Hello guys,
I have some problems configuring my 75/300 Vescs and need some help.
Hardware: 2x Hacker Q80 8M with following data from manufacturer:
1100gr
14 Polepairs - 28 Poles
24Coils
RPM Volt KV 180KV
Maximale speed 9000rpm
R= 0,013Ohm
Timing 25-30
5500w for 15sec
3500w Continuous
Idle Amps 1,5A/1,6A
8 Turns
58,8V Max
ERPM=147000
80mm Rotor
sensorless
Those motors are from an Aden Sport RaceKit with 2xQ80 Motors and 2 Jeti spin pro 200L controllers (RC Controllers) which are able to handle 170amps (peak 200amps) from battery at 8khz.
I was running them with 2 batteries with each 14s 16Ah Lipos which are able to deliver 192Amps continuous and peak 380amps. Each battery was running its own controller and motor.
I was not happy with the Jeti controllers because the 8khz and bldc mode were making the motors super noisy and they cut off because I took them to their limits of over 200amps. (measured at the battery side)
Here you can see the bike in action:
https://www.youtube.com/watch?v=wHoSp8RvKTg
So the motors are able to run 200amps for a short time (from battery) and thats what I want the motos to deliver in foc mode with the vescs.
After some problems with the vesc tool and trying a lot of configurations, I made them work. But when I reach a specific load/speed the motors are making weird noises and stop accelerate.
Driving slowly even little bit uphill is fine until I reach that weird noise/vibration. Motor and Vesc dont get hot, just a little bit warm when riding.
In the video, see link, the motors make a weird sound (happens twice in the video) and the bike starts vibrating. I dont get any fault code, when this happens.
https://www.youtube.com/watch?v=4xAMOP3uUXE
FOC detection was fine, I tried it often and with 4 same motors which had same values.
Directly after normal configuration, first I had problems with over current abs, changing Stator saturation compensation to 80% helped. (also tried with different lower values)
To solve the problem, I changed/tried:
- More than 100 Amps -> Problem appears earlier
- higher and lower DRV8301 OC mode Adjustment set to 31 -> nothing changed
- halfing and doubling Observer Gain -> not better
- Switching frequence higher and lower -> not better
Other values have not been changed because I am not enough into Vesc tool. I need more experience and more information what changing other values do.
Values in Vesc Tool for both Motors:
Motor Type FOC
Invert Motor Direction False
Motor Current Max 100
Motor Current Max Brake 0 (dont use Brake)
Absolute Maximum Current 450A
Slow ABS Current Limit True
Max Current Scale 100%
Min Current Scale 100%
Battery Current Max 100
Motor Current Max Regen 0 (dont use Regen)
DRV8301 OC Mode Current Limit
DRV8301 OC Adjustment 16
Battery CutOff Start 47,6
Battery Voltage Cuttoff End 42V
Max ERPM 150000
Max ERPM Reverse 0
ERPM Start Limit 80%
Maximum Wattage 1500000W
Maximum Braking Wattage - 150000 W
Acceleration Temperature Decrease 15%
MOSFET Temperatur Cutoff Start 85
MOSFET Temperatur Cuttoff End 100
Motor Temperatur Cutoff Start 85
Motor Temperatur Cuttoff End 100
Minimum Input Voltage 24V
Maximum Input Voltage 72V
Minimum Duty Cycle 0,5%
Maximum Duty Cycle 95%
Mimimum Current 0,1A
Fault Stop Time 500ms
Beta Value for Motor Thermistor 3380K
Auxiliary Output Mode Off
Sensor Mode Sensorless
Resistance 9,1 mOhm
Motor Inductance 3,5uH
Motor Flux Lincage 2,302 mWb
Current KP 0,0035
Current KI 9,11
Observer Gain 188,71
Openloop ERPM 700
Openloop Hysteresis 0,01 S
Openloop Time 0,01 S
Stator Saturation Compensation 80%
Temp Comp False
Temp Comp Base Temp 25
Switching Frequency 30KHz
Dead Time Compensation 0,080 uS
Speed tracker Kp 2000
Speed Tracker Ki 30000
Duty Downramp KP 10
Duty Downramp Ki 200
D Current Injection Duty 0,0
D Current Injection Factor 0,0
Sample in V0 and V7 False
High Current Sampling Mode False
Observer Gain At Minimum Duty 1%
Current Filter Constant 0,010
ADC
Positive Ramping time 0,1 S
Problem I think of: Cables from Vesc to motor are 70cm long
What can I change to make the bike work better?
Thanks and regards Oliver
Some pictures of the bike with vescs:
https://vesc-project.com/sites/default/files/imce/u7786/IMG_0938.JPG
https://vesc-project.com/sites/default/files/imce/u7786/IMG_0939.JPG
https://vesc-project.com/sites/default/files/imce/u7786/IMG_0941.JPG
It does sound like the controller is losing track of the motor under load.
To start with try setting sample v0 and v7 to TRUE and switching frequency back to 30khz. With sample in v0 and v7 set to true don't set switching frequency above 40 khz or the cpu will hang.
I see a couple of funny values.
Observer gain at minimum duty is 30% by default I think. Yours is 1%.
And current filter constant is default 0.1? You can click a button next to the value that will show the default setting.
DRV8301 settings will have no effect as the 75/300 does not use the chip.
Try setting openloop hysteresis and time back to 0.1s. Give the motor some more time to spin up, especially since you are on ebike.
The resistance and inductance of your motor cables will also be added to the detected motor values. Try playing around with resistance and inductance values with stator saturation set back to 0%. That you need such a high stator saturation compensation value is a sign that parameters are off.
Cool bike by the way
I agree with the previous reply, and will add the following:
In general I don't know if you can push the motors with too many amps using FOC, they are quite small and probably get way into saturation when approaching 200A. I'm planning to make a saturation mapping tool soon, which can help pushing the limits a bit.
Sounds exactly like what I was seeing, although i was using a revolt 160sh which really shouldn't saturate to badly at 180A or so.
Thank you very much guys, for your help! I really do appreciate!
Today I was able to try the modifications you suggested. First of all I did the motor wizard and input wizard again. New max current was 90amps for both motors. I left it like that till the end of my tests today.
With doing wizard, observer gain at minimum duty got back to default 30%, current filter constant is default 0.1 and openloop hysteresis time back to 0.1s
Then I changed switching frequency to 50kHz-70kHz with 10% saturation compensation because without those 10% it was cogging instantly with minimum load and giving me abs current fault codes.
From now on I used 15% saturation compensation.
At 60kHz it was little bit better than before (as seen in the video). At 50kHz it was worse. 65kHz was best option at this moment.
Changing the resistence of 13mOhm did not help. Tried 10 and 16 mOhm I think.
The best I did was changing the Inductance in small steps from 3.5uH to 5,5uH ! This was the game changer!
Now i can accelerate to about 40-50kmh without problems.
BUT the problem as seen in the video starts again higher speeds. With my other controllers I made 82kmh (GPS measured). So same problem like before but at higher speed. Riding like that did not bring fault codes.
Because of bad wheater I could not test more. Some ideas which values to change now?
I did not try "sample v0 and v7 to TRUE and switching frequency back to 30khz" yet.
Thanks !
Values from last setting I did:
Motor Current Max 90
Motor Current Max Brake -90
Absolute Maximum Current 450A
Slow ABS Current Limit True
Max Current Scale 100%
Min Current Scale 100%
Battery Current Max 250
Motor Current Max Regen -200
DRV8301 OC Mode Current Limit
DRV8301 OC Adjustment 16
Battery CutOff Start 47,6
Battery Voltage Cuttoff End 42V
Max ERPM 150000
Max ERPM Reverse 150000
ERPM Start Limit 80%
Maximum Wattage 1500000W
Maximum Braking Wattage - 150000 W
Acceleration Temperature Decrease 15%
MOSFET Temperatur Cutoff Start 85
MOSFET Temperatur Cuttoff End 100
Motor Temperatur Cutoff Start 85
Motor Temperatur Cuttoff End 100
Minimum Input Voltage 24V
Maximum Input Voltage 72V
Minimum Duty Cycle 0,5%
Maximum Duty Cycle 95%
Mimimum Current 0,1A
Fault Stop Time 500ms
Beta Value for Motor Thermistor 3380K
Auxiliary Output Mode Off
Sensor Mode Sensorless
Resistance 13,4 mOhm
Motor Inductance 5,5uH
Motor Flux Lincage 2,146 mWb
Current KP 0,0034
Current KI 13,42
Observer Gain 217,24
Openloop ERPM 700
Openloop Hysteresis 0,1 S
Openloop Time 0,1 S
Stator Saturation Compensation 15%
Temp Comp False
Temp Comp Base Temp 25
Switching Frequency 65KHz
Dead Time Compensation 0,080 uS
Speed tracker Kp 2000
Speed Tracker Ki 30000
Duty Downramp KP 10
Duty Downramp Ki 200
D Current Injection Duty 0,0
D Current Injection Factor 0,0
Sample in V0 and V7 False
High Current Sampling Mode False
Observer Gain At Minimum Duty 30%
Current Filter Constant 0,100
ADC
Positive Ramping time 0,1 S
Link to forum post: VESC help offered for private persons and companies
Website: www.electricfox.de
I wonder if it should measure the motor parameters at a range of currents and speed and have a mapping table to get the best performance. Like an advanced motor detection sequence where you set a range of currents and duty cycles to measure at.
Today I did some more tests, but this time with 130 Amps instead of 90 amps.
After changing that I directly get abs over current faults up to 700amps...
Still saturation problems as in the video seen, but at higher speeds.
I changed switching frequency up and down, Motor Inductance slightly up and down and Observer Gain doubled, but still problems and no fault code.
Without going to 30% saturation compensation I wouldnt even accelerate till 5kmh.
After spending 1400€ for the Vescs and 5 days work in it I start getting unhappy =(
Link to forum post: VESC help offered for private persons and companies
Website: www.electricfox.de
Its weird that your problems are at higher speed. Usually the saturation and poor rotor tracking is at low speed. At least that was my experience on my bike with a larger 80mm outrunner. But I didn't have problems until around 250A motor current and it ran well at all speeds provided the motor current was not too high.
Probably not the answer you are looking for but bldc mode will allow you to run high currents.
Also I would decrease the absolute current limit while you are playing around with settings. That gear system looks expensive and it can be quite rough on the drive train when the controllers lose track and high currents flow.
A gearing system or chain drive causes lash between the motor and inertia. That can cause problems because the motor detection will deliver bad results. Think of the rattling from a bad Motorcycle chain drive and the chaotic torque ripple.
You could try to eliminate that lash by removing the gears during detection.
I think I will try BLDC Mode once, and see if I can go with more amps. I will decrease them, thanks!
I did the motor detection with motors spinning free without gears attached. Thank you
Link to forum post: VESC help offered for private persons and companies
Website: www.electricfox.de
Today I had time to try out BLDC mode.
I wasnt even able to do the BLDC detection... I got an " bad detection result received"
I read the infos and tried in all kind of combinations:
2,3,5,8,10,20 amps etc.
ERPM 150,300,450,1000,2000 etc.
D 0,02 / 0,05 / 0,1 / etc.
Motor was running depening on values but still same error although I tried it on another identical motor. Battery Cuttoff was set correctly and no fault codes in terminal.
Cables are fine and motors too because I was running them after that with my old jeti spin controllers and had no problems.
Its time to give up...
Link to forum post: VESC help offered for private persons and companies
Website: www.electricfox.de
do you still keep working on your issue ?
i got similar issue just like yours
No, I dont accept investing more time in a product that does not work correctly with my motors.
Link to forum post: VESC help offered for private persons and companies
Website: www.electricfox.de
is it not because of too high ERPM?
CH