HPC Revolution XXX

Discussions related to motors other than hub motors.
This includes R/C motors, botttom bracket, roller and geared drives.
User avatar
circuit
100 kW
100 kW
Posts: 1431
Joined: Mar 19, 2009 11:43 am
Location: EU
Contact:

Re: HPC Revolution XXX

Post by circuit » Feb 22, 2016 11:24 am

Yes, high magnetic leakage and strong cogging is a little concerning...
smart and Tiny BMS solution
Visit my blog. Personal projects: Clean Peugeot conversion | Ultralight enduro
Any and all information, such as texts, specifications, drawings or other material that I post are to be used only as a reference for personal, non-commercial projects ONLY. Any republication for personal or commercial needs must be agreed on individually. Use of any information, that I post, for commercial purposes is strictly prohibited.

User avatar
macribs
1 GW
1 GW
Posts: 3334
Joined: Jul 22, 2014 5:59 pm

Re: HPC Revolution XXX

Post by macribs » Feb 22, 2016 11:36 am

circuit wrote:Yes, high magnetic leakage and strong cogging is a little concerning...
What is that in layman terms? Like net result of high magnetic leakage and strong cogging....not following.
Is this common for search motors or specific to Rv motors?

User avatar
circuit
100 kW
100 kW
Posts: 1431
Joined: Mar 19, 2009 11:43 am
Location: EU
Contact:

Re: HPC Revolution XXX

Post by circuit » Feb 22, 2016 12:03 pm

Magnetic leakage is somewhat common for cheap RC BLDCs. It just indicates that too little of rotor steel was used, or it is not the most optimal material in terms of efficiency.
As for cogging, well, also quite common in RC world... Probably the main reason to avoid it is noise and somewhat lower efficiency.
Don't get me wrong, it is still probably the greatest 'universal' motor in DIY world and I'm planning to mount one in bike as well, but there are numerous examples of cog-less and leak-less motors out there, like MXUS 3000W v2 for example. Not exactly middrive motor, but better in that aspect... It would be an absolute winner for middrive, if was easier to convert or at least smaller in diameter. Per my calculations, mxus is approx twice more powerful than RV-160s at same RPM.
smart and Tiny BMS solution
Visit my blog. Personal projects: Clean Peugeot conversion | Ultralight enduro
Any and all information, such as texts, specifications, drawings or other material that I post are to be used only as a reference for personal, non-commercial projects ONLY. Any republication for personal or commercial needs must be agreed on individually. Use of any information, that I post, for commercial purposes is strictly prohibited.

Jordan.1
100 mW
100 mW
Posts: 40
Joined: Feb 06, 2016 2:20 am
Location: Germany, Heidelberg

Re: HPC Revolution XXX

Post by Jordan.1 » Feb 22, 2016 12:58 pm

@Karolis im using aluminum so its not magnetic.

@madin88 he said 40Nm @ 200 phase Amps

User avatar
Miles
100 GW
100 GW
Posts: 11452
Joined: Mar 16, 2007 4:15 pm
Location: London UK

Re: HPC Revolution XXX

Post by Miles » Feb 22, 2016 1:14 pm

Jordan.1 wrote:@Karolis im using aluminum so its not magnetic.
I'm afraid that doesn't help you avoid eddy currents. The lower the resistance, the greater the losses....

Jordan.1
100 mW
100 mW
Posts: 40
Joined: Feb 06, 2016 2:20 am
Location: Germany, Heidelberg

Re: HPC Revolution XXX

Post by Jordan.1 » Feb 22, 2016 2:27 pm

Yeah but i dont see the point..? The whole frame ist made out of metal, the mounting plates -of course- aswell.
why should there be a problem then? eddy currents are everywhere but for me, theres no real problem or am i seeing this wrong?

User avatar
speedmd
100 MW
100 MW
Posts: 2820
Joined: Nov 14, 2012 12:16 pm
Location: new england

Re: HPC Revolution XXX

Post by speedmd » Feb 22, 2016 2:37 pm

Does it really matter if the rotor side outer shell is not fully - overly saturated or leaks just a little bit. Certainly not ideal if squeaking out max watts but possibly a good trade off having less mass to spin up quickly and heat transfer should be better from the magnets to the outside world with a thinner shell. Would be interesting to see how this motor performs with a custom higher flux capacity metal jacket with cooling fins machined in as a comparison to stock setup.
Last edited by speedmd on Feb 22, 2016 6:52 pm, edited 1 time in total.

Karolis
10 W
10 W
Posts: 85
Joined: Jun 01, 2009 8:03 am
Location: Lithuania, Vilnius

Re: HPC Revolution XXX

Post by Karolis » Feb 22, 2016 3:14 pm

I don't know if you can can call this a small leakage. You can almost lift motor from the ground (weight about 8Kg) buy putting some metal on its rotor surface.

Karolis
10 W
10 W
Posts: 85
Joined: Jun 01, 2009 8:03 am
Location: Lithuania, Vilnius

Re: HPC Revolution XXX

Post by Karolis » Feb 22, 2016 3:24 pm

Jordan.1 wrote:Yeah but i dont see the point..? The whole frame ist made out of metal, the mounting plates -of course- aswell.
why should there be a problem then? eddy currents are everywhere but for me, theres no real problem or am i seeing this wrong?
https://www.youtube.com/watch?v=E97CYWlALEs

https://www.youtube.com/watch?v=iABmUEH5s0k

know imagine your fast spining rotor is 2mm apart from your metal sheet as you said.

Leeleeducati
100 W
100 W
Posts: 192
Joined: Jul 06, 2015 8:38 pm
Location: Toronto, Canada

Re: HPC Revolution XXX

Post by Leeleeducati » Feb 22, 2016 5:51 pm

Very interesting videos. I have no doubt our home builds will experience inefficiencies and losses, but then again we're building for fun, they just have to go when we squeeze some throttle. It would be interesting to see the video experiments with magnets other than class N50. Also the tube experiment in a square tube (thin wall) rather than that perfect space around the cylindrical magnet. I don't think Alexey used N50s, I don't think we have that perfect spacing around our motors covered with metal... or even better the same experiment with a cylindrical "regular" magnet in a thin walled rectangle aluminum tube for uneven space around the magnet (motor). Might just drop through? just my thoughts...

Jordan.1
100 mW
100 mW
Posts: 40
Joined: Feb 06, 2016 2:20 am
Location: Germany, Heidelberg

Re: HPC Revolution XXX

Post by Jordan.1 » Feb 23, 2016 4:28 am

Thats also what im thinking. we just want to have fun. What I think is, that we dont even really use full motor power.

I guess what you guys are trying to say is that the eddy currents would disturb the motor from spinning, but really, i cant imagine this a big problem, sure its there but do we really care? We'll see if there are any problems, and if there are some, we have a great community to solve it ;)

BTW Im just making a clamp to mount sprockets on the shaft (without freewheeling), and I've calculated with the worst possibilities, and still its above the torque it has to handle. will post infos later

User avatar
Miles
100 GW
100 GW
Posts: 11452
Joined: Mar 16, 2007 4:15 pm
Location: London UK

Re: HPC Revolution XXX

Post by Miles » Feb 23, 2016 5:31 am

You're just adding to the parasitic losses... Less time to have fun, for a given battery :)

toolman2
1 kW
1 kW
Posts: 419
Joined: Feb 23, 2010 9:34 pm

Re: HPC Revolution XXX

Post by toolman2 » Feb 23, 2016 6:03 am

On paper (or screen) It seems like this could be i problem for sure, ive been concerned about it in a few builds with leaky (bell too thin) outrunners like the ca 120 at high rpm,
but for whatever reason it seems to have not been a problem that i can measure to be above ~20w of loss even with an alloy shell close to hugging the rotor.

A bit like the deadly and much feared low inductance monster, that appear to not be that big of a problem in the real world.

Do check though Leeleeducati, by measuring the no load losses of the motor at full rpm both in and out of the bike, and i must admit im partly saying that out of curiosity in this motor, wondering if it could maby be efficient etc and produce upwards of 15kw (measured output) at a way lower price than lighter high end motors, thanks :D .

User avatar
circuit
100 kW
100 kW
Posts: 1431
Joined: Mar 19, 2009 11:43 am
Location: EU
Contact:

Re: HPC Revolution XXX

Post by circuit » Feb 23, 2016 6:34 am

CA120 in notorious for deadly low inductance. Most likely associated losses are so high, that eddy related issues just drown in the whole mess unnoticed.
I know some people are running these and claiming good results. I had one, blew some s*** up, then did some math and realized I have to sell it. Its off topic, so I will not go in depth, but the forums are full of info.
As it was pointed out by Jordan.1, all is good when it is hobby, we can't really measure what is going on and most of our knowledge is built on bunch of strong opinions, like "it runs fine for me".
smart and Tiny BMS solution
Visit my blog. Personal projects: Clean Peugeot conversion | Ultralight enduro
Any and all information, such as texts, specifications, drawings or other material that I post are to be used only as a reference for personal, non-commercial projects ONLY. Any republication for personal or commercial needs must be agreed on individually. Use of any information, that I post, for commercial purposes is strictly prohibited.

User avatar
speedmd
100 MW
100 MW
Posts: 2820
Joined: Nov 14, 2012 12:16 pm
Location: new england

Re: HPC Revolution XXX

Post by speedmd » Feb 23, 2016 7:57 am

Is this concern just a matter of optimizing the iron to magnet ratio on this motor? Are the magnets thicker - more powerful than they need to be or back iron just a bit too thin to take full advantage of them? Is there more advantage to just using thinner -less powerful magnets (maintain the gap spacing) and not loose much power density and keep a lightweight power plant, or are you better off adding back iron and not have to deal with spacing the motor further from metal frame and protective covers. I can see some fiber glass use as crash plates until we see what the losses look like.

User avatar
madin88
100 MW
100 MW
Posts: 2581
Joined: May 27, 2013 2:02 am
Location: Austria

Re: HPC Revolution XXX

Post by madin88 » Feb 23, 2016 3:16 pm

Karolis wrote:I don't know if you can can call this a small leakage. You can almost lift motor from the ground (weight about 8Kg) buy putting some metal on its rotor surface.
WTF!
the revolt is more of a lower pole motor and has wide magnets. wide magnets require THICK back iron for optimal strength. thick back iron is heavy which is a big downside of such low pole count motors.
frankly speaking, after i have seen the vague performace graphs of these revolt motors, i was afraid that they have design flaws, but what would one expect from such rather cheap stuff?

edit:
there is a post with pics of the rotor where its clearly to see the wall thickness of the bell is much to thin for those magnets:

https://endless-sphere.com/forums/viewt ... 50#p929014
- Votec V-FR frame / MXUS 3T / 17" Mitas MC11 / Adaptto Max-E / 20s12p 25R -> approved as L1e moped and fully road legal in EU :)
- Vector white / MXUS 4T / 24" Hookworms / Adaptto Max-E / 22s11p 25R
- KTM Hardtail Pedelec / eZee V2 6T / 27,5" / Kelly KBSX @ 1,5kW / Torque PAS / CA-V3 / 13s4p Sanyo GA in Shark case

User avatar
speedmd
100 MW
100 MW
Posts: 2820
Joined: Nov 14, 2012 12:16 pm
Location: new england

Re: HPC Revolution XXX

Post by speedmd » Feb 23, 2016 3:55 pm

madin88 wrote:
Karolis wrote:I don't know if you can can call this a small leakage. You can almost lift motor from the ground (weight about 8Kg) buy putting some metal on its rotor surface.
WTF!
the revolt is more of a lower pole motor and has wide magnets. wide magnets require THICK back iron for optimal strength. thick back iron is heavy which is a big downside of such low pole count motors.
frankly speaking, after i have seen the vague performace graphs of these revolt motors, i was afraid that they have design flaws, but what would one expect from such rather cheap stuff?

edit:
there is a post with pics of the rotor where its clearly to see the wall thickness of the bell is much to thin for those magnets:

https://endless-sphere.com/forums/viewt ... 50#p929014
From the photo it looks to be much thinner in the edges that overlap the side plates and thicker at the magnets. I could see a nice slinky style magnet backer encased in a carbon or fiber glass shell. :shock:

Jordan.1
100 mW
100 mW
Posts: 40
Joined: Feb 06, 2016 2:20 am
Location: Germany, Heidelberg

Re: HPC Revolution XXX

Post by Jordan.1 » Feb 23, 2016 5:58 pm

So what are you guys suggesting? Would it help to put some metal around the motor shell to make it kinda thicker? What material would you suggest? Copper, aluminium, steel?

And what do you guys think about energy recuperation with that motor? I've planned a 3-4:1 gearing so the motor would spin quite fast even at low speed, and Ive also designed the gearing components to take on dynamic loads, so that would be no problem. Without recuperation, i guess,there would be a slight cog, or is it that hard? Does anyone know here?

toolman2
1 kW
1 kW
Posts: 419
Joined: Feb 23, 2010 9:34 pm

Re: HPC Revolution XXX

Post by toolman2 » Feb 24, 2016 10:49 pm

Hey Leeleeducati,
Others will chime in with the more complete answer but just to get you in the right direction a main contactor is a big relay (large switch contacts for heavy currents turned on by a low current coil) 7 and 6 are for the low current coil side.
6 definitely does NOT want connecting to B+ but 7 can be to turn the controller on.
The precharge resistor is for anti spark from battery + to controller to slowly charge the capacitors inside (a few seconds is enough) and yes you want it.

Once your up and running measuring stuff carefully, in this case no load losses of the motor accurately (and rpm if you can) b4 and after being on the bike is a good way to find the real truth compared to following the fears that others may have developed.
Keen to hear how you get on.

-edit, am i smoking crack or did your post with controller questions etc just dissapear ?

User avatar
spinningmagnets
100 GW
100 GW
Posts: 10916
Joined: Dec 21, 2007 10:27 pm
Location: Ft Riley, NE Kansas

Re: HPC Revolution XXX

Post by spinningmagnets » Feb 24, 2016 11:09 pm

And what do you guys think about energy recuperation with that motor? I've planned a 3-4:1 gearing so the motor would spin quite fast even at low speed
Some regen controllers have had to add "variable" regen because...max regen causes so much drag that the wheel can lock up. The maximum possible regen would require (at a minimum) fat tires and a significant ebike weight to create as much traction as possible. A light bike with skinny tires is the worst candidate. Also high C-rate batteries...or the batteries will get hot if they are being forced to swallow 7A or more.

Jordan.1
100 mW
100 mW
Posts: 40
Joined: Feb 06, 2016 2:20 am
Location: Germany, Heidelberg

Re: HPC Revolution XXX

Post by Jordan.1 » Feb 25, 2016 1:26 am

@spinningmagnets
What kind of controller do you all use? I'm here in Germany, we don't have much choice with the controller. I will use an 18 fet for my build, its a infineon style controller, so I can programm it. But the only regen available is from soft to medium to hard, there's no variable. Are there possibilities to do that with some special trick, or can the Cycle analyst V3 do this?

User avatar
izeman
10 GW
10 GW
Posts: 4630
Joined: Jun 21, 2011 8:25 am
Location: vienna, austria
Contact:

Re: HPC Revolution XXX

Post by izeman » Feb 25, 2016 3:24 am

Jordan.1 wrote:@spinningmagnets
What kind of controller do you all use? I'm here in Germany, we don't have much choice with the controller. I will use an 18 fet for my build, its a infineon style controller, so I can programm it. But the only regen available is from soft to medium to hard, there's no variable. Are there possibilities to do that with some special trick, or can the Cycle analyst V3 do this?
why should your choice of controllers be limited? you can buy/use any controller you want to? being located in austria, i ordered controllers from all over the world, us, china, australia, germany and i bought them in austria. but this was in the very beginning, and it had the worst price/performance ratio.
look for kelly, sabvoton or lebowski's controllers.
to answer your ca-v3 question: NO. that's not possible. the controller needs to do that. the CA can only limit/adjust the throttle signal not the regen signal.

Leeleeducati
100 W
100 W
Posts: 192
Joined: Jul 06, 2015 8:38 pm
Location: Toronto, Canada

Re: HPC Revolution XXX

Post by Leeleeducati » Feb 25, 2016 7:45 am

Wow you guys are quick! Thanks for the help and yes Toolman2 you did see correct. I did post a question and in searching the forums I found the answer. 6 to ground 7 to B+ and yes that main contactor is way to big to be usable so I will find some anti-spark type setup. Still no luck sourcing the 18650's but I will try a small power supply I have to get things spinning soon.

Jordan.1
100 mW
100 mW
Posts: 40
Joined: Feb 06, 2016 2:20 am
Location: Germany, Heidelberg

Re: HPC Revolution XXX

Post by Jordan.1 » Feb 25, 2016 3:35 pm

izeman wrote:
Jordan.1 wrote:@spinningmagnets
What kind of controller do you all use? I'm here in Germany, we don't have much choice with the controller. I will use an 18 fet for my build, its a infineon style controller, so I can programm it. But the only regen available is from soft to medium to hard, there's no variable. Are there possibilities to do that with some special trick, or can the Cycle analyst V3 do this?
why should your choice of controllers be limited? you can buy/use any controller you want to? being located in austria, i ordered controllers from all over the world, us, china, australia, germany and i bought them in austria. but this was in the very beginning, and it had the worst price/performance ratio.
look for kelly, sabvoton or lebowski's controllers.
to answer your ca-v3 question: NO. that's not possible. the controller needs to do that. the CA can only limit/adjust the throttle signal not the regen signal.

Cause i want to geht some warranty ;)

User avatar
macribs
1 GW
1 GW
Posts: 3334
Joined: Jul 22, 2014 5:59 pm

Re: HPC Revolution XXX

Post by macribs » Feb 25, 2016 4:26 pm

Cause i want to geht some warranty
You can buy ie Adaptto from Artur/Vector. Because you are then dealing with a German company. Check with him.
For all I know there are also various German sources for Kelly, Sabvaton etc.

Post Reply