Prius Gen2 Adapter Board for CHARGING

[NiHaoMike]

It is all a bit special with the Prius buck/boost converter because either the top or bottom IGBT are always on. Very annoying. When I get my new Prius inverter on the bench I think I will make a video about charging to make it clear.

Might there be an easy mod to disable one of them?

[johu]

That would be great indeed. You have to get inside the IPM, remove the gate resistor and short gate to emitter. So when bucking from DC bus to outside DC connection the low side IGBT would have to be disabled.
Of course not the preferred solution as it isn't really plug and play. But at least until we can come up with something better.

[RetroZero]

deleted

[RetroZero]

Contacters arrive to day, if all goes well thanks to COVID. I plan to remove jumper I put in place to HV Bus BEFORE trying out the charging process.

[RetroZero]

deleted

[RetroZero]

OK, before going blowing up stuff, had another read through steps for Charge Mode ( for the millionth time)

Step 2 - Precharge DC bus to rectified AC. That means precharge DC Bus to 320v rectified DC BEFORE connecting the Battery ?
If that is correct, the AC contacters are closed using the DC Precharge signal. The rectified 320 VDC come into the Inverter Motor phase for MG1. Using nice diagram I pinched and modified from Wiki.

Is this when the capacitors are "precharged" (in same way as we would precharge DC)?? I don't think so, but don't have the answer...
Because I understand that if we had capacitors charged up and then close the DC BUS - BOOM!!

Or is this where the Duty Cycle comes in and regulates the AMPS but not the Volts?

What important part of the puzzle am I missing?

[johu]

I will jump back in once I have my logic cable back (forgot it at friends place)

[johu]

So, I got my wire harness back today and immediately went testing (video maybe later). The most exciting finding is that function of the CSDN pin - Pin 29 of the inverter. So when it is left floating (or grounded?) the converter is ON. When CPWM is HIGH, the LOW side IGBT is on, when CPWM is LOW the HIGH side IGBT is on. When CSDN is HIGH both IGBTs are OFF.

So:

• I pulled high CSDN and CPWM, the latter via a 500R resistor
• I connected a 12V battery to the orange DC input and 20V to two phases of MG1.
• I started BOOST mode (always use boost mode with this converter)
• 1s later I disconnected CSDN

Now you could see something very interesting (see plot). At first the battery voltage is boosted from 12V to 90V as long as the duty cycle is low. The PI controller ramps up fairly slowly so it takes a while until the current flow reverses and current starts flowing from the power supply to the battery.

So right now the charging sequence looks like described above, maybe future versions of the control board should have the ability to control CSDN as well.

It is now a matter of finding a good time to activate the converter so that it doesn't boost so much. 12V to 90V sounds not so bad but what about 200V to 1kV - lets not. Toyota engineers will probably just click at us but lets not force them to

[LRBen]

That's very interesting. Does that mean that the potential max voltage the inverter can practically charge is now more than 320vdc rectified from 240vac?

[RetroZero]

Thanks for tesing Johu.I will have to go over that set-up from modified wiki diagram in First post. Am almost done wiring up relays on the bench, and was going to go connect it up to inverter....whew!
I was going to modify the 12v supply line to DCSW Signal - K2/1 (DC Main Relay) connected to K5/P4. Basically allow to close Main DC Contacter with AC Contacter.
Now looking at the plot you posted, would some sort of Precharge help, until PI Controller kicks in? I have a few 65Ohm/100W resistors...

[johu]

That's very interesting. Does that mean that the potential max voltage the inverter can practically charge is now more than 320vdc rectified from 240vac?

Will get to that next.

Now looking at the plot you posted, would some sort of Precharge help, until PI Controller kicks in? I have a few 65Ohm/100W resistors...

Resistors wouldn't help here, will think about it further soon

[RetroZero]

Thanks for the Patreon video !

Looking at my sketch in the first post, I'll rework the schema as per your findings along with my relay logic set-up. I'm thinking:

1 - CSDN and CPWM (470R) pulled HIGH (12V) when switching on Adapter Board & Inverter
2 - Precharge and Main contactor sequences as usual - AC and DC BUS at the same time (IGBT's OFF = no link between AC and DC)
3 - Switch to BOOST MODE - PI conroller controls charging (parameter protocole to set up & verify BEFORE) I understand Current IN is controlled.
4 - CSDN pulled LOW (disconnect) via 1 sec time delay relay - Problem of ramping up HV to potential 1KV to be solved first
5 - End of Charge Cycle (or charge plug disconnect), CSDN to be pulled HIGH first. (I'm thinking micro contactor on the plug that is actioned before removal)

I'll update the diagram in first post to make sense.....Thanks again for the Update. "Happy 12v time delay relays hunting "

[RetroZero]

deleted

[johu]

Well to make life simpler you can pull CPWM high as soon as you pull CSDN high, no delay needed in my opinion and one delay less to implement.

[RetroZero]

Deleted

[johu]

Forum release

[RetroZero]

1 - CSDN and CPWM (470R) pulled HIGH (12V) when switching on STM Board & Inverter
2 - Precharge and Main contactor sequences as usual - AC and DC BUS at the same time (IGBT's OFF = no link between AC and DC)
3 - Switch to BOOST MODE - PI conroller controls charging (parameter protocole to set up & verify BEFORE) I understand Current IN is controlled.
4 - CSDN pulled LOW (disconnect) via 1 sec time delay relay - Problem of ramping up HV to potential 1KV to be solved first
5 - End of Charge Cycle (or charge plug disconnect), CSDN to be pulled HIGH first. (I'm thinking micro contactor on the plug that is actioned before removal)

Been cleaning up this post, as I have realised I really did not know much about Charging, IGBT's and Inverters. Thanks for the Video Johu!!!
I am still far from considering myself knowledgable, however I think I am more educated on the subject.

So re-designing logic sequence as per the above steps. For step 4, where there is a good chance the voltage will be 1kV before PI Controler regulates system, what about inserting an inrush voltage/current limiter to AC INPUT?

[johu]

The AC input will never see the DC bus overshoot because there is the rectifier in between. I also think Toyota engineers will limit the DC bus voltage generated by the booster.

[RetroZero]

So ramping up the voltage (like in your video) is the best way forward?

[RetroZero]

Revised ChaDemo Prius Gen2 logic under way. I think I worked out a way to switch CSDN to HIGH before shut down of inverter.
Got me some delay relays, and will wire that up over next few days.

Light resistance will be home made ramp up voltage system once Boost mode is ON. (to solve over voltage spike)

Will write up diagram if it all works out....

Basically, AC will go through precharge and main relay sequence with DC BUS whilst CSDN is HIGH.
5 seconds later, CSDN pulled LOW ('disconnect).
AC ramped up to 230v with Variable Light Switch.
CSDN pulled high via microswitch before shut down....

[johu]

I will also add a minimum duty cycle to the software that will limit the unwanted boosting at startup. I think it will be more deterministic and maybe you don't need the time delay at all then.

[RetroZero]

That would be great, we're making progress. I'll try to implement what we have so far and manually ramp up voltage once in Boost mode.

The microswitch on the connector plug is the only way I see forward at the moment for pulling CSDN High before shut-down. Hope to try it out in next few days.

[Pcstreet]

This might sound stupid but understand everything apart from the pi charge controller. Is it hardware or software? In Johu video it looks like it is just software which he turns on using the touch screen. If so where can I get it and how does it work and connect to the inverter any help much appreciated.

[johu]

It is a software PI controller that comes in the normal inverter firmware.

[Pcstreet]

Thanks Johu I thought so but I can't find it in the foc how do I get to it please help