openinverter forum

What tool did you use to draw the wiring harness ?

modellfan wrote: ↑Sat Dec 13, 2025 4:52 pm What tool did you use to draw the wiring harness ?

I used a combination of Wireviz plus cpp (the C pre-processor) and GnuMake.

https://github.com/wireviz/WireViz

My truck wiring repo:
https://github.com/hpeyerl/evj55/tree/main/wiring

References a generic Zombie VCU repo:

https://github.com/hpeyerl/gs450h_ls600hl_zombie_vcu/

I'm not sure if I'll continue with this tool. It's more cumbersome than I had hoped. I might look at https://splice-cad.com/ before I do much more harness work.

HerbP wrote: ↑Sat Dec 13, 2025 6:38 pm I used a combination of Wireviz plus cpp (the C pre-processor) and GnuMake.

https://github.com/wireviz/WireViz

My truck wiring repo:
https://github.com/hpeyerl/evj55/tree/main/wiring

References a generic Zombie VCU repo:

https://github.com/hpeyerl/gs450h_ls600hl_zombie_vcu/

I'm not sure if I'll continue with this tool. It's more cumbersome than I had hoped. I might look at https://splice-cad.com/ before I do much more harness work.

starting to play with this.

the fact it uses yml as the file format is cool. very git compatible

Bratitude wrote: ↑Sat Dec 20, 2025 9:22 pm starting to play with this.

the fact it uses yml as the file format is cool. very git compatible

Yeah. I wish it had a native way to include files.

I briefly started playing with Wireviz-Web https://github.com/wireviz/wireviz-web but it doesn't really fit my workflow.

I liked the idea of having a generic gs450h/ls600hl repo that others can 'include' in their harness projects.

HerbP wrote: ↑Sun Dec 21, 2025 1:42 pm Yeah. I wish it had a native way to include files.

I briefly started playing with Wireviz-Web https://github.com/wireviz/wireviz-web but it doesn't really fit my workflow.

I liked the idea of having a generic gs450h/ls600hl repo that others can 'include' in their harness projects.

I’m starting build one for my generic zombie wiring interface.

I found this video

Pretty cool setting up auto build scrips in the git repo. Make a commit and auto gen a new png

Also found this rough project https://github.com/slightlynybbled/wireviz-gui

Still working on a bunch of stuff in parallel but trying to tick some sub-projects off the list.

First thing was a front axle rebuild. In order to accommodate a later model power steering box (that I will loop the hydraulic connections together), I needed a new relay rod, which required a new tie rod and tie rod ends. This led to replacing the knuckles with those from a newer model and thus a full knuckle rebuild (all new bearings/seals, brake backing plates, etc). I replaced the third member as well with a stronger fine-spline flange so I could match the bolt pattern on the front driveshaft. I realize this is all boring mechanical work and not really EV related, but it is distantly, as part of the Prius EPS sub-project. It's also reassuring to have brand new steering linkage as well.
Mounted my Rav4 rear electric brake caliper and butchered Honda rotor to the back of my tcase. This will be used for parking only since I've had a bad "parking pawl" experience in the past. The truck that this transfer case came from has a cable actuated rear parking brake. However, the rear axle on this truck doesn't have provision for a parking brake as there was a small drum brake on the original transfer case. My choice was to swap out the rear axle for a full-floater that I have and do a disc brake conversion, or limit scope creep and just mount an electric caliper to the back of this transfer case.
..and a new/relocated sway bar to make room for my battery box.
I've also received qty 20 Tesla Model S BMS slaves to prepare for 120s (though I'll probably start with 96s for now).
I've decided to use the Tesla slaves instead of the BMW slaves because I intend to go 120s and build my battery box for 120s. I learned yesterday that the various OEM OBC/DCDC chargers won't support >400-450v so I'll have to go with something else but I'll learn to walk first and run later. (Later I'll look at a Dilong or similar charger)

BMW was nice enough to number the HV lines in the modules which makes the pinout easier. I'm thinking about designing/fabbing an interface board from the BMW connector to the Tesla slave connectors.
Also decided to try out an external oil pump to give me more freedom for my front driveshaft to move around. So I sent this off to JLCCNC for fab. Should be delivered any day now. Did a test fit with a 3D printed adapter before sending it off.
...and I've ordered my LS600HL bellhousing lockout plate after cutting one out of polycarbonate on my CNC to check fitment.
I'm still working on iterating through my transmission<->transfercase adapter plate. I really do need to get a first iteration of this fabbed out of metal so I can get my driveline put together a bit more. It's blocking other work.

This is the frustrating part of each project. A whole lot of research/learning/procurement and not a lot of startling visible progress. When my neighbor comes over, it doesn't look like much is getting done.

Well, it fits anyway. Pretty good for $9.14 from jlccnc.com. I tapped it to 1/4" NPT. O-rings are #114 (3/32 x 5/8 x 13/16)

Once my connector/cable orders arrive, I'll confirm my wire routing etc and send this board off to fab.

It theoretically adapts the BMW i3 26pin .1" header connector to the 4 JST-JL connectors and 2 thermistor connectors from 2 x Tesla Model S BMS slaves. This should hopefully prevent me from a bunch of cut/strip/crimp/error iterations.

The local laser cutting shop that I use is pretty backed up so my friend Peter offered to cut out my transfer-case adapter on his ArcDroid CNC plasma. I had gone about as far as I could go with my 3D printer and using my CNC with polycarbonate so I wanted to see how much I'll need to modify it before I get it laser cut.

Many thanks to @jrbe who put a lot of work into converting 3D scans into models.
After cleaning up the dross and adjusting some holes this is what I ended up with.
I feel like it should work ok, there's a few M8 through bolts I still have to deal with but I think I'll be able to reassemble my Ravigneux section and hang the tcase. I've updated the design a bit based on my findings.

At last, a Toyota HF2A transfer case hanging from an L110F transmission without ratchet straps or welding clamps.

For a long time I've been working in a bunch of different directions, seemingly not progressing any one of them significantly. But this last few days, a lot of this work has come together resulting in something that actually resembles a vehicle.



My attempt to shorten the rear driveshaft on my hobby lathe did not result in what I would consider to be a nice concentric driveshaft. I will probably need to take it to a driveline shop near me.

Started wiring. First up I wired up the Toyota Center-Diff-Lock controller onto the transfer case so I can switch between AWD and 4WD (thus being able to lock the front and rear driveshafts together).

Next up was the external BMW oil pump. For now I wired it up with the inverter power so it comes on when I enable the ignition.

I had a 50/50 chance of getting the lines connected in the correct orientation. The first try had the pump spinning and I could hear gurgling inside the transmission but that didn't prove anything. It was suggested I try activating the parking pawl and try to run the MGs and sure enough, the transmission still spun but of course the output was physically locked. That suggested the clutches weren't active and the oil pump was connected backwards.

I reversed the lines and then the pump spun up but after a few seconds it locked up presumably due to backpressure and then it tripped the breaker on my inverter supply.

So I presume that this was the correct orientation and there needs to be some sort of bleed..

I noted this small passage on the OEM oil pump: It seems to be cross drilled into the left passage which I take to be the high pressure passage and I assume there's a similar passage on the transmission itself thus allowing the pump to continue spinning and presumably the OEM controller adjusts the PWM internally blah blah blah...

Does anyone know anything about this tree that I find myself barking up at?

That is looking like either a check valve or a pressure regulator / relief valve. Judging by the bleed port I'll go with pressure regulator / relief.

Any chance of getting that valve out and machined into your adapter?

Ok. So I think I got to the bottom of my external oil pump issue.

First of all, I originally had the plumbing backwards and hence the bubbling of air in the pan. However, when I then reversed the plumbing and experienced the 'hydrolock' situation, the problem was actually that the oil pump was dragging down my 15A linear power supply (that was also powering my inverter and Zombieverter).

Having brought in a car battery and powered the oil pump on that, the pump did not lock up but instead continued pumping and was strong enough to activate the clutches.

I even made another video to demonstrate.



I may still opt to run the OEM pump as an external pump because 19A at 12V is a fair amount of juice. I might also play around with a PWM motor driver I have.

I'm still moving ahead on several fronts.

I've started planning out the engine compartment. I have this fantasy that I'll be able have a bit of a Frunk and make it relatively easy to lift the body off the chassis in the future. I've also decided that I'm not building the ideal vehicle, but am going to learn by doing; as such, I'm making some short term decisions.

I've decided, for example, to put aside my thoughts on doing a 120s battery setup. I could fit it, barely, but then I realized, for my use case, I can get by with a 96s pack and maybe, for longer trips later, I can slide another one of my BMW i3 packs in the back of the wagon as a parallel pack. I'll have to do some BMS work probably but see "learning".

This is how the engine compartment is tending towards. I can fit 2-4 modules under the inverter and charger and hvjb. Lots of room for steering shaft and ibooster.
Drivetrain with shortened rear driveshaft and Rav4 electric parking brake.
Rear battery box. I'll put 4-6 modules back here. Currently leaning towards putting 4 back here for weight distribution, especially if I'm considering putting a whole other pack on top of this inside the cab. I might need some helper bags.
As an aside, my son and I discovered that BMW X5's with V8 and air-bags have a nice 12' long silicon sheathed 3/0 battery cable. At our PnP, they call it a "battery cable" and charge $6.49 CAD for it (as long as the battery fitting is on one end). Takes about 15 minutes to pull without cutting the cable. It's probably not actually 3/0 but it fits pretty much perfectly into some 3/0 lugs that I have and really doesn't want to fit in 2/0 lugs.

Hi Herb -

Very cool project. I see you referenced Splice CAD in the build log; not sure if you ever gave the tool a shot, but I'm beta testing a new Project mode that may be more amenable to a project as complicated as this. I took a go at replicating your WireViz for the ZombieVerter in the attached.

Projects like this are always the best test for improving the tool - if you're interested, I'd be happy to try to assist in documenting the project or share the design shown below in the beta app for you to edit/modify.

Congrats on the progress (apologies for cluttering up your build log here, I can't find a way to message directly. Please let me know if you'd like me to delete).





- Danny

I did briefly fiddle with SpliceCAD but I didn't invest enough time in it to get over any sort of learning curve. I do remember documenting connectors and decided it was tedious especially since practically none of the connectors I'll be using are going to be in any sort of parts library. However, that's maybe not a requirement in the tool.

I admit to having a certain fondness for being able to version control my project using Git and some public repo. Especially being able to separate out sub-harnesses like the Zombie-L110F harness, and the Toyota HF2A CDL harness, etc. This way, someone doing an EV conversion with a different transfer case can substitute their own tcase harness and release it on Github. However, I'm not sure this utopia would ever actually come about.

So, I'm interested in learning more about SpliceCAD and I've sent you a DM with my email address. I'm quickly getting to the point where I'm going to have to formalize my wiring.

@tom91 Have you ever tried/succeeded in reading/writing the firmware in a BMW i3 CSC?

The ones in my pack don't seem to speak either I3 or Mini-E. They seem to broadcast 0x1c0, 0x1d0 packets that appear to have cell voltages in them but don't match the cells in my module. (The cells in this module are all 3.879v-3.88v according to my meter)

0x1D0: 22 22 22 22 22 22 29 92
0x1C0: 22 11 11 22 24 22 42 22

It doesn't seem to want to respond to any of the I3 or Mini-E packets I send it.

The sticker on the CSC seems to match other pictures I've found on the interwebs.

HerbP wrote: ↑Wed Mar 25, 2026 2:19 pm @tom91 Have you ever tried/succeeded in reading/writing the firmware in a BMW i3 CSC?

Nope.

Can you share a log of it connected to a module spitting out CAN without a simpBMS attached and then one with?

tom91 wrote: ↑Wed Mar 25, 2026 2:29 pm Nope.

Can you share a log of it connected to a module spitting out CAN without a simpBMS attached and then one with?

I will have to re-create the log without a SimpBMS and I don't actually have a SimpBMS. I'm porting code to a Lilygo T-CAN485. I'm fine on the code front and appreciate your feedback on your observations relative to what my CSC's are doing.

These packs were bought at auction from a company that used them in the development of electric transit busses. As such, I'm suspicious BMW may have made some mods for that company. Gregski is sending me a stock i3 CSC so I can compare.

I'm away until sometime Saturday.

tom91 wrote: ↑Wed Mar 25, 2026 2:29 pm Can you share a log of it connected to a module spitting out CAN without a simpBMS attached and then one with?

Here are three files. With a BMS (based on SimpBMS) connected, with BMS powered off but wired up. With BMS powered off, wired up, and +5v turned off then on (no wakeup packet).

Further to that, every module I plug in after that, without Wakeup (no bms), broadcasts on 0x10n, 0x1Cn, 0x1Dn where 'n' is probably their module number. Here's a log for n=5. Plugged into the same module as n=0 above.

Ok, I've reconnected the SME and snooped the CSC CAN and have a treasure trove of data. I believe I can talk to the CSC's now.

Oh if you had an SME you can work out what changed. Did not realise you also got a matching SME.

Yes. This might also explain why I could never get any life out of the BatteryEmulator firmware talking to the SME. Thinking BMW did some custom stuff for this company.