VESC-Tool 2.04: FW 5 >> A BIG STEP FORWARD

kilow · March 7, 2021, 1:24pm

Frank, any good source on how to do this?

Sebas · March 7, 2021, 2:10pm

might aswell go with sensors at this point

xsynatic · March 7, 2021, 2:32pm

In the mobile version, is there no way to enable live packets (heart) so i can run foc_openloop?

ducktaperules · March 7, 2021, 4:02pm

The Issue with a broken motor sensor is not the risk or bringing the 5v rail down, its the risk of getting an incorrect position reading and trying to drive the wrong motor phases as a result. This cause cause huge currents through the ESC which can damage fets or blow the DRV.

Trampa · March 7, 2021, 4:31pm

In theory yes, but in practice the Hall PCBs and hall hall probe legs were more likely to short out than a simple probe that is pretty well insulated and uses thicker and more sturdy cables. Also there are 3 Hall probes vs. one temp probe. And there is a 10K resistor in the Temp line while PIN 6 of the SESNS port has a direct connection to 3.3/5V.
Second problem with sensors is the wrong signal reading they can give and that can cause potentially HW damaging effects.If we can save away causes for trouble, it is a good thing. One part less to worry about…

fessyfoo · March 8, 2021, 7:09am

sturdy wires probably doesn’t help with temp sensor noise, which is the reason a lot of us have disabled the temperature sensors.

I may have misinterpreted, but felt like jeff was just saying, with a lot of motors… we’re just f’d.

Known bad firmware versions VESC/UNITY [ SERIOUS ]

The motor temp problem stems mostly from the fact that not everyone is running the same analog LPF on the motor temp sense and the lack of control on routing of the sense wires in the motor. I think a lot of the hardware has higher cutoff frequency around 100 hz which means noise passes through up to 1kHz. The vesc 6 LPF definitely makes this issue less prevalent. If you want to know why noise is coupled into the motors temp sense all you have to do is look inside a motor to find all the terrible coupling you need:

We’ll continue trying to improve the code to get reliable performance but there is only so far digital filtering can go when aliased samples are taken. On the stormcore we ended up swapping the cutoff freq of the LPF to address this but there isn’t much I can do for all the people using unity’s out there, except to say get a motor with better temp-sense routing.

Trampa · March 8, 2021, 7:29am

Simply cut the wire and glue the temp sensor to the front side of the coils for example. Running the sensor cables above the spools is no good idea. Our motors had the hall pcb in front of the spools to avoid this issue. Fast spinning stong magnets above a wire carrying a signal is not the best design choice. Riding without temp feedback is also no good idea.

Trampa · March 13, 2021, 1:35pm

A new beta software is online: Firmware 5.03 | VESC Project

Sensorless startup should now be almost perfect if it’s a VESC HD or VESC6MKV or 75/300.
It also features a new wizard for e-skateboarding.

Andy87 · March 13, 2021, 1:58pm

I‘m not sure if it wasn’t already addressed before, but I very often have issues to setup my vesc based escs via can.
Mostly when starting the input wizard I get the „all vescs need to have the same FW“ message, which doesn’t allow me to go further with the setup.
Switching on of the escs, starting the vesc tool new or switching the esc I’m connected to usually fix that.
I do have this issue with all kinds of vesc based escs as well as with a set of vesc 6+.
FW is 5.1, not yet ready to change to a newer version

tipsy · March 13, 2021, 3:04pm

This is exactly what I’m getting trying to do detection with the new 5.03

tipsy · March 13, 2021, 3:21pm

So “HFI Start” mode is sensorless + a little HFI blip?
Do I have to do any of that annoying HFI tuning process for it to work as intended?

Trampa · March 13, 2021, 3:40pm

Nope, it should work without tuning.

b264 · March 13, 2021, 4:15pm

Sounds like riding ass is going to be easy. Huge improvement over hfi.

edit: I retract this statement. Doesn’t looke like ASS works on just 3 phase wires, you need sensors

pmg · March 13, 2021, 4:18pm

So ass is like a self configuring hfi? Any improvements regarding the noise?

Trampa · March 13, 2021, 5:14pm

Self configuring, no noise.

tipsy · March 13, 2021, 5:23pm

Went on a test ride with “HFI Start” mode.
Sometimes startups are smooth, sometimes pretty coggy. About 50/50 I suppose.
My gear ratio is pretty favorable I think, 14/44 @ 110mm wheel.

Measured and wrote the Offsets section before doing detection. Phase filters enabled. Temp compensation is on as well.

Trampa · March 13, 2021, 5:30pm

Just run the detection, do not measure the offsets.

Pedrodemio · March 15, 2021, 5:37pm

@Trampa did Ben pull out this beta download? I can’t download it, even when logged in

hexakopter · March 15, 2021, 10:39pm

For me the current download link on the vesc-project forum works fine.

xsynatic · March 15, 2021, 11:09pm

@Trampa

I’m having this issue with 2 vescs and 2 motors already.

Happened with Neobox v6 on hubs and 6374 motors. Was fixed after going from 5.02 to ack 3.103 to 5.02

Now having the same with the unity. Did not happen with the latest UnityFW - does happen with 5.02

Even setting duty cycle to 87% doesn’t stop it from causing this behavior

Unity was flashed minutes prior. Motor and Input were done completely and both settings were saved.

Almost everything is stock VESC configuration except the test with duty cycle, battery current max and what app to use.

this is how the graph looks like. The big red and Blue spikes are when the crackling occurs.

UPDATE: i flashed 5.01, redid everything and the sound does not appear anymore.
Update 2: Nope, it does still appear but later on the throttle.

Update 3:

Does not happen on UnityFW.
Happens on 5.02 (detection on PC)
Happens on 5.01 but later (detection on PC)
Happens on 5.01 (detection on metr)
Does not Happen on 5.01 (detection on Vesc tool android)