Crazy Idea - Could This Work Instead of Swapping Busbars?

[nubster]

I had a bit of a crazy idea and figured I'd post here in case someone knows whether this could actually work.

A client of mine wants to switch from OpenInverter to either EV Controls T-2C or Dynam Labs VCU+ for his LDU. I explained that he’ll need to reinstall the original Tesla logic board, and because his LDU is mounted in reverse, he’ll also need to swap two of the phase busbars (U, V, W) and the encoder wires. This is necessary because any setup using the OEM Tesla logic board is limited in reverse due to Tesla’s software, so you have to trick it into thinking it’s going forward.

While inspecting my own LDU, I noticed something interesting: it looks like you might be able to just swap the little white wiring harnesses going to the phase boards instead. In theory, this would allow you to swap two phases on the LV logic side rather than the HV busbars. Way easier, if it worked.

Obviously, I’m not keen on frying a motor just to test a theory, so I wanted to ask: has anyone tried this, or know enough to say whether it’s even possible? My gut says no, since those wires likely only carry sensor signals, and the actual phase sequencing is hardwired. But if this worked... wow, it would make things so much simpler.

[P.S.Mangelsdorf]

In theory, I think that could work. It's basically the same thing, changing which phase of the inverter is outputting to which motor busbar.

I've had the entire inverter apart a few times, so I know that the only connections to each phase are:
- the + and - busbars in the front
the phase output
the signal input that you're pointing out

The only other item you might need to change is that, if I remember correctly, the two outer phases have some sort of current sensor on their bus bar to the motor. I'm not sure if that can be accessed without removing the inverter from the casing. I imagine you'd need to swap those two as well to keep everything happy.

Edit to add: Large caveat - I have not done this so try at your own risk.

[midway]

I think this variation will work. But you still need to swap the current and resolver sensors for correct operation. And most importantly, the gearbox is not designed for operation in the opposite direction. The pump pumps only in one direction, and the gears have teeth tilted in the other direction.

[LRBen]

I tried this with an LDU a year or so ago. Inverter made some unhappy noises so I didn't try for very long as I didn't want to blow up the LDU either.
Decided not to pursue it much further given how old the LDU is these days.

[nubster]

I tried this with an LDU a year or so ago. Inverter made some unhappy noises so I didn't try for very long as I didn't want to blow up the LDU either.
Decided not to pursue it much further given how old the LDU is these days.

Did you also swap the encoder wires or current sensors?

[LRBen]

Did you also swap the encoder wires or current sensors?

Yes, although it is quite possible I missed something.

[MikeNotts]

Does anyone have anymore views on this approach to reverse a ldu ?

I’m at the point of thinking I want to cut my busbars, but obviously this would better if it works.

Any thoughts appreciated - Mike

[MikeNotts]

How does the Mercedes b250e and the RAV4 Ldus which are mounted in reverse differ from the Tesla’s model s ldu ?

[midway]

How does the Mercedes b250e and the RAV4 Ldus which are mounted in reverse differ from the Tesla’s model s ldu ?

The most important thing is that the gears are helical and have an inclination in the opposite direction.

The load on the gearbox bearings is distributed in the opposite direction.
The oil pump works in the opposite direction.

[nubster]

Does anyone have anymore views on this approach to reverse a ldu ?

I’m at the point of thinking I want to cut my busbars, but obviously this would better if it works.

Any thoughts appreciated - Mike

Confirmed this works well on the bench! Be sure to swap the two inner encoder pins as well.