When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Although I did not DIY (the focus of this thread), my PDK is now as good as new!
I towed the Boxster to Atlanta Speedwerks in October and they replaced the distance sensor while I worked remotely from quaint Gainesville, GA. Repair went well, 80 mile drive testing all modes and possibilities was good, ready to trailer back to Houston…
During the cold drive the next morning, the tech heard bearing noise while pulling out of the shop. (Previous day’s drives were all warm / hot after calibrating the PDK.) So, we towed home an empty trailer and put the trans on the list for a rebuild.
All new internals (bearings, solenoids, valve body, clutches, etc.) and I’ve now got a new PDK. Todd helped me out a bunch on the labor so I essentially paid for just the rebuild plus the new distance sensor. Although an expensive repair, I’ve now got a complete rebuild with the lifetime sensor from Vlad. I’m happy with the work and recommend Atlanta Speedwerks for any PDK repairs (they had two more out of cars for rebuilds when I was picking up). I’ve not had one other repair or issue with the car in 7 years, 30k miles, and 20+ track days.
Thanks for all the knowledge in this thread (and linked), I was very informed going into the repair.
Nice to read things worked out well and Todd and Vlad were able to get you fixed up. Can you tell us what bearings were replaced? We've talked a lot about the distance sensor and valve body but not much on bearing replacement. Were they going bad and if so which ones? Thanks.
Thanks for the replies on the necessity for PDK oil service. I have noticed the PDK oil temps get as high as engine oil temps at times so I'm assuming it can break down a little, although overall not as sustained temps like the engine oil. The DIY seems like a good option for those on a budget. FYI this was the cost of my last PDK/Gear oil service for parts only, add on labour at $150AUD per hour it does add up a bit. I had quite a few other items done so not sure what the labour component is. My car is more than half track car so that is the main motivation.
PDK filter kit $620AUD (~$420USD)
PDK oil 4.5 x $57AUD (~$175USD)
Gear oil 3.5 x $45AUD (~$107USD)
Labour ~$100USD/hr
He's didn't mention it himself so I'll do it for him. @jjrichar is the guy who figured out you don't need a PIWIS to change PDK clutch fluid and documented it in detail here.
Before I got to know him I tested this myself using his method then repeated the fill procedure using my PIWIS clone. The level was exactly the same, no difference.
People should reconsider paying a dealership to do this (unless you have a really good relationship) as they will likely put the screws to you in both labor and fluid costs. There was a thread over in the Turbo forum where not only did they charge him $45 a liter (more than double Sunset's price at the time), but they also charged him for 6 liters (the dry fill quantity) when only 3 liters comes out in a normal change. They will also likely want to replace the pan/filter whether it's needed or not. If you do go this route please get an itemized estimate up front and caveat emptor.
Great advice and thank you jjrichar and PV997. My last PDK clutch oil change was $590 at dealer two years ago.No filter or gear oil….just PDK clutch oil change. I agree with the opinion to change it much more frequently than Porsche recommends. Happy NewYear.
Good progress on the prototype sensor as bench testing is complete and it almost exactly matches the factory sensor. All programming is complete including temperature compensation and response linearization. The unit is potted and ready for the in-vehicle test that @jjrichar will conduct.
The factory sensor is in the background for comparison. The prototype housing is made from glass-reinforced nylon as we are still working on getting aluminum housings affordable in small quantities. The alignment screw and bolt hole bushings have not been installed yet but will be done by jjrichar. Note that the molex connector is temporary to allow switching out test units and that the sensor will be hard-wired in actual applications.
More detail over in the dedicated thread on this topic:
Hello!
New member here. Read this entire threat which has made my head spin from all the incredible data that's been compiled. I figured I'd add more to the mix. I purchased a 2012 Panamera S from an auction. The primary damage was to the drivers front side. I replaced the lower and upper control arms, the spindle, and the air strut and got everything working well in that regard. Then, when I tried to actually drive the car I encountered the dreaded trans fault error. So then I got a PIWIS 2 clone and began my journey of learning.
To start it was throwing out codes P0841 which correlates to a pressure sensor. I began by checking fluid levels in transmission, which seemed fine. Then tried to perform calibration, which terminated at approximately half way. I assumed there must be a part preventing calibration and stupidly threw money at a valve body, which i later find out on the panamera is actually not the most frequently fauled part but the distancd sensor is. Put the new valve body in and got same issue with calibration, this time the main error has been the dreaded and persistent P1872. Today's calibration terminated at 58% I have taken pictures of all the errors I have currently and would greatly appreciate any help!
A couple things I'm thinking.
One, I haven't gotten any characteristic rod codes that would point me to the distance sensor but from my understanding the pressure sensor is part of the distance sensor harness, along with several other sensors - so I'm wondering if this is the culprit.
Two, there are multiple communication errors with TCU it seems. I'm wondering if it could simply have been a module issue this whole time.
Three, I have been reading @PV997 thought process and am now wondering if there was an issue with me not using fill mode when I put the fluid back in which resulted in trapped air; or if my current TCU has the bad data stored in it and is incompatible with the new valve body.
A few variables complicating the matter:
- my piwis clone throws the low voltage warning every time I start it - I was told by a local indie mechanic that this shouldn't affect the calibration, not sure if this is true.
- the damage that was present on the car when I got it, I'm wondering if something else resulted from the accident that I haven't considered.
One last thing! I'm still learning a lot about PIWIS and am still a little confused. I have the PIWIS 2 and I'm not sure if I saw an option of calibration that distinguishes between part replaced vs not. If I'm missing something I hope someone can help me. Also, when it comes to programming the TCU how would I go about doing that?
Thanks to everyone! I appreciate all the incredible work being done. If it would help for me to pull values/data from PDK/DME I can do that tomorrow after work!
Lastly, this is probably a stupid question but if it is the TCU how does buying a used one work, assuming it has no similar faults stored would it presumably be fine?
Hi @mabdu007 - You probably already know this but the Panamera PDK is different from the units used in the 911, Cayman, and Boxster. They are similar in that they are both manufactured by ZF but much of the specifics of this thread and comments likely don't apply directly. So use caution in interpreting what you read here.
Do you have more details on the original P0841 fault? You are right that it's usually associated with a pressure sensor but there are different variants (open, shorted, not plausible, etc.). Did you do something to clear the fault or does it remain or go away on its own?
You mentioned P1872 but that doesn't show up in my 997 PDK reference list, maybe it's specific to the Panamera. Do you have more info? Based on your screen capture it seems like it's TCU related but I don't know more than that. The main one that has given some people fits is P1871 where the calibration aborts during the distance sensor calibration seemingly due to some sort of incompatibility. However, that's been seen after replacing the DS with an aftermarket unit. In addition, the Panamera DS is a completely different unit than the 911/Cayman/Boxster DS so I don't know how applicable that is here. It probably works on the same principle but I'm not sure.
None of your sensor readings look obviously wrong to me but then I'm not that familiar with the Panamera. One thing that does look a little funny is that "pressure regulator 7 output signal" reads 0.0 mA on picture 3 but then "input signals for pressure regulator 7" reads 500 mA on picture 4. All of the other pressure regulators match between output and input. I don't know if this is significant as I'm not sure how that regulator is used on the Panamera. It might be for an unused option and not significant.
On the 911, the PIWIS has an option for "calibration with part replacement" and "calibration without part replacement" (or something like that, don't remember exact words). Hunt around in the menus as the Panamera should have something similar.
Hi @mabdu007 - You probably already know this but the Panamera PDK is different from the units used in the 911, Cayman, and Boxster. They are similar in that they are both manufactured by ZF but much of the specifics of this thread and comments likely don't apply directly. So use caution in interpreting what you read here.
Do you have more details on the original P0841 fault? You are right that it's usually associated with a pressure sensor but there are different variants (open, shorted, not plausible, etc.). Did you do something to clear the fault or does it remain or go away on its own?
You mentioned P1872 but that doesn't show up in my 997 PDK reference list, maybe it's specific to the Panamera. Do you have more info? Based on your screen capture it seems like it's TCU related but I don't know more than that. The main one that has given some people fits is P1871 where the calibration aborts during the distance sensor calibration seemingly due to some sort of incompatibility. However, that's been seen after replacing the DS with an aftermarket unit. In addition, the Panamera DS is a completely different unit than the 911/Cayman/Boxster DS so I don't know how applicable that is here. It probably works on the same principle but I'm not sure.
None of your sensor readings look obviously wrong to me but then I'm not that familiar with the Panamera. One thing that does look a little funny is that "pressure regulator 7 output signal" reads 0.0 mA on picture 3 but then "input signals for pressure regulator 7" reads 500 mA on picture 4. All of the other pressure regulators match between output and input. I don't know if this is significant as I'm not sure how that regulator is used on the Panamera. It might be for an unused option and not significant.
On the 911, the PIWIS has an option for "calibration with part replacement" and "calibration without part replacement" (or something like that, don't remember exact words). Hunt around in the menus as the Panamera should have something similar.
Thank you so much for the reply! Admittedly this has been getting to me. I invested in this car thinking it would be straightforward. It just makes me feel better there are kind people out here willing to help.
The p1872 code from my understanding is equivalent to the p1871, basically it says "internal fault" and is due to incomplete calibration. The p0841 code strangely enough is still stored as a permanent fault code when I check the "permanent fault code" part of the PDK section in PIWIS. So it's still in there stored. It's description is: signal implausible.
Let me look into the pressure regulator. This is a stupid question but where would these be located?
The indie shop I took it to initially said that he's been able to successfully get 3 out of 4 pamameras back to operation and all 3 required a DS. But, as you said the pdk on this is different so maybe the rod codes I was expecting won't be triggered; thats the only thing stopping me from getting one. But of course it's all a guessing game sadly. At this point my thought process is the same. Maybe TCU. The communication errors are definitely odd. Only thing stopping me from throwing a new TCU at it is that the tcu seems to be reading okay from the values I sent. Or are these values read by something else?
Someone has a used TCU on ebay for 100. Maybe it's worth a shot at that price but im not sure if a used TCU would cut it. What do you think?
Notably when I get the car running for a while to get the trans oil hot enough to calibrate the TCU does feel very warm.
My calibration is also aborting during the shift rod part. I hear the clanking under there but it gets stuck at 58% or so. I'm guessing either the communication faults are stopping it from going further or the DS is bad and that's why it can't continue. Just lost as to what the next best step is. Let me know if I can provide any other data!
And again, my sincerest gratitude for your willingness to help.
Maybe I missed this last time but I connected the PIWIS again just to go over codes and did notice that the gateway, engine, and psm control modules all had TCU communication error codes. The wires around the TCU look fine. I know it's not super common but now I'm really suspecting the TCU is just not communicating.
The only thing opposing that is the fact that values are getting read. So I wonder, is the TCU the one responsible for reading things like pressure sensors and such or is it something else that's reading them? Or another long shot maybe the TCU's ability to receive data is intact but it's ability to output data isn't? At this point it's starting to sound like it has its own temperament.
Let me know if my reasoning is sound.
My 2012 Cayman R [987.2] PDK distance sensor failed while driving in a DE at VIR back in early November 2019. It hasn’t rolled under its own power since. Shortly after my experience, diagnosis at the dealership and offered remedy, PV997 started this thread and I have been a lurker ever since. I joined Rennlist in July 2012 shortly after buying the car, but never had a technical reason to post anything and I’m fairly antisocial, so this is my first post. I must apologize in advance for my longwinded nature, but I guess it’s been pent up for a long time.
When it comes to working on my stuff, vehicles and the like, I consider myself a DIY kind of guy unless I don’t have the tools, instruments or parts to diagnose or repair something myself. I have a technical background but I’m not a degreed engineer. I couldn’t design the complex melding of electronics and precision assemblies that comprise the 4 and 5-axis machining centers I and my team worked on for years, but I understand precision fit and finish, electronics and programable machine logic at a technician level.
I say all this not to blow my own horn but to encourage others in the DIY tribe to jump in and try their hand at this repair. (That said from a guy who only halfway through it currently.) From the jump, I’ve been so taken by the OP, PV997 and his idea that as incredible as the PDK technology is, this isn’t rocket science or neurosurgery that can only be serviced by the high priests of Porsche.
I totally understand those that don’t change their oil or work on their brakes. Time is money and the interest in driving their car is way higher on the list than taking it apart to see how it works or fixing it. I share his attitude that everything is fixable given enough time, parts and wherewithal. The absurdity of having a leaky seal (or a distance sensor module in my case) requiring the entire transmission assembly be replaced is an afront to the DIY tribe of self-reliant technical types out there. I understand that the guys taking these apart in their garages aren’t the average Porsche owner, but we are customers also. The company I just retired from took the same stance on major assemblies and in general all their technology in an effort to block the independent techs out there, so I understand, but still don’t appreciate it or the fact they won’t even let their dealership techs do it. That and I share PV997’s expressed traits of being stubborn and cheap.
I love to make stuff, so using a cheap MIG welder, portable bandsaw, drill press, bench grinder, belt sander and hand tools, I fashioned my own puller jig, a socket wrench to remove the lower shaft slotted bearing retainer nut from the lower shaft and the pusher assemblies to press the bearings onto the shafts and thus the cases back together. I configured a wire harness with USB 5.1Vdc power supply to bench test the distance sensor using my Fluke 87-V voltmeter, which measures frequency and duty cycle % of the PWM circuit output. Pictures of these items are in My Pictures under my username on the forum picture gallery, I think. I’ll see if I’m savvy enough to post them in this thread. I’ll also try and post my bench test results.
I want to thank PV997 and those who have plowed the fresh ground so remarkably and then reported on it with pictures and descriptions, especially jjrichar for the videos of late, taking things apart before our very eyes with great talking points. And of course, for all the encouragement in their posts. Well done all of you! Then there is the entrepreneurial spirit of T-Design and Vlad for designing an upgraded distance sensor with a warranty at a fair price, to my sensibilities anyway. See Porsche, that’s how it's done! Of course, the crowd sourcing you guys are pursuing of the distance sensor body, electronics and testing is simply extraordinary. A rare example of good things happening on the internet!
I am currently awaiting a distance sensor from T-Design and then we’ll see if I can get this thing back together.
Here's the pictures of the tooling I copied from others in this forum. Thanks to all for the concepts! This is my handmade Puller jig I used to split the PDK gear case. This another view (1 of 2) of the handmade slot spanner wrench I fashioned from a 1-11/16' 6-point impact socket. The original slotted bearing retainer nut is shown in the image. This another view (2 of 2) of the handmade slot spanner wrench I fashioned from a 1-11/16' 6-point impact socket. The original slotted bearing retainer nut is mounted to the socket in the image. These are the jigs I fashioned to press the bearings and gearcase back into position. M24-1.25 x 100 hex hd cap screw and M22-1.25 x 70 hex hd cap screw, mating nuts and handmade pusher cups This is the handmade wiring harness, the neodymium magnet, USB Power Supply and voltmeter I used to bench test the PDK Distance Sensor
Interesting, the voltage doesn't change on DS1 regardless of the magnet position. The TCU is looking for the change in DC voltage to measure the shift rod location?
Interesting, the voltage doesn't change on DS1 regardless of the magnet position. The TCU is looking for the change in DC voltage to measure the shift rod location?
Whilst it can be read as a change in DC voltage, it's a duty cycle change as the magnet moves that is being used by the TCU. This can be read as the Hz and % on your multimeter if it has this.
That being said, if the output as a DC voltage isn't changing like the other individual sensors, then the duty cycle isn't changing, which means that distance sensor isn't working. This is exactly the response I saw from Neko's broken distance sensor he sent to me. Whilst the circuits that receive the voltage are still working to send out the basic response, the detection coil or its associated circuits seem to fail so it doesn't give a response to magnet movement.
I'd be surprised if it wasn't measuring the 0-5vdc signal. While the sensor can be measured with the duty cycle, I'm willing to bet the TCU is using the 0-5v to determine the position. In my industry 5vdc systems are common for measuring feedback from sensors, whether that be thermistors, transducers, etc. But, I don't know anything about automotive so there's that. I could be all wet.
My understanding is they use the PWM signal because the engine bay is very noisy.
I also tested this a few months ago to see if I could get a response from PIWIS with a differing voltage. It gave nothing. Our conclusion from this is the TCU is looking for the PWM output.
12-31-2022, 06:39 PM
