Flying By Wire

[seatkid]

Driving up Woodhead pass and pulled out to over take a flatbed towing a trailer. 3/4 way past and......WHAT??!*!

 

Engine dies - power gone - Mind switches to "what to do? Don't panic, ffs, don't panic" mode

 

Dipped clutch engine still running but no response from throttle - nothing

 

tight sensation around buttock area

 

Fortunately - had **just** enough momentum and speed to drift in front of man and there was no approaching traffic. Also whitevan man pulled back instead of racing me, but must have thought me a right twat when I immediately slowed in front of him and switched the hazards on and pulled over.

 

Stood by the road - 1300 rpm - nothing on throttle.

 

So, resisted screaming "WTF I'm in the middle of nowhere", and switched off. Restarted and everything hunky dory again.

 

Could have been ever so nasty - makes me think twice about overtaking in those "it'll be close but I think I can do it" situations.

 

For those that don't know, Woodhead pass is in the pennines.

[Saif Rehman]

Very scarry and dangerous - so what caused this then?

[seatkid]

Intermittent throttle sensor - common problem reported several times on the forum before.

 

Just thought I'd report that the "fail safe" fallback when the the two potentiometer outputs of the accelerator pedal disagree (due to wear) could lead to leaving you alone in the lurch at a critical time. Just think that the software guys could have come up with a better solution. After all its all immediately restored after a restart, so why not "on the fly"?

 

Toyota have been in the news lately about faulty accelerator pedals (massive recall in the US following half a dozen implicated deaths) and I am sure other manufacturers also have had throttle sensor type issues.

 

And what did happen on flight 447? (the airbus lost over the atlantic)

 

Software....guys....smart software...... :o

Edited November 17, 2009 by seatkid

[DMJ]

Our Galaxy has been doing the exact same thing for awhile and i've been scratching head as to what it could be, it's gradually getting worse too (it did it 3 times to me today, once when over taking a slow moving van much like the op did).

 

Whats required to fix it? a new pedal or just the potentiometer? any body know part numbers and costs?

 

If I want to get it checked before shelling out for parts will it show up in the ecus memory or am I just going to have to change it and see if that fixes it?

[seatkid]

Our Galaxy has been doing the exact same thing for awhile and i've been scratching head as to what it could be, it's gradually getting worse too (it did it 3 times to me today, once when over taking a slow moving van much like the op did).

 

Whats required to fix it? a new pedal or just the potentiometer? any body know part numbers and costs?

 

If I want to get it checked before shelling out for parts will it show up in the ecus memory or am I just going to have to change it and see if that fixes it?

 

The pedal/sensor is a complete unit and costs between

Edited November 17, 2009 by seatkid

[Scorpiorefugee]

All this in the name of progress! :o :( :(

 

As you said, you have to wonder what the software engineers were thinking of. Worn potentiometers will almost always go intermittent so why go into shut down mode at the first sign. Some bl00dy imbecile probably thought that it was a good way to make sure you are aware that there is a problem. Prat! It is the same sort of thinking that is engendered in the European Parliament.

 

God help us!

[Mirez]

Not had it on the galaxy, but the focus did it twice and I can sympathise with you - its a twitchy moment! Both times were also when overtaking so i'm guessing its to do with a maximum value being reached/exceeded - software designers have a LOT to answer!

[Biscuit]

I use to have the same probs with mine so i removed the throttle pedal and sensor took the sensor cover off

and found out the contacts inside were covered in dirt.

I cleaned it off and not had the problem since!!!!!

10 minute job to do.

hope this helps.

[seatkid]

Good tip biscuit, might be worthy of FAQ listing if someone else can do the same.

 

I'll leave mine until I have a 2nd occurrence, because I'm lazy...

[Biscuit]

Good tip biscuit, might be worthy of FAQ listing if someone else can do the same.

 

I'll leave mine until I have a 2nd occurrence, because I'm lazy...

 

 

 

It really is an easy fix but it helps if youre a bit of a contortionist!!!!! lol

Where abouts on woodhead were you? i live about 2 mins from it.

[seatkid]

Where abouts on woodhead were you? i live about 2 mins from it.

About here, several hundred yards before the 3 lane climb. I was going to Manchester. If it happened on one of the hairy bits, I think I wouldn't be posting on here anymore. :o

Edited November 18, 2009 by seatkid

[Willie Krashitt]

I think it is a special feature for SK! The PCM sensed that he would get nowhere near 56mpg if he was overtaking other vehicles, so it shut the throttle down to bring him back to his senses!

 

Seriously though, the "different reading between sensors = disable throttle" really is there for your safety. If it wasn't, then there would be nothing to stop it if it failed in the other direction, ie it could leave you at full throttle. I know which scenario I would prefer.

 

It is my understanding (but don't know for fact) that this is also in the legal / certification requirements for 'fly by wire' throttle systems.

 

Progress? I guess it is no different from the old days when your throttle cable broke, leaving you at tickover...

 

George.

[Scorpiorefugee]

I think it is a special feature for SK! The PCM sensed that he would get nowhere near 56mpg if he was overtaking other vehicles, so it shut the throttle down to bring him back to his senses!

 

Seriously though, the "different reading between sensors = disable throttle" really is there for your safety. If it wasn't, then there would be nothing to stop it if it failed in the other direction, ie it could leave you at full throttle. I know which scenario I would prefer.

 

It is my understanding (but don't know for fact) that this is also in the legal / certification requirements for 'fly by wire' throttle systems.

 

Progress? I guess it is no different from the old days when your throttle cable broke, leaving you at tickover...

 

George.

Sorry mate, I have to disagree strongly. What you say is correct to a point but the safe solution is much simpler. i.e failsafe if there is an open circuit but return to normal if you get a within range reading. Easier still, use a bit of simple analogue design and strap a fixed resistor to the control wiper and cold end of the control would eliminate all of the fancy digital stuff. The trouble is that most digital design engineers know little or nothing about analogue or have blocked it out of their thinking.

 

Sound a warning and put a message on the display but never remove power without warning. That is criminally irresponsible. This goes back to my earlier comments about the dipsticks that make the rules.

 

As far as driving is concerned, I did have a similar but unrelated problem and I simply switched the ignition on and off while still in gear and moving and the reset cleared the problem. No doubt there will be comments that this will damage the engine or something but I never had a problem.

 

One thing is certain. I think Seatkid did us all a great favour by bringing this up. I think every Galaxy driver should be made aware of this problem.

Edited November 19, 2009 by Scorpiorefugee

[tim-spam]

George is quite right about fly-by-wire throttles and his comparison with a breaking throttle cable (which, by the way would not reset itself...). People who have never designed machinery such as cars tend to be very quick to criticize the design engineers, as they always have, usually (but admittedly not always) without much justification.

 

Another example of this type of situation is switching into limp mode when the turbo VNT sticks, causing excessive boost. Again, this is done for your, and the car's, safety, even though it is somewhat disconcerting at the time. The alternative is complete turbo failure, no power at all and potentially expensive damage. If you regularly get this problem, you really need to either clean the VNT mechanism, or replace the turbocharger.

 

Similarly, problems with the throttle potentiometer usually show up on VAGCOM before the type of failure described above. Again, the DTC should prompt replacement / cleaning before failure occurs - otherwise, who's really to blame?

[Mirez]

Similarly, problems with the throttle potentiometer usually show up on VAGCOM before the type of failure described above. Again, the DTC should prompt replacement / cleaning before failure occurs - otherwise, who's really to blame?

 

I would say the software engineers, but not for it failing - I've thought for a long time that the galaxy ECU is extreamly lax in putting the engine warning light on - I reckon if we did a pole hardly anyone would have seen there MIL light on. A failing TP SHOULD bring a warning light with it - I'm sure a fair few people won't like that idea and see it as scaremongering but if its failing its failing and should be replaced or repaired - simple. :)

[Scorpiorefugee]

This is generating a lot of interest and a lot of good comment.

 

I stand by everything that I have written on this.

 

As far as broken cables go. There was nothing that could be done about it and it is therefore not a fair comparison. In this case there are a whole lot of possible simple and safe alternatives and the one chosen is the worst possible choice. To remove all power without warning when it is not necessary is irresponsible! The other situations like over boost etc. only result in a reduction of power.

 

In mitigation, the software is probably getting on for 20 years old now.

 

As far as criticizing designers goes, what about the climate control panel? Whoever thought of putting the most unnecessarily complicated bit of control gubbinry in such a daft place must qualify for some sort of prize.

[big_kev]

In this case there are a whole lot of possible simple and safe alternatives and the one chosen is the worst possible choice.

 

Its the only possible choice available.

 

The system has detected a fault and is unable to detect the position of the throttle pedal.

 

Can you imagine the consequences if it "guessed" the position using faulty or incorrectly reading sensors.

 

After a restart it no longer has to guess the throttle pedal position as its has the off position as a reference to measure from.

Edited November 19, 2009 by big_kev

[Scorpiorefugee]

In this case there are a whole lot of possible simple and safe alternatives and the one chosen is the worst possible choice.

 

Its the only possible choice available.

 

 

As I tried to point out earlier, any half competent electronics engineer could add a couple of components on the ECU board to eliminate this possibility. The possibility of any situation more dangerous than sudden total loss of power just shouldn't exist.

[big_kev]

As I tried to point out earlier, any half competent electronics engineer could add a couple of components on the ECU board to eliminate this possibility. The possibility of any situation more dangerous than sudden total loss of power just shouldn't exist.

 

Components are always prone to failure at some point.

 

If the throttle pedal position sensor ( whatever type ) fails then what should the control unit do ?

 

Should it assume full throttle position ( i think this is a bad idea ).

 

Should it assume mid throttle position ( I think this is a bad idea because if the pedal is not actually being pressed, or lightly so, then when the driver presses the pedal halfway down this would be seen as the control unit as full throttle ). Also if the pedal was pushed past this it would be seen as "more" than full throttle !

 

The only safe option is to cut of the throttle altogether.

 

 

 

Maybe their should be a fixed service interval where the pedal sensors are renewed or at least inspected for wear and contamination to minimise this problem.

[gregers]

Scorpiorefugee wrote

As far as criticizing designers goes, what about the climate control panel? Whoever thought of putting the most unnecessarily complicated bit of control gubbinry in such a daft place must qualify for some sort of prize.

 

close,but still think its got to be the numptys who put a join on the rear washer pipe above the ecu on a mk 1,

now really we all know why the desingners do this is repeated custom,from peeps who do not have the savvy or inclination to seek out a bit of knowledge and get there hands dirty.

[Scorpiorefugee]

The system can detect the failure. Why does it have to shut down permanently? Why can't it pick up again when the fault clears. Most problems of this type are very intermittent. Ask Seatkid if he would have preferred to get the power back after he released the throttle and pressed it down again.

 

I have no statistics on sudden total failure of the control but, based on reports on other users of this site, I would imagine that it is very much less than the sort of problem that most users get. If this does happen then it is very much like a broken cable and shut down is the only possibility. However, I should be suprised if this were more than one chance in one thousand and that would leave the driver with only unmistakable symptoms of a faulty throttle control and no dangerous situations for the vast majority of situations with still no danger of an uncontrollable full throttle condition.

 

Other possibilties are available with only minor circuit additions but the simplest solution would need only a minor software change.

 

This is all getting to feel like arguing against the nanny state that this bl00dy government is trying to impose. :) :) :)

[big_kev]

This is all getting to feel like arguing against the nanny state that this bl00dy government is trying to impose. :) :) :)

 

Don't get me started about this f&*#ing Labour Government :) :D :D

[Scorpiorefugee]

Don't get me started about this f&*#ing Labour Government :) :) :)

 

I think it would be nice to end this topic on something we can all agree on.

 

:) :D :D

[tim-spam]

1) The system can detect the failure. Why does it have to shut down permanently? Why can't it pick up again when the fault clears.

 

2) I have no statistics on sudden total failure of the control

 

1) You can be absolutely certain that this system will have been very carefully considered by very knowledgable and highly trained engineers, and I am certain that they would be able to answer your question which was bound to have been asked many times during the system's development. If you had any experience of automotive design at all, you would realise this, as well as the raft of legislation that has to be complied with together with the many other constraints that the engineers work within. All design has to be a compromise between function, appearance, manufacturability, quality and cost.

 

2) The manufacturers will have this data, and if there really was a significant safety issue (ie: if this type of failure actually resulted in real accidents, rather than just brown underpants), it would have been addressed long ago, probably with a product recall. After all, this system is used on millions of cars, a significant proportion of which are now quite old.

Edited November 20, 2009 by tim-spam

[Scorpiorefugee]

Tim spam,

 

I agree with what you have said but how many times has the logic and decisions made by these super intelligent and trained people been proved to be wrong?

 

I cannot accept that a responsible person could leave a system which turns off the power suddenly and without warning and then allows the user to continue driving without understanding why.

 

You have only to look around at the world today to see the results of bad decisions by these trained and skilled people in all walks of life. So many of these decisions are made for wrong reasons or without a wide understanding of the implications.

 

I'm sorry to have to say it but I still believe that this is just another case of someone making a decision and then it being widely accepted without anyone re-examining the evidence. I also still maintain that there must be simple electronc means of reacting to the problem without killing the power completely.

 

However, the only possible action is to make the case and I am sure, or at least hope, that someone is already on that. As far as we are concerned, I think that SeatKid has already done the job in making everyone aware of the problem.

 

I'm done now.

 

:)