Page 1 of 1

Inverter Firmware 5.20.R

Posted: Sat Sep 10, 2022 5:11 pm
by johu
You're not gonna believe it, after almost one year there is a new release.

It seems the pesky field weakening/unwanted regen/high speed tripout issues are now solved, good parameter tuning assumed.
I want to stress at this point that I'm not sure whether the rev limiter will limit motor speed in field weakening. Don't rev up the unloaded motor beyond base speed

https://github.com/jsphuebner/stm32-sin ... /v5.20.R-2

Sample parameters for Leaf motor can be found here: viewtopic.php?p=45280#p45280
  • Hill hold functionality
  • Bidirectional throttle support, throttle sets torque and direction (e.g. for boats)
  • Throttle speed mode, throttle sets speed rather than torque (also for boats)
  • Regen parameters received hopefully more intuitive names
  • Pull down PWM pins already in bootloader when active high PWM selected
  • Never run idle speed controller in off mode, caused some issue, don't remember which one
  • Less ADC averaging in FOC mode for better transient response
  • MTPA motor variables are now parameters
  • Field weakening is now based on around limiting PWM amplitude
  • Torque command is limited in field weakening mode
  • Improved PI controller anti-windup
Sine version is untested as always, please leave feedback here if you tested it.

Re: Inverter Firmware 5.20.R

Posted: Sat Sep 10, 2022 7:03 pm
by Bigpie
Might be worth noting, your syncadv will likely need to be decreased significantly from 10 and theres some new parameters for FOC

Re: Inverter Firmware 5.20.R

Posted: Sat Sep 10, 2022 7:15 pm
by johu
I removed syncadv :) (huh, again?)

I have also updated the tuning guide: https://openinverter.org/wiki/Using_FOC_Software

Most of all I no longer think that there are two valid syncofs that are 180° apart. Rather there is one value that attracts the internal magnets and one that opposes them. I think the latter is the correct one but you can't tell from outside which one that is. The correct one will yield smooth rotation in both forward and reverse direction.

Re: Inverter Firmware 5.20.R

Posted: Sun Sep 11, 2022 2:01 pm
by Bigpie
It drives really well. Had my motor up to 10k rpm, regen ramping seems a lot better. More torque for sure.

Re: Inverter Firmware 5.20.R

Posted: Sun Sep 11, 2022 2:09 pm
by johu
10k, wow! Thanks for posting back.

Re: Inverter Firmware 5.20.R

Posted: Sun Sep 11, 2022 2:22 pm
by Bigpie
My motors the front Outlander, supposed to be rated to 14k rpm

Re: Inverter Firmware 5.20.R

Posted: Mon Sep 12, 2022 11:51 am
by Dusty4
I've tried this firmware out briefly in my beetle and leaf motor using your parameters Johannes. It is great! Seems to have more power down low and didn't have any unwanted regen. Great work on the simulator Pete and Johannes and others! I didn't have time to grab any logs if you need them but should be able to on the weekend.

Re: Inverter Firmware 5.20.R

Posted: Mon Sep 12, 2022 11:54 am
by arber333
I will try this for sine with my Mazda :). Report back when i do...

Re: Inverter Firmware 5.20.R

Posted: Mon Sep 12, 2022 1:36 pm
by UStas
Hi, folks.
I`m try 5.20 firmware.. well, what can I say - the unexpected recuperation seems to have really gone... but the behavior of the inverter itself has become completely unpredictable, it seems that the inverter has been torn off the throttle and it lives its life 1. When pressed throttle, it accelerates very sluggishly from the spot, but when passing 30 km/h, it suddenly accelerates, even if the throttle is not completely pressed... Setting lqminusld=1 make better with start from spot, but acceleration worse than my current firmware v.5.06.
2. When driving at a speed of 60-80km/h with full release of the throttle, it continues to far to the full even when you press the brake, it continues to turn with a large flow rate, that is, to stop, you need to suppress the motor by brakes!!! :oops:
3. There are some new settings in the firmware, but configuring them did not improve the behavior...
4. The motor is still blocked when the neutral is turned on during movement, last time on the previous firmware (5.12) this broke of the hinge...

Re: Inverter Firmware 5.20.R

Posted: Mon Sep 12, 2022 1:42 pm
by johu
sounds not too good but not too surprising either:
1. MGR apparently needs large lqminusld, maybe 5. Try it
2. Please check curkifrqgain=0 and please post the parameters with new firmware (the ones you attached are your older ones). Also make sure syncofs is well tuned
4. damn, forgot to keep this. Ok, no neutral for now.

Re: Inverter Firmware 5.20.R

Posted: Mon Sep 12, 2022 7:30 pm
by UStas
johu wrote: Mon Sep 12, 2022 1:42 pm sounds not too good but not too surprising either:
1. MGR apparently needs large lqminusld, maybe 5. Try it
2. Please check curkifrqgain=0 and please post the parameters with new firmware (the ones you attached are your older ones). Also make sure syncofs is well tuned
4. damn, forgot to keep this. Ok, no neutral for now.
1. I`ve try it, is not help :(
2. Try curkifrqgain=0 - not help



UPD: May be isn`t clear from video, but after 80kmh I release the throttle, but motor continue to drive and when I press brake it still provide power (see ampermeter on dashboard) I need continue to press brake for stop a car. :twisted:
OVERCURRENT was because I set "idcmin": -100 for save my transmission.

About: "Also make sure syncofs is well tuned" - with 5.06 firmware works perfect. If only we don`t need tune it for each fw version?

Re: Inverter Firmware 5.20.R

Posted: Tue Oct 18, 2022 9:54 am
by Romale
Hello, Johu!
Please, could you return the neutral to the new release?

Re: Inverter Firmware 5.20.R

Posted: Tue Oct 18, 2022 10:14 am
by arber333
Romale wrote: Tue Oct 18, 2022 9:54 am Hello, Johu!
Please, could you return the neutral to the new release?
I think we would be better off if we could use "Neutral" function as a NOTAUS switch. What do you think?
So a single function of "Neutral" should be applied when either/or conditions are in effect:
- Notaus button on pin Mprot is pulled to GND
- Both FWD and RWD inputs are floating

Would that be doable?
I think the benefit would be greater explainability to TUV inspector how/why we dont trip the main contactor on the road...

Re: Inverter Firmware 5.20.R

Posted: Tue Oct 18, 2022 2:01 pm
by Romale
johu wrote: Mon Sep 12, 2022 1:42 pm
2. Please check curkifrqgain=0 ...
I also had it with the new firmware (gas reset, but the motor kept going and even the +12 input brake did not stop the motor) I set this parameter to 0 and the motor now responds perfectly to the gas pedal. there is only one problem, a very large idle current and the controller is very hot. it was tested on a bundle of "small motor generator" Nissan note e-power + Honda IMA controller.
Screenshot_20221018-164848_1.png
Screenshot_20221018-165156_1.png
Screenshot_20221018-165254_1.png

Re: Inverter Firmware 5.20.R

Posted: Tue Oct 18, 2022 2:46 pm
by Romale
johu wrote: Sat Sep 10, 2022 5:11 pm I want to stress at this point that I'm not sure whether the rev limiter will limit motor speed in field weakening. Don't rev up the unloaded motor beyond base speed
by the way, the speed limiter works over the base speed.
this motor has a speed of 1000 rpm from 36 volts without field weakening, however I rotate it 3000 rpm and the limiter is running (200 hz)


Re: Inverter Firmware 5.20.R

Posted: Fri Jan 20, 2023 4:42 pm
by Boxster EV
arber333 wrote: Mon Sep 12, 2022 11:54 am I will try this for sine with my Mazda :). Report back when i do...
Hi Arber

Did you report back on this (sine version)? Would appreciate your feedback.

Re: Inverter Firmware 5.20.R

Posted: Fri Jan 20, 2023 7:31 pm
by arber333
Boxster EV wrote: Fri Jan 20, 2023 4:42 pm Hi Arber

Did you report back on this (sine version)? Would appreciate your feedback.
No testing yet as it is -2deg C outside and i took Mazda battery in house to power my solar. At least it is warm.

Re: Inverter Firmware 5.20.R

Posted: Fri Jan 20, 2023 8:20 pm
by Boxster EV
arber333 wrote: Fri Jan 20, 2023 7:31 pm No testing yet as it is -2deg C outside and i took Mazda battery in house to power my solar. At least it is warm.
Wow. That's what you call a hybrid battery! :D

I will try 5.20.R tomorrow. If anyone else has experience with this sine version I'd love to hear your feedback.

Edit: now updated and write-up here:

viewtopic.php?p=51689#p51689

Re: Inverter Firmware 5.20.R

Posted: Sat Mar 04, 2023 8:24 am
by gesso
^^

A bit late, but 5.20.R (sine) is working great for me in my limited testing so far.
But that's not surprising as my setup is fairly similar to yours (reverse rotation LDU, single Pacifica hybrid pack, fairly light car)