to normal operating temperature (holding the revs at 1000-1200 for 10 seconds to see whether the revs jumped to 2000), and it didn't happen once. It's also not happened since. (The second driving lesson therefore went much better!).

 

So it seems that although disconnecting the battery for from 15 minutes to all night and doing the chicken dance does not trigger a throttle body adaptation - as many people have already pointed out - it nevertheless solves at least one type of throttle problem - which many of those same people do not accept despite the evidence. For them, there is only one solution to the problem: the throttle body adaptation process (which requires a £270 VCDS etc). However, it seems that there is another solution. It doesn't trigger a full throttle body adaptation but it nevertheless does do something e.g. clears some faulty extra calibration data whilst retaining the calibration data from a previous throttle body adaptation.

 

Disconnecting the battery all night may have fixed the problem without the need for the chicken dance. However, that doesn't explain why the problem occurred in the first place, because I disconnected the battery for more than 15 minutes when I removed the airbag. After that I drove the car without waiting 40 second / 4 minutes etc. The experiment needs to be repeated to determine this. However, I don't want to do this in case the problem happens again and I can't fix it.

 

I hope this helps someone who, like me, does not have VCDS and faced with the cost of new a throttle body and / or buying VCDS, would otherwise scrap their vehicle. I hope it also clears up, or at least helps clear up, the 10+ year debate on whether disconnecting the battery and / or the chicken dance triggers the full throttle body adaptation process.

 

Cheers,

 

JB.

Edited 20 December, 201510 yr by JB2015

Nice work sorting that out.

I think that for your age car, the option exists of doing TBA using VCDS Lite, registered for $99 (£66.50 toady), with a third-party cable which can be had pretty cheaply. This combination can do the job on my 2003 Polo for instance.  I think only newer cars need the full £270 version, not sure on the details.   Obviously you'd want to confirm with other owners of Mk1 Octavias that I'm not misleading you.

 

I guess you don't need it right now, but it's a really nice tool to have in your toolkit.

I think that for your age car, the option exists of doing TBA using VCDS Lite.

 

OK, thanks, I'm looking into it. I read somewhere that the full version is required and so didn't pursue it further. From http://www.ross-tech.com/vag-com/vag-functions.html it looks as though it should work on my vehicle.

 

The full version appears to be less finicky that the lite one i.e. no interface compatibility problems, COM ports to fight with etc. I would pay the full £270 if I knew that I was staying with VAG cars. However, I didn't want to pay that much in case I was still unable to fix the light throttle problem and then moved onto a non-VAG car for which the software / interface was of no use. (I'd also prefer a bluetooth / smartphone version of the full VCDS software to avoid the need for a cable and laptop).

 

What do full VCDS owners typically charge for doing a simple job like TBA? (As I'm a new member, I don't think I can PM my local ones).

 

Anyway, thanks again.

 

JB.

Edited 21 December, 201510 yr by JB2015