If only we could enlist Nick as our agent in these proceedings.

 

Software??

Have you posted this before? I knew this was a possibility, but the first time I've heard you say it.

A software fix would be easy you'd think. If it was software, why wouldn't it be the same thing in every car? More of a mechanical problem wouldn't it be?

 

While that's true that as a general rule, software operates based on the data received from hardware, in this case, sensors. MAF, O2, knock, pressure, throttle, etc. So maybe the O2 sensor in my car falls within the standard parameters as does DUX's, but mine is at the upper end of acceptable and his is at the lower end. Conversely our mass air flow sensors are slightly off. And that difference, maybe along with others, combined, is enough to confuse the computer enough to make it diddle around a bit with the fuel/air mixture, causing his car to stumble and mine to run like a train.

No mechanical device is perfect, and no mass produced item is going to be exactly the same as the last one (although that's sometimes counter-intuitive). So for every parameter fed into the datastream, the system needs to allow for variation. As does quality control. You can't insist that every single O2 sensor is accurate down to 0.0000001 micrograms. It becomes cost prohibitive. So they allow for the standard Bosch "close enough" sensor, and most of the time, that is good enough.

Or it may just be bad gasoline.

 

I've ruled out the gasoline theory - that was my first thought. I have tried many different brands with the same results.

Also, in my naive assessment, I don't think it's a purely mechanical issue. At RPMS above 3000 my engine is rock-solid and powerful. I would expect mechanical problems to get *worse* not better as RPMs increase.

I do think it's a bad sensor. If it were completely a software issue like you said everyone would have the problem. But I believe it's a problem with the software doing wrong things because a sensor is feeding it incorrect data.

 

I'm starting to believe that it does in some way have to do with gas or octane... I'm starting to believe it's marginal with 91 oct... Maybe in conjunction with 91 octane and the anti-nock capabilities... My problem is only intermittent but I started to notice a pattern where when I occasionally fill a a particular station in a pinch which is a different brand. Who knows maybe they fill will a lower grade..

Can't prove any of it, but I've stayed away from that station and haven't noticed any issue lately..

 

This makes the most logical sense to me.

 

What happens if you use a high octane racing fuel. Is anyone on this thread from europe and have the problem.

 

Let me say this again. I believe that the stumble is software related (not caused solely by the software) and here is why.

First of all, I have checked with non-stumblers and we are using the same software. However, if us stumblers happen to have a borderline faulty component / sensor in our engines which is not functioning quite as it should (from a faulty batch etc), the software will attempt to correct or compensate for the fault, thereby resulting in the unintended consequence in the form of a stumble.

Without the shadow of a doubt my problem goes away every time they reset my DME, which has now been 5 times and counting.
The following happens every time: My DME is reset, I drive away happy. The software starts to gather data (including data from the potentially flakey component) and after a period of time (usually ~24hrs) the software attempts to compensate for the faulty component (which is operating within its predetermined thresholds) and the stumble reoccurs.

This is why I say s/w related. The software has been configured to allow this component to operate within unacceptable parameters.
The car is basically a laptop on wheels, where the software monitors and controls the hardware, which in turn feeds hundreds of parameters back to the DME, which sends commands to the components, and so the cycle continues.

The software can obviously be configured to allow for different tolerances and thresholds. However, if these thresholds are set within a generous enough range to accommodate a defective component, neither we nor the service tech would ever actually see an error code being generated.

I have been at this for a year, going through every permutation possible - from extra weights (to simulate load), to ramps and hills, to fuel types (yes also racing fuel), to temperate variances, to PSE, Sport+, A/C, throttle resets etc etc.

THE ONLY THING THAT HAS WORKED (temporarily) IS A DME RESET.

I may well be completely wrong, but just sharing my thoughts and experiences.

 

^^ That would explain why some have it and others do not. Now what and how does porsche CPO this problem.

 

Based on your evaluations why can't the faulty sensors be replaced? Wouldn't that solve the software issue?

 

Which sensors should be replaced? And therein lies the beginning of this absurd problem.

Because if the software is not setup to recognize the faulty sensor(s) (because the sensor is operating within the threshold) and the software "corrects" the problem, the problem really doesn't exist.

I have a question for my brethren - Does resetting the DME also reset a faulty sensor? If not, then we're back to the software (after 24 hrs of learning) compensating for a less than optimally operating component.

 

I made a few audio files in an attempt to demonstrate what my stumble sounds like:

https://www.dropbox.com/s/qjjgy5y04m16khl/gp1.m4a?dl=0
https://www.dropbox.com/s/wigxtn9rt5txx1d/gp4.m4a?dl=0
https://www.dropbox.com/s/nekn59iqdzzc7q6/gp6.m4a?dl=0

All done with a smooth acceleration and PSE on. No abrupt throttle changes.

I think you'll notice a point at which there's a distinct change in exhaust pitch, and this of course is accompanied by a noticeable stumble/hesitation.

 

This post from almost 2 years ago has me thinking about the camshaft and/or crankshaft position sensors as possibly being behind this mess. I haven't had either worked on (I believe there's a current campaign to add a shim), but I have suffered the no-start and stalling scenarios in the past. Merely a coincidence between the stumble and no-start/stalling, or is this indicative of cam/crankshaft sensor issues also causing the stumble?

 

^^ This baby has long been the suspected culprit, but yet to be confirmed by anyone.
Spoke to the dealer last year and they said no error = no swap.
Heck, I'll swap it out myself as I've done will my M cars. It's usually one measly bolt and a pain to get to.

There must be a DIY somewhere...

 

I know of two people who have started LL cases and had their other issues resolved. Has anyone turned in a LL case on this one? It's really quite simple.

ChuckJ