no HALL sensors, help with BLDC braking

DIY Builds
1

using 2 motors on BLDC as without hall sensors FOC seems off the table, having a dangerous braking issue with PPM.

While Riding:
whenever I apply the brake (neg throttle) i feel the board stop, then go, then stop, then release brakes.

while on bench:
negative throttle stops the wheels if they’re moving, can easily turn them by hand while applying brake

I’ve set braking regen to -99A and max brake current to -60A and on “current control no reverse with brake” control scheme.

pleasee help. it’s the finishing touch here.

2

there’s your problem

i wonder why there’s brake weirdness :thinking:

set it to one of the following:
current
current with brake mo reverse
current with brake
current smart reverse
current hyst reverse

3

Oops typo, fixed. it’s on the no reverse with brake. I want to use smart reverse but that seems to make the throttle worse too, while also having no brake or reverse.

settings haven’t changed, just my ability to convey information

4

What ESC are you using? I think most people are using FOC these days, with sensors or without. Exactly due to weird issues like this :smiley:

5

tenka. (and before anyone rants, yeah IK the tenka acts weird for a vesc)

if you can give me a hand getting the sensors connected int he right order, or getting HFI working, or getting sensorless FOC working… honestly any of these seem like good options but I haven’t been able to figure them out on my own, the hall sensors really don’t wanna work for me (bad hall detection literally every time)

6

There’s nothing to it really, just set the motors to FOC and “sensorless” under the FOC tab. If you want smooth start you can search for “HFI” on the forum here to see how to tune it.

7

I’ll try that again but I got terrible results last time.

Edit: after watching Jeff’s video I now understand why. I have to tune this. Alrighty.

8

do you have any help with getting HFI working? I’m following all of the steps and threads I’m finding but things only look kinda right for one motor, the brakes are still super weak.

9

got RT data on motor 2, it looks very different to motor 1. straight vertical lines instead of dots forming curves. solder joints look ok

10


11

Have you tried swapping the motors between the ESC sides? That would help determine if you have a faulty motor, or a faulty ESC side.

1 Like
12

When swapping motors, both start acting weird, though the stuttering is happening on the same VESC side regardless of the motor.

Are you certain something here is faulty? That sounds like my worst possible outcome. I really want to hold onto the hope that the correct settings can fix this.

I’ve also tried swapping phase wires around on the MR60 plugs.

13

This shouldn’t do anything other than make you run detection again, which you should do I you’re swapping sides and testing

14

Yes of course, I run detection religiously after any motor related physical changes

15

Well, HFI is very hit or miss, especially on older hardware.

Remember that in sensorless mode it’s expected that the motors might spaz out / refuse to spin before the ESC can detect their position correctly. I.e,. need to push to start, and then apply throttle and test.

Anyway, if one side of the ESC is giving you trouble, then it’s pretty likely that it’s faulty hardware. Unless you aren’t programming the sides identically or something. After watching your video, I’m pretty sure that’s an ESC issue. Are there any reported faults by any chance? Go to the terminal tab in vesc tool and type in “faults” after you reproduce that stuttering issue.

16

Nothing besides a weirdly usual bunch of faults after the auto-shutdown timeout (set to 30M). But that also happens on my other board with a different VESC. No faults while doing the tests or riding it. (Metr unity)

Would the tenka or the evolve gt 5055 motors be the "older hardware? (AFAICT same motors they used in the GTX and GTR)

18

I never got the Racerstar 5065 motors to work worth a fuck on HFI.

Try the ld_lq_diff measurement (measure_ind 0.2 in terminal) like in the video. What do you get? If it’s less than 10% of total inductance, just forget about HFI on those motors.

15% to 25% is good, the higher, the better.

19

I get about 25%. honestly I’m starting to think the tenka is being fucked, which probably sucks a lot for me

20

25% should be really good. My experience with those motors on HFI has been really bad though with multiple motors and multiple controllers.

Try the same controller on other motors to see which detail is causing the hiccups.

21

I gotta get the VESC to co-operate first, I reflashed stock FW and it started behaving worse after detection. sudden halting, one side not responding anymore, even louder noised when there shouldn’t be,…

and this is the 2n’d tenka after the first one blew while braking downhill. I’m so inclined to just think MassiveStator shipped me garbage twice in a row.

I’ll test with new motors but my hopes are pretty low for this VESC at this point.