Doing useful things with the Prius Gen2 buck/boost converter

[JeremyMakesThings]

Reporting back....I gained access to the MG1 current sensors, and ran a piece of wire though them, hooked it up to my benchtop power supply, and ran varying amounts of current though it to see what it would do. Put the inverter in boost charge mode and got some really odd numbers back. I've attached a screenshot of the graph. I started at 10 amps on the power supply, and stepped down in 1 amp increments to zero. At 10 amps it was reading about 17 amps, and stepped down proportionally, but at 0A, it bumped back up, which seems odd. I'm not sure if I like what I'm seeing, and figured I should seek adult supervision.

I also had one of those "lying in bed" realizations that I wouldn't be able to charge my battery below the rectified mains voltage...but is that Vpeak or the average Vdc? If it's peak, I'm not sure there's much point in me continuing on with this, since my minium battery voltage would be ~250V.

[johu]

yeah it's peak, so 320V for 230 VAC folks... Works just about with 96S packs if you don't take them too low

[JeremyMakesThings]

I've got a 100s pack, so I decided to give it a go. I've got everything wired up (including my charge port!). But when I plug it in, I'm not getting any current to the pack, and just a slight rhythmic click from the buck/boost converter.
I'm wondering if I need to run through the other current sensor? Right now i believe I'm running though il2.

[johu]

Doesn't matter if all 4 are actually connected. It must be something else.

[JeremyMakesThings]

dang, I was hoping it would be simple.
I checked to make sure I was getting power into the inverter- I have AC voltage going into the rectifier, and dc coming out (if it's disconnected from the inverter). With everything hooked up, I'm seeing the low side of the buck/boost cycling between ~350V and ~400V on the multimeter. I'm getting the same reading with the inverter in run mode. my battery voltage is sitting about 408V right now, so it seems like the booster isn't boosting. Attached are my charging parameters and spot values. Any direction would be greatly appreciated, I feel like I've hit a wall and don't know what to try next.

[JeremyMakesThings]

huh. I'm also not getting output from my DC-DC converter. I swapped in the converter from my other inverter with the same results. some clicking, but it doesn't seem to be working.

Edit so I’m not just replying to myself: I went back and checked my wiring (and soldering) and I have good continuity all the way from the pin on the IC on the control board to the blue wire on the connector that goes into the buck/boost converter. In run mode, I’m getting 2.5v at the connector to the buck/boost. I tried disconnecting the wires that run down to the ac inverter/dcdc converter, no change in charging. I tried disconnecting where I have my bridge rectifier hooked up, and no change to getting the dc-dc converter to come to life. I’m at a loss as to what to try next, or what I’m missing.

[JeremyMakesThings]

I'm pounding my head against the wall. I've read through the forum, I've read through the wiki. I've hooked 12v to CSDN via a 470R resistor and also tied it to DCSW. I can get the converter to shut down at start up, but once the DCSW pulls it down, it's still bouncing between 350v and 400v. I've set pwmgain=0 and ran pwmofs up and down the entire range of values with no change. I think I've just about run out of things to check, but it feels like I'm missing something simple.

[johu]

uh oh... Have you pulled up CPWM to 12V with the likes of 1k? The control board is open collector but as opposed to the inverer PWM inputs the converter doesn't have its own pull-up

[JeremyMakesThings]

The good news is the needing the pull up was the part I didn't understand. I added a 470R (because that's what it said various places in the wiki and forum and I couldn't find a 1k). And now I have signs of life.

The bad news is they aren't very healthy signs of life. If I plug in the charger with the buck/boost shut down, when it starts up, the low side voltage still goes up to ~350V, and i get a high pitch whine as the voltage drops back down until it gets to my rectified voltage (I'm using a 120V charger right now, I can try my 240v tomorrow if it stops raining since it's outside). As it drops back down to the rectified voltage, the whine stops, (edit) the BMS sees something it doesnt like and shuts down the charge signal. I do see the AC current rise slightly right before the BMS kills it
I disconnect CDSN, when it plug in the charger, it blows my circuit breaker. (edit) I got out my trusty Openinverter certified pre-charge lightbulb, and with that in the mix, it will behave the same as above.
Putting the inverter in Run mode, either with or without the CDSN/DCSW wiring connected, I'm getting about 0.35V on the low side bus. I played around with pwmofs with no (or very minor) change.
I did plug in the charger with my HV disconnect pulled, and the DC-DC converter came to life! So I can confirm that works.

Edit- I'm not sure what, if anything I did, but I can now get the buck boost working in run mode, and I figured out the DC-DC converter needs to have the proper voltage before getting the 12v signal to turn on. I'll wire in a relay to trigger that with DCSW.

Getting closer.

[moebelschmitt]

Oh man, I'm so excited to see if it works!
Please keep us updated.

[JeremyMakesThings]

I’m charging!!!!!!!! (Negative pack current on the bms means power is flowing in). I’m using a 1.4kw charger, so 100% efficency would give me about 3.5a at the pack, so I’m hoping I can tune it a bit better.
I went back and watched Johu’s parameters video last night and when he got to charging parameters he said “hardly anyone uses these so we’ll skip them.”
Can I get a brief explanation of tuning chargekp, chargeki, and chargeflt? It looks like my current sensor is really noisy, so I adjusted chargeflt until I got 12a on my clamp meter on my ac input. Not sure if there’s a good process for tuning chargekp or chargeki other than….FAFO?

[johu]

Great
You can find some parameters here. Unfortunately the frontend puts most of them into "Regen"
https://openinverter.org/parameters/view.html?id=29

chargeki is not so critical, leave it at 10.
chargekp in conjunction with chargeflt tunes the ability of the controller to keep up with the sinusoidal shape, i.e. do some PFC. More chargekp and less chargeflt results in better tracking but also more noise. High chargeflt and low chargekp results in low noise, poor power factor

[JeremyMakesThings]

Great
You can find some parameters here. Unfortunately the frontend puts most of them into "Regen"
https://openinverter.org/parameters/view.html?id=29

Ah! I’d looked at that, but didn’t see the relevant parameters under charge, but I see them cleverly hidden now.
I need to get some of my wiring cleaned up (and probably also get a cooling system) before I really try to tune this thing, but I’ll give them a shot then.

I’m also trying to understand how the derating works for charging. Does it start derating at udcmax, or before that? I’m wondering because with a lower charegflt value, I’m seeing less amperage going in on the ac line. If I tweak chargeflt I can get it to match my charge current value, but too much chargeflt and it goes even higher.

[JeremyMakesThings]

I’ve made enough progress to make a video one it: Boost Charging with the Gen 2 Prius Inverter


I’ve also now got a cooling system, so as soon as I can run the battery down a bit, I can try to do some more tuning.

[johu]

Brilliant, cool to see someone actually having some success in pursuing that path
Speaking of derating it will cut back before reaching udcmax just like it would under regen. You can also setup some CAN mapping between Orion and the inverter to modify the chargecur parameter.

As for efficiency I'd expect it to go up with higher power. Right now the idle draws eat into it quite badly. Since in the Audi I never noticed any significant warming it must be quite efficient at the 3 kW I limit it to.

Speaking of documentation it'd be much appreciated if you add your findings here: https://openinverter.org/wiki/Toyota_Pr ... Controller (also add that resistor to the diagram)

[MattsAwesomeStuff]

Neat!

Someday I'll get around to finishing my Prius Gen 2-based car.

I'd very much appreciate diagrams of what you did, to be added to the wiki. It was hard to keep track of what you ended up doing.

I'm on the fence about the limitations of re-purposing the Gen2 as a charger. I really like the all-in-one aspect of the Gen2, but, if it's more compromise than it is luxury, I'll probably just go the dumb way.

A lot of the limitations are battery voltage based. For example, the 12v DC-DC really confines what you're allowed to use for a pack voltage. The A/C inverter is a nice touch, but, you can just take the A/C off a Gen3 instead which has an inverter onboard the A/C itself and then just feed it PWM.

I like your use of just an add-on rectifier.

[johu]

A lot of the limitations are battery voltage based. For example, the 12v DC-DC really confines what you're allowed to use for a pack voltage

Nooo Matt. As described in the entry post which also has the diagrams you asked for you will see that you can move the DC/DC converter BEHIND the buck stage and thereby reduce battery voltage to one the DC/DC can work with.