openinverter forum

Ok, i bit the bullit.

I bought myself a Porsche Panamera!

The idea is to to add a small rear drive unit to the rear and to add one to the front not sure yet if i will succeed with that or if i leave it RWD.

Wish me luck!

My first question: Does anyone know if this car has been done before?

Goal Component list:

- VCU from Damien but with own software;
- DCDC from Gen 1 Chrevolet;
- Charger from Gen 2 Tesla
- Heater from Outlander
- 2x SDU from Tesla MS one front one rear model;
- 32x TayCan battery module;
- 2x my own Can Gateway for battery comms;
- AC Unit of Tesla;
- 3 and 4 way valves from Tesla M S;
- TayCan fuses and relais;
- ID3 CCS2 chargeport;
- BMW Lim CCS Charge Interface;

Obligatory Doug DeMuro review:




Finding CAN data on this car will likely be difficult. But, definitely doable if you have experience with that.

-Matt

Well the donor choice was all about whats available and tempting to do / use

I saw his review, don't know if its positive or negative; anyway i like the car.

I wanted to have proper 4 seats and a 911 just doesn't do that. Also the maximum allowed total weight of the car is quite high and that's very nice if you want to fill it up with batteries and electric motors to get a proper range. Also there is a 4S version which means AWD parts are available...

We also manged to get the engine running again, it had the typical V8 broken camshaft bolts problem, and i made the CAN loggings to have some thing to replay. I think i will be able to tackle the CAN-Bus, saying that it will be a challenge since this car has already a gateway meaning that it's a web of networks tied together.

If it's like other Prosches you will not really be interacting with the gateway much. Once you have the Powertrain bus solved (basically just replicating the DME data) you're good. Having friends with a a PIWIS II tester will be very helpful. The only other gotcha is the security system, which may or may not be a pain.

-Matt

Yes, so i thought if the security is a pain i will do a MIM device. and if its simple (like Tesla was) then i could just emulate the DME.
Yes the gateway itself is not the problem only the translations it does are an extra factor and the web it creates, like no access from OBD2 port, is what i don't like.

but hey can-bus play is like mid-game. First; i need to mount the engine in the existing rear sub-frame. That's phase one.

Well some progress..

I removed allot of floor panelling and i finally lowered the sub frame and took the diff on the way out, out...
We where then able to take the fuel tank out (with its fuller neck) and that meant i could see the space around (and above) the rear sub-frame.
in the panamera the fueltank is below the passengers rear seats and flows over the differential on top of the rear subframe.

Looks like enough room for me to fill up with the Tesla SDU.

I ordered another, extra sub frame, so i have a spare in case everything goes t*ts-up...
This has for me the advantage that the car can stay on its own wheels.
'

Progress:

Ok so then you have a Sub-Frame, a motor and a car.

Now its time to make a plan. And since we only have one good go at it, whats better then mocking it up with some help in 3D and play around with the parts and situation.
Next step is mocking it all up and decide how to mount it.
I think i will split the inverter from the motor, so also that part need tobe taken care of.

Then we need to decide how to make the modifications since the frame is AlSi7Mn, welding is an option but perhaps bonding & rivets is easyer and less risky and with enough surface it should be strong too. (Any advice is welcome).

Parallel on this i'm working on getting into the canbus of the car .
In the end it needs to be perfect harmony between the new VCU and the car.
For that i'm increasing my tools with more capabilities.

Nice progress so far.

I'll offer this advice - it should respond to OBD2 PIDs when directly connected on the powertrain can bus. You can use that to find out scaling of certain items, as they are standardized values for the OBD2 side. I was unaware of this / was working with cars without CAN OBD2 so I didn't have this luxury starting out
https://en.wikipedia.org/wiki/OBD-II_PIDs

If you post a capture I can potentially give some other hints.

Aside from the security aspect, since you have no gearbox to worry about, the rest isn't so bad aside from traction control.

-Matt

Well my english left me there. I was already in the can-bus and made logs. What i need todo next is figure out what is what and who sends to and what comes from the DMU. But i might do a permanent MIM VCU if its difficult.

Ok, since The Netherlands went into lock-down and my EV garage is not at home, i have lost some time.

This saturday i proceeded with the replay and analysement of the CAN-Bus data.
Since i also work on the MIM device itself (and the application to give it a GUI) not all the time was tinkering time.

But the first results are there and i expect it to start catching up speed now..

0x101 = RPM (atleast)
0x10E = Oil pressure (atleast)

I also now have a list of ID's comeing from the DME.

Ok i got the compressor pumping now when i turn on the vehicle...hoorayyy..

So next thing is i relocating the compressor (making an mount for it etc.) and i did some more CAD work on the engine model to get into the cad program.

The CAD loaded up in the end..the IGES went from a non working 2.7GB to 4MB if i remember correctly. (see the model thread how i did that a bit).
The new place for the compressor is where the rear exhaust was.
Originally it was hanging under the rear differential.

I always try to use as much parts from the vehicle itself in this stadium.
In this case i used the exhaust bracket and welded up the compressor bracket to it + some more stiffening strips.

Ok not to much progress as i like but:

First step is done:

Ok, since the inverter is removable and I don't have the room, i need to consider mounting the inverter on a different spot.
Anyone any ideas and do's or don'ts about that?

Keep the phase wires short and of equal lengths, make sure to shield the encoder wires, all of which you probably know already.
That is much easier on the SDU since the cooling loops are explicitly separate.

How far off center will your unit end up if you're needing to take off the inverter?

Isaac96 wrote: ↑Sun Mar 28, 2021 10:43 pm Keep the phase wires short and of equal lengths, make sure to shield the encoder wires, all of which you probably know already.
That is much easier on the SDU since the cooling loops are explicitly separate.

How far off center will your unit end up if you're needing to take off the inverter?

Thank u for your response!

Yes, those two things where indeed in my mind. But i had no idea if the bus capacitors would complain, but i guess its uncharted teritory...
I'm trying to come up with another solution but as one can see the HV connectors are like hidden behind the pilar of the frame.
I will be investigating mounting the cover with the inverter 180 degrees (upside down) or shifted to the rear.

I'm mounting it centered but with the inverter off it will be like <100mm.
If i leave the inverter on it will be way more, basically the offset of the output shafts in the design.
But i thought that with the newer drive shaft technology the difference is very small. Remember these aren't old school rotating U joints.
Which have angle speed issues.

Ok lots of progress but not as many pictures as i would like.

First i mocked up the 3D design a bit:

This is the end goal... I bought myself a nive present : an AC/DC TIG welder so i can weld Alloys...

Sources the material and off i went welding...
To weld the AlSi7Mn frame you need AlSi5 or AlSi12 weldingsticks and a construction material which also is weldable with this alloy.
First i started todo some tests on a piece on the frame which would be cut away..
It went quite well honestly i needed quite some amps 140ish, a very thick electrode (green or blue tip) (3.2mm) and a 50/50 AC balance (very important).

Progress!

Update:
- Lots happend: inverter mount, air suspension fixed, gearbox almost ready to take out but then i got the jiggles about software....

So:

- I bought the Porsche diagnostic interface (official one) and implemented a 3-WAY man in the middle situation in my desktop canbus manipulator studio (3 PEAK interfaces).
One leg to the transmission ECU, One leg to the DME (VCU) and one leg to the car.

- Es expected now it became clear who sended what.
- Next step was cracking the CRC's; luckily there are allot of smart people on the world (and on internet) and i found out how todo it and integrated it in my canbus studio.
- Next step was implementing manipulation options in my 3WAY Canbus MIM studio. Now i could flip bytes and overwrite messages and i could try to find some pieces of the puzzle.

Those 2010+ cars are nasty. Each module gossips about each other one. So getting to a clean sheet with no errors is a thing and i feat it will stay a thing... if someone has any idea what a ESP module needs, canbus wise, i'm all ears.

- Next step now is generating VCU code from the project file from my own 3WAY Canbus MIM studio.
My goal is that i can click on a button and it generates the .h and .cpp file with alle the bytes and functions etc. there filled in.
I like to see a pattern in my codeing and generating automaticly the code will give me just that.

After that next step will be disconnecting the Porsche DMU and start hooking up the VCU, verify it works.

Then i will start pulling the gearbox out and the engine;

Then i will start on my battery box / battery system;
(since we have great lead time on components i did start with sourceing critical parts and PCB design) and with this big second side project.

Ok Battery gateways are ordered. (see that seperate thread: viewtopic.php?f=24&t=1646).

Leaves me with converting my CAN project to a VCU program.
For that i want to use code generation script.
That's what i'm building right now.

preview:

after some more fiddling with the scripter:
mhhh i see i mis a ';' now thats easy fix....

Long time no see!

Thought i updated more.
Software is running.
Inverter relocated (case made)
Pictures say more then 1000 words:

Software Highlight:
- Gearstick is added to the application; so changing stick should update dashboard and give VCU something to send to the drive unit/inverter;
- Air suspension is working (height control buttons);
- Gearbox MCU and DME is emulated both disconnected right now;
- RPM, Temperature, Oil pressure work.
- Dashboard is error free when "engine is running";

Water temperature is a challenge tough. i know the displayed values per bit change but they don't make allot of sense to me.

Not sure if i can take the PDK and MCU out; (i don't fancy voiding the diagnostics of Porsche tools for example).

As you might have seen i have obtained data access to my battery modules (TayCan) so next step is engine and gearbox out and start working on the battery box.

Well this weekend was a memorable step:

Its out: and its brother..:
Further more i made 2 driveshafts and tried/fitted them as a check to be able to send them off to the maker....
Will let you guys know how that went.

Ok this week i started with battery pack construction.
I use Taycan modules.

My idea was to use an extrusion profile for the cooling, by using the channels als coolant channels and mounting the flatside with a thermal pad to the battery...

The pads are on Ali Express 200mmx400mm

The profiles I bought from Amazon where 160x16 I 8


Pictures speak a thousand words....

Good find this profile. Rock on