VESC-Tool 2.0 and Firmware 4.0 - The beginning of a new era - (SERIOUS)

I just got HFI working no problem with my Flipsky Dual FSESC4.20 Plus and TB 6355’s. I can start from a stop in my room on the thick carpeting with no cogging at all running the Torqueboards AT Wheels. Setup was very easy. Looking forward to getting out to ride tomorrow.

4 Likes

Sorry for getting back late on this one. The serial port error is expected for a short period after a FW update.
However, if the device does not reconnect, or only reconnects after cutting power to the VESC and then re-powering it, the bootloader is broken or not existent.
If the Bootloader is broken, SWD Prog will resolve the issue.

Hi there, when the FW update it is going to be available please? I have the wobbles problems…the T=1000uS did not worked for me…Any advice?

Thanks

5 Likes

The 100OuS should work. You need to use the desktop tool, set the time constant and then run detection in the FOC Menu, not Wizard. You could also use FW 3.65 or 3.67 FW and then use the Wizard.

1 Like

Why not use the app?

4 Likes

App also works, but be sure to manually detect.

3 Likes

The desktop tool is also an application.

23 Likes

You have achieved peak Brian with this comment.

33 Likes

I used the phone app manually, did not work…Hope this will be solved soon, I am just riding with caution (a bit boring not being able to go faster)…

1 Like

@DerelictRobot SWD proged both VESCs last night. Everything works fine (other than the motor detection wizard - have to do the detection one at a time through the Motor Settings tab) until we try to configure App communication over CAN. As soon as we start trying to do anything over CAN, the firmware and/or bootloader corrupts.

Next question, why does VESC tool 2.02 load hardware 6 firmware onto my MK3 VESCS instead of the hardware 6_MK3 firmware? Andrew and I grabbed the correct firmware from Vedder’s GitHub, and I didnt even realize that the tool was putting the wrong firmware on until then.

4 Likes

VESC Tool looks for the FW version on the device and then it matches the FW it uploads.
Once MK III FW is on the device it will only load MKIII FW.

1 Like

Have loud speakers playing music. If you’re against that, have it play the BLDC whine XD

7 Likes

I went and loaded the 60_MK3 firmware directly using an stlink & swd cable towards the end of our troubleshooting but it had no effect.

Some context, @BenjaminF has a pair of brand new VESC6 MKIII but has not been able to get them setup. He’s a pretty advanced DIY builder, nothing else in his build was indicative of being a cause for the issues we were seeing. He has encountered a variety of issues trying to get setup completed & the pair of ESCs responding normally to his remote throttle.

Trying to run the setup wizard initially returns one detection with some pretty strange values, and ultimately fails to complete due to comms issues over canbus, coming back with the ‘old firmware’ error.

Here’s the first setup attempt:

Both show sensorless & the first VESC is returning 1000mWb flux linkage. Sometimes this fails to complete the detection, sometimes it shows success but still has these values.

Manually running detection on each VESC individually produces different results, we updated T to 1000 vs 4000 and ran each detection via USB individually:


Hall sensors responded normally during these manual setups, and we received correct values & foc detection results.

Going through the Input setup wizard would do one of two things:

Fail on the ‘old firmware’ error when trying to communicate over canbus, or it would detect both VESCs on canbus and complete the Input Setup Wizard to the end (with proper PWM min/max values recorded and set).

However, only the master VESC spins up/responds to throttle after the input setup wizard is completed. Attempts to manually configure input on the VESCs resulted in loud violent cogging/lockup behavior. Checked canbus cable, it is canH/canL only and well made.

All around, pretty strange behavior coming from 2 brand new MK3s.

We updated the firmware to 60_MK3 after going through the above steps, and repeated to the same exact results.

10 Likes

Dont make me blush :blush:

@Trampa I emailed Dave and Ted about this already, if you could would you direct them to the robot’s post? I feel that it explains the issues more concisely than my long rambling email :joy:

4 Likes

Could be a HW problem with one VESC. Dave will instruct you and if the problem isn’t solvable, we will swap the VESC.

1 Like

HFI on a Monoboard, no sensors involved:

8 Likes

There is a lot I don’t like about newer @Trampa firmware. A lot. Like a shit-ton.

But seriously - credit where credit is due - HFI sensor mode is the motherfucking bomb ass shit. It’s so fucking awesome. It really is the bee’s knees.

It’s like, I don’t know, the only single reason that I would for a moment strongly consider NOT using Ackmaniac 3.103.

It makes me really conflicted.

15 Likes

It’s rare that something new gets you so jacked. I’ll try it when the firmware updates don’t lag and time out.

9 Likes

I would feel comfortable running it if it didn’t make one of two identical vesc stutter that works on every pre 4.xx version for no apparent reason
also why is power usage noticeably higher on 4?

3 Likes

Yeah bugs and lack of trust are big giant turn-offs, that’s for sure. And I miss the max erpm (top speed) limiter. Had to turn that up to an unreachable value because of bugginess and jerkiness.

But the HFI is awesome. So awesome

4 Likes