You are here

VESC with auto calibration of grace one motor (videos included)

11 posts / 0 new
Last post
ligi
ligi's picture
Offline
Last seen: 10 months 3 days ago
Joined: 2020-08-21 08:21
Posts: 19
VESC with auto calibration of grace one motor (videos included)

Hi,

 

I am trying to get the VESC (VESC 6 MKIV) working with a grace one motor for an e-bike project.

 

6abbc5ba-a814-42b0-b5c6-14f57d01925f.jpeg

Unfortunately I cannot get any movement out of the motor - seems to be totally unaffected. Detecting gives me success but there is no difference if I connect the motor or not. Is it possible that the voltage is not high enough? I am currently using 38V - any hint on what I could try would be much appreciated!

Currently I only connected the 3 phases - not yet the HAL-sensors.

 Selection_249.png

 

But somehow the motor seems to be connected. If I turn the wheel I see spikes in this graph. There is a direct correlation between me turning the wheel and the ERPM spiking to 1

 

Selection_250.png

 

And one more problem (but just secondary currently - no need for BT until the motor spins) - I can connect via USB - but when trying to connect via Bluetooth (the BT-MAC gets detected) - I am getting this error:

 

Selection_252.png

ligi
ligi's picture
Offline
Last seen: 10 months 3 days ago
Joined: 2020-08-21 08:21
Posts: 19

Update: 

I got some movement from the motor by connecting a PSU with 48V *YAY* - unfortunately not yet much as the PSU only has 0.5A - so it resets in the test as the voltage drops. But really happy the motor seems to work ;-)

Was fearing a bit that the motor or the VESC is broken - but it seems it was just because I was to hesitant starting with a high voltage. Will now build a battery with the correct voltage and do more testing.

The bluetooth problem persists - if anyone has an idea what to do there - would be much appreciated.

---
friendly greetings,
ligi
https://ligi.de

ligi
ligi's picture
Offline
Last seen: 10 months 3 days ago
Joined: 2020-08-21 08:21
Posts: 19

I now attached some batteries so I have more amperage at this voltage. Unfortunately I still cannot make it work. The motor does do some movement but not a full rotation when autodetection. I made some videos - maybe this gives hints on what I could try:

https://youtu.be/QmVxjRLJ99U (running autodetection)

https://www.youtube.com/watch?v=_vje5P29ZCs ( graphs when just turning the motor)

I also tried it with lifting the bike so the resistance on the wheel is lower - but can't film this - the bike is quite heavy and I only have 2 hands ;.)

Any idea on what I could try are most welcome!

 

 

---
friendly greetings,
ligi
https://ligi.de

TheFallen
Offline
Last seen: 5 months 3 weeks ago
VESC Original
Joined: 2017-09-11 11:46
Posts: 222

It ought to work from 10s unless those (Makita 5s) power tool batteries have any smart circuitry that's clamping the output when it sees a current. I've always found auto-detection to be a little terrifying and it only moves the motor a little. I've also found using the keyboard to control motors a bit hit and miss, you don't know if the duty cycle setup is quite right. Have you tried an eBike throttle yet?

ligi
ligi's picture
Offline
Last seen: 10 months 3 days ago
Joined: 2020-08-21 08:21
Posts: 19

Thanks for the Answer @TheFallen

No I did not yet try a eBike throttle. Wanted to try this after I got the motor working. Thought I do one step after the other (also did not yet connect hall sensors). The makita should deliver a lot of current before shutting down - but I also tried with plain LiPo's. 

So you suggest not using autodetection? In videos I have seen autodetection make full rotations - so you say it could also work after movements like shown in the video? I can attach the throttle - just did not think it is time for this yet.

---
friendly greetings,
ligi
https://ligi.de

TheFallen
Offline
Last seen: 5 months 3 weeks ago
VESC Original
Joined: 2017-09-11 11:46
Posts: 222

If auto-detection passes then it passes. It doesn't necessarily make full rotations.

ligi
ligi's picture
Offline
Last seen: 10 months 3 days ago
Joined: 2020-08-21 08:21
Posts: 19

I now attached a potentiometer - but still face the same issue.

Now I searched a bit more and found the old / more manual method with this help:

Selection_280.png

Wonder if what you see in the video is more the first one (weak) or the second one (cog)

I am guessing more cog. Unfortunately decreasing ERPM did not work for me. Tried a lot of values.

Also interesting: even though the controller should support 60V - when I use 59 the wizzard fails (fail to detect sensors) - and measurement of R and L fails for the old method.

Would really love some advice what to try next. I am kinda out of ideas on what to try unfortunately.

---
friendly greetings,
ligi
https://ligi.de

TheFallen
Offline
Last seen: 5 months 3 weeks ago
VESC Original
Joined: 2017-09-11 11:46
Posts: 222

Yeah, don't exceed more than ~55V battery voltage for a 60V controller. You get voltage spikes from the motor which could take you above 60V which will then destroy the ESC.

ligi
ligi's picture
Offline
Last seen: 10 months 3 days ago
Joined: 2020-08-21 08:21
Posts: 19

Thanks for that info! I thought 60V safe means I can have max 60V at the input. What exactly get's destroyed here? And is it a "noticable" destruction or is something working badly afterwards?

---
friendly greetings,
ligi
https://ligi.de

TheFallen
Offline
Last seen: 5 months 3 weeks ago
VESC Original
Joined: 2017-09-11 11:46
Posts: 222

The destruction depends on what dies. The MOSFETs are 60V max and the DRVxxxx is a very hard 60V max. The DRVxxxx chip in charge of driving the MOSFETs and reading the current sensors as well as converting the battery voltage down to +5V. The DRVxxxx chip is probably the more fragile of the two but if you've had big voltage spikes anything may have popped. Sometimes it's obvious when you open the VESC mkiv's lid but that will definitely void your warranty so maybe email Trampa support?

I've only ever run VESCs at 10s because of this, early v4.12 hw was notorious for blowing up under rapidly changing loads at 12s. I believe the newer mkvi VESCs will handle 12s very well but running 14s, it's entirely possible that a tiny voltage spike of just 1.2V has killed it.

ligi
ligi's picture
Offline
Last seen: 10 months 3 days ago
Joined: 2020-08-21 08:21
Posts: 19

I got another motor now and this one works as expected:

https://youtu.be/vbRhLKyVTaw

So it must be something with the other motor. And it seems that the ESC is OK. Would still be interested what's up with the other motor and happy about tips on how to get it going. Unfortunately it is too late now to send back the motor.

 

---
friendly greetings,
ligi
https://ligi.de