Guide to Repairing a PDK Transmission
#1246
Instructor
I think the constant here is the P0700 DTC; believe that may point to the TCU. I did a very basic Google search, however and saw someone else with a similar issue on a 991.1, and the problem there ended up being bad wiring, so maybe double/triple check all the wires and connections as well?
since the t-design distance sensors don't come with a new harness - how commonly do we see the distance sensor harness being an issue?
#1247
Instructor
I would check the pins at the TCU side connector. I've seen these wearing out to where they don't make good contact with the pins on the TCU. This would be more likely if you've had the connector on and off multiple times. Using a meter to probe can also cause this (don't ask me how I know). You can try to wiggle the wire bundle and or the connector while monitoring PIWIS for faults. This type of pin is easy to remove from the housing without a special tool (just need a small screwdriver). If you need to replace them you can get new ones here (this is the smaller size, the larger ones you can adjust the clamping pins with a small screwdriver if weak):
https://www.mouser.com/ProductDetail/571-5-963715-6
You will need a good crimping tool though.
https://www.mouser.com/ProductDetail/571-5-963715-6
You will need a good crimping tool though.
#1248
Rennlist Member
I mentioned this before but I'll state it again. The PDK transmission control unit (TCU - inside the car) stores adaptation values that are matched to the PDK. These adaptation values are characteristics that are developed over time using an algorithm that matches the PDK performance to the driving style and PDK wear. If you run the Vehicle Analysis Log you can see them, there are literally thousands of values unique to your PDK. I'm not certain but I don't think these adaptation values are completely cleared/updated by a calibration as they have different purposes. (Edit: see amet's comment below about two different types of calibrations, it's relevant.)
There are many reports of shops claiming the need to replace the TCU ($800) after a PDK replacement in order to get things to work. When they do this they are switching to a TCU with a clean slate (no stored adaptation values) so the synchronization, calibration, and adaptation goes smoothly. Do some searching around and you'll find multiple reports of this.
I do not think a new TCU is needed but I do think it may be required to clear the adaptation values and reinitialize the TCU. The reason for this is that when the transmission was malfunctioning the TCU may have developed unusual adaption values in an attempt to cope with the problem. Now that the problem is fixed the adaptation values are so out of whack that the TCU senses things are beyond the normal range and throws an error.
T-design is almost certainly right when they state this is not caused by their unit. Your P1731 codes are gone, this is a software fault as the document T-design posted shows. The PDK itself has no software whatsoever, it's simply a dumb box with solenoids and sensors. Software errors are in the TCU. Search around in your PIWIS for initializing the TCU to get it back to a clean slate. Your comment, "I have some contacts do Germany and will check but I have the suspicion that the sensor delivers values outside of the tables in the control unit." (deleted)
Edit: After thinking about it you might be right in that the sensor is supplying values outside the table range, but the problem is not with the T-design unit. The problem is with the tables in your TCU. I've edited out the sentences I had above as I probably read more into your comment than I should have.
BTW, I'm an electrical engineer that knows a lot about the PDK but in no way am I qualified to tell you how to use your PIWIS on your car. You can do some real damage if you are not careful so please do your research and tread lightly. Most of the "software faults" in the Porsche T-shooting document state that the fix is to replace the TCU. If it were my car I would try TCU initializations with the PIWIS before replacing it. That being said, a new TCU is not very expensive in the big scheme of things.
Good luck and please provide updates.
There are many reports of shops claiming the need to replace the TCU ($800) after a PDK replacement in order to get things to work. When they do this they are switching to a TCU with a clean slate (no stored adaptation values) so the synchronization, calibration, and adaptation goes smoothly. Do some searching around and you'll find multiple reports of this.
I do not think a new TCU is needed but I do think it may be required to clear the adaptation values and reinitialize the TCU. The reason for this is that when the transmission was malfunctioning the TCU may have developed unusual adaption values in an attempt to cope with the problem. Now that the problem is fixed the adaptation values are so out of whack that the TCU senses things are beyond the normal range and throws an error.
T-design is almost certainly right when they state this is not caused by their unit. Your P1731 codes are gone, this is a software fault as the document T-design posted shows. The PDK itself has no software whatsoever, it's simply a dumb box with solenoids and sensors. Software errors are in the TCU. Search around in your PIWIS for initializing the TCU to get it back to a clean slate. Your comment, "I have some contacts do Germany and will check but I have the suspicion that the sensor delivers values outside of the tables in the control unit." (deleted)
Edit: After thinking about it you might be right in that the sensor is supplying values outside the table range, but the problem is not with the T-design unit. The problem is with the tables in your TCU. I've edited out the sentences I had above as I probably read more into your comment than I should have.
BTW, I'm an electrical engineer that knows a lot about the PDK but in no way am I qualified to tell you how to use your PIWIS on your car. You can do some real damage if you are not careful so please do your research and tread lightly. Most of the "software faults" in the Porsche T-shooting document state that the fix is to replace the TCU. If it were my car I would try TCU initializations with the PIWIS before replacing it. That being said, a new TCU is not very expensive in the big scheme of things.
Good luck and please provide updates.
Multiple attempts to run the calibration (with or without parts replacement) failed and the alert remained. A TCU reset also had no effect. I then re-programmed the TCU (I used the option with no customer complaints, not sure what the difference is). After doing that and coding it as a USA vehicle I was able to clear the P1872 code and successfully run a Calibration after Parts Replacement. I cleared a few DME CAN faults that resulted from the Cal and now the car is happy again.
I think this supports PV997s theory that old adaptation / calibration values can interfere with the calibration procedure after parts replacement. In the future I'll just re-program first after any parts replacement to ensure I start from a clean sheet.
Last edited by stjoh; 04-29-2023 at 04:24 PM.
#1249
Rennlist Member
Haven't touched the TCU connector (at least not under my ownership, and I doubt if previous owners did), but I'd love to double-check the connection. But gaining access to the TCU seems like a pretty involving process of pulling multiple trims? Or there is an easier way to gain access?
#1250
Instructor
Here's another experience supporting PV997s suggestion that one should start with a clean slate software wise after position sensor replacement. After replacing the position sensor in a 2010 Cayman Base with another OEM unit I tried to run Calibration after Parts Replacement. Not only did this fail but it resulted in a P1872 fault. The car that previously drove fine despite not having a calibration was now a paper weight.
...
Multiple attempts to run the calibration (with or without parts replacement) failed and the alert remained. A TCU reset also had no effect. I then re-programmed the TCU (I used the option with no customer complaints, not sure what the difference is). After doing that and coding it as a USA vehicle I was able to clear the P1872 code and successfully run a Calibration after Parts Replacement. I cleared a few DME CAN faults that resulted from the Cal and now the car is happy again.
I think this supports PV997s theory that old adaptation values can interfere with the calibration procedure after parts replacement. In the future I'll just re-program first after any parts replacement to ensure I start from a clean sheet.
...
Multiple attempts to run the calibration (with or without parts replacement) failed and the alert remained. A TCU reset also had no effect. I then re-programmed the TCU (I used the option with no customer complaints, not sure what the difference is). After doing that and coding it as a USA vehicle I was able to clear the P1872 code and successfully run a Calibration after Parts Replacement. I cleared a few DME CAN faults that resulted from the Cal and now the car is happy again.
I think this supports PV997s theory that old adaptation values can interfere with the calibration procedure after parts replacement. In the future I'll just re-program first after any parts replacement to ensure I start from a clean sheet.
#1251
Rennlist Member
@stjoh Is there a post/thread that documents what is the procedure of "re-programming" the TCU to its clean state and subsequently coding it as a US car? Tempted to try it on my car to see if it will help the intermittent errors that come up after replacing the distance sensor.
It will ask for a 40A charger but in my experience it works fine with a fully charged healthy battery (perhaps backed up by a regular charger). I like how they left a bit of German in there.
The programming doesn't take very long.
Selecting "USA" is done under the "Codings adaptations" tab. Only selection there.
Last edited by stjoh; 04-30-2023 at 01:51 PM.
The following 2 users liked this post by stjoh:
byroncheung (04-29-2023),
tony9189 (05-03-2023)
#1252
2012 Panamera S - PDK fixed
Just wanted to come back and post an update in case anyone else can benefit from my experience. I bought an auction panamera about 6 months ago that would not shift out of neutral. Trans fluid was low on initial check and it was topped off. No luck. Continued by changing distance sensor, valve body, trans oil pressure sensors, nothing. Finally took it to a local indie shop and decided to put a used trans in. Paid 2500 for a trans with 70k miles. After calibration, it's running perfectly.
when looking at the "bad" trans I noticed the input shaft seal was gone. Fluid had been leaking and I'm assuming system wasn't pressurized at all. I had noticed this leak prior but didn't think a leak could prevent the transmission from moving at all.
So, in short. Trans problems with obvious source of substantial leak could be a culprit in preventing successful calibration. I realize mine is a very specific problem. Leak was from bell housing.
Hope this can help at least one person. I'm not sure why the input shaft seal would go bad but maybe just wear and tear? I don't think the original accident caused any damage to the seal directly, I suspect it was already bad.
Thanks to everyone for all your help. I truly appreciate it.
I also now have a valve body, wire harness, distance sensor, oxygen sensor and a rebuild able transmission if anyone is interested.
again, my many thanks!
when looking at the "bad" trans I noticed the input shaft seal was gone. Fluid had been leaking and I'm assuming system wasn't pressurized at all. I had noticed this leak prior but didn't think a leak could prevent the transmission from moving at all.
So, in short. Trans problems with obvious source of substantial leak could be a culprit in preventing successful calibration. I realize mine is a very specific problem. Leak was from bell housing.
Hope this can help at least one person. I'm not sure why the input shaft seal would go bad but maybe just wear and tear? I don't think the original accident caused any damage to the seal directly, I suspect it was already bad.
Thanks to everyone for all your help. I truly appreciate it.
I also now have a valve body, wire harness, distance sensor, oxygen sensor and a rebuild able transmission if anyone is interested.
again, my many thanks!
#1253
2013 981s PDK fix
Hey All.. So in the middle of fixing my PDK ZF.. I have it all done for one slight peice.. How do you close it. I see tons of threads on how to open with custom tool. Well, what are the thread size at the end of each shaft??
#1254
PDK Tool for closing case
Hey all... in the thick of it and now I'm at a cross road! How do I close the case? I have seen many tools and description on how to open case but very few on how to close. Any directions/tools out there to close case? What is the size thread on end of both shafts? Also, what type of oil is recommended for the gear side? Thanks in Advance!
The following users liked this post:
Dynamicp (04-30-2023)
#1256
Rennlist Member
Hey all... in the thick of it and now I'm at a cross road! How do I close the case? I have seen many tools and description on how to open case but very few on how to close. Any directions/tools out there to close case? What is the size thread on end of both shafts? Also, what type of oil is recommended for the gear side? Thanks in Advance!
The following users liked this post:
Dynamicp (04-30-2023)
#1257
Instructor
another question - i've only tried calibration without part replacement before - i imagine the with part replaced calibration is largely similar?
#1258
Rennlist Member
hey @stjoh thanks for your detailed response with the pictures! think i will give it a shot - clear the tcu and do a calibration with the part replaced.
another question - i've only tried calibration without part replacement before - i imagine the with part replaced calibration is largely similar?
another question - i've only tried calibration without part replacement before - i imagine the with part replaced calibration is largely similar?
The following users liked this post:
byroncheung (05-01-2023)
#1259
Hi, So I replaced both speed and distance sensors with the T desing units, but im running into some issues with the calibration.
I have tried more than 20 times no matter which PIWIS i use (tried 2-4) without luck. the Error shows when the procedure of the calibration goes into calibratimg of operating distances, it goes up to 94% then it stops, some error shows then it says cannot be completed
I was told that I need to do a bleeding for the PDK? I cannot find that option on PIWIS (maybe the fill option is the bleeding?)
you can drive the car and it will shift into all gears, but sometimes it grinds 4th gear and after that it goes into limp mode. the error that its on the TCU is P1870 something about the calibration not done correctly.
attached images of PIWIS measurements
I have tried more than 20 times no matter which PIWIS i use (tried 2-4) without luck. the Error shows when the procedure of the calibration goes into calibratimg of operating distances, it goes up to 94% then it stops, some error shows then it says cannot be completed
I was told that I need to do a bleeding for the PDK? I cannot find that option on PIWIS (maybe the fill option is the bleeding?)
you can drive the car and it will shift into all gears, but sometimes it grinds 4th gear and after that it goes into limp mode. the error that its on the TCU is P1870 something about the calibration not done correctly.
attached images of PIWIS measurements
#1260
Instructor
well just to report back - same error (p1732 p0700) came back in driving 10 miles or so after reprogramming TCU and recalibrating, so didn’t really help, but hey didn’t hurt to try…