VESC FW 5.x - Beta Testers wanted [SERIOUS]

Everybody’s impression of what feels intuitive is different. That mode does not feel intuitive to me, I really like smart reverse.

4 Likes

I have always been a “current with reverse” kinda guy but since switching to fw5 iv been trying to give smart reverse a fair trial.

My one niggle with smart revers that always catches me out is if im rolling backwards a little but want to roll backwards faster. I instinctively pull down to reverse, but that engages the break applying power and forward acceleration, then i stop and it suddenly ramps into reverse, giving reverse acceleration.

i feel like if your rolling back and want to go forward you would push forward, so if you want to go back surely you pull back? why does this then effectively add positive acceleration, for a split second then add negative acceleration. the whole thing just seems janky and inconsistent.

My proposition would be that if your already rolling back when you pull reverse then you go straight to the duty cyle ramp, and this happens from your current duty rather than from 0 duty.

TL:DR;
forward should be forward power and backward should be backward power.
backward should never be forward power.

11 Likes

so my wiring was always correct and it was a firmware issue which is now fixed :partying_face:


8 Likes

I think this could be an improvement potentially. I’m coding it up now to test.

11 Likes

It would be… I’d appreciate this functionality too! Would make smart reverse :+1::+1: rather than just :ok_hand:

Just picked up a unity and successfully uploaded FW5.
Guess I better get the rest of the build ready to do some testing. Wondering if it’s still worth buying another of these unity’s

2 Likes

Vedder will probably have some opinions but I added this snippet of code to the smart reverse function which produces the described behavior. I think in my opinion it is an improvement. Just went out and ride tested it and was pleased with the behavior.

12 Likes

Is it just me or this last release has no PPM tab under APP.

Edit:
Nm. Had to switch app to use to PPM.

4 Likes

Hey All,

Im a little confused with my unity and FW5.

Setup everything using the can forward feature (new to me but ok)
I basically have to treat the unity like two can connected escs, sure no problem.

But now my unity wont work, and neither will my telemetry on my remote.
Usind metr pro unity and feather remote with unity fw.

metr turns on and connects, but no data gets through.
Feather can gas it but gets no telemetry.

Reverting to unity fw worked and telemetry came back.

1 Like

Settings and clicking on the blue gear near the module name.

Click Upgrade and wait until firmware update over BLE is complete.

4 Likes

I’ll try that tomorrow and get back to you.
I assumed it was vesc related since the feather is out too

The old unity firmware had a slightly modified packet structure. Because of this many Arduino style remotes have modified library due old unity firmware.

My guess would be you more need to update your remote firmware and set it up using standard vesc library rather than unity library.

@Deodand thanks for the super responsive changes. If vedder is ok with it then i look forward to another beta firmware to test.

4 Likes

Damm, the speed of implementation :heart_eyes: I’ll try it out too when available!

has some wierd behaviour today, dont know if its FW5 related but its not something iv not experienced before.

Basicaly when going up a long hill i started to notice a weird “pulsing” feeling with the power. here is the log section https://metr.at/r/L17ox?zoom_start=6320&zoom_end=6600

so at the point i marked with blue i started to feel the power drop away, you can see that the motor and battery current start to drop off even tho the PPM is stable. Shortly after you start to see the motor and battery current start to oscillate sharply, again whilst ppm is smooth. At the yellow mark i notice the weird feeling and let off the power.

Interestingly at around the same time the motor temperature seems to get really noisy. all recorded values are below the motor thermal limits but im wondering if there extra samples between the graph data points that were not logged. Its starting to get warmer here so it could well be a combination of warmer motors and noisy temp sensors but why did they get so noisy all of a sudden at the same time i started to loose power?

any ideas? any way to smooth the temp sampling?

6 Likes

Nice (or not) to see someone on the same situation

I have this happen since forever, when near the temperature limits and at high phase currents, they induce noise in the temperature measurements and the current limiting kick in and out all the time, depending on how noisy it gets really dangerous

3 Likes

There should now be some heavier filtering on the motor temperatures. Which ESC and which Beta version of FW?

2 Likes

Unity and I think I’m on beta 3 or 4 but I will check tomorrow. Also looks like I can only see 1 motor temp sensor on the logs although this could be a hardware issue or an issue with metr logging. Again, I will hook up the vesc tool tomorrow and see if I can read both temp sensors from there.

Yeah Beta 5 or 6 has increased motor temp filtering. So make sure to run latest. Let me know which Beta version you were on when you update please.

Having the high impedance temp sensor run parallel to phase leads can cause some interesting coupling so heavy filtering is needed I think.

6 Likes

Would wrapping the sensor wire around the phase lead run help?

1 Like

Wrapping metal foil/mesh around the temperature wire and connecting the foil (or wire mesh) to ground can definitely help.

Or just using a very small coaxial cable all the way from inside the motor to the ESC could help, with the shield connected to the Sensors or PWM (“PPM”) ground pin.

1 Like