PowerVelocity controller review

Status
Not open for further replies.
Exactly my point, the difference is that you can't afford to push hardware with something like that. For example, if you are flying an expensive aerial photography machine you want the utmost reliability, not some tinker toy in a critical component. If the cheap Cyclone controller, or the PV controller fails while riding my trike, well the worst is I'll be sweaty as heck from pedaling the lead sled back home... whereas if the cheap tinker toy fails while your 8000 dollar helicopter is in mid-air, with a 500 dollar gimbal holding a 1500+ dollar DSLR camera attached to a 2000 dollar tele-lens, then you'll certainly regret not buying the Kontronik when you see it taking a plunge from 300 feet... I can attest that tele-lens and DSLRs don't like falling from 300 feet... If you can afford to spend 12000 dollars for the helicopter/camera package, I think you can easily afford a 900 dollar Kosmik controller too...and while my scale heli isn't as expensive as others I've seen at the field, ppl who build turbine helis spend more than 20 grand on those things, (not counting the hours in building it) and when you look inside everything you see is top of the line stuff. Otherwise there would exist the concept of top of the line.

The VESC is probably decent for eSkates and other non mission-critical stuff... just like most of the eBike stuff these days, its all meant to be tinkered with b/c it if tails the worst its usually you having to pedal home (or call for a ride home)

G.

ElectricGod said:
The original VESC worked pretty well, but then the newer VESC 6 is taking advantage of lessons learned on the original model so that they are more robust and reliable. When the VESC 6 becomes more readily available, I'll buy at least one to "play" with.

Controllers burning out...anybody that has played with enough hardware and tried to push that hardware sufficiently has burned things out. I've destroyed a couple of controllers and a motor that way myself! I'm not sure I would call that "bad design" as much as I would call it abuse and exceeding the design limits of the device. I readily accept that when I blew up those 2 controllers and the motor that I was most definitely exceeding any reasonable limits for those devices. I bet your friend with the blown VESC, probably caused it's demise.
 
ElectricGod said:
Controllers burning out...anybody that has played with enough hardware and tried to push that hardware sufficiently has burned things out. I've destroyed a couple of controllers and a motor that way myself! I'm not sure I would call that "bad design" as much as I would call it abuse and exceeding the design limits of the device. I readily accept that when I blew up those 2 controllers and the motor that I was most definitely exceeding any reasonable limits for those devices. I bet your friend with the blown VESC, probably caused it's demise.

If you mean:
abuse -> pushed over the specified limits (hacked firmware, higher current settings)
than i agree with you^^

But if we are talking about the lack of features like short circuit protection (quick measuring of current on each phase), temperature monitoring or protections against watersplashes and such, than it is a differnt story, namely CHEAP made.
I personally spend more money for a reliable device in which i can trust 100% instead of having something that is prone to failures.

ElectricGod said:
VESC's are small and light and could be used in an RC heli for 30% the cost or less of these crazy expensive RC ESC's. I'd go that route first.

The secret is the governor mode.
this means the controller runs at lets say 70% PWM duty cycle and tries to keep the RPM of the Rotor as constant as possible.
similar to the cruise mode on e-bike controllers.
Kontronik ESC's have the best governor mode on earth, and i think no VESC can be compared with that.
 
Until gman mentioned the Komtronics products, I can't say I've ever heard of them before.

Clearly you build based on your needs. I've never gone the uber expensive route on helis. All my stuff was 3D or inexpensive 2D helis and for that sort of thing going with the best of the best isn't called for. I wrecked several 3D helis for various reasons. There was no way I was going to spend a fortune on them.
 
I bought two of the newer 12F AOT with dual sensored / sensorless mode and want to share my experience.

The app in the play store already contains the dash board function (the BT modules of my controllers unfortunately do not support the dash board), and it works stable and flawless.
Only thing i noticed is that sometimes, after flashing the settings, the controller does not respond to the throttle. It needs to be turned on and off to make it work again (the ignition).
Apart from this it works well.

The quality and packing:

This was a bit dissappointing. On both controllers the BT modules were flying around inside the case like they came loose during transport. The mounting bracket of the sidecovers on both controllers was bent and the case was scratched. It probabaly was not handled well during transport, but better packing could have avoided those damages.

So please, pack it better in future!

For making it work, i was forced to disassemble the controllers put the BT modules back to where it belongs.
than i noticed this:

mDHk7Y.jpg


who sees the flaw too?

About the function:

So far i have tried it with MXUS 3k and BionX D-motor. The controller works well and the motors are efficienct and silent.
The sensorless kicks in by simply unplug the halls. It changes to trapezoidal control than.
On the D-motor it looses sync very quick, but this motor is 44pole direct drive so more of the unusual type. Other motors i have not tried with sensorless yet.
In summary i am happy with the purchase.
 
madin88 said:
who sees the flaw too?

You got me stumped. I see a little Kapton tape on the end but doesn't look like a problem.

I see the FETs are upgraded, but I wouldn't call this a flaw. From what I can tell, those are Alpha&Omega AOT290L:
http://www.aosmd.com/res/data_sheets/AOT290L.pdf

These are significantly better than IRFB4110s based on the specs. These have very high pulse drain current and power dissipation ratings.

Edit:
I see it now, sort of hard to see in the picture. Extra layer of Kapton jammed in the wrong place. Not sure if that would cause a failure, but not ideal for sure.
 
fechter said:
Edit:
I see it now, sort of hard to see in the picture. Extra layer of Kapton jammed in the wrong place. Not sure if that would cause a failure, but not ideal for sure.

Yep thats it. The fourth FET from the left is sitting as a whole on a double layer or Kapton tape which is one of the thicker sorts (the roles i have are about half as thick).

I removed it, added a kerafol thermal pad in combination with kapton tape from top of the heatsink just until the bolt holes and not further (only for avoiding shorts from the pressure of the bolts on top of the FET's).
As i like to mod things, i also added two low ESR caps (the brown ones) and three ceramic caps at the backside :mrgreen:

The battery and phase wires have adequat cross section, but they are quite rigid. More flexible ones would be nice.

F0P0lX.jpg
 
madin88 said:
fechter said:
Edit:
I see it now, sort of hard to see in the picture. Extra layer of Kapton jammed in the wrong place. Not sure if that would cause a failure, but not ideal for sure.

Yep thats it. The fourth FET from the left is sitting as a whole on a double layer or Kapton tape which is one of the thicker sorts (the roles i have are about half as thick).

I removed it, added a kerafol thermal pad in combination with kapton tape from top of the heatsink just until the bolt holes and not further (only for avoiding shorts from the pressure of the bolts on top of the FET's).
As i like to mod things, i also added two low ESR caps (the brown ones) and three ceramic caps at the backside :mrgreen:

The battery and phase wires have adequat cross section, but they are quite rigid. More flexible ones would be nice.

F0P0lX.jpg

The BT module had come completely loose? Vadym has been using that exact connector for a while now. This is the first I've heard of this happening. I never get a controller with the module secured, but that's me. He knows I'm going to take it all apart anyway and hack and mod things. He has told me that usually they get some silicon on the side of the connector so they can't come apart. If you look at my controller pictures after I'm done modding them, I use s a JST connector. They lock together a bit better than pin headers do. BUT as long as the whole thing gets secured with some silicon, it won't be a problem either way.

The final telemetry module isn't released yet that makes the dials and gauges work. It's still going through some development. I have 2 versions of it so far and I know Vadym is currently working on a third version. You have the first version which does motor speed and battery current monitoring, but can't pass that to the app yet. We talk about stuff to add and new functionality pretty often so it's still a moving target. However, I can say that the second iteration of the telemetry module works pretty well and all those pretty dials and guages are in fact functional. The app is going to see some additional changes as well. Currently it doesn't show any temperature information, but the v2 module supports dual temp sensors. Also, the v3 module does a bunch of new stuff. I'm not going to say what all that is, but lets just say that it will be great. Vadym and I have been discussing user profiles as well. The idea is you have multiple PV products and the app will store parameters for each one separately. The app currently has a single profile. That's fine if you only have one EV, but if you have 4 of them...well that gets a bit confusing. I think profiles that auto-select based on the BT MAC address or something like that is coming soon too. All you should have to do is connect and maybe select the correct profile. Exactly when or the specifics of how that plays out is up to Vadym.

Regarding the "make a change and then needing to power cycle the controller" issue. We know about this and in future releases it is fixed. I've already tested it. I have a stack of controllers form various sources and about half of them need to be power cycled after a programming change. We didn't think it was a show stopper, but it's taken care of now. You make a change and it should take effect immediately.

I personally have 5 of these controllers and none of them came with an extra bit of kapton like that. I bet Vadym didn't even notice it was there. Until you said what the problem was, I missed it too and I'm probably a bit more familiar with these controllers than most folks. Was there something wrong with original kapton insulator? Why did you replace it? At most, I would have removed that extra bit of kapton and then kept the long piece in place. Sometime ago I bought a 12 fet sinusoidal controller from Grintech and the kapton insulator was damaged under 2 mosfets. I was bench testing the controller and suddenly it was dead. I soon discovered a couple of mosfets were fried. When I replaced them, I discovered the kapton insulator underneath them had a couple of holes in it that allowed current flow from the mosfet to the heat spreader. I had to replace the kapton insulator and later, Grintech...being the great folks they are... replaced the controller for me despite me working on it. Anyway, I was curious if you had found that it was damaged.

Regarding the wires on ALL of these controllers. They are teflon coated for all the large wires. Teflon coated wire tends to be stiff. This is common practice on any controller I have ever seen that uses this type of form factor. I've never seen any controller anywhere that uses silicon wires which is very flexible. Me personally, I replace the teflon wires with the largest AWG silicon wire I can cram in the holes, but that's my choice and isn't anything I've ever seen done by a manufacturer. I believe that if you ask, Vadym will replace the wires when you order one of his controllers for an extra charge. Otherwise, the factory wiring is adequate for the job.
 
On both controllers the modules where completely loose even if there was some glue on the connector and module to fix it as you described.
As mentioned, the package must have been thrown around ALOT so it wasn't the fault of Vadim alone. I am happy that both controllers work fine.

About the kapton tape i also thought the lower one might have got damaged, but it wasn't the case.
Why did i remove everything?
Well, as said i like DIY and improving things. It wasn't hard to add the kerafol thermal pad, and soldering additional caps - at least the large one in the middle because there are already holes on the board for it :)

Looking forward for the new modules. Good to know you keep on going with the development of new features. Thumbs up!
 
USPS priority shipping (within the US) is usually OK. They are more or less gentle with packages. I rarely have issues in US.
But this one went to Europe and I suspect that's where it took most of the beating. It's in a strong aluminum box, so it has to be hit really hard to damage it but who knows what those Europeans are doing with it? ;) I double pad them in bubble wrap envelops and that's how I can offer affordable and fast international shipping. If I boxed it, it would cost 2x more for shipping. Larger and heavier controllers ship out in boxes though with lots of padding.
 
Powervelocity.com said:
USPS priority shipping (within the US) is usually OK. They are more or less gentle with packages. I rarely have issues in US.
But this one went to Europe and I suspect that's where it took most of the beating. It's in a strong aluminum box, so it has to be hit really hard to damage it but who knows what those Europeans are doing with it? ;) I double pad them in bubble wrap envelops and that's how I can offer affordable and fast international shipping. If I boxed it, it would cost 2x more for shipping. Larger and heavier controllers ship out in boxes though with lots of padding.

What?! My 24 fet came is a plastic bag with no bubble wrap at all! It was beat to hell and all bent up. I spent an hour just making it look good for pictures. OK...not really...it was boxed up and arrived in good shape...exactly like in the pics. Ha ha.

madin88 said:
On both controllers the modules where completely loose even if there was some glue on the connector and module to fix it as you described.
As mentioned, the package must have been thrown around ALOT so it wasn't the fault of Vadim alone. I am happy that both controllers work fine.

About the kapton tape i also thought the lower one might have got damaged, but it wasn't the case.
Why did i remove everything?
Well, as said i like DIY and improving things. It wasn't hard to add the kerafol thermal pad, and soldering additional caps - at least the large one in the middle because there are already holes on the board for it :)

Looking forward for the new modules. Good to know you keep on going with the development of new features. Thumbs up!


I've gotten a 12 fet with a bent tab on it before when it came in a bubble wrap envelope. It was no big deal to bend it back. A minute with some pliers and it was good to go again. I think sometimes stuff gets dropped or whatever in transit. No real harm done. I didn't even bother to tell Vadym about it...there was no point. It sounds like your controllers got a good bashing to knock the modules loose. Dang! Use more silicon Vadym!

Modding is one of my biggest personal reasons for getting involved with these controllers. I saw them as having potential for lots of various and sundry done to them. Since you got the dual mode controllers, they already have thicker shells on them, but if you haven't, look at the beginning of this thread and what I did to my first couple of 12 fet controllers...added internal heat spreaders. All of my 12 fets have the 2 original shunts removed and replaced with 4 identical .005 ohm shunts, bigger wires and so on. My first 2 controllers didn't come with AOT290's I had to convince Vadym to get away from the IRF4110 so my first 2 controllers got several hours of love getting a mosfet swap out to AOT290's. Now they are standard equipment. Say, "thank you EG".

I couldn't find a pic of my shunt swap out, but I used 4 .005 ohm shunts. Two go in place of the factory shunts and 2 more are on the bottom of the board. Ignore those copper wires in place of the shunts. It was a bad idea and didn't work. The MCU couldn't detect current across the copper so it wouldn't run reliably. Anyway, notice the beefed up 10 awg silicon wires and JST connector. I also add a cap in that empty spot, but it's not in this pic.

Controller%20upgrades%201.jpg


NOTES about shunts:
1. You can't go below .001 ohms or else the MCU can't detect current draw. Four .005 ohm precision shunts is .00125 ohms. That's about all the lower you can go and run reliably.
2. Once you swap out the shunts, the current setting in the app will be wrong. Adjust conservatively!
3. If you look at the sticker Vadym puts on the side of the shell it will say something like 2.02 micro ohm. There is absolutely nothing wrong with that value, but 1.125 micro ohms is better. :)
4. 4 shunts instead of 2 is like buttering up your shunts with solder which I refuse to do. I want know what my shunts are actually doing. and 4 shunts allows for more current flow into the controller.
5. WARNING!!! doubling up the shunts like I do increases the chances that you will blow mosfets. Don't expect Vadym to warranty your controller if you do! I sure don't, but that doesn't stop me from telling him about my blown stuff. Information about maximum capabilities means we know more about what these controllers can do and where they die.

BTW...I've been running one of my 12 fet controllers at 192 phase amps and 60 battery amps at 66 volts for the past 30 miles or so. That's pretty close to 4000 watts continuous at the watt meter. I'm not sparing the controller at all. WOT from a dead stop to full speed, up long steep hills, full speed for a mile straight, whatever. The last ride was 12 continuous miles of that. The controller go to 80F or so in 65F ambient. I'm now turning the phase amps down incrementally to the point where the motor starts losing torque. Then I'll know what the real phase amps are. I'm also waiting for parts to arrive so I can build and actual phase amp meter. It will measure all 3 phases individually and display 3 separate gauges. Then we will know if the app and reality are in sync or not. It will be able to measure up to 200 phase amps. More to come soon. I don't think I told Vadym about that yet!

There's also these solder points that do stuff too that isn't implemented. I've messed with most of them at some point. Probably the one I wish was standard is an LED at P2. It's the status LED. It blinks and stuff. Actually P1 to P4 are all LED ports. At one point I had all of them wired up just to see what they did. One just turns on and doesn't do anything else. Unfortunately the manufacturer hasn't been particularly forthcoming on the LED's.

12%20FET%20controller%203_zpss7rjxq3f.jpg
 
I've spent waaaaaaaaay to much time getting the ASI controller running on my C80100! 6 hours at least...WTF! About a year ago, I spent many more hours just learning the general ins and outs of this controller so I could set it up for an HLD big block. Well finally the C80100 outrunner motor runs with halls. I'm sure there is a place in EV heaven saved up for me after that much effort to learn this controller and to get a motor to run! I now need to set it as close to my PV settings as possible. So far I've been focusing on just getting the controller to run my motor with halls, never mind bothering with anything like parity with the PV 12 fet settings. I still don't know how to set up regen on the ASI controller. It's in there somewhere. Honestly, it would be easier to turn off regen in the PV controller than it will be to turn it on in the ASI controller.

I've messed with 15 or so unique controllers now and most of them had some kind of set up app for them. My first 2 controllers were Kelley controllers and they have set up apps and NOT the same app or settings since they were very different controller types. They were a breeze to configure compared to the ASI controller. Every motor setup is something new with this thing. Some motors you put the pole pairs in the field for that. Other motors you enter how many magnets there are and others...just guess because nothing logical will work. Then there's the battery settings. Nothing about the hardware of this controller says "maximum input of 72 volts". In fact, the filter caps and mosfets are 100 volt components. However try running this thing on more than 74 volts or so and it will refuse, throw errors and disable itself. You have to fiddle with the battery voltage and set it 76 volts i think and then it runs at 82 volts. OK...that makes loads of sense...Ummm no NOT really! 82 volts is a very reasonable upper continuous limit for a 100 volt controller and 90 volts max is totally safe too. Brakes...geez brakes...by default, with or without brake sensors of any kind, the controller is set to disabled or rather to assume that the brakes are on and to disable itself. REALLY!? What kind of brain fart was that? So...after much turning stuff off that is irrelevant and just in the way or just plain wonkie settings, then you can actually do a motor discovery WITHOUT sensors. The motor may run smoothly and you might not get more controller errors that disable it. That's highly unlikely if you don't have the inductance and resistance of the motor set right...among other things. Then you can move on to discovering the halls. To date one motor EVER that I have tried had its halls discovered correctly on the first try. It's not like I've tested 1 or 2 motors on this thing either. LR big/small block inrunners, Astroflight 3220, BOMA inrunners, 3 different outrunners, a couple of no-name hub motors and a partridge in a pear tree. My AstroFlight 3220 that I recently added halls to is the only painless motor discovery process so far. AND keep in mind, "painless" is a relative term. It's always painful to set up this controller with a motor. Everything else required standing on your head while drinking water through a straw that you snort up your left nostril! GIANT, ROYAL PAIN IN THE BUTT! The C80100 took 6 hours to get it running. At least the motor poles were not a guessing game and I didn't need to fudge the battery setting since the Currie runs at 16S/66 volts.

Considering the pain factor in setting up the BAC2000 under the best of conditions, PV controllers sure are easy to use! I honestly wouldn't wish the BAC2000 on anyone that didn't have loads of patience and a strong background in controller configuration! I thought I was at least mediocre at this stuff, but then I met the BAC2000 and I was quickly pushed into the the complete noob moron category.
 
I've been running a 12F for over a year on my A2B. Only 52v so with my motor I can only get about 60A max battery draw (limited by the resistance of the motor windings). No issues with the controller, just the motor overheats if I run it hard for too long. Over 2kW into a 500W rated motor, so this is not unexpected. At road speeds, there are no heat issues at all, only when I'm climbing steep trails at 5mph.
 
fechter said:
I've been running a 12F for over a year on my A2B. Only 52v so with my motor I can only get about 60A max battery draw (limited by the resistance of the motor windings). No issues with the controller, just the motor overheats if I run it hard for too long. Over 2kW into a 500W rated motor, so this is not unexpected. At road speeds, there are no heat issues at all, only when I'm climbing steep trails at 5mph.

Right....I've had similar experiences. It's pretty impressive how much you can over watt a motor and it will still live.
 
So, E.G.... any thoughts about making the controller startup using halls then run in sensorless trapezoidal mode automatically after certain eRPM as an option? The Cyclone motor seems to have issues with the PV controller due to something... unknown.

Have you guys thought of a timing table for different RPM ranges and throttle settings, and linearly interpolate between the timings and throttle settings? that way you could control RPM through timing too as well as with PWM.

So there is zero current sensing on the phases? Is there any way to add that?

G.
 
Cyclone is the only motor I am aware of that has problems. Every other motor I've tested or heard from customers about is working fine.
Even Crystalyte Crown that is not supported by expensive controller brands, works just fine with PV controllers.

I suspect that issues with Cyclone is around sensor signals. Cyclone must be sending something out of ordinary. I am a having a customer that owns a Cyclone try it with a dual mode 18F controller in sensorless mode. Chances are, if we take the sensors out of the equation, it will work fine. We'll see.
 
Seems like a popular motor kit these days, and its one of the longest threads in E-S... if you can get it to work would be certainly good.

Thanks.

G.
 
I only had 3-4 customers running it. Might be popular before but, at the face value, internal reduction maybe adding unnecessary complexities. Motors can be wound for particular speed and/or torque belt or chain can be the only reduction necessary. Adding reductions on top of reduction seems like asking for a trouble. Plus added noise. I've never owned this motor though, so I don't have a first-hand knowledge of how good it is. Speaking purely from a theoretical standpoint.

gman1971 said:
Seems like a popular motor kit these days, and its one of the longest threads in E-S... if you can get it to work would be certainly good.

Thanks.

G.
 
Powervelocity.com said:
I only had 3-4 customers running it. Might be popular before but, at the face value, internal reduction maybe adding unnecessary complexities. Motors can be wound for particular speed and/or torque belt or chain can be the only reduction necessary. Adding reductions on top of reduction seems like asking for a trouble. Plus added noise. I've never owned this motor though, so I don't have a first-hand knowledge of how good it is. Speaking purely from a theoretical standpoint.

gman1971 said:
Seems like a popular motor kit these days, and its one of the longest threads in E-S... if you can get it to work would be certainly good.

Thanks.

G.

I own three of those motors on three different EVs, and the only reason I don't try your controllers is b/c it has issues working with it.

G.
 
Okay. Maybe send me one for a test and I'll try to see what the issue is.

gman1971 said:
Powervelocity.com said:
I only had 3-4 customers running it. Might be popular before but, at the face value, internal reduction maybe adding unnecessary complexities. Motors can be wound for particular speed and/or torque belt or chain can be the only reduction necessary. Adding reductions on top of reduction seems like asking for a trouble. Plus added noise. I've never owned this motor though, so I don't have a first-hand knowledge of how good it is. Speaking purely from a theoretical standpoint.

gman1971 said:
Seems like a popular motor kit these days, and its one of the longest threads in E-S... if you can get it to work would be certainly good.

Thanks.

G.

I own three of those motors on three different EVs, and the only reason I don't try your controllers is b/c it has issues working with it.

G.
 
I can try talking to the guy at Cyclone to see if he would be willing to send you one.

G.

Powervelocity.com said:
Okay. Maybe send me one for a test and I'll try to see what the issue is.

gman1971 said:
Powervelocity.com said:
I only had 3-4 customers running it. Might be popular before but, at the face value, internal reduction maybe adding unnecessary complexities. Motors can be wound for particular speed and/or torque belt or chain can be the only reduction necessary. Adding reductions on top of reduction seems like asking for a trouble. Plus added noise. I've never owned this motor though, so I don't have a first-hand knowledge of how good it is. Speaking purely from a theoretical standpoint.

gman1971 said:
Seems like a popular motor kit these days, and its one of the longest threads in E-S... if you can get it to work would be certainly good.

Thanks.

G.

I own three of those motors on three different EVs, and the only reason I don't try your controllers is b/c it has issues working with it.

G.
 
gman1971 said:
I can try talking to the guy at Cyclone to see if he would be willing to send you one.

G.

If you have an in with LunaCycles, then yeah get us a motor. I bet it's the halls or EMF interference. The one customer I know of supposedly replaced them, but then never tested the halls before trying out the motor. So who knows??? And we've never seen a scope signal from them either. Dirty hall signals will be a problem for any sinusoidal controller.
 
ElectricGod said:
gman1971 said:
I can try talking to the guy at Cyclone to see if he would be willing to send you one.

G.

If you have an in with LunaCycles, then yeah get us a motor. I bet it's the halls or EMF interference. The one customer I know of supposedly replaced them, but then never tested the halls before trying out the motor. So who knows??? And we've never seen a scope signal from them either. Dirty hall signals will be a problem for any sinusoidal controller.

No Luna... something else. :) Let me see what I can do.

G.
 
Status
Not open for further replies.
Back
Top