Tesla Model 3 Rear Drive Unit Hacking

[jrbe]

I've ordered the temp sensor boards and holders from JLC. thanks again to jrbe. I'll uprev to V3.3 and see how we go.

The temp towers you screenshot are for the through hole temp sensors. Just want to make sure you got the ones you need.
The ones for the PCB are here, https://github.com/jrbe/Tesla-Model-3-C ... Style.STEP just delete the .txt at the end

[Jack Bauer]

Order in with JLC3d. SLS(Nylon).
Material 1172Pro Nylon.

I'll try and find a ntc that works and a toroid for the current sensor when the bits get here.

Edit : 0402 NTC found. Panasonic ERT-J0ET473J. Ordered at JLC.

[Davidk]

Exciting weekend ahead. Thank you Damien, and everyone else who contributed.

(if it is not appropriate please delete post, wanted to show delivered board)

[Davidk]

All so far went well.

Regarding TDK power supply (should be on future boards) i decided to order new, mine us covered by coating on one site.

Issue i faced are:
Hole H10 mentioned that being slightly off is expected and it is not issue, mentioned in video.

I have issue with H7 where i can start bold and after few turns i feel resistance from side of the hole (pictures in attachment) decided to stop for now.

I can not put black clip over 30 pin connector, top hole is just too small (pic attached).

[davefiddes]

Congratulations on progress so far. I don't think that H7 has given any problems before. Have you tried loosening all the fasteners and giving everything a bit of a gentle jiggle from side to side and back to front? There is a lot of springiness in the MOSFET pins that can push things out of line.

H10 is a known problem with the 3.2 and earlier boards and documented in Damien's videos and the install instructions:
https://openinverter.org/wiki/Tesla_Mod ... he_Chassis
Hopefully to be fixed in future boards.

[Davidk]

Have you tried loosening all the fasteners and giving everything a bit of a gentle jiggle from side to side and back to front? Th

Thanks.

My board doesnt have room to wiggle (maybe 0,2mm), board is pretty much hold still by top right locating pin and HW loop.

https://share.icloud.com/photos/085cyiy ... Rrod-ZDksg (was not able to upload video here)

To fit it better i would need rotate board counter clockwise board a bit.

Maybe i soldered 30 pin connector too much to one side of holes and that creates this issue?

[Jack Bauer]

Shouldn't be a problem. My apologies for the hole misalignment. Its really frustrating but I am reducing the error with each revision.

[Davidk]

Shouldn't be a problem. My apologies for the hole misalignment. Its really frustrating but I am reducing the error with each revision.

No apologies needed at all, one of first version available for public so any errors are expected. There is also good chance that i messed up something.

Is there any way how i can help? documenting each hole and to which direction it is misaligned for me? I have also company around corner with CT and/or professional 3D scanner, should i scan both boards so you will have exact location of the holes? maybe also with my invertor as each can be bit different (wanted to do this anyway, i want to relocate it)? I used them for scanning before and it was 0,01mm precision.

[Davidk]

ok, issue was me. After unscrewing plastic shrouds around 3 phase and using some force to move board to correct place i succeeded and all bolts are in place. Some are still bit miss aligned but bolts can be screwed in without chewing board.

My invertor has extra white plastic around 2/3 phases, most likely making it tighter fit compared to invertor Damien had in videos.

One another issue i discovered is missing connection for E12 which was missing on my inverter from factory, not sure if it will be issue or not.

[jrbe]

The white plastic is a many years later isolation fix I guess? Looks like they forgot or couldn't do a few areas.

For the hole issue, if 2 holes are perfect you can swap in flush head / cone seat screws to align the board to those 2 then fine tune things from there for the next batch.

[Jack Bauer]

Am speechless. Will let the pictures talk. Now to find a gapped toroid:)

[Jack Bauer]

Temp sensor standoffs. Not got the boards yet but hopefully next week.

[Jack Bauer]

This popped up on my discord last night : https://github.com/techn0zahrt-ctrl/c2000-inverter-M3-Y

I'd have been delighted a few years back. Anyway, as this now effectively negates the board design I'll fulfil current orders then remove it from sale and terminate development.

[davefiddes]

Lovely. I'd noticed they'd forked my old (archived) repo last month. Thought they'd hit the brick wall I did and give up. Apparently not.

You appear to grossly undervalue the solution that we have created. There has been so much engineering and testing that has gone into the hardware and firmware of the replacement PCB. It is a completely different proposition. It builds on the past 15 years of johu's inverter development on the STM32. The C2000 is extremely weird and is effectively throwing away much of that knowledge and experience and starting again.

Competing solutions will always exist in FOSS/OSHW. That's kinda the point.

Anyway I'll keep supporting the M3_DU firmware as long as anyone else is interested in running it.

[jrbe]

I will also continue supporting this in ways I can help.

[Jack Bauer]

Thanks guys. Much appreciated.

[ted.walsh]

I too am getting on with the installation of my board, Ive de-solderd the old one (without the use of the blow torch, just a hot air gun) and cleaned everything up with lashings of liquid flux. had to order some heat sink paste but everything seems ready & OK. having done a few solder jobs in the past where parts required alignment Ive pre mounted the header into its housing and located it with the plug from the management loom so hopefully itll be in the correct place once soldered. ill let you know how it fits shortly.

as a community question, typically the conductors from the battery to the DC plug are heavy duty and massive, im paranoid about keeping things large diameter and not restricting the current path, you know, its meant to suck 800a this thing. Then I notice the size of the conductors on the inside of the inverter DC socket and what the current path look like and start to wonder why Im worried? Anyone got some reassuring noises that the 8mm x 2.5mm cross section on DC -ve and the 11.5 x 2.5mm on the +ve wont turn to liquid, I know thats the way Elon has made it but all the same seems a wee bit skinny? thoughts anyone...

p.s.
Please keep up the good work everyone. this is a well supported community with great ideas and dedication.

[Jack Bauer]

Just waiting on the last few parts to arrive in at JLC before next board order. The temp sensor boards are on the way.

[Jack Bauer]

Thanks to Jrbe we now have OEM like temp sensors for the V3.3 Thank you so much.

[Jack Bauer]