10S custom skate ESC: testers wanted!

[lizard]

Does 'anything below 300kv' mean, that you run something over 300kv at the moment?
If my calculations are right, 300kv would translate to about 200kph topspeed on 12s with 86mm wheel diameter.

Actually, I find it quite remarkable that the VESC can start the motor _at all_ with such a high gearing

 

[erwincoumans]

The current hub motors are 149kv, and I'm happy with their performance with VESC, aside from the lack of sensors.

If you read the introduction topic of this forum the recommendation for motors, see http://endless-sphere.com/forums/viewtopic.php?f=35&t=53506

Other than size, the KV of the motor is important. This can be as low as 138kv and as high as 270kv.

I'm happy with the 138kv - 270kv range, that's what I meant by lower than 300kv.
Anyway, I don't want to get into a KV discussion, I just want to get Hall sensors working with VESC with fairly standard low-KV values.

 

[pkondratyuk]

@Shep_pard:

Hey pkondratyuk the active breaking is adjustable. You can adjust the max breaking current and the max regenerative breaking current(which goes back into the battery). I did not have a chance testing the regenerative breaking yet, but the normal active breaking is really nice and smooth. And as a warning be careful with it. It really homes to a halt fast and brutal if you want it to.(almost flipped a 4kg alternator)

Thanks for your response. When you say the active braking is adjustable, do you mean you can control it in same sense you control torque (by PPM, for example), or is it a setting that can only be set through the BLDC tool? After all, braking is just negative torque ))

 

[Shep_pard]

Thanks for your response. When you say the active braking is adjustable, do you mean you can control it in same sense you control torque (by PPM, for example), or is it a setting that can only be set through the BLDC tool? After all, braking is just negative torque ))

Yes if you use the PPM control with forward only with breaking it breaks accordingly to how far you pull the stick(if using a radio) back. So yeah it basically is just like using the current control in reverse(And just like with the current control, you can set how much "negative current" is maximally applied). And I think I was wrong about the actual active breaking as lizard posted. It only works as regenerative breaking, but still I think there isn't too much of a need to just actively break with no energy going back into the battery. That would just waste a ton of energy.

And just as a quick disclaimer, I could be totally wrong , due to my lack of in depth knowledge about the VESC and only played around with one for like half an hour before I killed it. But tried to read as much about it as I can.

 

[lizard]

For a skateboard I think active braking is enough (you can't brake hard anyway because you'd fall off) for an onroad R/C car I am not sure. From what I understand, with active braking the currents are reversed, so there is the same energy available for braking and acceleration. But I need more energy while braking, my car brakes quicker than it accelerates. Acceleration to topspeed is about 100meters, braking to zero around 40 or so.

 

[vedder]

So there are 8 ceramic capacitors that need to be replaced on onloop's VESCs, is that right?

Yes

For a skateboard I think active braking is enough (you can't brake hard anyway because you'd fall off) for an onroad R/C car I am not sure. From what I understand, with active braking the currents are reversed, so there is the same energy available for braking and acceleration. But I need more energy while braking, my car brakes quicker than it accelerates. Acceleration to topspeed is about 100meters, braking to zero around 40 or so.

As far as I know, there are no sensorless RC ESCs that can do active braking. They can either do regenerative braking or just shorting the motor windings, which VESC also can do.

Lets look at two similar motors (I don't know if the specs lie though):
http://hobbyking.com/hobbyking/store/__24847__Turnigy_TrackStar_1_8th_Sensored_Brushless_Motor_2350KV.html
http://www.myrcmart.com/leopard-lbp427425y-2000kv-4poles-inrunner-brushless-motor-18-car-no-vents-p-4928.html

Both state that the resistance is about 0.006 ohm. The voltage difference required to get a current of 120A is 120 * 0.006 = 0.72V. Since you run the motor on 25V, you will be able to brake with 120A down to about 3% of its operational speed with just regenarive braking. 200A braking will work down to 5% of its operational speed. More than that will blow the VESC up. So, if you really need full braking below 3% of the motor speed, active braking would be required. Otherwise, you are completely fine with just regenerative braking. In my experience, on my RC car, regenerative braking alone is more than enough.

 

[lizard]

Ahh, this is so complicated. I thought there is just active and regenerative braking from reading this thread. Now there is also just shorting the windings Not sure what my Mamba does, all I know is, the brakes make this squealing sound and it is powerful and smooth. Patrick from Castle said somewhere on a forum, that the Mambas have regenerative braking, but with low efficiency, only 20 percent or so are fed back to the battery.

Having full braking power above 5% motor rpm sounds perfectly fine to me. That would mean I can brake hard down to 8kph. Actually, I have the feeling that is about the speed the Mamba brake brakes hard, below that it feels like if the controller is trying to drive the motor in reverse or doing something differently. Not sure if my observation is right, though.

Is the maximum braking current I can use safely really 200A? I thougt the shunts are only designed for up to 150A current?

Considering my lack of understanding, I guess I'll just start with small values like 50A and then carefully work my way up from there if neeeded

Oh and I forgot to order higher voltage caps, I have 3x 1000uF 35V low esr caps here, can I use them (all in parallel) as long as I stay on 6s?

 

[vedder]

Ahh, this is so complicated. I thought there is just active and regenerative braking from reading this thread. Now there is also just shorting the windings Not sure what my Mamba does, all I know is, the brakes make this squealing sound and it is powerful and smooth. Patrick from Castle said somewhere on a forum, that the Mambas have regenerative braking, but with low efficiency, only 20 percent or so are fed back to the battery.

Having full braking power above 5% motor rpm sounds perfectly fine to me. That would mean I can brake hard down to 8kph. Actually, I have the feeling that is about the speed the Mamba brake brakes hard, below that it feels like if the controller is trying to drive the motor in reverse or doing something differently. Not sure if my observation is right, though.

Is the maximum braking current I can use safely really 200A? I thougt the shunts are only designed for up to 150A current?

Considering my lack of understanding, I guess I'll just start with small values like 50A and then carefully work my way up from there if neeeded

Oh and I forgot to order higher voltage caps, I have 3x 1000uF 35V low esr caps here, can I use them (all in parallel) as long as I stay on 6s?

I think the mamba just shorts all three phases with a certain frequency and duty cycle, which is why you hear the squealing. Doing regenerative braking with synchronous PWM is quieter, smooth and better, but is more challenging for the ESC when overcurrents have to be measured and prevented.

You are correct, 150A is the maximum that the shunts can measure without saturating the ADC. I wrote 200A since that is where you can blow the VESC up in a fraction of a second, even if you change the shunts.

The capacitors you have will work fine as long as you don't use more than their rated voltage.

I look forward to see some test drives with you rc car. It looks really impressive

 

[okp]

hey, just to keep you posted. As I had no reply from Jason on this thread; I send a PM to onloop for next steps on the BETA VESC.

As a customer of enertion, I want to know if all the BETA VESC are sent back to him for fixing / or get the money back (or whatever is needed to get it fixed including getting a new VESC from the supplier) or to whoever is designated so we can get the full product in line with Vedder specs... in quality. I think this is what we all bought (at least the BETA Onloop testers) as it seems we have now real evidence that the caps are wrong one which has been confirmed by Benjamin and also highlighted for quite a while.

I am not really happy with the way things are going and this huge fog and multiple contradictions around the BETA VESC. Sorry guys, but I always like things clear, and accountability when we are talking about Business. We all can make errors, this is human life... but I am not expecting someone to fix my errors; so I am not doing it either.

Of course, I could go ride with my existing setup... but well... would you do it when the creator of the product states that you shouldn't... no way; especially because at the end, he's the most reliable guy in the topic.

 

[onloop]

Attention all BETA VESC customer who purchased from enertionboards.com

It has now been determined that the product, VESC BETA, manufactured by Shenzhen Flier Electronic Co., Ltd. Purchased online at enertionboards.com has been made using a variation of components as listed in the BOM at http://vedder.se/2015/01/vesc-open-source-esc/

The variations identified are:
》different capacitors
》different shunts
》different dropout regulator

There has been much discussion about this supplier and the quality of product they are supplying to this community.

As it stands currently, the product they supplied DOES function and I have successfully tested this on my own board several times despite the different parts.

However it has not been built using the recommended parts as requested by enertion and therefore enertionboards.com is now taking action to rectify this error. It is our mission to develop, promote & supply only the best quality parts for the DIY eboard community, so we must now prove our commitnent and resolve this issue so we can all continue to focus on the long term advancement of the eboard industry.

We would like to emphasise how much we value this community and will not tolerate inferior quality components or dishonest suppliers who don't value our goals.

Enertionboards is officially boycotting Flier as an electronics supplier from now.

If you are unhappy with your purchase of VESC BETA or not willing to power up & test the VESC because you are concerned about the build quality you may return the item/s for a store credit equal to the value of the purchase, shipping charges not included in credited amount. You now have 30 days from today to return the vesc to enertion HQ. Return shipping charges will be paid in full buy the returnee.

》》 Due to ongoing demand we will be manufacturing another batch of vesc, We have just purchased 100 sets of components directly from mouser to ensure parts substitution cannot occur during assembly of this next batch. We will be assembling these 100 units with a strict quality assurance protocol, loading the latest firmware and offering a 300 day money back guarantee on all vesc purchases.

Yours sincerely
Jason Potter
Founder
Enertionboards.com

 

[torqueboards]

Nice... A+++ Customer Service.

This is why I'm hesitant with promoting things which I have not tested and doing pre-orders without testing. It's never truly guaranteed.

Sucks Flier has to pull this crap too. Everything to cut costs.

 

[elkick]

Tested the VESC first time today, everything went fine until the VESC stopped responding in the middle of a fast ride and kept the motor speed where it was. It was just continuing at the same speed, not stopping at all, completely out of control.

There was no reaction on braking or anything, Nunchuk had optimal connection. The only way to stop the motors spinning was to run after the board and to switch it off. Since this can be very dangerous I'll need to find out the cause for this (I was able to jump off the board without accident, but nevertheless, if this happens somewhere with traffic or a crossing it might be fatal).

I tried to replicate the effect, but the only other behavior I experienced was that the VESC was not responding after a couple of minutes driving (no reaction on any Nunchuk movements, but at least not keeping the motors running) and I had to switch the board on and off again and it worked for another couple of minutes. Motor was not hot at all, VESC was cold too, batteries fully charged.

My setup: 7S2P Lipos (35C) with 50A BMS, Alien Power Switch, 5060/270KV Motor, Kama Nunchuk, newest FW 1.3 & BLDC Tool, did all the steps necessary in BLDC without changing anything after motor was recognized and data saved , the only change I made was enabling reverse on the Z-Button and the Nunchuk remote in the App Tab.

About the Nunchuk: the receiver is located outside of the box to ensure reception and it never lost reception (I am quit aware about problems wich might occure with the Nunchuk, but it this case the problem must be at the VESC.

Any ideas how to solve this? Is there anything I could change to avoid this in the future in the BLDC Tool?

Please don't get me wrong, it's just fun to drive the VESC and it is by far the best ESC I ever had! And testing should anyway always be done in deserted areas!

 

[pf26]

Hello,
I received my VESCs and tested one on the bench. It seems to run fine so far. A collegue compiled a windows version of BLDC-tools for me (revisiting some Qt serial port library) and this worked fine for setting the parameters and showing real time graphs.
Unfortunately, then I was more confident and tried to upgrade firmware from V1.2 to 1.3.
It seemed to work fine, showing FW upload done (really fast) - but the progress bars mentioned only 98%. And now, the VESC seems stuck somehow, not even enumarating on the USB. I guess it is time to look for an STM32 programmer...
I have some programmers here (AVR Dragon, ICD3 for Microchip), is there some compatibility in JTAGs ?
Or is it possible to use a STM32F0DISCOVERY kit as a programmer, using the SWD port ? It seems like the cheaper option for me.
Otherwise, I saw there is an integrated serial bootloader in STM32F405 ROM, it may be possible for me to use it (I have USB to low voltage serial interface here)

Anyway, I am happy so far. I am confident the VESC will do a great job for my Ebike. Thks ! Pierre

 

[Vanarian]

Just when I was happy to get these two folks in my mailbox Flier sucked this time, fortunately you are there for us guys.

Attention all BETA VESC customer who purchased from enertionboards.com

If you are unhappy with your purchase of VESC BETA or not willing to power up & test the VESC because you are concerned about the build quality you may return the item/s for a store credit equal to the value of the purchase, shipping charges not included in credited amount. You now have 30 days from today to return the vesc to enertion HQ. Return shipping charges will be paid in full buy the returnee.

》》 Due to ongoing demand we will be manufacturing another batch of vesc, We have just purchased 100 sets of components directly from mouser to ensure parts substitution cannot occur during assembly of this next batch. We will be assembling these 100 units with a strict quality assurance protocol, loading the latest firmware and offering a 300 day money back guarantee on all vesc purchases.

Yours sincerely
Jason Potter
Founder
Enertionboards.com

This is a strong backup. I don't have a proper supply to test them yet, and I don't want to bother with firmware issues because of Flier. I'll PM you.

 

[Tartopom]

Tested the VESC first time today, everything went fine until the VESC stopped responding in the middle of a fast ride and kept the motor speed where it was. It was just continuing at the same speed, not stopping at all, completely out of control.

There was no reaction on braking or anything, Nunchuk had optimal connection. The only way to stop the motors spinning was to run after the board and to switch it off. Since this can be very dangerous I'll need to find out the cause for this (I was able to jump off the board without accident, but nevertheless, if this happens somewhere with traffic or a crossing it might be fatal).

I tried to replicate the effect, but the only other behavior I experienced was that the VESC was not responding after a couple of minutes driving (no reaction on any Nunchuk movements, but at least not keeping the motors running) and I had to switch the board on and off again and it worked for another couple of minutes. Motor was not hot at all, VESC was cold too, batteries fully charged.

My setup: 7S2P Lipos (35C) with 50A BMS, Alien Power Switch, 5060/270KV Motor, Kama Nunchuk, newest FW 1.3 & BLDC Tool, did all the steps necessary in BLDC without changing anything after motor was recognized and data saved , the only change I made was enabling reverse on the Z-Button and the Nunchuk remote in the App Tab.

About the Nunchuk: the receiver is located outside of the box to ensure reception and it never lost reception (I am quit aware about problems wich might occure with the Nunchuk, but it this case the problem must be at the VESC.

Any ideas how to solve this? Is there anything I could change to avoid this in the future in the BLDC Tool?

Please don't get me wrong, it's just fun to drive the VESC and it is by far the best ESC I ever had! And testing should anyway always be done in deserted areas!

Hi Elrick,

I had the sale problem as you but the cause is not the vesc! I have an alien esc. The problem come from the link between the nunchuk and the wiiceiver. I turned back to my faithful gt2e and no problem since then

 

[elkick]

I am aware of the problems with Nunchuk and Alien ESC connection and I solved that a while ago (no problems anymore for months while driving daily). Also, I tested the connection to the VESC countless times and it's absolutely ok (after the incident I even removed batteries from the Kama while driving slowly, the VESC behaved like it should and the board just rolled out without motor).

The incident today was different: Kama was in normal connected status and the motor just went on autopilot...

 

[okp]

nice to see things handled with such a professional approach. I am sending back my two VESCs and putting back the APS ESC meanwhile.

 

[akiraEC]

Same here.
My VESC will travel with okp's ones.

 

[elkick]

About my incident above: I went through the BLDC procedures step by step again and realized that I might have entered the two values from motor detection not accurate enough (integrator limit should have been 70 -> was 140, BEMF coupling should have been 500, was 700).

Corrected this and removed reverse from Nunchuk as well. Everything else remains on the default values provided by the motor or Benjamin!

Will give it another try tomorrow with the new values.

 

[okp]

great, take care and wear protection gear ! our eboard are dragsters!

 

[silviasol]

Tested the VESC first time today, everything went fine until the VESC stopped responding in the middle of a fast ride and kept the motor speed where it was. It was just continuing at the same speed, not stopping at all, completely out of control.

How long is the wiring from the remote reciever to the vesc? I had a cable that was about 10" and this happened to me when testing if I gave it alot throttle quickly. I shortened the cable to about 4" and never had the problem again. Same problem with loading the firmware, if the cable is too long it will not work. Try shortening the cable.

 

[silviasol]

Attention all BETA VESC customer who purchased from enertionboards.com

It has now been determined that the product, VESC BETA, manufactured by Shenzhen Flier Electronic Co., Ltd. Purchased online at enertionboards.com has been made using a variation of components as listed in the BOM at http://vedder.se/2015/01/vesc-open-source-esc/

The variations identified are:
》different capacitors
》different shunts
》different dropout regulator

There has been much discussion about this supplier and the quality of product they are supplying to this community.

As it stands currently, the product they supplied DOES function and I have successfully tested this on my own board several times despite the different parts.

However it has not been built using the recommended parts as requested by enertion and therefore enertionboards.com is now taking action to rectify this error. It is our mission to develop, promote & supply only the best quality parts for the DIY eboard community, so we must now prove our commitnent and resolve this issue so we can all continue to focus on the long term advancement of the eboard industry.

We would like to emphasise how much we value this community and will not tolerate inferior quality components or dishonest suppliers who don't value our goals.

Enertionboards is officially boycotting Flier as an electronics supplier from now.

If you are unhappy with your purchase of VESC BETA or not willing to power up & test the VESC because you are concerned about the build quality you may return the item/s for a store credit equal to the value of the purchase, shipping charges not included in credited amount. You now have 30 days from today to return the vesc to enertion HQ. Return shipping charges will be paid in full buy the returnee.

》》 Due to ongoing demand we will be manufacturing another batch of vesc, We have just purchased 100 sets of components directly from mouser to ensure parts substitution cannot occur during assembly of this next batch. We will be assembling these 100 units with a strict quality assurance protocol, loading the latest firmware and offering a 300 day money back guarantee on all vesc purchases.

Yours sincerely
Jason Potter
Founder
Enertionboards.com

Sorry for our little argument before. If I offended you in anyway I apologize. If you can't find a company you can trust to build your boards I would be happy to do it. I will offer warranty services also if anything goes wrong with the boards so everyone can be happy.

 

[elkick]

How long is the wiring from the remote reciever to the vesc? I had a cable that was about 10" and this happened to me when testing if I gave it alot throttle quickly. I shortened the cable to about 4" and never had the problem again.

Thanks silvasol, that helped! Shortened it and also made sure with a lot of hot glue that those cables stay in place in the micro-JST plug.

Might have been the reason for the strange behavior of the VESC: the blue cable (data) might have become loose, the others stayed in place corecctly. So data coming from kamareceiver were inconsistent while the receiver itself was still working fine, but the VESC just got the last command and not the ones that followed because the data wire became loose.

 

[vedder]

How long is the wiring from the remote reciever to the vesc? I had a cable that was about 10" and this happened to me when testing if I gave it alot throttle quickly. I shortened the cable to about 4" and never had the problem again.

Thanks silvasol, that helped! Shortened it and also made sure with a lot of hot glue that those cables stay in place in the micro-JST plug.

Might have been the reason for the strange behavior of the VESC: the blue cable (data) might have become loose, the others stayed in place corecctly. So data coming from kamareceiver were inconsistent while the receiver itself was still working fine, but the VESC just got the last command and not the ones that followed because the data wire became loose.

The data cable coming loose might explain the behaviour that you have been seeing. The VESC has a timeout that will stop the motor if the data received from the nunchuk has been the same for longer that the timeout period (configurable from BLDC Tool, default is 1 second). The noise from the accelerometers is enough to not trigger the timeout when the joystick is held in the same position. So when the throttle gets stuck with the nunchuk (and does not switch off), one of the following things can happen:

1. The timeout is disabled in BLDC Tool and the signal is lost/battery runs out etc. The timeout is in the app tab.
2. The data cable comes loose and valid i2c packets are received now and then, but often enough that the timeout does not trigger (I never had a loose data cable, so I don't know if this can happen in practise).
3. The nunchuks fails somehow (e.g. the pot in the joystick gets stuck while the accelerometer still gives valid data).

I have been driving very long time with a short cable to the receiver with glue on the wires, and the throttle never got stuck for me longer than the timeout period. I have been testing this on two of my longboards and some friends longboards.

 

[furp]

I am planning to use my VESC (by jacobbloy) on a 5065 alien motor but I just can't decide what voltage to go for. I will be using recycled power tool cells sold by doctorbass here on the forum, either in 6s4p or 10s2p.
The cells can deliver up to 10c, so my max amp would be 60 (6s4p) or 30 (10s2p), on 10s I think I would have to limit the max speed to about 75% for safety. In theory, even 30 amp should give me about 1kw out of that motor, that should be enough, right?
The overall Wh is not much different, after all 6s4p is just 4 cells more. Would the higher voltage offer any advantage? I am really not sure and can't decide for now. When I get all the parts, I can check if I might squeeze 10s3p in there, which would be the best option I think.
I am hopeing to use a LiIon ebike charger for 10s and they usually come in 2A versions which should result in a acceptable charging time, or is there any reason I can not use one of those chargers (the cells dont need any bms)?

and lastly, does anyone know wheter the alien esc programmer can be used on the VESC, if so, what cables go where? Thx