mabdu007

On panamera you remove the valve body to get to them. They're visible after pan is dropped. The plug in to them is connected as a sensor pack that also includes speed and distance sensors.
I agree! I don't think the electrical problem is causing any leak, I'm wondering now if this is a totally unrelated problem. Now the question is why aren't the gears engaging? Is it possible a bad pressure sensor could cause the system not to let gears engage? At this time the error codes are the pressure sensor and the p17d1 or gear can't engage codes - its a couple others that have to do with hydraulics.
My only thought now is to swap pressure sensors since theyre the same part and see if the code switches to sensor 2.

mabdu007

Here is a pic of the codes after failed calibration.




​​​​​​​

jjrichar

None of those errors are currently present. Prior to starting the engine to start the cal did you delete all codes? Any cal I've done before always needs codes deleted prior to starting or it will always fail. There are a few codes that it will complete the cal, but only a handful, and it's lists them when you start the cal procedure.
Also, if it's easy to get to the sensors, then swapping is an easy way to diagnose if one is bad.

mabdu007

So I'm not 100% sure at this point. I gave up for the night, but when I tried deleting them initially it told me they couldn't be deleted.
The procedure for the Panamera might be slightly different because it doesn't list anything before cal. But those codes popped up right after calibration failed.
Also maybe the issue is I tried to delete them with the car on. I can try again and see.
are you saying they aren't present because they're greyed out?

byroncheung

Another update - same errors (P1734 & P1764) came up yesterday, only this time I can't seem to clear them and they reappear as soon as the car start, with those errors there I can't do a transmission recalib. Also, this time seems like it's having trouble getting into forward gears as well, not only the reverse. It's still not acting very consistently - yesterday I managed to get it into gear occasionally and made it back home by starting the car/putting it in gear multiple times, going to play with it more when I got time but I think the sensor is really quitting for good this time and I might need to have the car transported to the shop that will do the sensor replacement for me instead of driving there

One side of my brain knew this is what likely would happen if I kept driving the car before having the sensor replaced, but the other side wanted to keep pushing my luck and experiment, and this is what it is now - intermittent sensor error turning permanent.

mabdu007

Two things I noticed today that're very odd.
One, even though the car isn't moving the input signals for vehicle speed is at 2.74 km/hr with engine off and will sometimes jump to 655 and a fraction when engine on. Clearly this seems to be an issue since the system thinks the car is going, right?
Two, the pressure regular 7 I see will jump from 0 to 1 mA as the car is running, all other ones are receiving more current.

I'd very much appreciate your insight here: where do I go to check the pressure regular 7 output signal? Is this a valve body thing? And finally I'd love your input on these things I noticed. Am I overthinking lol or is it possible these issues could mess with vehicle going into gear?

12gauge

I agree. Yeah, my kingdom for the design change to a snap ring, right? I'll file down the longer tab to match the shorter one in some attempt to maintain balance and get on with it. If it ever does come loose, I can imagine it would chew into the webbing on the inside of plastic part Porsche calls the Seal Ring. As you pointed out before, the area is "dry" as the bearings are greased and sealed, so Loctite will be the belt while the staked tabs will be the suspenders for this threaded piece. Still, I'd like to know how others dealt with this oddball size nut and those staked tabs.

I would love to hear some guy somewhere tell us something like "a 1948 Dana spindle nut socket fits those nut slots perfectly" or some such thing, but a hook spanner can't be used because it's back in the aluminum bore of the case and using a chisel and hammer to the slots just would be akin to an unnatural act. But there doesn't appear to be many DIY repairs being done at this time and the Indies have remained silent. So, I fashioned a socket.

The proprietary nature of certain parts is fascinating to me; why Porsche made the design choices they made. I mean really, M8x50 aluminum oval head screws they don't use anywhere else in their product line - or anyone else's product lines that I can find - and an oddball M33-1.5 stake-rim slotted bearing retainer nut, they changed to a snap ring the next model year? And why their Procurement Managers made deals with the suppliers to keep certain parts from the public standing on the other side of the Parts counter. I understand the buyer's deal more than the designer's but will never appreciate the business managers that orchestrated the whole thing. OK, admittedly I might need therapy to get over it. Rant over. I'll press on.

But really, if I had 10 cents for every time as a technician I asked the designer, "what were you thinking?" They usually had a good explanation for it in the industry I came from, but the reason was not apparent at first or second glance.

Keep up the good work and thanks for the prodigious contributions, Jeff. It spurs me on. ~Cheers!

jjrichar

During our testing of our new sensor and then installing the OEM again, we've learned a bit about what the 1764 code is about. When doing a cal, it pushes each shift rod into each gear and sees what the corresponding response from the distance sensor. It stores this and then is looking for this each time a gear is selected. It also seems to look at the response with other possible gear pre-selections, as adjacent magnets can affect the response.

So for example when doing the cal 1st is selected and the response from the distance sensor is 9.0mm. Afterwards when driving you put it into gear, it selects 1st, but for some reason the distance sensor response is 8.0mm. The TCU says 'that's not enough', so it skips the gear and moves to the next, and will continue to do so until it finds a gear that it's happy with. With our prototype the response was too sensitive. When I installed the OEM sensor after the prototype testing, I didn't do a cal and selected 1st, the response wasn't enough. After the cal from the prototype it was looking for a distance of about 10mm, but the OEM sensor response was only 7.5mm, After selecting 1st, I rapidly saw it automatically select 2nd, 3rd and then finally 4th where it found a gear where the response was enough compared to the cal from the prototype.

What was interesting is that the TCU logic seems to be that it's looking for a distance that is not less than what is stored from the cal. eg. If looking for 7.5mm and it gets 10.0mm, it seems to think this is OK. If looking for 10.0 mm and gets 7.5mm, it says that's not enough, the gear hasn't engaged hence the failure.

So your symptoms point to doing a cal and getting all the numbers in the TCU, but later when gears are selected the response is different. The 1734 will occur when you get an implausible response from distance sensor 4. To me this all points to a distance sensor issue, with the response being either at times inconsistent or at other times implausible.

Have you tried unplugging/re-plugging at the transmission and also the TCU? I've had issues with this type of signal with other transmissions that was just a contact problem at the plug. Moving the contacts to get a good connection immediately fixed the issue.

jjrichar

Can you confirm all those errors in the image appeared after you did the cal and it was clear prior to starting?
If so, there seems to be a significant electrical or TCU issue if it's not talking to other control units in the car. It might just completely shutdown due to the low pressure problem. It's hard to know the follow on effects when something significant fails. My approach would be to find the source of the leak and fix that first. This is the known problem, and if it's causing a low pressure throughout the remainder of the transmission, this might be causing the issues.

If you turn the car off, disconnect the battery, reconnect and then get into the TCU with PIWIS without starting the engine, can you clear all the codes? The TCU and DME can't have codes cleared with the engine running.

Noah Fect

Porsche just buys the gearbox from ZF, so it's likely a part from ZF's standard in-house catalog.

It won't be true quite yet, but at some point in the future, it will be likely that the contributors to threads like this one will know more about the 997-era PDK's internal operation than anyone at Porsche AG. ZF is likely keeping secrets from their customer, just as that customer keeps them from us.

PV997

Pressure regulator 7's current seems odd which I noted upthread. That being said, I have no idea what #7 does or how it's used in a Panamera PDK. I'd look for more info on that prior to blaming the problems on it. It could be there for an option (like a locking differential) that your car doesn't have. If that's the case then it would make sense that it's reading is zero.

Sorry I can't be more help but the Panamera PDK is a much different beast than that used on the sports cars. I don't have much info specific to it.

byroncheung

No I have not tried unplugging the transmission or the TCU - are those plugs easy to access to?
Also, yet another update - i left the car overnight, tried again, managed to clear 1734, and did a calibration, and now it's working again.
I can try to unplug those if it's easy to get to, see it would stop 1734 from coming back...

jjrichar

Other 997 owners will help on where the TCU is in your car. PV?
Plugs on transmission are pretty obvious. The one you want to find is at the rear of the transmission (towards the front of the car in a 911). You don't have to completely disconnect. It has a lug on the side of the plug that you just rotate, which will move the plug rearwards and move the contacts against each other. Just move the lug back and forth a few times and then return to the original position. If you can see any oil or fluid that might have gotten into the plug I'd disconnect and clean with contact cleaner.

It's a long shot doing this but the simplest thing to start with when getting inconsistent output from an electrical item.

PV997

In US 911's it's directly behind the driver's seat under the door. You'll need to pull the back seat side interior trim panel, the trim piece that runs along door sill (with the switches in it), and the plastic piece that is hidden under the sill trim. With that removed you can peal the carpet back and access the TCU.

This picture shows the side trim still in place but it's easier to just remove it IMO (I have a Cab, may be different on Coupes). If you unhook the strap over the TCU it will lift out and make accessing the plugs easy.

byroncheung

Yeah, I agree that it is a long shot, one pattern in my case is the error always pops up when the car start, but never comes up in the middle of driving / if the car has already started - I imagine if it's connection not tight it would tend to come up during driving from vibrations disturbing the connections.

But I will try to see if I can give the connections a checkup...