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

[johu]

Why you do not use original audi mount? will you have a rubber bushing there? or EV motors are so vibration free that you can mount it directly to frame?

Argh! NO dont bolt your motor directly to frame!

No no, there will be rubber in between! I couldn't make do with the original Audi mount, it sits too low.

EDIT: screenshot added

[geduxaz]

I made it on original mountings wanted to make pictures but they are ugly. Its test version and later i plan to rebuild to more nice ones.

[geduxaz]

Can Leaf BMS work on not full pack?
I think full 96cell pack is too much for lightweight A2....

[johu]

The inverter is bulkier than I thought.
@geduxaz: can you close the hood on yours?

On the plus side I fixed the torque arm

[johu]

Nevermind. Removed that silly air duct, now it fits comfortably.
I think the duct was to prevent sucking engine room smells and heat into the interior. With these gone, no need for it

[geduxaz]

The inverter is bulkier than I thought.
@geduxaz: can you close the hood on yours?

On the plus side I fixed the torque arm

1. Hood closes perfectly, but i still didnt install air duct. it was removed during engine removal. But from the uncalibrated eye it should still fit nicely.
2. From the pictures you have gearbox rotated more and MG1 wiring cover is higher than my.

[geduxaz]

the lower mounting point could not hold. i did made something like you. it was destroyed soon... made a solid fix from old Lada rear trackarm (have some midsize warehouse with lots of old cars/parts/metal stuff)
Next car will be 1973 2103 with GS450h or Audi 100 quattro RWD

[johu]

the lower mounting point could not hold. i did made something like you. it was destroyed soon... made a solid fix from old Lada rear trackarm (have some midsize warehouse with lots of old cars/parts/metal stuff)
Next car will be 1973 2103 with GS450h or Audi 100 quattro RWD

Ok, at least I have a template for a proper one now

Yes transmission sits kindof high and also more forward than the original

[geduxaz]

Also I see that I moved all gearbox much closer to the stabiliser (to the rear) and made it lower.
I was thinking that it must be as low as possible, in the center for the weight distribution.

[crasbe]

A little hint about Nylock nuts: The screw must protrude at least 2-3 threads out of the nylon part for the Nylock to work properly.
In your case it's about flush (hard to tell from the picture), which would be too short.

[RetroZero]

Transaxle support with front & rear mountings

On my golf, I decided against using the transmission bolt holes to fit a support for the torque arm. With heavy regen or torque from standstill, I am concerned about breaking those bolt and where they are fixed. Landing up with a damaged transaxle is not ideal. I used the engine to transaxle lugs and fitted the torque arm in front (on a reinforced front sub frame support)
I also made a transaxle mounting that went all the way around and bolted to the original top mounting support from where mountings could be fixed (front and rear). Don't know if it all makes sense sense, but golf setup is different as well so maybe not relavent.

Torque arm support

[geduxaz]

Don't know if it all makes sense sense, but golf setup is different as well so maybe not relavent.

A2 has soft front, no attachement points.

[RetroZero]

I was thinking of bolt on bumper support

[johu]

Ok, that looks much beefier than mine, but also compared to the stock A2 torque arm

[MattsAwesomeStuff]

Re A/C: It's probably quite easy if someone with an actual Prius can sniff that 6-pin connector.

Lacking technical expertise, I'll try doing some outreach on Prius forums or something.

What exactly would be the task I am asking?

Let's presume the people I'm reaching out to are nerdy EEs who drive Priuses. Would any moderate EE hobbyist have the tools needed? Is it CAN-based? What are they using to sniff, what format would the data be stored in, etc? I'm trying to figure out how to break this into tasks that I could ask specific people to perform.

[johu]

Will get back to it later.

Now, new video:

[johu]

Summary:
- Putting a V3 mainboard into the ECU enclosure
- Reusing original wire harness without cutting off connectors
- Preparing the inverter

[AdvancedEnergySys]

Re A/C: It's probably quite easy if someone with an actual Prius can sniff that 6-pin connector.

Lacking technical expertise, I'll try doing some outreach on Prius forums or something.

What exactly would be the task I am asking?

Let's presume the people I'm reaching out to are nerdy EEs who drive Priuses. Would any moderate EE hobbyist have the tools needed? Is it CAN-based? What are they using to sniff, what format would the data be stored in, etc? I'm trying to figure out how to break this into tasks that I could ask specific people to perform.

If it's any help, I drive an 08 highlander hybrid and the main inverter is just a 250kw version of the 80kw gen2. For another couple months it will stay in stock form and I can grab any data needed.

[johu]

Maybe lets offload that to a new thread (in the Toyota section) as it has nothing to do with this build

[hugow93]

Summary:
- Putting a V3 mainboard into the ECU enclosure
- Reusing original wire harness without cutting off connectors
- Preparing the inverter

Great video!
I was wondering about how the original HV connector on the inverter could be used.

Just checking, should the positive be rerouted to the contact point I've circled in the picture below?

I gave it a try this evening, but my precharge bulb lit brightly instead of flashing momentarily.

[johu]

Yes, thats where I connected it. Not sure why that creates a short on yours

I tested charging via the A/C inverter today and already blew some of the IGBTs/diodes by some transient, even though I just used my 16A lab supply. I might just throw out the IGBTs and replace them with a bridge rectifier. At least the chill plate is reused then

[johu]

Opened the A/C inverter today and indeed two high-side IGBTs have failed shorted. They are rated 30A continuous and I loaded them with a 16A lab supply, so not sure what happened there. They are GT30J324 https://toshiba.semicon-storage.com/inf ... e=GT30J324

Will swap over from the remaining phase as I'd like to know if they are good for charging.

[johu]

Tested again, now they are holding up. Must have been some experimentation transient that killed them.

[hugow93]

Tested again, now they are holding up. Must have been some experimentation transient that killed them.

Ok nice, glad you didn't have to do any surgery.
I also note now that my ~96v lead acid test pack seems to just be measuring 25v on the DC bus!
I'll get another pack on the weekend and see if that was the problem.
Strange though as the short circuit isn't replicated when I reconnect battery connection back up the bus bars near MG1.

[johu]

Today I reinforced the torque arm, will just test whether it holds up.

Yesterday I modified the inverter to accommodate an additional current sensor on the cable down to A/C inverter and DC/DC converter. Otherwise there would be no current sensor on the charging path. It is normally a +12V/-12V supply type but when it is just fed with +12V and GND it simply cannot measure negative, which is not required anyway. So instead of connecting the MG1 current sensors to the control board, I connected this. To keep it from interfering with normal operation, I only switch it in when actually in charge mode via a small relay. I have shorted the forward and reverse pin and also installed a Schottky diode to 12V. So now I can connect a single wire to 12V to
- Put inverter in charge mode (both forward and reverse high)
- Close the current sensor relay
- Supply all relevant circuitry for charging

Of course the additional current sensor has a different gain, I'm thinking about adding a "chargegain" parameter that switches to a different gain in charge mode.