Porsche 718 Forum banner

i made the computer mad...

6K views 31 replies 14 participants last post by  nineball 
#1 ·
this past weekend i was at daytona for a DE event. was cruising around 155-160 (gps, not speedo) on the banking/straights and suddenly started getting all kinds of warnings popping up on the screen. the auto-aim headlights failed, the lane change assist failed and i had a PSM failure. kinda freaked me out to say the least. luckily there was a local shop at the track with some customers and they specialize in cayman race cars.

i talked with the owner and found out this is a somewhat known problem with caymans for the past few years. apparently if the car is at an angle of greater than 30 degrees the computer freaks and starts shutting down some systems. the banking at daytona is... you guessed it... 31 degrees. he said they have had porsche factory engineers at daytone for testing with his shop twice and they can't find a way to prevent this from happening.

the fix? park the car, turn it off, lock the doors and wait 30 mins. the computer will reset itself and i was good to go by the time the next session came in.

so word to the wise - if you do any track events on nascar tracks and the banking is over 30 degrees you now know what to expect :)
 
#3 ·
That's my bet: It's a safety feature built into the ECU to prevent really bad stuff from happening in the event the car tips or rolls.

Interesting that the angle is set at 30 degrees; there must be a really good reason for that number to be used, such as fluid levels or load intolerances (i.e., too much weight on a corner of the suspension = SNAP! goes them connecting rods, or somesuch).
 
#5 ·
can't see why it would be any different.
 
#7 · (Edited by Moderator)
I just did this on my 718 Boxster on the approach to my local bridge. I was driving in an *ahem* spirited fashion around this awesome almost 300 degree slightly banked loop and I must have gone over 30 degrees at some point.

The car threw up 3 different failures PSM, LCA and Engine Fault. I did a quick stop and restart which got rid of all the warnings except the Engine Fault.

I remembered this thread, and looked it up. Now I am hoping my car does the same as nineball's and resets itself after 30 mins of being parked/locked/shut off.

Edit: same as nineball, all faults and warning messages cleared themselves after leaving the car parked/locked for an extended period (i left it for 5 hours)
 
#8 ·
Useful info. Thanks.

I've been around Daytona (riding shotgun in a pace car) and recall how steep that banking was. It seemed that if you went too slow, you would slide downhill off of it.

The only similar track that I might do a DE day in is Pocono (it's about 90 minutes from me). But Pocono is one of the "flattest" NASCAR tracks. The banking angle is about half that of Daytona. So that shouldn't be a problem for my Cayman.
 
#10 ·
I think the issue is more to do with a potential out of range value then anything else. The stability control module likely sees 30+ degrees for x amount of time and decides it's an implausible signal.

It's highly unlikely the scheme for responding to a potential out of control situation is to turn off the stability management.

Purely speculation on my part.
 
#11 ·
How does the car know its on a 31 deg bank? IT would seem that the forces that are holding the car up on the banking in the curve would also act on any onboard sensors.


imagine a plumb bob hung from your mirror. on a hard right turn on a flat road it would swing toward the driver side. but on a high banked turn the car doesn't see the turn as its forced up on the banking. so in this instance the same plumb bob would point toward the console or floor. since that is where the artificial gravity would force it.


OK physicists and engineers, wade in...
 
#14 ·
With a 3axis acceleromter you're still seeing 1g (at sea level) acceleration on the Z axis regardless of the lateral acceleration.
I've written algorithms to deconvolve the tilt vs g-force effects on accelerometers. If I can do it I'm pretty sure Porsche can do it :)
 
#13 ·
Android phones have an accelerometer and a gyroscope. Also a magnetometer (can find North).

I can't be sure how these devices interact in a phone but it's likely the gyro could maintain Z-axis, even if phone is tipped (I think this is how the software application called "Harry's Laptimer" works....I use it in my racecar).

I would guess the car uses similar technology, but again, I'm only guessing. Interesting question.
 
#15 · (Edited)
I remember this thread from when it first appeared. Mind-blowing then and still is -- wrt both ECUs and smartphones.

btw Talladega exceeds Daytona with 33 degree banking. Just reading in the book Porsche Turbo about Mark Donohue's mythical world-record-setting lap there in a modified turbo 917 https://www.motorsportmagazine.com/magazine/greatest-laps-mark-donohue-talladega

Donohue called that early Porsche race-car turbocharging: "the unfair advantage".
 
#17 ·
well anyone can speculate, and most all of it is beyond my abilities :) all i can do is report what happens.



since last year i have done 10 track days at mid-ohio and pittrace, two challenging courses but no where near the banking. haven't had a single problem. i'll be leaving here at the end of november to go back down to daytona and i will report back what happens this year. i am predicting the same thing as last year but that's just my guess. i just put on new rotors, upgraded pads, fluid change and a fresh set of tires. can't wait to hit that track!


btw, got to a new low a couple weekends ago. some guys measure de weekends at top speed, some at lap times, me - i do it based on how low i can get my mpg, 5.621. i was quite happy :)
 

Attachments

#18 ·
since last year i have done 10 track days at mid-ohio and pittrace, two challenging courses but no where near the banking. haven't had a single problem. i'll be leaving here at the end of november to go back down to daytona and i will report back what happens this year. i am predicting the same thing as last year but that's just my guess. i just put on new rotors, upgraded pads, fluid change and a fresh set of tires. can't wait to hit that track!

I'm jealous! I have to wait until next April at the earliest to get back to the track.


I did about 10 days this year at Watkins Glen, Lime Rock and Palmer.
 
#19 ·
I'm jealous! I have to wait until next April at the earliest to get back to the track.


I did about 10 days this year at Watkins Glen, Lime Rock and Palmer.

pm me if you would be interested in doing daytona (or anyone else interested) this year. the dates are 11/20, 12/1 and 12/2. there are still spots open!
 

Attachments

#20 ·
pm me if you would be interested in doing daytona (or anyone else interested) this year. the dates are 11/20, 12/1 and 12/2. there are still spots open!

Daytona would be great. But right now I don't have the time. It would take me a few days just to drive there. :)


When you replaced your brake pads what did you do about the brake wear sensors? The only ones I've found are from Suncoast parts and they are very expensive (I think around $80). On my 987 I could get them from Pelican parts for about $5.
 
#23 ·
This is curious as to how it would sense the 31 deg. tilt.
The artificial horizon suggested by @phroenips is driven by a pneumatic gyroscope. (You can hear them spin up as you start the plane.) Quite certain we're not running gyros.

But.. accelerometers such as those in the phone wouldn't notice the actual bank unless you stopped the car. As you travel around the bank at speed, the vector is still largely going to be down through the floor. By example, when I used to lay my phone sideways under the radio in my old car, I used to get the orientation to change with every hard turn due to the lateral g's making it think it had been rotated. The activation angle would have to be MUCH lower than 30 deg.
 
#24 ·
But.. accelerometers such as those in the phone wouldn't notice the actual bank unless you stopped the car. As you travel around the bank at speed, the vector is still largely going to be down through the floor. By example, when I used to lay my phone sideways under the radio in my old car, I used to get the orientation to change with every hard turn due to the lateral g's making it think it had been rotated. The activation angle would have to be MUCH lower than 30 deg.
Not quite. Think of it this way: If you parked your car on a 45deg slope you'd have ~.7g on the z and +/- ~.7g on the x (1/sqrt2), start moving and those will change but no matter what you're going to have gravity baked into the x,y or z axis plus your actual acceleration for whatever you're doing.

Without digging into the java code in my tablet I'm just going to assume the table decides whatever axis has the highest acceleration is up and orients the screen to it. Just a guess though..

Per my original post,
"I think the issue is more to do with a potential out of range value then anything else. The stability control module likely sees 30+ degrees for x amount of time and decides it's an implausible signal.

It's highly unlikely the scheme for responding to a potential out of control situation is to turn off the stability management."

It's entirely possible the system is not aware of angle but rather just acceleration on various axis and when they are out of range for too long the system shuts off.
 
#27 ·
well i tried setting up both my phone and my gopro but i couldn't find any way to mount them to get a clear shot of my dash.


took it easy on friday never getting over 150 and no problems. rained on and off saturday. sunday was clear and took it easy the first session. came in and noticed the wheel weights on my front left were moved and some missing. turns out i drove the weights off all 4 wheels in varying degrees and had to remove and run them over to pepboys for a rebalance. i just had the tires mounted and balanced before the trip but the shop never taped them down and i forgot to check.



went all out floored on the banking and tri-oval during the 3rd session and as soon as i exceeded 155 the same thing happened - lost the 3 systems. came in, parked the car, waited 30 mins and it was fine. went out for the last session and around the 3rd or 4th lap floored it again and once again the codes flew and that was that. i did managed to get up to 164 this year :)


ps> in an earlier post i mentioned PSE when i should have said PSM.
 
  • Like
Reactions: joshua@arcflashllc
#32 ·
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top