hi
ich have the same problems like @rich.
The Motors are cogging when starting in offroad conditions, slightly uphill only cogging, no start up is possible.
On flat concrete nice start up with slight cogging.
My Motors are 118kv Trampa and runnig in FOC sensored.
After my last tour i print the Faultcodes of both Vesc6 and the Motor config.
All Files are at my onedrivefolder
https://1drv.ms/f/s!Arg8WO10ccy4iqkSuNZxqtVD8u745A
Here are the datalog of the Tour
And a Vid of the cogging.
https://1drv.ms/v/s!Arg8WO10ccy4iqkTmASG7cshnH3Anw
I find that after about 10 min after the tour the temperature of the motors and the fets is still very high.
T Fet 35,60°
T Motor 51°
For the cogging, double check your hall sensor configuration. You may not have enabled it under the FOC tab. You should have a constant smooth torque at 0 RPM or even during forced reverse. You can test by setting duty cycle in VESC tool to 0.05 or lower and stopping the wheel with your hand.
Regarding the ABS_OVERCURRENT faults, set your abs current max higher. It looks like you have it set at 50A since it faulted at 51A. You should have this much higher than what you actually give the motors for peak current. Try bumping it up to 100A or keep it at the max 150A.
Cheers!
Hi Josh, frank
thx for your quick help.
i think we fix the the problem.
i use the Wizard again without any belttension. After this the cogging are 99% eliminated, only at very deep soft dirt is a little little bit cogging at the Start. I find that is not worth mentioning. But the fault erros are the same. I only change Battery current max from 60 (the wizard) to 100 and everything Fine.
mcm 50a
mcm brake -50a
amc 150a
bcm 60a -->100a
bcm regen -40a
So iam little Bit confused...
@frank i read other posts of you and you wrote motormax = battmax
i change the battmax from 100a to 50a and no faultcodes. At 60a battmax 15 faultcodes per Vesc, at 100a and 50a no faultcodes. o_O
New stats
mcm 50a
mcm brake -50a
amc 150a
bcm 50a
bcm regen -40a
The VESC can step-down the voltage and thereby step-up the current to the motor. For example when you use a e-skateboard to go 5km/h, you'll have about 14% dutycycle, so only 14% of your battery voltage is effective on the motor. If you have 42V battery, the motor will be seeing only 6V. Now if you need 35A in the motor to go this fast, you'll be drawing only about 5A from the battery.
But the VESC cannot boost the voltage. Therefore the upper limit for the battery current is the same as the motor current (this would happen at 100% dutycycle, but VESC is limited to about 95% dutycycle)....
Good Morning
Using little higher values actually for extreme Mountainboarding :
MC 130A
BC 100A
AC 150A
Getting this ABS overcurrents by hard acceleration...
For my understanding the MC and AC is to close together !?
Can i move higher with AC ?
Or is there another reason ?
Have a nice Day
Jenso
You can try using 160A. It would also be useful if you can connect a USB cable after getting the faults and print them in the terminal to see at which current they occurred. It might also be necessary to increase DRV8301 OC Adjustment to something like 18 as the DRV could be cutting the current making the control loop unstable.
If nothing helps you can also increase the abs max current to something like 180A, which effectively disables it for portions of the motor revolution, but before doing that you should print the faults and see at how much current they are triggered.
Good Morning Ben
Thanks for super fast reply.
I have to work actually but will do more testing this evening... will report
Thx
Jens
@Necromenz, good 2 hear that your cogging is gone!
I want to mention that after some rides there was sometimes (very seldom) cogging at startup in sensored FOC. After more rides it got worse but fortunately I found the reason for that. It was the sensor wire connector or extension connector or wires, it seems very sensitive regarding vibrations. I put duct tape around the connectors and wires to stiffen it up and that solved my problem. Now I understand why @frank suggest to solder extension wires instead of using JST connectors, if the problem come again I'll follow this suggestion.
changed the values as you suggested
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170824_164432.jpg
here comes the fault code from about a minute
... from both VESC :
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170824_165959.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170824_170005.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170824_170013.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170824_170035.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170824_170039.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170824_170043-Ko...
How much influence have the used battery on this problem ?
Just wanna make sure that not the battery is bad...
Thanks
Ups it dont shows the pictures, in the editing menu i see them.
Addet Links to my website, please click on the links above to see the Faultcodes and Settings
Thank You
Jens
It looks like the faults are triggered just above 160A, so things are probably not too unstable. You can try using 175A and see if it helps. If something breaks I can repair your VESCs again. This is a good test to see how much HW6 can handle :-)
Some technical details in case someone is wondering about the higher then 165A readings
Regarding the current measurement capabilities, 165A is actually not the upper limit if the method used for sampling is taken into account. I realized this yesterday while trying to figure out why so many current samples are above 165A while the ADC never reads something on a single phase that is higher. The reason is that when three shunts are used, my implementation uses the two shunts that are in the zero vector for the longest duration during the modulation. This means that when the phase lag of the current is low, and thus the currents and voltages are almost in phase, the phase with the highest current will selectively be avoided for calculating the total current and the truncation by the ADC has less influence. I'm working on a simulation to figure out exactly how much current can be measured for a given phase lag and will post some graphs later in a new post.
Okay i will do
175A AC
Will test in the afternoon.
Report back.
Thanks
Sounds good, let me know how it goes!
I will make a simulation of the currents in the next days and post the results or make a video. After a quick test in my svm simulator I have found that it is possible to optimize the current sampling combinations of the three shunts to measure up to around 190A without distortion, but that will require some adjustments in the firmware. Not too difficult though, so expect an update soon with higher current capability.
Okay here we go with 175AC :
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170825_185137.jpg
and new Fault codes :
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170825_184948.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170825_184935.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170825_184859.jpg
http://www.e-toxx.com/wp-content/uploads/2017/08/K800_20170825_184851.jpg
Uncutted Video with ackmaniac live data from this fault codes :
https://www.youtube.com/watch?v=uWTh3MWPerY&feature=youtu.be
There is also another thing you can try: when calculating FOC motor parameters, set the time constant to 500 µS instead of the default 1000 µs. It looks like you are getting the faults when the wheels loose traction and regain traction, and the current loop has to be really fast to keep up with that. I will try to do some testing with the firmware and see if I can implement a high current sampling mode that can go up to around 190A without too much distortion, then you can try increasing the maximum current even more.
Hey Ben,
i will try any settings you suggest
(-: crash test dummy ---- yeah
will do this FOC detection thing tomorrow... to be honest let me say that my "fault codes" dont appear by "normal driving"...
what you see is by super hard acceleration on meadow with backward shifted bodywheight...
its UN-SENSORED also.... so by normal uphill riding NO faults apear... check this uncutted video from right now
Let me say Ben that this is the SICKEST esc i ever used...
Thank you so much for your afford... remember if you ever need anything made out of Aluminium or CF let me know..,
Cheers
Jenso
Nice to see updates coming to the firmware for more current Ben! I think all the mountain board folk will really appreciate that. How do you guys keep those motors cool at those currents? My 6355's approach 80C after a few minutes at just 70A.
Thanks for the testing and reporting Jens, it really helps a lot. I have implemented the high current sampling mode in the latest firmware that comes with the next VESC Tool update. You have to activate it in the advanced page of the FOC settings:
With this mode the VESC6 should be able to measure up to around 190A without saturation distortion. The default maximum currents are increased to 120A and 160A, so to test this you still have to use the no limits fw. You can try setting the abs max to 190A and also the trick with the time constant and see if it helps. I look forwards to hearing about the results.
Hey
I installed the new FW on the first, got problems with motordetection... tryed to install FW again... now no USB connection possible... blue lights on... did i messed it up?
The second one worked good, i think the problem was that i set "High Current sampling mode" first... it dont works with Motordetection IMO...
Anyway one success one error... shit
Any chance to get it running again?
Regards
Jens
When you installed the firmwate the second time, did you unplug power directly after uploading? Doing that can mess up the firmware as that is when the bootloader copies it from the temporary location to the main location replacing the current firmware. If that process is interrupted the firmware will be incomplete and not run. So after the upload has finished power must stay connected for at least 10 seconds. I should probably write a warning about that... The only way to fix this is to connect a programmer and upload the firmware over swd. I can send one of my stlinks, but I don't know how to use them from windows. I think there is a post somewhere with a guide on that.
Regarding the detection, it should not fail from using that option, but I will give it a try when I get home. Maybe I missed something.
Yes thats it for sure.
Second time uploading i reacted as the other times after "first" uploading... powercycle ... none (-;
Yeah a friend got one STlink, hope i can meet him today... ordered one already, should be here on thuesday
Necromenz, the guy i stole his tread (-; has written an step by step guide and also a YT Video... THANKS Badti
By setting the "High Curent" , the Foc detection gives me no "R" it always says "R" is =0 ...
Also played with the 500ms versus 1000ms ... but i think its the "High Current" which not works with Fod-Detection
The good is i will have more time for working today (-; No problem shoul be running again soon
Thank you Ben for implementing this new feature... looking forward to 190A AC ((-:
BTW should i still " increase DRV8301 OC Adjustment to something like 18 " ???
Have a nice Sunday
Jens
I just gave it a try, and for me detection works in high current sampling mode. I wonder if something else was wrong. If it happens again I can have another look.
You should still increase the OC adjustment :-)
Yeah maybe i messed up anything else... i´m good in it (-: You KNOW
First ST-Link shipped, should arrive on Tuesday hopefully (second backup st-link orered on mouser)
Will report
Jens
Spend hours on trying to connect the St Link to the Vesc...
Followed the sequence "necromenz" posted, i dont get the step :
7. Open the tap TARGET and klick Connect --> the program show you the infos of your VESC
The ST Utility Tools always says :
"Can not connect to target!
Please select "Connect Under Reset" mode from Target->Settings menu and try again. If you're trying to connect to a low frequency application , please select a lower SWD Frequency mode from Target->Settings menu.
No target connected"
Its the same @boesila mentioned... but he made it to work....
Tryed a original ST-Link and a China one...
In theory the SWD pins are just GPIO pins that default to the SWD alternate function. So an application could at startup configure those pins to become GPIOs and no longer work as SWD pins. The VESC firmware does not reprogram those pins.
My first guess would be that you have something wrong with the wiring between the VESC and the STLINK. Did you hook up the 3.3V from the VESC to the STLINK? Congrats you just blew up the VESC! Don't do that (on the chinese STLINK). Did you swap SWD and SCK ? Did you power the VESC externally? Connect a battery?
Thanks Roger for your reply.
No 3.3v connected.
Battery powers Vesc
IO and LK are NOT swapped
??? I have to swap them???
Okay tested again...
the problem is the VESC itself... if i connect to the my other VESC it works in a second !
So my connections are okay but the Vesc itself NOT...
Did someone have an idea what could be wrong? The blue LED is on, no reaction on USB and SMD !
What i have done ? What could i do (-;
thanks in advance
When the CPU doesn't react to SWD I usually consider it "dead" and "burrial" is the only option.
One thing you can test is if the board has 5V and 3.3V. (I don't know where the blue power-led connects). 5V should be available on the hall sensor connector. The 3.3V is on the SWD connector. Check the voltages....
@Roger Wolff
Thanks for your reply.
I checked the voltage, it seems fine... can measure 5V and 3.3V on different ports
3.3V for example on SMD Port
5V on CAN and COMM
on both VESC i measure the same on the ports, except on HALL port :
On the one which not work is 5V on the other 3.3V
But i think it depends on the "jumper" or "switch" next to it... Only wanna mention this too !
What check next ?
I'm not an expert on VESC6 repairs: I haven't even opened up mine. I'm told there is a sensitive heatpad that will break if you open it up.
If your powersupplies are OK, I don't know what to check next. As I said: It looks as if your CPU is fried somehow. I have a too-big pile of "doesn't work, need to figure out exactly what needs to be replaced". I tried it a few times and usually after swapping out a few components, it still doesn't work and I give up.
Your best bet would be to get hold of Benjamin. Maybe he can do some more remote debugging that I missed, and/or offer to fix yours in an attempt to debug "what goes wrong" in the field (to try and improve the whole system).
@Roger
Thanks for all your input and afford.
I decidet to replace the faulty Vesc and send it to Benjamin for science.
@Benjamin
So here we go with new Vesc and new Firmware with the High Current Sampling.
15minutes Run with Metr.at Log: :https://metr.at/r/kEr84
Settings and Fault Codes from this Run: http://www.e-toxx.com/2017/09/02/high-current-cycling-ac190/
Hope it helps,
Cheers
Jenso
Another Run with same Values... except another Throttle Curve
Metr.at : https://metr.at/r/pnS7Z
and Fault Codes :
http://www.e-toxx.com/2017/09/04/same-except-throttle-curve/
I finally had a chance to test the "high current sampling mode". I notice from Jens metr.at logs, that he didn't reach the programmed motor current of over 160A (his max was under 110A motor current), so I thought maybe it would be useful to have someone who really tests using over 160A motor current and so I did that today.
The VESC 6 in my eBike now has a heat sink and today I was finally ready for the test and flashed the no limit firmware. I set up an absolute maximum current of 185A, set the DRV8301 OC Adjustment to 18 and enabled the high current sampling mode. Then I increased both the battery and motor currents in steps. Because my battery can't candle that much current I stopped increasing the battery current at a maximum of 120A. I was able to push the motor current to 170A without overcurrent faults when accelerating. You can see logs of it here: https://metr.at/r/n4jds and https://metr.at/r/MCi1q. Setting the motor current to 180A results in overcurrent faults when accelerating hard. I will write the faults at the end of my post with numbers. Faults 1-3 is from accelerationg with 180A motor current.
Faults 4 and 5 are with a 160A motor current, but with the wheel spinning freely without a load (tire in the air). With a load I had no problems with 160A motor current. Maybe that are some useful infos for you Benjamin, to determine the absolute current limits the VESC 6 is able to handle. If you want me to test more on that let me know. With a lot of effort I might also be able to test more than 120A battery current.
One thing I didn't understand yet is the decreasing motor and battery current while I am still using full throttle and I am not near the topspeed. For example in this record https://metr.at/r/n4jds the currents are going down beginning at 15:56:23 where I am just at 36kmh. The next 20kmh to 56kmh are a lot slower because of the decreased currents. You can clearly see the inflection point in the speed graph where the slope decreases after the currents get lower. Is it because of some PID control loop that trys "smooth out" the acceleration? Maybe it is possible to get a faster acceleration with tweaking some parameters. Let me know if somone of you have an idea?
Fault 1 : FAULT_CODE_ABS_OVER_CURRENT
Current : 187.5
Current filtered : 184.9
Voltage : 48.23
Duty : 0.363
RPM : 851.6
Tacho : -27
Cycles running : 7344
TIM duty : 3045
TIM val samp : 2
TIM current samp : 4200
TIM top : 8400
Comm step : 0
Temperature : 33.46
Fault 2 : FAULT_CODE_ABS_OVER_CURRENT
Current : 186.9
Current filtered : 185.0
Voltage : 48.43
Duty : 0.287
RPM : 361.0
Tacho : -32
Cycles running : 3292
TIM duty : 2414
TIM val samp : 2
TIM current samp : 4200
TIM top : 8400
Comm step : 0
Temperature : 34.35
Fault 3 : FAULT_CODE_ABS_OVER_CURRENT
Current : 189.9
Current filtered : 185.1
Voltage : 48.14
Duty : 0.390
RPM : 954.4
Tacho : 3
Cycles running : 13516
TIM duty : 3277
TIM val samp : 2
TIM current samp : 4200
TIM top : 8400
Comm step : 0
Temperature : 34.30
Fault 4 : FAULT_CODE_ABS_OVER_CURRENT
Current : 212.0
Current filtered : 182.4
Voltage : 47.61
Duty : 0.498
RPM : 1924.6
Tacho : 84
Cycles running : 1824
TIM duty : 4184
TIM val samp : 2
TIM current samp : 4200
TIM top : 8400
Comm step : 0
Temperature : 35.37
Fault 5 : FAULT_CODE_ABS_OVER_CURRENT
Current : 187.4
Current filtered : 176.0
Voltage : 47.73
Duty : 0.513
RPM : 1595.0
Tacho : 3263
Cycles running : 1624
TIM duty : 4312
TIM val samp : 2
TIM current samp : 4200
TIM top : 8400
Comm step : 0
Temperature : 35.57