Successor to the infineon clone controllers in the 1000-2000w power range?

[neptronix]

Yep. The block is more silver/aluminum on mine.

Oof, i should have ordered the heatsink to be sure. But i'm only running 2.5kw. Curious to see what i can get away with, hah!

Dual Ubox V2 75V and 100V Heat Sink for stronger cooling

 

[neptronix]

Ok, so..

This spintend is kind of a pain in the ass to setup. There's some uncertainty in the documentation and tons of articles to read to figure out what to do from scratch. There's also tons of parameters.. reminds me of working with an ASI ( at least it can auto detect! )

Pros:
- This series controller would be badass for a 2WD setup and seems to have ABS and traction control, makes me want to build a 2 geared hubmotor setup..
- It basically has every controller/vehicle feature you could dream of.
- Diagnostic lights are awesome and i wish every controller had these.
- Looks and seems well built

Cons:
- Documentation is scattered among blog posts, videos, and some unlisted wiki with no domain name, initial learning process is bad and only the devoted will make it through.
- Included throttle wire ( red, black, green ) is pinned incorrectly, negative and positive reversed, you have to pop the pins out and redo them.
- No idea if the 12v connection even needs to be hooked up, i did it anyway, had to make a cable becauese the bag didn't include one.
- I bought the suggested thumb throttle and brake lever. Throttle works great, brake lever is cheap and uses a tiny switch ( a little worrysome, i think a hall based switch would be safer/more durable )
- Can't get regen working no matter the settings but the ebrake lever does respond and cut off the throttle. I have to contact tech support already.
- The EDC v3 adapter has no protective backing or coating to prevent it from a short, you have to DIY a solution. I'm using zip ties and foam. This thing deserves a plastic case.
- The finished assembly is a mess of wires and the packaging of it is lacking even more now with the EDC v3 adapter.
- I can't get the suggested procedure to calibrate the brake and throttle to work, maybe this suggestion was for an older version/doesn't apply anymore.
- They don't make a heatsink for this model so without DIYing one together, i suspect the continuous current handling is much lower than advertised.

I think this VESC is very nice ( feature and quality wise ) but i think they have some refinements to make to make it newbie recommendable. The learning curve is steep and some unnecessary DIYing is needed, and it looks like a science project.

Will get impressions on the road in the near future.

 

[neptronix]

Disappointed with the end-user feasibility and packaging, i decided to also buy the Flipsky FT85BS v2.0 ebike controller.



The complete package with accessories is very appealing. They are obviously made for a scooter, not an ebike, but will work for my case.

Comes with a 53 page manual that's well written ( attached )

3 speed switch limits based on speed and not current ( good! )

No field weakening feature ( bad! )

Downloaded the flipsky software. Really simple. Has auto detect, all main settings you need are on a single screen.
It's ugly but i love the simplicity of it. This reminds me of the infineons.

 

[scianiac]

I think this is perhaps the biggest issue with VESC, not enough documentation which makes it very complex for new users. But it is an issue that can be fixed by us, the community, as it is an open source project. Something like, I don't know, a wiki page on how to setup a VESC for an ebike would probably make the whole process as easy as following a step by step guide in most cases. I'm working on compiling and testing as many setup tips for VESC ebike use as I can now and while I'm certainly not expert enough to write it entirely I will hopefully be able to help.

 

[neptronix]

Yeah, i notice the little ( ? ) button next to certain options is useful though lately.
Just the learning curve is too high and it's very complex for what is an otherwise simple task.


Here's another issue.
I setup the VESC again with the phase wires hooked into an alternate configuration ( this should be okay )

I run the input wizard and motor configuration wizard, and now, regen works at the brake lever but it's on all the time... the rear motor doesn't freewheel anymore, it comes to an abrupt stop.

Quite confused on why the behavior is different and can't find a setting that changes this.
I disconnected the halls and i'm back to no regen and can't find a way to get it back. I've contacted spintend for help.


I also found out, by using sensorless mode, that my original motor has hall problems ( weird because it's brand new ). Sensorless mode on the VESC in VSS mode works great. As expected, we get some mild stutter from a startup, and 'tick' sound when re-engaging the throttle while moving... but otherwise very smooth, predictable, and quiet.

Should have a RH212 soon.

A couple miles of abusing the motor at 92.5A with 5A field weakening barely made the controller warm, and the controller was tucked in my case. Not bad!



This is a nice controller but it's very much a nerd's toy and can't win the title of infineon controller replacement i think.

I look forward to trying the FT85BS v2.0, our next candidate!

 

[Cowardlyduck]

Your comparison here is interesting to me, mainly because in my experience over the years I still needed to whip out XPD to get infineon controllers working half decent.
Yes they more typically worked out of the box, but I found that when they didn't the only way to fix was with XPD. That and the stock config was pretty nerfed most of the time.
XPD was pretty rubbish and getting the USB->serial cables working could also be a huge PITA, particularly if you had an older cable and tried using it on anything post Windows 8.1.
Compare all that to the Adaptto and Nucular controllers which can also be quite finicky when it comes to using the right software, firmware incompatibilities and usage of old SD cards to make things work with your setup.
Even the Grinfineons and Phaserunners have their issues in setup. I had major trouble getting a Grinfineon to work with an older Golden motor because the default throttle signal doesn't work when you use a motor wire adaptor....go figure. Then there's all the issues with the TTL ports on the Phaserunners over the years and having to use JST -> TTL adaptor cables to bypass.

I guess my point is, yes the Infineons worked ok, but so far in my experience (with a lot of different controllers) none of the higher power more advanced controllers have been plug and play or easy to setup/use...even the Infineons.
If we are looking for a modern equivalent to the Infineons then I think we should expect a low/moderate level of pain in setup/config.
If we want something that is like the infineon in power/simplicity, but easy to setup/config (the unicorn?), (and I want this too) then I think it's not quite an Infineon replacement.

Anyway, just my 2c.

Cheers

 

[neptronix]

I've never used XPD, i have an old em3ev copy with em3ev infineon controllers and it works great still ( must run in compatibility mode in win10 and have some 2000's era .dll and .ocx files for running visual basic 6 programs alongside ).

The Phaserunner is a peach to program compared to the VESC, and the documentation is excellent. I would call it the #1 controller today. We just need to find that #2 controller!

I think the problem with new controllers is the complexity of FOC. This complexity oozes out into the tuning, troubleshooting, and setup process by making it works, unless some effort has been made to decomplexify it ( Grin did a good job )

Honestly, sinewave with torque-emulating throttle ( literally just some lines of code added ) would be fine for 90% of setups but everyone is reaching for FOC lately. If i could find such a controller that was programmable i'd be playing with it instead of a FOC.

 

[Cowardlyduck]

Have you tried a Nucular? They may not fit the 'cheap' category, but in most other ways they are pretty good.
I have a 12F in my LMX and while it has had some issues, once I worked through those and got it running, it is pretty nice!

Cheers

 

[neptronix]

Nucular controllers look badass but yeah, expensive and also you can't get the 6F/12F version currently... sometime in 2024 they say.

I feel like, for high power controllers i would happy to use a VESC, i don't know of any extra features on the Nucular good enough to get me to pay extra.. i also really like that the software that runs the VESC is open source & open to modification. That really adds value points for me.

 

[scianiac]

I run the input wizard and motor configuration wizard, and now, regen works at the brake lever but it's on all the time... the rear motor doesn't freewheel anymore, it comes to an abrupt stop.

It sounds like you just need to check the control input setup which should be pretty easy depending how the ebike adapter board works. Generally I just wire the throttle to ADC1 and the regen brake to ADC 2 and I'm guessing that's what the adapter board is doing but I don't know for sure. I'm guessing the regen as just a switch and not a hall or resistor type variable input is confusing the setup wizard but that should be very easy to fix (although personally I would just use another hall thumb throttle as the regen so I have variable regen).

If you go to App CFG > General the APP to use should be ADC or ADC and something else.

Then under App CFG > ADC > Mapping you can set the input voltage range. On that page at the bottom if using the PC app (if using the phone app you click the "..." at the bottom and hit ADC mapping) you can see the voltages so if you press the brake one will change and just make sure that voltage change makes sense for the ADC start and end voltages set. So if it's on all the time just make the start voltage higher than the voltage when it's not pressed and make sure the end voltage is about what is is when pressed or slightly lower.

 

[pwd]

Quite confused on why the behavior is different and can't find a setting that changes this.
I disconnected the halls and i'm back to no regen and can't find a way to get it back. I've contacted spintend for help.

I recommend trying out the desktop application of VESC tool or at least just opening it to read the tootltips if you haven't already. VESC does have a higher than normal learning curve but there is a help/info button for most settings. The realtime data/logging is also useful for troubleshooting. Both the VESC Tool for mobile and desktop are great; I prefer the layout of the desktop app for configuration though. I usually use my smartphone as a bluetooth/tcp bridge (done through VESC tool) since my desktop doesn't have bluetooth and connect that way.

 

[afzal]

I run the input wizard and motor configuration wizard, and now, regen works at the brake lever but it's on all the time... the rear motor doesn't freewheel anymore, it comes to an abrupt stop.

If control type(AppSettings->ADC->ControlType) is duty cycle, change it to current control & check

 

[neptronix]

I only use the desktop app.

Here's some screenshots of my settings and my XML file.
The control type is set according to what the spintend suggests for the ewheel v3 adapter in the manual. I tried a lot of different control types but the behavior is the same.

 

[scianiac]

Yeah just as I expected, on that 2nd sceenshot you see the start and end voltage for ADC2 (which how it's setup is the brake) are set at basically the same voltage. When you pull the brake lever you should see the CH2 bar on the bottom move. You just want the start voltage to be above what the CH2 bar shows (so set to like 0.1V) when the brake is off and the end voltage to be below what the voltage shows when the brake is pulled (set set to something like 3V). So the regen braking then will just be 100% the max you have it set in Motor > general > current.

It's just that the wizard is expecting a variable regen braking input not a switch, which honestly is more spintend being lazy and just using an off the shelf brake lever with a switch when one with a hall sensor would provide a much nicer braking experience.

To explain a little more, the control type means that ADC1 is the throttle which is in current mode and there is no reverse on the throttle and that ADC2 is for the brake. It's worded a little confusing since it doesn't say ADC1 but does say ADC2, but I guess it just always assumes ADC1 is the primary control input.

 

[neptronix]

Did the modification as suggested, including measuring the voltage. No change in the behavior.

 

[scianiac]

Weird, is that screen shot before you hit "write app cfg"? Because CH2 says 31.6% when it should say like 0% at that voltage.

 

[neptronix]

Yeah, the math doesn't add up.

 

[scianiac]

Are you sure it saved the app config, if you hit the read app cfg (the "up arrow" "A") are the start and end voltages what you set? The bars should read correctly as long as the realtime app data is on which I can see it is because the RT APP is lit up. Maybe you hit write motor cfg ("down arrow" "M") instead of write app config as they only write their own category settings not all the settings.

 

[neptronix]

Oh interesting! i am only writing the motor config.. i didn't know it was seperate.. so you have to write both when updating the controller eh?

 

[neptronix]

I have emails going back and forth from spintend right now.

I also have emails going back and forth to Justin at ebikes.ca to remotely diagnose this halls issue while i temporarily run sensorless. Apparently the phaserunner can spit out some useful diags, really cool. Bought the tools i need to open the motor and repair it too - wiring issue is suspected.



I did ride the bike hard with the spintend today for 12 miles and i loved it. Oh my god i love being back on an ebike! Grin all axle 24" STD and 52v handles ~30mph all day up mild ( 1-4% ) hills and is only warm-hot to the touch at worst.

To my surprise, even though the spintend sits in a closed frame bag, it actually runs cooler than the hottest part of the hub motor at 35A batt / 92.5A phase.

I think that the current/voltage rating of the spintend is actually the battery current, like most ebike controllers... instead of being rated on the phase amps, like many RC controllers ( and also kelly controllers - shame on you ).



Gee, that explains it.. i barely touched the controller's capabilities. I bet i could run 40A batt / 120A phase and a buttload of flux weakening and get away with it.

The VESC is growing on me.

 

[scianiac]

I think the VESC's interesting way of splitting the config into two parts, having the read settings and read default settings buttons easy to get to and such is to help prevent accidentally setting settings you didn't mean to, so you can for instance just hit read settings right before you change something to make sure you didn't change something on another page you didn't mean to. Not as much of an issue on a PC but much easier to accidentally do on a phone.

If you turn on the Hall sensor mode in motor cfg > general > sensor mode and then go to data analysis > realtime data > rotor position (with the RT data button on the right activated) and spin the wheel by hand it should show a reasonable output if the halls or working or something weird if they are not/the halls are not detected properly.

 

[afzal]

If you turn on the Hall sensor mode in motor cfg > general > sensor mode and then go to data analysis > realtime data > rotor position (with the RT data button on the right activated) and spin the wheel by hand it should show a reasonable output if the halls or working or something weird if they are not/the halls are not detected properly.

To add, there is a VESC terminal command, IIRC, 'hall_analyze' that provides details about Hall setup. Typing help on terminal would provide some helpful info

 

[neptronix]

Are you sure it saved the app config, if you hit the read app cfg (the "up arrow" "A") are the start and end voltages what you set? The bars should read correctly as long as the realtime app data is on which I can see it is because the RT APP is lit up. Maybe you hit write motor cfg ("down arrow" "M") instead of write app config as they only write their own category settings not all the settings.

Thanks to this my regen is working and i'm SO happy!
Thanks a gazillion, i would have never thought of that.

Even better - the regen works all the way down to 0RPM and with my tune, tends to get stronger the closer you get to 0RPM. In that way it acts very much like a bike brake.

This is the best regenerative braking i've experienced.

Also, i bumped up the power to 40A batt, 100A phase, 10A of flux weakening. Nice boost, a 6 mile ride finally started to get the controller warm-hot in the bag, So, this is approaching or already at the limits of what the controller will tolerate without any kind of ventilation.

 

[A-DamW]

Also, i bumped up the power to 40A batt, 100A phase, 10A of flux weakening. Nice boost, a 6 mile ride finally started to get the controller warm-hot in the bag, So, this is approaching or already at the limits of what the controller will tolerate without any kind of ventilation.

Couple things come to mind... (like $10 on Amz, eby, etc)

 

[gobi]

maybe a thick alu plate attached to the bottle holder with the same screws, and then the controller heat sink in contact with said plate?

PS: I have a been following this thread, just place my order, thanks to all who contributed!!!