Hi Benjamin,
I was trying to analyse the impact of the analog current filters in the control loops. It seems there is only a little amount of lowpassing on the current sensors of the VESC 6 MK III. I tried to find the R and C values in the schematics, but was not able to find these for the MK III.
I was looking at the schematic of the MK V, but I expect that it must have a different design. Those filters should definitely be visible in my measurement.
Can you please give me the MK III schematic, or the R and C values of the lowpass filter?
Kind regards,
Elwin
Based upon my measurement I would estimate R = 1k and C = 4.7 nF:
Pretty sure its this one, https://vesc-project.com/sites/default/files/Benjamin%20Posts/VESC_6.pdf
100R 4.7n
That would surprise me. In the code I see that the mk3 also uses the switching of the current sensors. That schematic does not show any switching of the low pass filter.
Measured with the current filter turned on and off.
You clearly see some minor low passing behavior at the high frequencies. I would not be surprised if the wigely lines at the now frequencies is due to aliasing, which becomes worse without the filters as expected.
I think the MKV has stronger low pass filtering, which seems like a good idea.
The Mk3, MK4 and MK5 have the switchable low-pass filters on the current measurement with the same values as in the MK5 schematic, that is 1k and 15n (+1n). They are switched off during HFI (as you can see in the code).
The main difference with the MK5 are the switchable phase voltage filters which help track the motor close to 0 speed. One difference in the later versions of the MK5 is that C49, C57 and C59 are 220n instead of 100n.
Elwin, if you PM me your address I can see if I can send you an MK5. I appreciate what you are doing, just let me know if you need any hardware.
Mmm that is interesting, I would expect more attenuation and phase lag with those values. Always good to be wrong, opportunity to learn more.
Thanks a lot, I will send a pm.
What was reason behind changing 100n capacitors on the voltage filters to 220n? Does that make a material difference?
NextGen FOC High voltage 144v/34s, 30kw (https://vesc-project.com/node/1477)