FLIPSKY new 20s 100A tiny controller (vesc based)

Battery cells should be the LG MH1 (10A). They advertise 17,5ah but that cell is 3200mah so that would be a 16ah if it’s a 13s5p battery. Cells should be able to provide 50A though. Correct me if I’m wrong, I’m new to this. I don’t know about the BMS but I’ve seen other people daily driving this 35A peak controller with this bike with no issues. I should have some margin for improvement.

I looked at other, easier to setup options, but all sine wave controllers that I could find were either too big to fit in the bike's frame (it’s a folding bike) or too expensive. (I’m looking for a sine wave controller because I want a quieter and smoother ride also). The next best option would be to modify the case of a Sabvoton 7545.

I’m ok with learning the vesc interface. My concern is that no matter how hard I try, I might not be able to make this run properly because some settings are very technical and I saw that even some pros struggle with it. I’m wondering if the reported issues are intrinsic to this particular controller or are related to specific battery-motor configurations that push this controller to its limits / improper vesc setup. Or maybe they only affect the first version of this controller which was prone to overheating.
 
Basically new cells would be able to provide that 50 amps, so if you have put it through some charge cycles, it would have dropped a bit. If you have witnessed 35A being used with no problems, it's probably fine to start there. If you have power cutting out, then you have to figure out if it's the BMS stopping the controller from asking too much or the controller overheating/hitting upper limits of amperage or temps.

It does seem like the initial version suffered the most from the temperature problems since the aluminum heatsink on the revision is pretty big. Also if you are willing to dive into sabvoton settings, I am betting the vesc will be easier to interact with (no direct experience with sabvoton I just see a lot of questions and frustration from users online).

I've been following this other thread since the controller came out, pretty extensive testing, troubleshooting and firmware releases on there. How to update firmware on the Flipsky 75100 & 75200 FOC ESC


TL;DR: yeah, the small footprint is probably worth it for the aluminum heat sink version and some patience with settings unless you can find a slightly cheaper option. Make certain you have your battery current setting and phase filtering set to off properly as described in that esk8 post
 
IsaacLarsen said:
I’ve read the entire thread—a lot of technical stuff that went over my head. I still can’t understand if one can make this work as a daily driver with a cheap Chinese ebike. I’ve read people complaining about motor noise, poor acceleration from 0km/h and other issues, but they all use big, powerful motors. I have a 20” fat bike with a 750W geared hub motor, 48v battery, 20A max, 10A rated controller. I’m looking to replace the original controller with something that will give me more power and a quieter and smoother ride.
Click to expand...
With your underpowered battery you won't, and very likely you will ovearheat your motor. If you can, Sell your battery and motor and get a bbshd, 72v battery and VESC or get the BBSHD with 52v battery and use it stock. Get a small charing on the front because it's a fat bike and you are good to go.

This threat is for mostly overpowered, over-the-top, very fast bikes, scooters, etc... Based on what you said, doesn't look like that what you are looking for.

If I am wrong, please forgive me :)
 
brazilianboy said:
With your underpowered battery you won't, and very likely you will ovearheat your motor. If you can, Sell your battery and motor and get a bbshd, 72v battery and VESC or get the BBSHD with 52v battery and use it stock. Get a small charing on the front because it's a fat bike and you are good to go.

This threat is for mostly overpowered, over-the-top, very fast bikes, scooters, etc... Based on what you said, doesn't look like that what you are looking for.

If I am wrong, please forgive me :)
Click to expand...
That’s way over budget. I’m just looking to gain some extra performance while staying within the thermal limits of the motor. In a different forum, I’ve seen people use a cheap 35A peak controller with this bike with good results so there should be room for improvement from stock.
 
E-HP said:
So why aren't you doing the same thing, if that's what you want?
Click to expand...
I'm looking for a sine wave controller (I want a quieter and smoother ride) with a "legal" profile (no throttle, 25km/h top speed) and the ability to switch between legal and off road on the fly.
 
IsaacLarsen said:
I'm looking for a sine wave controller (I want a quieter and smoother ride) with a "legal" profile (no throttle, 25km/h top speed) and the ability to switch between legal and off road on the fly.
Click to expand...

Me too!
 
VESC's themselves (most that I've found) don't really support toggle-mode-switches.

You can always hookup a CA for control and have profiles that way.

Depending on how "easy" you want modes, the Voyage Mini (previously "metr") add-on for VESC's has a mobile app with swappable profiles, including the ability to set a profile "temporarily" (till the next vesc restart). It's, of course, polite to turn off the bike if stopped -- and when it turns on, of course it's at legal limits.
 
chuyskywalker said:
VESC's themselves (most that I've found) don't really support toggle-mode-switches.

You can always hookup a CA for control and have profiles that way.

Depending on how "easy" you want modes, the Voyage Mini (previously "metr") add-on for VESC's has a mobile app with swappable profiles, including the ability to set a profile "temporarily" (till the next vesc restart). It's, of course, polite to turn off the bike if stopped -- and when it turns on, of course it's at legal limits.
Click to expand...
The normal vesc tool app has profiles just the same as you mentioned for the metr app. The flipsky "pro" version had Bluetooth built in that you can use the app and profiles with or a cheap nrf51 adapter for the older controllers.
You can set a % of current or watts and a speed limit for each profile. I mainly use it for a quick current adjustment so I can have full power for playing around off road and another for my commute without the bbshd getting too hot after a few miles at wot.
 
Is there a way to quickly switch between profiles with a button or do you have to reload the profile each time with the app?
 
Hello, I keep reading that the Flipsky signals are very noisy. Does anyone have a scope capture of the output of the current shunt amplifiers and/or voltage senses?

EDIT: Ideally, with a comparison with genuine/high end VESC.
 
I know some of the early flipsky and related brands 75200s (and possibly other versions) were very noisy because if I recall they used the 5V to supply the opamps but newer versions don't seem to have this issue. Also I don't think you need to scope anything as if you go to "data analysis" and do a single sample with the motor disconnected and check the raw box it will tell you how noisy they are at baseline. From what I've seen a few amps or that range range is good enough and that's what I'm getting from a new flipsky 75200 pro V2. I haven't tested other models and of course there is always the possibility of noise from other sources when actually running. Still working on getting it to run stably at the power levels I want but I don't think that is a shunt noise problem, I think it's the inductance or linkage values are off.

I'm sure they are sill noisier than a high end VESC just due to them cheaping out of filtering but they seem to work after some tuning. Are they as stable when pushing them hard, probably not, is the risk higher of them blowing up, probably, are they much cheaper, yes. So I think it's mostly a matter of personal prefrence, I have one of the earlier 75200s on a bike that has been running great for awhile now (I haven't got around to checking the noise on that one but it's probably bad) but it took quite a bit of messing around with the tuning to get it to run nice and I'm still not pushing it that hard (130PA), mostly becaue I have no need to it and more power would probably just make it harder to control.
 
Been tuning the bike with the 75200 pro V2 some more and there is some noise at high currents although how much of a problem it is probably depends how how cautious you are being. It's powering a Leaf 1500w at 72v and at 140PA with a 160 max amps limit I was getting some overcurrent faults at around 6000ERPM from some noise I think. Now I could have easily set the max amps much higher like it should be but I'm tuning for good stability here before I turn up the PA.

Now here is the important bit, in 6.05 "all sensors combined" current sample mode was added specifically to address noisy boards and it does what is says on the tin and does seem to help with high current noise. Turning it on was able to get to 160PA again with 20A of overcurrent headroom no problem before I ran out of daylight.
 
scianiac said:
Been tuning the bike with the 75200 pro V2 some more and there is some noise at high currents although how much of a problem it is probably depends how how cautious you are being. It's powering a Leaf 1500w at 72v and at 140PA with a 160 max amps limit I was getting some overcurrent faults at around 6000ERPM from some noise I think. Now I could have easily set the max amps much higher like it should be but I'm tuning for good stability here before I turn up the PA.

Now here is the important bit, in 6.05 "all sensors combined" current sample mode was added specifically to address noisy boards and it does what is says on the tin and does seem to help with high current noise. Turning it on was able to get to 160PA again with 20A of overcurrent headroom no problem before I ran out of daylight.
Click to expand...
Alternatively; you can re-tune using a higher amp value. The default motor tuning setup in VESC uses a really low (like 15) amp value for finding the FOC parameter. If you go the FOC tab, you can re-run that portion of the motor tuning, but you get to pick the Amps the tune runs at. Nearly every DD motor I've done where I run high amps, I've needed to re-tune it this way with VESC controllers to make it behave better at higher amps.
 
I ran the wizard at a pretty high current, actually I know what the variables of the motor should be pretty closely, tested the resistance with a PSU, the inductance with an LCR meter and the flux linkage with a scope. Interestingly on 6.02 inductance values were reasonably close and linkage exact, resistance was different but that is probably my power supply being off a bit and it never had low RPM instability so it's probably correct. But 6.05 the inductance measurement was much closer to what I measured, could have been I ran the wizard at a higher power or something else in the detection changed but seems to work better as previously the Lambda comp observers were unstable but now they seem to be working.

I expect though I might need to start tweaking some things as I'm probably starting to saturate the motor so we'll see how well the saturation compensation works. I wonder if maybe sometimes the reason why detecting at high current levels works is not because the numbers are more accurate but because they are more accurate at high current levels where there is some level of saturation which changes the inductance and flux linkage. Which in an ideal world those would be adjusted on the fly with a lambda comp observer and the saturation comp.
 
Is this noise likely why my 75100 cuts out at a full speed light load steady cruise? Its done it with 2 different low powered motors.

It happens to quickly for me to glance down and catch what kind of fault it is.

Im using the 3.3v throttle in and the adc 2 as a brake which is a simple on off switch.

I was wondering if noise was triggering a value on adc 2.
 
I haven't had any noise on the ADC inputs and I believe those are decently filtered anyway (easy to filter when the signal is very slow). It will tell you if and what the fault was if any occurs after a poweron somewhere I think, I think in the app on the last page of live data maybe, I know I've seen it somewhere but can't recall where. It could just be an over current fault though if the observer loses track of the rotor position due to noise.
 
I'll have to have a look around some more. I looked for places where faults were reported and it was not obvious. It definitely seems to have more to do with erpm than current draw which made me think of noise or losing synch with the motor.
 
You must log in or register to reply here.

Similar threads

TrotterBob
Anyone use a Flipsky VESC based controller?
Replies
3
Views
1,610
amberwolf
amberwolf
mxlemming
    • Brainy
New ground-up ESC - MESC_FOC_ESC
5 6 7 8 9
Replies
222
Views
30,162
mxlemming
mxlemming
Back
Top