openinverter forum

Bigpie wrote: ↑Thu Apr 20, 2023 7:57 am Cars been lowered
IMG_20230420_084940017.jpeg
Unfortunately it wasn't desired.
IMG_20230420_085010726_HDR.jpeg
I heard a clunk when I set off then the ride was pretty terrible.

Oops!
On a positive note, they're pretty cheap and easy to replace.

Springs replaced.

Love how easy they are to change on the back of those cars 2 jacks 2 bolts done in under 10 mins I bet! Probably even easier with no exhaust!

Not quite, one nut was well buggered. Done in an hour though.

So coming to Farnborough now?

Not likely, it's a five hour drive without charging stops, not managed to get extra battery in yet. Would have like to, but a bit far for a day trip with 17kwh.
Plus I've got the plague now, kids bring al sorts home from school

Finally feeling a bit better after Covid again, so I got the 3rd and final pack in the car, mounted in the boot for now while I decide if I can be bothered to make a box to mount it in the old fuel tank space. The HV and LV wiring isn't done yet, and then I'll make a new floor for the boot to finish off.

It leaves me with quite a small boot and they're not massive to begin with.

Nice job! I know all about loosing boot space to batteries! Does this mean spaceballs bms is imminent?

No, not had time to get the code done. Will hack the simpbms some more.

Pete9008 wrote: ↑Tue May 09, 2023 1:02 pm Did you ever try changing the resistor values or is that the next thing to try?

Continued here as to not barge in on nathaniel's project thread.

I've not, I've dug out this https://github.com/damienmaguire/Prius- ... _curve.jpg suggesting the current sensors can read +-600V I guess these are the values that need changing.

Sorry, just noticed back here viewtopic.php?p=52911#p52911 that I offered to work out what needed changing but then forgot all about it

What current gain parameter values are you using at the moment?

4.56 and 4.5

OK so just to check I'm not losing it a 12-bit adc gives 4096 codes or plus and minus 2048. 2048/4.5 = 455A which is roughly where you start to trip so the explanation of the ADC clipping at around 450A does make sense.

What current would you like to be able to read up to, 500, 550, 600A? The higher you go the poorer the measurement resolution, and the higher the noise, you actually get.

600A I wonder if it's worth logging the raw ADC value to confirm if it's getting close to the boundary

OK, think I now understand how that circuit works. Three resistors need to change to get the new gain and keep 0A in the right place. Give me a few minutes to work out the equations.

I think you need to change R44 and R49 from 62k to 47k and R48 from 3.16k to 2.7k. That should keep 0A in the same place but give around an extra 33% range (450A to 600A). The current gains will then need to change to around 3.42 and 3.38 to compensate in software.

Spreadsheet used to calculate attached.

Note - first order only, it doesn't take account of the load of the R45 and R44 on the input potential divider (which should be pretty small and since I've aimed to keep zero and the end points in roughly the same place as the standard values it should cancel out).

Note2 - This does assume that the Prius CTs can cope, it will need the output to go to -1.67V and +11.66V which should be possible but only if the neg supply to the CT is really -5V (iffy if it's an early charge pump version of the logic board)!

Note3 - It's a slightly unusual circuit in the way the offset is removed. Wondering if it leads to a slightly non-linear transfer function?

Edit - The other possibility is that you are already getting 500A and it's the il gains that are wrong!! Might be worth checking how it was originally calibrated if you have any way of contacting Damien?

Bigpie wrote: ↑Tue May 09, 2023 3:03 pm I wonder if it's worth logging the raw ADC value to confirm if it's getting close to the boundary

Wouldn't do any harm and would prove whether or not it's an issue.

Thanks

I'm suffering with the brain fog and exhaustion too since my recent covid so not sure when I'll get to the motivation of removing the board for modding but will put modifying the firmware for a raw il1 and il2 parameter for logging on my todo list.

I'll ask Damien on Discord. And it is the early board, first batch.

Sorry to hear that, hope it improves soon. There are actually some studies going on at the moment to see if B12 supplements help with long covid link, early days but might be worth a try??

Thanks, I'll mention it, I need to go to docs anyway, lungs are in state and I've got 15k tough mudder in 8 weeks :S Glad to see the B12 seems to be working for you.

Damien's response: Dc first then ac rms x1.414 to peak values

Bigpie wrote: ↑Tue May 09, 2023 4:20 pm Thanks, I'll mention it, I need to go to docs anyway, lungs are in state and I've got 15k tough mudder in 8 weeks :S

You'll probably need blood tests before they could prescribe anything. If you want to try something sooner then sublingual methylcobalamin tablets seem to be the best option. My wife had covid about the same time as you, no fatigue but she's still trying to shift a persistent chest infection/cough too, not nice.

Had to look up what a 15k tough mudder was - not sure I could have ever managed one of those - good luck with it!

Bigpie wrote: ↑Tue May 09, 2023 4:20 pm Damien's response: Dc first then ac rms x1.414 to peak values

Sounds like the current gains should be right then.

Don't suppose you know whether the MG fault lines from the inverter are active high or low? I've either misunderstood how they work or got a duff inverter here.

I see you've got that fault signal figured out.

Finally got the 3rd pack wired in and setup with SimpBMS, though I have a bug somewhere in the balancing code/offsets, I now have 26.4 kWhs
Delta is larger than I'd like so will get the balancing code sorted next week.

The get this to work, I've got 1 can bridge changing a single module as I had duplicate IDs in the 3rd pack, then another can bridge for the entire pack, so quite a few places to check for the can messages to get messed up.

Last job is to get some wood cut and some boot carpet then I can move on to something else.

That should help with the range How many miles do you expect now?

Are the strings always paralleled or do you have separate contactors for each?

Also realised that the current input circuit block that you posted, and that I used for calculating the resistor values, was for MG1 so the component references I quoted will be wrong. It's the equivalent ones in MG2 that will need changing instead (assuming it is the MG2 CTs you are using?)

I'm hoping for 80 miles maybe. Though I've got some issues with the can bridge code that I need to address.
I have contactors for each and a fuse for each, so when the car is off the packs are all isolated, ideally the BMS would decide if it's safe to parallel each pack, but at the moment this isn't the case.

I hope to drop in and show you the Beetle over summer if you're up for it. One time we go without the kids

Ahh right, good job you mentioned before I go removing resistors

Bigpie wrote: ↑Mon May 15, 2023 6:13 pm I have contactors for each and a fuse for each, so when the car is off the packs are all isolated

That seems like the safest way to do it

Bigpie wrote: ↑Mon May 15, 2023 6:13 pm I hope to drop in and show you the Beetle over summer if you're up for it. One time we go without the kids

Sounds good, drop me a PM when you want to sort something out.