[jjrichar]

I haven't put this in the DIY section with the other videos as it's just a request for help.

Background:
The sensor that fails most in the PDK is the distance sensor, which is the one that senses the position of each of the shift rods. I'm trying to gather the required information so it's possible to fix the sensor yourself.

Information below regarding aftermarket sensors is current as of Dec 22.

The only replacements that seem to be available are aftermarket. Of the two I'm aware of, only one lists the price of the replacement sensor. It's made by a Canadian company XemodeX, and the list price is US$2199. It's a company that seems to specialise in aftermarket automotive products. The sensor they provide comes with a new plug, but you will need to splice your existing speed sensor into the new wiring harness they provide. I have no information regarding this sensor's performance.
Porsche PDK Shift Fork Position Sensor New | XeMODeX Inc.

The other aftermarket option, T-design, has a price of US$1890. You will need to contact the company for the pricing, as it's not listed online. Later in this thread T-design give this as the sensor price. Unlike the XemodeX sensor it doesn't come with a new plug, and requires you splice the six wires of the distance sensor onto your existing harness. I have heard good reports on the sensor's performance elsewhere online.
PDK dispacement distance gear position sensor for Porsche 7DT45 7DT70 transmission

From the research I've done it seems the required sensors to fix it would cost a few dollars only.

The OEM sensor from what I understand uses old school technology (apparently an inductive coil), and the aftermarket option uses hall sensors.

The sensor detects the position of a magnet that is attached to the shift rod. Maximum distance detected is 12.7 mm either side of a central position. Movement of the shift rod to select a gear seems to be about 9 mm either side of the central position. PIWIS shows a 7.5mm movement when a gear is selected, but it's actually about 9mm, which is easily measured via numerous other means.

The sensor gets a common 5V supply and ground with a single output from each sensor that gives the position.

EDIT: Response from the sensor is a PWM signal either apprx 1000 Hz or 860 Hz. Alternating sensors have either a 1000 or 860 Hz response. The TCM seems to measure duty cycle of the return, and is frequency agnostic. The TCM will perceive a valid and correct distance regardless of if that channel receives a 1000 or 860 Hz response.

1st video below shows my initial testing, where I thought the output was a simple DC voltage. That has consequently been found to be incorrect. It's definitely a PWM signal. Also, the TCM won't respond to a simple DC voltage. It needs to be PWM within certain parameters.

2nd video shows consequent testing and the PWM response.
3rd video shows how to test sensor outputs at the TCM
4th video shows replacement of the sensor.

 

[jjrichar]

I guess the question is, what are you looking for? I presume you're looking for a way to replace/fix this unit? Since it's potted, I don't think fixing an existing one is in the cards for most people (though you may have the skills to do this, from what I've seen). You mentioned that the factory one uses coils... do you have detail on what those look like?

Hall sensors are certainly a better way to go, as they are off-the-shelf, usually far more durable. What you'd need is a linear hall sensor, which is designed to measure distances.

I'm only really interested in getting out as much information as possible, and hopefully a solution or two might spring out of it. Whether that's enough information for others to fix their own sensors, someone creates a kit so you can fix your own, or someone says 'send me your sensor' and I'll fix it.

If there is something in this transmission that will fail the most likely thing is this sensor from all reports. Whilst it's good to at least have aftermarket options rather than replace the transmission completely, this seems to be priced far higher than what it would cost to do it another way.

There is a good thread on Rennlist that has been running for a number of years about how to fix the PDK. It seems to be based on information been provided from a number of sources, including workshops that work on the transmissions. Workshops are pretty cagey about giving a lot of hard earned information away, so a lot of it is speculative.

Using a linear hall sensor looks like a a very simple option. It's just finding the correct specification so the sensor response from the movement from the shift rod magnet is correct. It also needs deal with the temperatures normally seen in the transmission.

I've done a bunch of searching to find a sensor that might fit the bill, but for someone who has no experience in the field I have very little idea what I'm looking for. I don't know how to test the shift rod magnets effectively for strength so I can then find a sensor that gives a correct response over the range that is needed.

Link here to a few sensors that to my eyes look might be suitable. No idea if they are or not.
https://www.digikey.com.au/en/products/detail/allegro-microsystems-llc/ALS31000LLHALX/6624074

This one looks like the SOT 23 version mounted in the center of the sensor movement range might be what is required.
https://www.ti.com/lit/ds/sbas639c/sbas639c.pdf?ts=1668980295818&ref_url=https%3A%2F%2Fwww.google.com.au%2F

 

[jjrichar]

More than happy to do the heavy lifting.

I'd like to keep the discussion here so others can contribute if able.

Below is a snip from the texas instruments sensor. Any idea which of the 5V varieties might be suitable regarding sensitivity? The shift rod magnet is a cylinder sitting on its side and running parallel to the sensor. Its 10mm diameter and 6mm long. I suspect it's either N50 or N52. It seems to be slightly less magnetic than the other N52 magnets I had on hand that I was using to test. They were N52, 12 x 14 x 3 mm. Closest the magnet will get to the sensor is about 3mm directly above. Furthest away about 12mm.

 

[jjrichar]

The original sensor is reported to be a coil, not a hall sensor. We're trying to make it a hall, as there is an aftermarket company who has done this. Our effort is simply to see if we can get there, publicly, so others can do this work themselves as well.

@jjrichar As for the spec. First step, are you sure you're getting 5V from the car-harness for these coils? If so, then yes we can run a 5V sensor... otherwise we may have to put in some voltage regulators or something. Can you confirm what voltage the car delivers?

As for actual components, I would get a few of them (they are low cost) and see if anything gets us close to start with. I think the key spec to watch is the temperature, to find something that is well tolerant. I don't think we can guess at the specs otherwise, since we're coming from non-hall to hall-sensors anyway... but after testing a few you may find one that works, or something that gives us guidance as to what is most important. For example, you mentioned the magnet feels weaker, so a "longer range" one may be needed, so that it's more sensitive to the weaker magnetic field. That's why it's best to grab a few and try them, I think. Be sure they're "linear" hall sensors, of course.

If you still want to dive into specs, then we can consider trying that... I'd need to know what distance we're dealing with between where the magnet is and where the sensor "would be" inside the potting.

It might be good to get a bad assembly, and clear out the epoxy so you can see what space you have to work with inside. This would also give you a way to build test assemblies and double-stick-tape sensors at varying heights or something, so you can try various distances as well.

I think this needs to be an iterative exercise, as we are going off-script for what the original part did... but I can dive into math if we really have to go that way.

Definitely 5V. It has this on the wiring diagram and it also gives the voltage value in PIWIS.

The texas instuments sensors are designed for automotive applications, with a rated temp or -40 to 150 C. I haven't been able to find anything better than this so I assume this would be fine. It's definitely a linear sensor. So it might be just a case as you say of getting some and having a play to see the response.

 

[jjrichar]

-40 to 150C is great... yeah, I'd say get a couple of those and run with it.

As you noted, initially speaking, it may not operate in the "direction" you want, and we can fix that. The first test would be to see if you can get it to have a similar "span" for voltage (regardless of direction/polarity) when at an appropriate distance (hence digging out space in an old part so you can space it as necessary).

Keep in mind that you may be able to change the direction by orienting the sensor in different directions - ie flipping it over or spinning it 180-degrees may fix that as well... but if not, it's easy enough to fix that with circuitry.

Ordered some sensors from mouser. Can you have a look at the doc in the link below. Stuff I don't really have a full understanding of is the specifics of the circuitry. It seems a capacitor (1 microfarad) needs to be placed in parallel with Vcc and ground and it also talks about an additional resistor. I'll be using 4 sensors in parallel. Do I need a single capacitor of the same capacity or do I need something larger/smaller. From my understanding the capacitor is to control the initial buildup of current when it's turned on. Correct? Can you give me a basic description of what is required and why. Thanks

https://www.ti.com/lit/ds/symlink/drv5055-q1.pdf?ts=1669240411148&ref_url=https%3A%2F%2Fwww.ti.com%2Fproduct%2FDRV5055-Q1%3FCMP%3Dconv-poasamples

 

[jjrichar]

For now, I wouldn't worry about the "optional noise filter"... chances are, there is filtering present in the car already, and if there isn't we can visit that later. The resistor is for "fault detection" which we don't need either.

The cap at the voltage input is just to provide some basic filtering from noise on the input voltage. On a bench, you won't likely have much of this anyway, but it doesn't hurt to have a cap to start out with. Note that it says minimum of 0.01uF. Since you're using multiple sensors, you can certainly go higher and use a single to cover them all. Whatever you have above 0.04uF would be fine, I'm sure... don't overthink it, haha.

Be sure the sensors are all wired parallel to the Vcc and ground... don't wire the sensors in series. (You said you were doing this, but I just want to make sure you keep doing it, haha.)

A few other comments from the manual:
1) Note that the device "switches" internally between two different operating voltages... and that changeover happens right at 4.5V. For this reason, I'd recommend getting a true 5V power supply instead of the 3x 1.5V batteries you've been using so far. A typical USB charger should be able to get you this (get a USB cable and cut the end, if needed)... you may have other options, of course. Ideally, if you can get a true benchtop power supply, that might be ideal... because the USB chargers can be VERY noisy.
2) Note there are 5 different versions of these sensors (A1-A5)... not sure which one you are getting, but be aware of this.
3) Section 8 in the manual is worth a quick read... they give you some guidance on how these sensors are intended to be used, orientations to the field, etc... something we should keep in mind as we work forward.

Our first step goal is to get the proper voltage range for the car's expectations. This is where spacing and orientation is important, so that's the first challenge - can we voltage movement at the ranges we have with the existing magnets. Once we get movement, we can adjust the voltage to meet what the OEM delivers via external circuits... so let's focus on the first thing - getting voltage movement.

One thing that might help is to test with the OEM part and see what the range is for the voltage. So, while you're waiting for sensors to arrive, put the OEM part back in, and see what you get for voltage ranges at varying shift positions. I don't recall if PIWIS will tell you this, but you can also probe this with a volt-meter if that's easier for you. See if you can record the min/max in actual operation for each sensor (remember, we want actual shift positions, not the full range of the sensor... so maybe you just do this with your working trans in the working car setup?) If you can't get them all, that may be ok... get what you can.

Lots of good stuff for me to think about while the sensors are in the mail.

I ordered a few of each of them. A5 is just a reverse version of A1, so I didn't bother with it.

You don't get the voltages for each of the sensors in PIWIS, just the position. You do get the input voltage, which was a little over 4.8V from TCM in the project car. I'll see what I can get.

Figure 21 in section 8 is exactly the way it will be installed. It seems that it's just a case correct orientation reference the magnet and then finding the correct sensitivity to get the correct voltage response.

 

[jjrichar]

Apologies for the poor initial information. After testing using different modes on my multimeter, this is what I found it's clearly a PWM signal. Making a quick video to show all that I found. It's a bit wacky how it's all setup.

 

[jjrichar]

Video here of what I found

 

[jjrichar]

Exactly, that's why I was suggesting taking one apart and trying to identify the actual component, if possible. Guesswork isn't what's called for here.

FWIW, if the outer shell is aluminum, I can X-ray it if someone wants to send one up here (Pacific NW). That could potentially identify the part, or at least narrow it down.

Appreciate the input but we aware of the internal components of the OEM sensor. Go to the link below and then click on the 'detailed failure information' tab. It has an X ray of what is in the OEM sensor as well as what's in the aftermarket version.

The OEM is from what I understand seriously old school, with sensors superseding it since that are vastly cheaper and more reliable. Hence why the aftermarket solutions use these and why we are investigating this route.

Porsche PDK Shift Fork Position Sensor New | XeMODeX Inc.

 

[jjrichar]

Just tested the car again and they are all either 1040 Hz or 860 Hz plus or minus a bit, same as the other sensor by itself. PV, If you have accurate equipment I suspect what you are seeing is correct and I have an instrument error.

Also, duty cycle is either 21%, 50% or 79% for all plus or minus 1. Had the car in rolling test mode on the lift where I was able to run the car all the way through the gears and test. I'm not sure how accurate this would be considering the possible errors above.

For testing of the new HE sensors I guess it shouldn't make any difference. Just as long as the existing sensor and new do the same thing.

 

[jjrichar]

@jjrichar (and everyone else): I emailed my best-friend/EE over the weekend, and just called him this morning... he was in a meeting, so I couldn't dive in deeper, but he says that there are other electronics in that chip already and he believes that we can generate a 1kHz PWM from that chip directly. I certainly didn't see that in my quick reading, but I will chat with him later on and get more detail on it... so don't toss out those chips just yet!

EDIT: He may have been talking about a different variant of that product line, and maybe I misunderstood, but we'll see... I'll report back as soon as I have a chance to chat.

EDIT2: Ok, so his statement was that there was probably a timer chip (eg a 555 timer) inside the epoxy as well, that generates the PWM. He said it's stupidly-easy to recreate that for our 1kHz PWM if we wanted. But he's feels it's silly that Porsche would go through all that trouble, and he thinks they are probably averaging the PWM on the car-side of things, and using that DC level to determine position. If that's the case, we can skip the PWM and just use the analog DC-level ourselves, saving ourselves the hassle.

So, to test this, the suggestion is to feed DC voltages into the car, and read the positions from PIWIS, to see if they correlate linearly. If they do, then that's what they are doing... and we, too, can skip the PWM completely. So, if you have a bench-top variable power supply, you can simply use that to feed voltage into the car at a few levels (eg 1V, 2V, 3V, 4V) and record what the PIWIS readings are at each level. I can detail the wiring setup further if needed.

Alternatively, if you don't have a variable power supply, then you can use single, double, and triple-battery setups and do the same (ie 1.5V, 3.0V, and 4.5V). Again, I can detail the wiring if needed... but basically you want the negative on the shared-ground-wire, and the positive will go to whichever signal-line you want to test.

I this test shows a linear relationship, then we have our answer, and can skip PWM. If they are really using the PWM in the car, then these would all read "0" or some max value, since our battery/power supply is only putting in DC voltages (ie, there is no pulse).

DISCLAIMER: All of this should be safe to do, as it's low voltage and well within the input levels of these signals. However, we are testing and doing this "off the reservation" and mistakes can happen, so I don't want anyone coming after me for damage caused to anything. I would feel comfortable doing these tests on my own car, but if you mis-probe or miswire, I don't know what could happen and bad things might happen. I just don't want someone coming after me for damages (the USA can be very litigious!)...

Thanks for the input. I went and tested but unfortunately it gave an invalid signal with a DC source at the sensor output. No harm trying.

PV and I have been emailing offline about the problem and one thing we were concerned about was the differing PWM freqs of alternating sensors. So I swapped output wires for sensors 3 and 4 at the TCM (these have different PWM freq outputs) to see how the TCM responded. It read each as a perfectly valid signal within PIWIS and the distance was correct, just swapped around.

My conclusion from this is the TCM is looking for a PWM signal within certain parameters that is wider than the 860-1046 Hz range that I've seen from sensors. It then measures the duty cycle of the signal to determine magnet/shift rod position.

PV has ordered some sensors and programming kit that look to be suitable. Link below. They are good to 170 degC, and have programmable PWM and sensitivity. Hopefully it's going to be a relatively simple task of achieving the correct sensitivity and placement.

By PVs estimation there is less than $15 of electronic components in there total. He's designing the circuit board, which he thinks he can get pretty cheaply, and then we need a housing and epoxy fill. The only thing where I think we need some real help is the housing. PV is thinking that a 3D printed version would work well, but we need the input of someone with knowledge in this area to help out. Anyone?

Also, I'm going to change the thread title so it reflects where this thread is going, which is a community project to create an affordable distance sensor.



https://product.tdk.com/system/file...linear_hall-effect_sensor_with_pwm_output.pdf

 

[jjrichar]

Otherwise it really looks like some kind of crusade against small business, that makes parts superior to Porsche in quality and warranty, while comparable in price! Hope everybody understands that our resources are incomparable to Porsche, and it costs Porsche 10 times less (per unit) to make, support, and market parts.

My apologies if any offence has been taken with the posts above. I've amended those you refer to so any speculation has been removed. I've tried to be as objective as possible about the information I've provided. If there is something here that you think is incorrect or simply speculation, please let me know. Please understand though that the availability of information on anything related to the PDK is pretty thin, and the community is trying to gather what it can from what little is available. I was just repeating what I had read elsewhere online.

I'm certainly not on a crusade against small business. One of the reasons that forums like this one exist is for people to find alternative solutions to fix their car. What we are doing here is just that.

My personal opinion is that US$2199 for a replacement XemodeX sensor is far more than it should be. From our research it seems that a price that is a small fraction of this would be appropriate considering the cost of components, even at very low production numbers. We are just attempting to find an alternate solution.

To clarify, if I wanted to purchase one of your sensors to fit it to a car myself, what is the price? Your website seems to list the price of the other products you make but not the distance sensor. Your comments above seem to suggest your price is far lower than the XemodeX sensor. I think it would be valuable to the discussion to have this information.

I also want to clarify a few things for the general audience. This isn't speculation, but is based on my first hand experience. I bought a car to completely disassemble and make videos of everything I could. I was then very kindly donated a PDK to completely tear down and again make videos of every step. My aim was primarily to lift the veil of knowledge from these things and show that it's just not that hard.

The Porsche sports car is designed like nothing else I've ever worked on. If there is anything the project has taught me it's that the engineers who designed these cars did stunning job of making maintenance tasks easy, especially considering the engine is stuck in the middle of the car. I laugh at how easy sometimes. Having owned and worked on other cars made by the VW/Audi group, I'm amazed they all come from the same parent company as the contrast in how they are designed could barely be greater.

ZF transmissions are the same. I've pulled apart a few and I'm always amazed how easy they make something so complex.

Being able to complete tasks on the car is just a case of having the information and parts. All we are trying to do here is fill the information void that would otherwise be quite empty.

So to my points of clarification:

1. This is a 981 thread. For the 987, 981 or 718 the transmission doesn't need to be removed. Access to the distance sensor is easy. Exhaust, subframe, coolant and oil cooler (981/718 only). Remove rear casing. Remove sensor. Splice 6 wires. Install sensor and then put it back together again. Fill with coolant (981/718 only), start car, calibrate. Definitely a weekend DIY that's possible for someone who can turn a wrench in my opinion.

2. In the 911 the transmission needs to be removed. Yes, more work but not onerous.

3. There is a post on the Rennlist thread I referred to above by a workshop that say they have installed the T-design sensor. They quote a minimum of 28 hours of labour for a distance sensor replacement. Up to a maximum of 45 hours. Threads like this one here are to simply provide a possible alternative to this. Not just the labour cost, but also parts alternatives.

If you disagree with anything I've written here please let me know. I want to be as open, objective and honest as possible.

 

[jjrichar]

It's true that on Boxter/Cayman it's possible to open PDK without taking it off the car. But, all serious shops still do. Not to gauge the price, but to provide a quality service. Opening PDK without damaging it beyond repair is not a trivial thing and it's much easier off the car. To ensure problem-free separation some shops use techniques that are impossible on-the-car. I mean you have a 60-80% chance that everything will be fine, but then you have 20-40% chance that case will crack trying to pry it open in limited space under the trunk... and then you are in for $20k. What do you do?

I guess that's DIY in a nutshell. We take that risk every time we work on our own cars. Transmission, engine, convertible roof, it's all the same. But with care, patience and some decent information it's not hard. Everyone is a grownup and knows the risks.

I'm sure there are other weirdos like me out there who think that the pleasure of owning these cars isn't just the driving experience, but maintaining it as well. I simply have zero interest in someone else working on my car. As time goes on, these cars will be owned far less by people with deep pockets who are willing to throw money at any problem. Like the 986, they will be owned more and more by enthusiasts, who want to tinker and do it themselves. Forums like this with the information they provide are a lifeline for these people.

Even if I wanted someone to do the fix, finding a shop that knows what they are doing is difficult. Here in Oz I get nothing when I do an internet search. Just LN engineering in the US and Becks in Europe. The Rennlist thread I refer to above is littered with people having to truck their car interstate to fix the issue. And that's in a market as huge as the US. Putting my car on a float for a long trip and then be charged 28 hours of labour minimum is something I'm not interested in. I'm convinced there are others like me.

I'm not sure if you saw the question I asked in the post above, but do you sell your sensor to individuals, and what is the price? If you only sell to shops I can only assume the only DIY option that currently exists is from XemodeX for US$2199.

 

[jjrichar]

The current price is $1890 on simple order to non-partner individual. But there are very few people who have the space, tools, software, and will to work on a PDK gearbox. Despite all the information for DIY repairs freely available and referenced right on our website - I think we had 2 people in 2 years who took on a task fixing PDK in the garage.

By the way, we do have partners in Melbourne, and in Brisbane. Not sure how they advertise but they surely do provide quality PDK repairs in Australia. No need to ship the car overseas

Thanks for the information. I'll update the original post to reflect this.

Are you able to provide the names of the shops in Melbourne and Brisbane that you are partnered with? As an exercise I spent an hour ringing every non-dealership Porsche workshop in these two cities and came up with nothing.

 

[jjrichar]

After the discussion above I had to go and find the answer for myself. Can the distance sensor be removed safely with the transmission still in the car, and how long would it take? Was 28 hours minimum reasonable? Was there the serious risk of cracking the casing compared to having the transmission removed? So I pulled the distance sensor from the project car as though I was doing a replacement to see for myself.

I did the whole job easily in a day, and that was with me taking video footage etc, which always consumes a good amount of time. I’m pretty familiar with the car and transmission so I guess it would take someone who isn’t a little longer. You would easily knock it out in a lazy weekend. There just isn’t that much to do. Honestly, if you were doing this as a job and really familiar, I think you should be able to fully complete it in well less than a working day.

There is more space than you can poke a fat stick at to do the job with the transmission installed in the car. I also removed the bumper, including the metal bar behind and the heat shields. This all took less than an hour to get to the point where I was working on the transmission. Time well spent in my opinion. Have a look at the photos below. You can make your own decision on the amount of workspace available. If someone cracks the casing I suspect it has far more to do with technique and tools rather than it being still in the car.

In my opinion, working on a transmission doesn’t get any easier. If I had a preference I’d keep it in the car rather than remove. It’s like it’s stuck on the most solid transmission mount you can get. If the car was on jack stands you would just sit on the ground and it would be right in front of you. You wouldn't even need to lie down. Just sit there beer in one hand and wrench in the other. If you were a mechanic doing this all the time and very familiar you might leave the bumper on, but removing it and the heat shield above gives excellent access.

As you might be able to tell I’m a little stunned at how easy it was considering the noise from shops and others about how difficult and time consuming this task is. Trust me when I tell you there are many far more difficult DIYs in this car.

If it was a 911 it would take longer as the transmission needs to be removed from the car. I'm pretty convinced there isn't another 3 days work in that though. I might be wrong.

There was no need to drain the coolant. I just clamped the hoses, removed and plugged.

Test drove with the car on the lift using rolling test mode. Easy to just run through the gears back and forth over and over to test all is good.

I shot a bunch of video and I’ll put that together sometime soon. It will include the additional information I learned while working with it in the car.

 

[jjrichar]

It's funny with auto transmissions, but there has been a reluctance from car makers to crack them open for a few decades now from what I've seen.

I did a project like this PDK one about 10 years ago, where I stripped and documented the transmission from my old E46 BMW. This was the 5HP19 also made by ZF. There was more information available than the PDK, but very little DIY work being done. The fluid in the transmission, was said to be 'lifetime', which is clearly rubbish, but in researching why I found the reason was that car makers didn't want dealerships opening up auto transmissions. They had seen so many warranty claims due to poor work being done by their staff they realised the best option was to not allow any work on them at all. They would do no transmission work at the dealership and just suggest an auto workshop to do the job.

Fortunately due to these transmissions being in many cars, including the original Boxster, parts were readily available from ZF. Also, due to the numbers of transmissions out there, companies like ATSG, Sonnex, etc provided information and alternative part solutions.

It all seemed ironic to me, as I also stripped and documented the BMW engine. The skills required for that were far more extensive than the transmission, and like the Porsche engine, there was extensive documentation and parts available for rebuild.

Whilst I think Porsche locking down the parts availability shows contempt for their customers, I also think there is more to it. At the risk of being flamed, part of me (a small part) understands why they have done it. I don't think it's just greed or malice. To me there a few things at play here.

1. Porsche don't do the training to allow to allow the work. Why? It should be clear from the what I've shown that the tools and expertise required are nothing special. I suspect the decision many years ago to stop work on auto transmissions now has created a void of knowledge that would be very difficult to fill. If Porsche were to supply the parts, then there would be the assumption they can provide the expertise to do the work.

2. The transmissions are extremely reliable, and so the need for corrective work is low. Whilst we are digging into the distance sensor problem here, when I was ringing dozens of Porsche specialty workshops the other day all around Australia, I didn't find one that knew immediately what a distance sensor was or its function. It clearly isn't a failure that happens often, even though it's the most likely failure point in the transmission.

3. Porsche licensed ZF to close down the availability of parts and information. This has clearly been discussed at length before. The thing I find strange about this is one of the best marketing tools Porsche have in my opinion is old models still driving the roads and being kept in good working condition. I was waiting the other day on the sidewalk of a shopping area and out the front was parked a 1980s 911 SC in really nice condition. It's not something you see very often where I live. A young couple, maybe early 20s and not obvious petrol heads, walked by and spotted the car. They knew exactly what it was. The looks of desire on their faces as they walked around and took it in was just plain funny to watch. I didn't expect this reaction because we are constantly being told the younger generation want an EV. You could just about see the 'I want one of those' thought bubbles over their heads. The MBA driven bean-counters at Porsche would prefer that their most desirable cars fail and be unfixable after a certain time period. Try and tell me that a normally aspirated flat 6 of the last 15 years isn't going to invoke the same emotion and desire in the years to come. The Porsche strategy of deliberately trying to get these cars off the road long-term makes little sense to me.

4. The number of PDKs out there is very low. Compare this to the latest range of 8 speed ZF autos that are in millions of vehicles. The motivation of companies like ATSG, Sonnex and others to provide information and parts for a PDK is close to zero because there's no return in it. Workshops are in the same position. It's only those that see a lot of these cars that have the throughput and knowledge that will confidently tackle the job. You will pay the price for it though.

So like is happening here on the forum, the knowledge and solutions are going to come from the community. Maybe Porsche will have a change of heart in the years to come and allow the sale of spares. I'm not holding my breath.

 

[jjrichar]

Video of the replacement procedure is in the OP for those interested.

 

[jjrichar]

A question to the masses about sensor design. See picture at bottom.

There are grooves along the side of the aluminium casing that are aligned with three of the sensors (red arrows). The shift rod magnets are on this aluminium side for those three sensors. There's another groove that doesn't seem to align with anything (green arrow). The magnet for shift rod 4 is on the other epoxy fill side, and there isn't a groove here as it's just flat epoxy.

PV and I have been discussing offline what the purpose of them might be, and we can't seem to find a reason they are there. Anyone got any ideas? It will just make the CAD drawings easier if not required. CAD drawings should hopefully be reasonably simple. Nicely rounded edges around the end connecting holes aren't required as there is loads of spare space here. It's just the center connecting hole that needs to be the nice shape as one of the shift rods gets to about 2mm away from it.

I'm also interested in some feedback on the material of construction. The OEM and other aftermarket sensors are made of aluminium. It seems this is going to be the most expensive part of a sensor. We have the thought of 3D printing it from glass or carbon reinforced nylon. This is the stuff lots of parts in the car are made of. Transmission pan, intake manifold, and lots of sensors are examples. My guess is the transmission speed sensor is made of it.

We ask the question why the OEM sensor is in an aluminium casing. We think it might be because it needed to be a potted sensor, and to try and pot effectively into an extruded plastic casing wouldn't have been possible. The epoxy wouldn't be secure enough in this environment. The beauty of 3D printing is that those problems wouldn't occur. It would be easy to 3D print a casing that has loads of loops/jagged edges etc that when potted will hold the epoxy very effectively. You could even 3D print an inside overhang in the casing edge so it was physically impossible for the potting to come out. The temperature capability of these materials is the equivalent of the sensors, as they have better heat resistance than standard nylon.

3D printing of this material doesn't use a particularly expensive/bespoke machine compared to aluminium. There would the need to reinforce the screw holes. Maybe a stainless tube. It doesn't need to be particularly strong, as the screw is M5 and will take 8 Nm. We are thinking a sleeve could be just pressed into it like they do for the transmission pan or any other sensor made of it.

Can anyone see a good reason for not pursuing this? Strength, resistance to using a plastic sensor, anything else? Ideas on which, carbon or glass would be the most appropriate? Carbon = stiffer and stronger. Glass = better impact resistance.

Some additional info for reference:

Both of the aftermarket sensors have the grooves. The XemodeX website has pictures of their sensor internals, and the sensors themselves are right up one end well away from the grooves. I don't know if they just put them there because the OEM did.

Magnet distance from sensor side is about 1.5mm on the aluminium side, and about 2.5mm on the epoxy side. The grooves are 0.25mm deep.

When I removed the distance sensor the other day from the project car, there was a load of ferrous sludge on the magnets. This is normal, but there was so much it was touching the sensor. The sludge looked like it was being pushed to the sides of the magnet as the gap reduced to nothing and more sludge gathered over time. This car only has about 60,000 km. I would imagine that if you opened up a transmission that had a lot of miles you would have a really good accumulation of sludge on each magnet, with most of it being pushed to the side and the gap between magnet and sensor remaining filled.

Of interest the transmission I pulled apart had only about 15,000 km, and the level of magnet sludge was minimal compared to the car. The 15k transmission is from a Cayman GTS with a bunch of track time, the 62k transmission is a base Boxster with just street usage. I thought there might be initial wearing in of the gears that would accumulate quickly and then settle down to nothing as time progressed. This seems to indicate otherwise, with progressive wear seeming to occur. Very small sample size to come to any sort of firm conclusion though.

 

[jjrichar]

Snip from here for 30% glass filled nylon (30% is pretty typical from what I've seen).
Thermal expansion and strength highlighted.

Link to document
https://www.theplasticshop.co.uk/pl...s/glass_filled_nylon_technical_data_sheet.pdf

Snip of 6061 aluminium below. This is what the XemodeX sensor is made from.
Thermal expansion highlighted.
Link here:
6061 Aluminum: Get to Know its Properties and Uses - Gabrian

But then if I go here, it tells me they are very similar:
Thermal Expansion - Linear Expansion Coefficients

Epoxy from this site has an expansion co-efficient of 45-65.

If I use the information that gives the greatest expansion of the GFN, it's about 2.5 that of the 6061 Al, which would equate to an expansion of 0.003 mm (in height) and 0.01 mm (in length) for the sensor that is 50 x 180 mm. The relative expansion compared to the contact materials is going to be a bit more than half of this. I'm not sure if this is a lot or a little, but certainly isn't going to move the sensors enough to have any significant effect on sensor output.

If the screw holes had a sleeve like they do with most items that are made of GFN, I would think the problem of expansion along the height (where the screws are) would go away. It would then be the screw expanding equally with the stainless sleeve. Steel thermal expansion is about 15, so would expand less than either the Al or GFN.

My only real concern (assuming the expansion of GFN is about 60) is upon significant cooling. Think cold temps of Canada that I've never experienced. In this case the GFN will contract a lot around the stainless sleeve. I see either stainless or brass sleeves embedded into GFN in a lot of places on the car where there is this level of thermal range and it doesn't seem to be a problem.

Maybe the sleeve could be a 'light fit' into the holes, so this contraction wouldn't be an issue. Like with most sleeves into GFN, the screw end is wider than the sleeve (screw head is 12mm wide). So the GFN body can't slide off if it ever came loose from the sleeve.

More than happy to provide more photos and dimensions if required.

Anyone with experience in the field please don't be afraid to chime in.

 

[jjrichar]

I think ZF have already done the R&D for us. Image below of the inside of the pan.

Filter is Nylon 30% glass fibre.
Pan is Nylon 35% glass fibre.

I knocked out one of the aluminium sleeves. Comes out with little effort with a punch and hammer. I could push it back in by hand. It's clearly a relatively loose fit and the wide screw head over the black plastic is what is keeping it in place. The sleeve is just there to torque against and ensure the screw can't come out.

The pan is designed to lower the transmission, so I'm convinced of the material's strength.