Throttle becomes unresponsive

[mops.mad.house]

Hi guys,

I am new to PCGB so hopefully I have done this right, I only joined the beginning of the year as I intend on tracking my 2005 Boxster S 987.1 this year. I was kind of hoping that my first post would be in relation to this, however 2 weekends back I took it off axle stands after replacing all the coolant pipes at the front end and servicing it over the winter and went for a test drive, unfortunately as soon as I left my drive the throttle became unresponsive when I lifted my foot and reapplied the throttle the car came back to life it then happened several more times during a 2 mile test drive.

Last year a few months before I put the car away for winter I had experienced similar symptoms, but back then it seemed to happen only after the brakes had been used, the next time you accelerated it would initially accelerate then the throttle would go dead and you would maintain the speed you were at, sometimes it would spring back into life on its own if you kept your foot where it was or if you released the accelerator then reapplied it it would then be back to normal the other way to reset it was to dip the clutch, it was intermittent but was happening enough for me to identify these traits. After some research I decided to change the brake micro-switch and this appeared to fix the issue. This time the symptoms seem different I don't believe it is related to the brake pedal and when it does happen the car slows down like it has lost all power but if you dip the clutch the engine is still running, the only thing that seems the same is if you take your foot off the accelerator it returns to normal, unfortunately I have not done more than 6 miles in it around our village at 30/40 mph since this has happened so it is a little difficult to say if this is the full symptoms, the only other thing I could potentially add is it has only happened in 1st and 2nd at slow maneuvering speeds.

I decided to get a diagnostic machine as I was sick of trying to guess what issues I have, I ordered a Thinkdiag+, I read the fault codes and there were not any related to the DME or PSM systems. Initially with it seeming to happen at slow speeds and turning I thought I would look at the PSM, I monitored Car speed, all 4 speed sensors, Steering angle sensor, brake light switch and the PSM Active signal. When the fault happens all the speeds match nothing wacky seems to happen with the SAS deg, the brake light switch is not active and PSM stays inactive and the dash traction control light is not illuminating, so I don't believe its PSM related.

I then decided to monitor the throttle response when stationary, i get the extremely similar results as Mac who used a durametric on his own car when it was serviceable granted it was for a 996 but I am hoping it is close enough, Mac got the following:

Hi plugged in the durametric today, no fault codes logged

the first number is with ignition on and foot off the accelerator, the second with foot to the floor
off / full
pedal value 0% / 99.99%
pedal encoder potentiometer 1 0.76 volts / 3.3volts
pedal encoder potentiometer 2 0.38volts / 1.65volts
specific throttle plate angle 4.2% / 99.99%
actual throttle angle 4.2% / 99.9%
throttle position sensor 1 0.78volts / 4.35volts
throttle position sensor 2 4.25volts / 0.67volts
throttle plate angle potentiometer 1 6.2% / 99.6%
throttle plate angle potentiometer 2 5.8% / 99.6%

Thanks to MAC996t for posting this on 911uk.com

I decided to take the car for another test drive and monitor these values, I have attached a screen shot of the graph. The fault occurs at time stamps 30, 60, 105. I am hoping the picture quality is good enough, on my phone the screen shot is really clear however looking at it on this laptop it is not clear. Anyway at those time stamps you will see the dark blue line (Data Pedal Value) goes flat line at around 4.5%, however the red and yellow line (Data Pedal encoder potentiometers 1 & 2) keep on rising, on the 2nd and 3rd time I floored the pedal and then lifted it slightly and pushed it back to the floor to see if it kicked back into life which it didn't and hence why they look like an 'M', and when all this is happening the turquoise line (Data Nominal throttle plate angle) returns to almost closed.

I am not entirely sure how the Data Pedal Value is calculated, as far as I am aware there are only 2 potentiometers in the pedal which would correlate with the 6 pins on the sensor of the pedal, I would assume these are the encoder potentiometer values. Any way I believe as this Data Pedal value drops to 4% and the other values are still rising the computer doesn't like it and closes the throttle plate. I did find on the internet someone with the old 986 cable pedal rather than drive by wire, have symptoms where the potentiometers were reading fine but the Pedal Data Value range was only 0-30% all of the time, he replaced the potentiometer sensor and it fixed the issue. My thinking is that on the 987 drive by wire system the Accelerator Pedal is the same component so will probably fix the issue.

Anyway I am hoping that someone on here is far more in the know how the system works and if there is a possibility another system like the PSM is causing this value to drop to 4%.

One final thing has anyone taken the Accelerator Pedal sensor apart before, I have had a look and it looks feasible or better still knows the pin out so I can try and do a resistance check will operating it?

Sorry for the really long post but hopefully I have managed to put enough detail in for someone to help.

Many Thanks Paul

 

[Snowball171]

Hi Paul,

just saying hi and welcome! I can't help you with your problem, but I'm sure one of the guys in the know will be along shortly.

Hope it's nothing serious

Philip[]

 

[mops.mad.house]

Thanks Phil that is very much appreciated.

Whilst sleeping last night or lack of , I thought I would email Hella who make the sensor to see if they could give any advice especially for the pin out diagram so lets see what they come back with.

 

[hotfrog1]

I am anything but an expert on Motronics, but I have studied the descriptive material extensively. I found a very useful Porsche branded manual about Motronics on the web a couple of weeks ago. It refuses to attach, maybe too big. If you send an email to me at dene99@gmail.com I will forward to you. It gives a good view about the E Throttle on Motronic from V7.2 to V9.1 There is also a lot of other gold dust about Porsche engine control technology.

 

[mops.mad.house]

hotfrog1,

That stuff you emailed over although I have literally just flicked through 1 of them it seems spot on for me trying to understand the system and how it works. Unfortunately from that quick flick my theory of it having 2 potentiometers and a variable resistor are not right, it only has 2 potentiometers, which surely must mean the Data Pedal is calculated in the ECU, this is potentially starting to sound expensive [&:].

Anyway this is a super find I shall definitely be adding this to my Boxster folder for the future, although I am hoping I won't be having to read it that often, the next few days will involve a bit of reading to see if I can glean any information to point me in a direction on the current fault

Hotfrog1 you are a superstar thanks for the info

 

[Motorhead]

Paul,

It does seem to be a very odd problem, and unfortunately I can only add moral support.

As I understand it, the system appears to be functioning correctly when the engine's not running, which I suppose indicates that all the electro-mechanical components are working correctly? On that basis it does sound as though the problem source is with the electronics. I guess you need to find out where the Data Pedal value is calculated, but the DME would appear to be the obvious place (unless there's a separate throttle module to throw a spanner in the works!), although I have to say you'd have to expect the DME state to be the same both when the ignition is switched-on and when the engine is running; but maybe not?

Hopefully the material Lawrence has sent you will throw some light on the matter. Good luck, and keep us posted.

Jeff

 

[mops.mad.house]

Jeff,

I read the book that Lawrence sent me yesterday, it has been a great help for enlightening me on how the system has evolved over the years and how it operates, I would have preferred if it told you how you get each reading from a diagnostic machine but it doesn't. That being said it has certainly gave me enough information that I think I am either looking at the throttle body as it gives an example of if both potentiometers fail inside it it would give me my exact symptoms of the throttle going dead as it does not know the position of the valve, therefore it shuts it to the electrical stop and apparently at this point you have a slightly elevated idle point and you are using the timing angle only to control the RPM.

What it doesn't say is if you would get a fault code, however if something in the throttle body is to fail as far as I am aware it will give you a fault code, so I re-analysed the data and as far as I can tell all 4 potentiometers are working so that is probably why I am not getting a fault code, unless the fault keeps occurring between time frames.

So this is now leading me to think it is the brake micro-switch again, it should be quick to identify as I can just unplug it and drive it round the block to see if it happens or not. I am currently working from home so I will try later to get my nose in the diag tool to see if there are other parameters to monitor in regards the brake switch, I am also going to look at the MAF as potentially it could be faulty it has a K&N filter fitted so it could have put oil on the MAF. If none of that seems to be leading anywhere I believe I saw on the diag tool I can force it to relearn the throttle position just in case it is out of sync or something.

I will get to the bottom of this [8|] I just need the time to get to work on it, once I do I will post an update.

 

[Motorhead]

Sounds as though you're on it Paul. Probably I'm missing something here, but it just seems odd that everything appears to be working OK when the engine's off. But what do I know? I'm a mechanical engineer whose specialisation is sound and vibration, so electronics aren't my forte. []

If it's of any use to you, just to mention that I seem to recall that the engine will still run if you disconnect the MAF - I think it just goes to a default setting.

Jeff

 

[hotfrog1]

I am intrigued by your issue. Will be nice to know what the final outcome is.

You can test the bottom end of the MAF range by plotting air flow against engine revs with car stationary. It should plot a fairly straight line roughly proportional to engine speed. So if OBD reported air flow is say 18 Kg/hr at 1000 rpm then it should be about 36 Kg/hr at 2000 rpm etc.. Obviously this is not exhaustive as it is at no load but it helps to nail another factor.

 

[mops.mad.house]

I have news..... It is the brake micro-switch!!!!

I worked this out this afternoon, like I said I decided to have a bit more of a look at what the Diag Tool can monitor and found in the PSM 'Read Data Stream' part under 'Signal Input' rather than 'Signal Value' there are 2 parameters it can monitor when it comes to the brake pedal, one is the 'Brake Light Switch' as is available in the DME section but the second is called 'Brake Test Switch', when I compared the 2 as before the 'Brake Light Switch' said 'Not-Active' however the 'Brake Test Switch' said Active, I found this rather strange so I pumped the pedal a few times to see if it changed when you applied the brakes, eventually when the pedal was not pressed they both said 'Not-Active', I tried to get the fault to reappear but it didn't however just a minute touch on the pedal and it would cause the 'Brake Test Switch' to go 'Active', so I went for a test drive whilst monitoring these parameters. Funny thing the fault didn't reappear as it had been doing [], anyway I literally brushed the brake pedal and the Diag tool registered the 'Brake Test Switch Active' and the car died I reset the throttle and the engine went again but the Diag tool still says 'Brake Test Switch Active' but the car drives normally, when I actually used the brake they both went 'Active' and then 'Non-Active'. I went to the petrol station and it did not do the fault itself, until I left the forecourt when it did it I quickly looked at the Diag tool and there it was 'Brake Test Switch' 'Active' eureka, it was only there briefly so it was a good job I saw it, I carried on driving and it didn't do it again, I was about to pull into my street when I decided to go round the block one more time to give it another chance of happening, I was glad I did as shortly afterwards it did it again but this time it stayed on once again I reset the throttle by going off then on and it was working again even with the 'Brake Test Switch' saying 'Active' still, I guess the computer decides it can't be valid and ignores it but who knows, don't you love drive by wire!

When I got home I had a look at the switch again and took it out by fiddling around and a bit of stripping down, turns out there are 2 elements to the switch of which the computer monitors. When you are not pressing the pedal it makes a circuit (Pins 3-4) which is the one labelled 'Brake Test Switch', It also has another circuit which is open and this is the one labelled 'Brake Light Switch' (Pins 1-2), when in this state the Diag tool should say 'Not-Active' for both, (reverse these for brake on). On mine the 'Brake Test Switch' element was super sensitive so the slightest bit of movement caused it to Activate. I tried using the adjustment but this had no effect it would just auto reset to the original position. My next idea was to shorten the spring inside to try and get the contact to make further down therefore take longer before it would disengage, I slowly trimmed away at the spring making it shorter each time, initially I was refitting and testing but after the switch had been in and out a few times I noticed it was becoming loose so I used my multimeter to try and gauge if it was improving the situation. Eventually after trimming I reckon 50% of the spring away it worked, just be really careful I could not have removed anymore I don't think as the 'Brake Light Switch' would not have worked anymore, plus the setup says that the lights should come on after 5 - 10 mm of travel but before the perception of pressure, mine was very close to pressure being felt with the engine off and the servo having no pressure, what the manual doesn't say is what condition the car should be in before feeling the pressure obviously in my condition you get the least amount of travel before it is felt and it operated at 15mm, I don't know if that is good or not however when do you only press the pedal 1.5cm! When I put it back in and using the diag tool to monitor they now pretty much operate at the same time. As a side note like I said when i first started this post I did replace this switch at the end of last year, I bought it from GSF Car Parts I believe I was in a hurry to get the part and they were the only ones locally that had one. Perhaps if I had this Diag tool the old one was operating in the same manner and I could have done this to rectify it, or on the other hand this was made cheap as chips and does not have the same tolerances unfortunately I will never know. What I would like to add though is if I had this tool I could have done a proper setup in the first place and saved myself all this heartache but I certainly didn't know there were two elements to the switch I assumed it was using the 'Brake Light Switch' to assess the position, hopefully this will prevent someone else from making the same assumption and think the switch is fine and can not be causing engine issues!

Just took it for a quick test drive round the village no signs of issues but like I said about the test drive beforehand it had become a lot less frequent only time will tell, however I am feeling quite confident that this was the issue, if it does return I shall re-post, obviously not quite sure when I will actually get to test it properly due to COVID-19. Two bonuses to this, it didn't end up costing me a fortune to fix and now I have a Diagnostic Tool, just need to see if it can program some extra features like global opening of the windows and the main one I want is the doors automatically locking when I drive off, having to press the button every time is a pain.

 

[mops.mad.house]

Jeff,

I am an Avionics Tech so I am use to looking at scopes and wiggly lines, what the issue is there though is you can sometimes get to engrossed in trying to use the data to find a fault you forget to do the easy stuff first, like simply looking to see if the brake switch is on or off or in this case both elements to the switch .

Lawrence I don't suppose you can remember where you found that document as I am going to try and find one for each system of the car as that was a great document!

 

[hotfrog1]

I was very pleased to learn of your success. Very interesting tale. Well done.

I found the said document by a Google search that took me to: https://www.academia.edu/

This site looks too good to be true and I am now getting daily emails trying to get me to make a commitment and I am completely unsure about how much of a trap this is. It seems to have good credentials. I have a sneaky feeling that they make material available that has been superseded. I recently bought a book called "Advanced Automotive Fault Diagnosis", it was recent publication and cost about £14 for a Kindle download. I subsequently found a free copy on the Academia website but it was about 15 years old. Anyway I hope that you find it interesting and find what you want, if it is there.

 

[hotfrog1]

Sorry, but I meant to ask, what sort of diagnostic OBD tool do you use?

 

[mops.mad.house]

I bought it last week, it's called Thinkdiag+, basically according to their website it has all the same features as the £700+ diagnostic stuff but it costs you £84 to buy but you basically rent the software for each manufacturer for the year and it connects to your phone. Currently if you buy the Pro it costs a pound more than the normal but you get 2 manufacturers instead of 1 plus apparently if I refer anyone they will get $10 off and I receive 10 points. Each manufacturer you download costs $40 so what ever that translates to in £'s, even though I have quoted it in dollars it came from the UK and arrived pretty quick.

I bought this because I wanted the pro features to be available, for example mine has been lowered and I keep getting a headlight aim error if I go quickly over undulating ground, apparently the back sensor picks up it is outside of limits, so I am going to try and use this to re-calibrate it to see if that resolves it, as well as I have already said about auto locking the doors. I am probably going to buy at a minimum the Icarsoft Por V1 maybe the CR pro, so I can reset the service light on the boxster every time I do an oil change rather than paying $40 for the pleasure, unless somehow I can do it without paying the yearly fee.

One thing I am not liking about it so far and as this is my first diagnostic tool I don't know if this common across all of them, but I would like to be able to select any parameter on the car and log it, for example on this snag I would have liked to monitor the Accelerator Pedal Data and the brake switches together but I couldn't get it to do it, currently I have only been able to select the parameters within each system and even then you can't have signal and values at the same time. I am going to email them as it might be possible but I haven't worked it out yet. I have sent a couple of emails and the guy seems to email back pretty quick so their customer service seems ok.

 

[mops.mad.house]

forgot to add the other thing I thought was better about it was those ones that cost £700+ the yearly fee to update them is normally just as expensive, this way you pay a fraction of the cost to buy and a dam site less every year to keep it up to date assuming you only have a couple of cars to maintain!

 

[mops.mad.house]

One final one for Lawrence.

Sorry I completely missed your message with the links for the documents, I am a little tired I haven't slept properly for the last 2 nights with all this rattling around in there, plus having to run a house with 3 kids whilst working from home and the wife is living away it's fun times, as the saying goes what doesn't kill you makes you stronger. Anyway enough of me babbling on, thanks Lawrence for the links, I have also found a link to some potential documents so when I have time I shall look at these and post the link if they are any good.

 

[Edwill997]

So glad I found this thread. Mine has been doing exactly the same. I have tried two switches from euro car parts and they both made my car do weird things, it's a 2005 tip btw, cruise wouldn't work and the gears were all to pot. I've ordered an original switch now so am hoping this cures the issue. It originally started twelve months ago then went away when I cleaned the MAF. The gremlin is back again so hopefully an original switch will sort it. Thanks for the thread. Ed

 

[mops.mad.house]

Ed,

Fingers crossed it works for you.

Paul

 

[Edwill997]

Cheers Paul! Will let you know.

 

[911hillclimber]

Have to add, thanks to all for this thread and many like it.

I'm a humble mech engineer and glad these complexities are not in my life, but reading this haseducated many, esp me!

Thanks for posting in such detail.

Graham.