Erratic throttle response when cold

[HPsauce]

Tricky to do that as it only normally bogs down near the start of a journey and at a low speed when moving off. And I'm sure resume won't work until you're over 25mph anyway, same as set.

If I lift off, as I typically do when setting or resuming cruise, it will clear the problem anyway as it always does. If I don't I would expect the same rocketship effect as today.
But I'll see if I can manage that test somehow. I have an idea how....

[PsYcHe]

You're on a fly by wire throttle, so the cruise control sends its own signals to the throttle/ECU.

Can't remember.. did you do the new throttle sensor like Jim had fitted recently? I'm sure there's a VCDS measuring block for pedal position, might be worth just playing with that when cold to see if it gives the figures you expect.

[HPsauce]

I have a replacement throttle pedal sensor on hand, awaiting fitment.
It's been on hold until the catalyst was sorted in case that changed symptoms - it hasn't.

I've tested the throttle position readings several times with VCDS and it's always appeared perfect.

The intriguing thing is that, as explained above, if you are on cruise you can override speed upwards by pressing on the loud pedal, which is why I did the test today. Basically the pedal was already pressed a lot but the car was acting as if it was only slightly down, which is the main part of the problem. As soon as I "set" cruise (without moving my right foot) it acted as if the pedal really was where it was and shot off!
So effectively, turning on the cruise control, or more precisely "setting" it, cancelled the problem and restored "normal service". Lifting my right foot then putting it down again does the same, but less violently.

I'm trying to understand what this "reset" effect from setting the cruise tells us about the problem.
It does suggest to me that the pedal sensor is OK, but that the signal from it is being processed incorrectly - a software problem?

[ainarssems]

It does sound like a accelerator pedal sender issue or associated wiring. There are two potentiometers in the sender for redundancy and safety. Both potentiometers go in different directions, for one resistance increases as you press pedal, for other it decreases, it is this way for safety and for more reliable reading. If both resistances does not match same pedal position engine will operate with reduced power output. If the difference is too great it will trigger fault code, otherwise it is reset when pedal released and if next time values are within specified range it will return to normal operation. Looks like setting cruise also resets pedal reading.

[HPsauce]

Interesting Ainars. All my tests with VCDS show both potentiometers behaving properly, but what you describe is pretty much exactly the symptoms though definitely no fault codes.
I guess replacing the sender is the logical next step, for which I am prepared!

[Delboy]

That description of operation describes exactley how your car can show these problems, if one pot is slighty iffy then you would get the reduced power that he mentions that clears in release of the pedal, is it possible to log the pots in VCDS and go for drive, then once you have a fault scenario you can timestamp I vcds so the log shows when the fault was so you may find it easier lookng through the log later.

[HPsauce]

As I already have a replacement I'll just swap it, checking readings before and after sitting on the drive.
It's not reproducible enough to set up with the laptop and go for a drive.

Might have a look at how easy it is to swap this afternoon as it's sunny, if it looks tricky I'll get the garage to do it.

[ainarssems]

I don't think VCDS show direct individual readings from potentiometers only calculated pedal position but could be wrong

[HPsauce]

It shows both (block 62, items 3 and 4 IIRC), but I'm going out in a minute to check again what the values are and how they change with the pedal movement.

[HPsauce]

Confirmed. Block 62 shows 4 values.
The first two are the throttle valve position and they "oppose" adding up to about 100%.
The second pair are the pedal position and one is roughly half the other.

All behaving as expected.........

Now going to take the knee panel off and have a look at access.