[DRIVING] Audi A2 /w Prius Gen2 Transaxle & Inverter [FINISHED]

[johu]

is there somewhere that you have tabled your optimum MG2 parameter values?

Yes: https://openinverter.org/parameters/view.html?id=29

[johnspark]

Awesome!

[johu]

I 'roughly' converted this total Nm (MG2+MG1) to power/RPM graph.
You can see constant power region from 1500>rpm is around 95hp - 70kW!

Indeed on a test drive I was able to pull 150A from the battery when accelerating uphill at low speed. That is 54 kW from MG2 alone at 360V It can't be sustained in the constant power region though. Maybe MTPV will eventually improve that.

[johu]

Some more Audi progress today

Thomas designed a holder for the vacuum pump:

And I installed it

He also made some pole covers for the 12V battery and a bracket to connect the cross member to all 3 bolt holes of the motor mount (just visible under the battery)

Now just need to cover the +12V connector and then take the car to Göttingen TÜV again next Monday.

[johu]

Changed the front wheel bearing today and went for a spin. What a nice absence of noise!
TÜV on Monday and then I consider this project done

[geduxaz]

Changed the front wheel bearing today and went for a spin. What a nice absence of noise!
TÜV on Monday and then I consider this project done

How it went? Was it done? Is it street legal? Can you share any documents required for legalising it into road legal?

[geduxaz]

How does 12V battery feel with out BMS?

[johu]

I picked up Audi at TÜV with 60% SoC or so. I then drove it home (50 km) and drove another 20 km to empty the pack. By that time the lowest cells had sagged to 3.4V and the BMS issued a quite low power limit. I then took a snapshot of all cell voltages.

Then while I was away I charged the car 2h a day with a timer relay, after roughly 4 days or 16 kWh it was full. Then I took another voltage snapshot.

I then calculated the distance between the lowest and highest voltage. The lower the difference, the more life should be left in the cell:

Summarized I'd say the pack is still not well balanced, now some cells sit at 4.05V (90%) and some at 4.1V (100%). The varying difference in full and empty voltage might not be as dramatic as it looks at first glance as it includes the sharp drop off below 3.6V for some cells.
The BMS still thinks we're at 72%, which sort of aligns with the 16 kWh the battery accepted. I hope it improves as the BMS does the balancing.

How it went? Was it done? Is it street legal? Can you share any documents required for legalising it into road legal?

Have to go one more time, hopefully next week.

How does 12V battery feel with out BMS?

In Polo I used the setup for 5 years and they did not drift, I checked from time to time

[Romale]

johu: quote redacted


Смотри, что есть на AliExpress! Балансировочный активный эквалайзер для литий-ионных и литиевых аккумуляторов А, BMS 2S 3S, Φ 6S 7S 8S 10S 13S 14S 16S за 278,55 ₽ - уже со скидкой -20%
https://sl.aliexpress.ru/p?key=y7aIZKy


6 such pieces of 17 lines will allow all 96 cells to be made perfectly as one voltage. each channel 17 needs to be hung parallel to the first of the next part of the new 16 cells.

[arber333]

johu: quote redacted


Смотри, что есть на AliExpress! Балансировочный активный эквалайзер для литий-ионных и литиевых аккумуляторов А, BMS 2S 3S, Φ 6S 7S 8S 10S 13S 14S 16S за 278,55 ₽ - уже со скидкой -20%
https://sl.aliexpress.ru/p?key=y7aIZKy


6 such pieces of 17 lines will allow all 96 cells to be made perfectly as one voltage. each channel 17 needs to be hung parallel to the first of the next part of the new 16 cells.

I was not impressed by those inductive balancers. They had random inaccuracy. Capacitative balancers were much better.

[MattsAwesomeStuff]

Jeez, under $1 per string? A whole 96s BMS for under $100?

I've seen a few cheap options people have been talking about lately, all sub-$200. For someone that planned on not having a BMS, that's cheap enough I would just spend the money.

When it gets a bit closer to time for me to build my packs, I'm going to try to poll around for the best cheap one.

[johu]

All sorted

I asked again about EMI certificates. He now confirmed that the car can be any year as long as the drive components (motor, inverter) belong together.

Oh and this time I did some bookkeeping.
Car: 2300€
Batterie 24 kWh Leaf: 3750€ (would be much cheaper now, this was back in 2018)
All supplemental stuff (cable, driveshafts, metals,...): 2000€
TÜV: 635€
ICE stuff sold: -337€
Total: 8348€

Keep in mind Prius inverter, transaxle and Leaf A/C compressor were free

[Bigpie]

Awesome, so has this got working A/C? Did you do anything to flush out and decontaminate the old system components with the old oil?

[johu]

So far the A/C isn't working as we haven't adapted the pipework to the Leaf compressor. I don't know about rinsing... We can flush each pipe individually and the dryer will be swapped for a new one anyway. Only components remaining would be front and inside radiator

[johu]

Wrote all about the temporary CCS in the CCS thread.

Lets continue with some Toyota "invincible" Prius stuff.
I tested the new firmware in Touran today all while the Audi was charging. Both inverters have SSID "inverter". And some point I must have logged into Audis wifi und updated that. I found the breaker tripped and charging stopped.

When done with Touran I did the same in Audi, even drove around a bit all seemed fine. Only the windows wound down remarkably fast but I thought I'm just imagining it. I think at that point I had more than 14V on tap...

After some turning off and on suddenly something in the front went up like a fire cracker. I just wouldn't stop! Until I turned the ignition off. I then stopped the buck converter that supplies the DC/DC converter and found I can still drive. So inverter ok.

Disconnected the DC/DC and no more fire crackers. Buck converter outputted a steady 200V.

Ok, lets disconnect everything, drain the coolant and take the assembly out and open it:

Oops. I think when I resetted the inverter while charging the DC/DC must have put full pack voltage (380V) on the DC/DC and apparently it doesn't like that. The damage then showed later.

Still had a DC/DC board handy and dropped it in and all seems to work fine again. Full reassembly tomorrow.

[Ev8]

Mmm crispy

[johu]

Now put everything together, started, DC/DC works but can't go in forward. Inverter fault signal trips, can't see why. Of course coolant already refilled after testing the DC/DC. I need a break.

[RetroZero]

Sorry to see that Johu. Need to put some kind of safety code in place so updating can only be done in basic start up mode

[Romale]

did I understand correctly, the controller running the version 3 board was in charging mode and when updating in this state, something went wrong?

[johu]

The Prius controller was in charging mode and I accidentally updated (or tried to update it). I think resetting mid charge put the buck module in full-on (because PWM stops) mode and dumped too much voltage onto the DC/DC converter. Something along those lines. All I know for sure is that the house breaker tripped.

I have now swapped in the guts of another Prius inverter. It was easier than swapping the entire assembly. Now we have a new problem. I can go into drive and give it a tiny bit of throttle and it will pull forward. Give it a bit more and it stops with the non-existant "OVERCURRENT" error (Prius board has no overcurrent comparator). Also I can't see the desat signal go low, so no idea what triggers it.
On the other hand if I give it full throttle right away it pulls forward strongly and doesn't trip. WTF??
Same software and controller that worked fine yesterday

UPDATE: put the failed inverter module on the table and indeed high side of phase 2 is shorted! Wow, I killed a Prius inverter!

[Ev8]

Did you swap everything from the new inverter? Capacitor, main inverter brick? Or just the damaged dc-dc side?

[Ev8]

Just read update, welcome to the I killed a Prius inverter club, if you were local I’d have another couple for you to try

[Ev8]

Still wondering about the Main cap though they are hard to see they are damaged

[johu]

Actually main cap and buck/boost converter are the only thing left I didn't swap (apart from the enclosure...).

Good news is: back on the road. I simply chose to ignore the fault signal MFIV. Just trusting the inverter will protect itself even if I don't stop PWM. Drives completely fine now.

Have 3 inverters and 4 DC/DC modules, so still 2/3 to go

EDIT:
and I also fixed this disturbing problem:

When transistioning to strong regen at speed field weakening would collapse and cause really strong regen. Does catch itself but still not good.
I simply increased vlimmargin from 2000 to 4000 and that fixes it.

[Romale]

When transistioning to strong regen at speed field weakening would collapse and cause really strong regen. Does catch itself but still not good.
I simply increased vlimmargin from 2000 to 4000 and that fixes it.

still, an adequate description of all these parameters would not hurt, then the users (including me), knowing what is what, would not be afraid to experiment by changing the default settings and we could all come to quality management sooner.