Revolt RV-series motor review and comparisons

Electric Motors and Controllers
User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 11 2019 10:19pm

Avi wrote:
Mar 11 2019 4:05pm
But after a week i realized I was not interested וn out -runner motors any more
Sorry for cutting up your post and taking you out of context...

I did have a few things to point out about outrunners in general. I'm sure this will start a whole new debate and line of arguing and that I'll get nit picked all over again. LOL!

1. Outrunners are generally wired delta. I have yet to see one ever wired WYE. I have yet to see an inrunner wired delta.
2. Outrunners use a longer "lever" than inrunners. The armature is the lever that makes you go. A really small lever has to spin super fast to create the torque of a much slower and longer lever. This helps with simplifying gearing and motor controller eRPM limits. All else being equal, for the same wattage in you should get more torque from an outrunner than an inrunner thanks to the longer lever.
3. For the weight, size and cost an out runner is much smaller, much cheaper and weighs less than an equivalently powerful inrunner. For example I have an HLD inrunner. it is the same approximate cost, size and weight of an AP 12090. The inrunner is good for about 6-7kw while the 12090 more like 10-12kw.
4. Inrunners heat soak the magnets since they are trapped inside the stator. Outrunners don't and keeping the stator cool is not difficult in an outrunner.

There's other factors that play into this too, but this covers the big details. I'll probably never buy another inrunner again. OK...if an Astroflight 4535 dropped into my lap for a really good price, I'd buy it. It's not for lack of trying them either...I have 2 Astroflight 3220's, 3 HLD inrunners and 3 BOMA inrunners. Outrunners smoke inrunners in just about every way. The few ways inrunners are better are not IMHO worth switching back to get them.

If your choice is hubs, I'm not overly thrilled with them either. Power to weigh ratio in a hub is way worse than even an inrunner. They are a heavy spinning mass exactly where you don't want it...in the middle of a wheel. They are a massive unsprung weight in the middle of the wheel. Ugg! And again...this is not for lack of trying them. I have 6 hub motors of which 2 are capable of 40kw (hubmonsters). Having said that, Once I drop the RV-120-regular in my moped build, I'll be building on a hubmonster. Yes John...you are right...it's about bloody time too!

IMHO, for the cost, weight and size for the wattage output, you can't beat an outrunner.

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 11 2019 10:38pm

larsb wrote:
Mar 11 2019 2:53pm
AP motor with same nol oad losses but a third of the resistance losses is a superior motor to the RV-120. A diagram of the noload losses per rpm plotted for all the motors would be nice.

The saturation point of the steel will only be known when you put some load on the motors, then you can say something about power potential of the motors.
The AP motor has 3X MORE phase current under no load than the RV-120-regular. That doesn't sound like it's "better" to me. My test was simple. Connect the motor to the controller, crank the throttle to WOT and measure the phase current. I tried all 3 phases on both motors. They were consistent across their phases and always the AP 12090 had 3X more phase current under no load. How is that better?

Saturation and loading...I agree...you are dead right! The only way to know for sure is to load test. Good thing I'm getting the 120-regular set up so I can run it on my scooter! I'm pretty certain (95-98% sure) it will out perform the HLD inrunner on it now.

Well...my working 12090 has a problem...grrr! I posted phase inductance and resistance back a few pages for a bunch of motors I have. The inductance across the 3 phases is really inconsistent in the 12090. No other motor I have has that much variation between phases. This got me curious so I slid the bell off and there's a partial short against the stator on one of the phases! THere's no visible signs of damage, this is underneath the windings. Since it's just one phase, this probably explains why the motor still runs and no one phase shows more phase current. It probably also explains the wacky inductance on the one phase. Looks like a rewind for 6 phases on a 12090 is in my near future! The irony is the motor has seen bench testing only. Otherwise it's brand new! I bought it purely to have for messing around with and never intended to use it anywhere. Once I rewind it for 6 phases, you can be sure it will see use after it is bench ran a good bit.

I think I'll use 14 awg. That hair fine wire on it now can't possibly have a very high copper to insulation ratio.

I won't be running the 12090 again until I get it rewound. I'm sure that stator short is absolutely unkind to mosfets!

Punx0r   100 GW

100 GW
Posts: 5238
Joined: May 03 2012 8:16am
Location: England

Re: Revolt RV-series motor review and comparisons

Post by Punx0r » Mar 12 2019 6:08am

ElectricGod wrote:
Mar 09 2019 7:43pm
AP 12090
Battery amps: 3.2A
Phase amps: 22A
I'm not looking for an argument here, but as a spectator to this thread it looks like it might help to point out the crux of the problem here:

If 20S li-ion (call it 72V) then controller power in is: 3.2 x 72 = 230W

Controller efficiency is going to be about 95% worst-case, so power out (to motor) would be approx 230x 0.95 = 219W

You said testing was done at WOT, so we can assume 100% duty cycle and Vout = Vin

If total average phase current was 22A, Pout would be 22 x 72 = 1584W i.e. power out was 6.9 times power in.


As true phase current is hard to measure, why not compare battery current/power for the different motors instead? Yes, if the motors are running at different speeds then switching losses in the controller will vary, but we're proably talking about a few per cent error on the final measurement.

User avatar
madin88   100 MW

100 MW
Posts: 2944
Joined: May 27 2013 2:02am
Location: Austria

Re: Revolt RV-series motor review and comparisons

Post by madin88 » Mar 12 2019 6:35am

ElectricGod wrote:
Mar 11 2019 10:38pm
The AP motor has 3X MORE phase current under no load than the RV-120-regular. That doesn't sound like it's "better" to me. My test was simple. Connect the motor to the controller, crank the throttle to WOT and measure the phase current. I tried all 3 phases on both motors. They were consistent across their phases and always the AP 12090 had 3X more phase current under no load. How is that better?
It is the battery current which is relevant for comparing no-load losses, and not the ripple current in the motor phases (this is what your clamp meter probably was reading).
Even if it features true RMS AC current measurement, it is still might have been designed for measuring it in a range of 50-60Hz, so it's accuracy might be off if the frequency changes. I did look at the manual of your clamp meter and it says 50-60Hz AC current measurement.
As punxor wrote, true RMS phase current is really hard to measure.
ElectricGod wrote:
Mar 11 2019 10:19pm
1. Outrunners are generally wired delta. I have yet to see one ever wired WYE. I have yet to see an inrunner wired delta.
Only because you never have seen an outrunner wired in star, or an inrunner wired in delta, it doesn't mean that they do not exist.
Most Hubmotors are wired in star btw.

Your signature says it all: It's is a joke, don't take things to seriously, so we should do with your statements which are based on assumptions most of the time and not on facts.

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 12 2019 11:35am

Avi wrote:
Mar 12 2019 3:18am
I am into in-runner motor
All the big shot use them
KTM - freeride
Zero bike
Alta bike and so on.....
You can not compare outrunner to YASA motors. new saietta motors Magnax motor and so on ......
They all use inrunner...
You gone to tell me that they are big motors , Not suitable for bicycles.
It's a matter of time antill they do ....( that lead me to the Crazy idea i have got , will post soon)
Out runner o.k too .
I just like inrunner more....
 I do not rule out the out runner at all.
 That's why I'm participating in this forum.
Inrunners: (IMHO, the 3 big advantages of inrunners)
1. Having the stator on the outside of the motor helps keep it cool. In an outrunner, that's usually pretty easy to control with forced air flow or a radial fan in the motor. The Revolt Pro and E series motors have a radial fan built into the top of the armature. I don't know how good it is yet, but I will be using my anenometer (sp? Wind meter) to find out.

2. They are typically enclosed which is good if the motor will be exposed to mud and lots of dirt. In an outrunner, sealed bearings takes care of one issue. Laquer dipping the stator and spraying down the internals with a decent electrical paint protects the motor from damage to water and dirt incursion. I would not be afraid to run an outrunner in the mud after a little protection is applied to it.

3. Inrunners tend to have far fewer magnets in them than outrunners. For example an HLD inrunner or 3220 have 8 magnets. The AP 100 series outrunners have 14 magnets and the 120 series have 28. More magnets means higher pole counts and that always means higher motor eRPM's. Typical EV controllers are 40-50,000 eRPM's so you have to be more careful to not exceed the eRPM limits of the controller on an outrunner. Of course there's faster controllers too...so that's a way that the pole count can be made to not matter. This is an area that I really like about Revolt outrunners. They keep the pole counts in their outrunners down to 7 on the entire product line.

I didn't say there aren't good inrunners. There are...just the cost for them compared to a similarly powerful outrunner is significantly more, weight is more and size is more. There are exceptions for some of those things. For example the AstroFlight 3220 is rated for 6kw and is the same size, weight and power as a C80100 outrunner. It also costs nearly 3X more. It can be done, but at considerable cost. I have yet to see an inrunner that costs the same as a similarly powerful outrunner that also was the same size and weight.

For the price, size, weight and power density, outrunners usually win over inrunners.

Yasa, Saletta and Magnax motors are not inrunners. They are all axial flux. Axial flux is all the best of an outrunner and inrunner combined and uses either both sides of the magnets or both sides of a single stator. This is a very different animal from an inrunner...and yes they are generally better than either inrunners or outrunners. I'd like to see a bicycle with any of these motors on it! They would probably twist the frame in half even at 10% of full power. I have a Motenergy axial flux motor. It's capable of 80kw so it will be a while before I use it anywhere.

district9prawn   100 W

100 W
Posts: 252
Joined: Jul 20 2013 10:46pm
Location: Sydney

Re: Revolt RV-series motor review and comparisons

Post by district9prawn » Mar 13 2019 3:05am

I had a play around with a small HK outrunner and a cheap UNI-T UT204 clamp meter. The meter is supposed to have true rms capability.

I tried pwm frequency from 5 to 30khz in foc. At 30khz phase current was 6a. Dropping the switching frequency resulted in higher measured motor current. At 5khz it was over 15a. In trap mode it was about 7a. Controller was a vesc.

Someone suggested earlier in this thread that ripple currents are likely the cause for the high readings. Seems likely as ripple will be a lot higher at low fsw.

EG I'm still a bit baffled why you still believe your measurements when many of us have pointed out they blatantly violate conservation of energy.
Neu 8057 6kW left hand drive hardtail with 18 fet Vesc: viewtopic.php?f=6&t=96754

User avatar
madin88   100 MW

100 MW
Posts: 2944
Joined: May 27 2013 2:02am
Location: Austria

Re: Revolt RV-series motor review and comparisons

Post by madin88 » Mar 13 2019 4:44am

district9prawn wrote:
Mar 13 2019 3:05am
I tried pwm frequency from 5 to 30khz in foc. At 30khz phase current was 6a. Dropping the switching frequency resulted in higher measured motor current. At 5khz it was over 15a. In trap mode it was about 7a. Controller was a vesc.
That makes a lot of sense.
If you would run this motor with 5kHz PWM, the phase current peaks would be always by those 9A higher (6A at 30kHz vs 15A at 5khz).
Or as an example if the current under load would be 100A, it would peak actually at 109A.
The problem is that those additional 9A (or ripple current generally) will just produce more heat in windings and add nothing to the torque output.

I guess you have set it to 30kHz then for this motor, right?

Punx0r   100 GW

100 GW
Posts: 5238
Joined: May 03 2012 8:16am
Location: England

Re: Revolt RV-series motor review and comparisons

Post by Punx0r » Mar 13 2019 4:52am

ElectricGod wrote:
Mar 12 2019 12:00pm

Help me understand this: Vout always equals Vin at any RPM or duty cycle. There's no high current voltage regulator or step down circuit in a controller. The mosfets either apply full battery voltage to the phases or they don't. Speed and torque control is not done by adjusting the voltage, but by adjusting the PWM or the frequency of the current sinusoid going to the motor. This is typical Class D amplifier stuff.
Kinda. The inductance of the motor windings plays a part, though. It's smooths out the PWM so motor sees a continuous, lower voltage rather than pulses of full battery voltage. So at 50% PWM the phase voltage is roughly 50% of Vbat. The current flow (and so motor torque) is then limited by ohm's law. That way, with a throttle that only controls the PWM we get speed and (some) torque control.

IIRC it's also the motor inductance that enables the controller to function like a buck converter and so generate phase current that is greater than battery current when phase voltage is lower than battery voltage.

As to why the Alien Power motors seem to have larger peaks in the phase current I can only guess. Maybe it's fairly normal and no big deal, maybe similar peaks exist on the Revolt motors but aren't being tested, maybe it's because there's something nasty about the BEMF of the AP motors. No idea.

Rube   100 W

100 W
Posts: 251
Joined: Jun 25 2016 10:18pm
Location: Brisbane, Australia

Re: Revolt RV-series motor review and comparisons

Post by Rube » Mar 13 2019 5:36am

The discourse about measurement accuracy and precision is fascinating and worthy of a standalone thread with examples/ industry standards. I appreciate EG's effort and those skilled in analysing the real world charcteris of motors, controllers and measurement devices.

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 13 2019 11:32am

Punx0r wrote:
Mar 13 2019 4:52am
ElectricGod wrote:
Mar 12 2019 12:00pm

Help me understand this: Vout always equals Vin at any RPM or duty cycle. There's no high current voltage regulator or step down circuit in a controller. The mosfets either apply full battery voltage to the phases or they don't. Speed and torque control is not done by adjusting the voltage, but by adjusting the PWM or the frequency of the current sinusoid going to the motor. This is typical Class D amplifier stuff.
Kinda. The inductance of the motor windings plays a part, though. It's smooths out the PWM so motor sees a continuous, lower voltage rather than pulses of full battery voltage. So at 50% PWM the phase voltage is roughly 50% of Vbat. The current flow (and so motor torque) is then limited by ohm's law. That way, with a throttle that only controls the PWM we get speed and (some) torque control.

IIRC it's also the motor inductance that enables the controller to function like a buck converter and so generate phase current that is greater than battery current when phase voltage is lower than battery voltage.

As to why the Alien Power motors seem to have larger peaks in the phase current I can only guess. Maybe it's fairly normal and no big deal, maybe similar peaks exist on the Revolt motors but aren't being tested, maybe it's because there's something nasty about the BEMF of the AP motors. No idea.
You are referring the averaging effect the motor phases/large inductors create in circuit.

I have never checked this myself so this is purely supposition on my part. Minus the large inductors, looking at the voltage on an oscilloscope, the PWM varies, but the voltage does not. If it weren't for the inductors (motor phases), there wouldn't be a sinusoidal pattern at all. The same can be said for class D audio amplifiers. They need the inductors to even create the sinusoidal current wave form, but without them, you get PWM pulses at essentially full voltage swing. For example the mosfet drivers don't partially turn on or off the mosfets. This would be inefficient and the mosfets would heat up rapidly. They turn them on 100% or off 100% all the time. All that changes is the duration of on vs off. In this regard, the mosfets are merely a light switch that gets flipped really quickly ins combinations that change the direction of current flow through the inductor. It would not be a very good motor controller and would quickly break down, but you could create commutation with relays that are arranged like an H bridge. The reason this would work is the mosfets are just highly reliable switches.

Phase current is an area motors that has always held some mystery for me. Getting significantly higher phase amps than battery amps seems impossible. I've read about it many times, measured it multiple times, I understand this conceptually, but in a part of my mind, it is also "magical". Kind of like over unity devices...magical power out that wasn't there originally. I've thought many times about how I could tap all that crazy current in a motors phases and use it somehow for running lights or whatever. I'm not delusional and I do understand that this is just fantasy. Never the less, how cool would that be?

Anyway, yes...there are possibilities here to why the AP motor shows much higher phase currents. Without a lot more effort, finding out is not likely. And frankly, most people won't ever go even as far as this discussion! Personally, I'm content to read my meters and leave it at that for now. On my "list" of things to study eventually in a lot more detail is motor theory so that I can be a lot more knowledgeable in this area. So far, I've read many articles and white papers, but not bothered to spend time really getting "deep". There's way too many things I can spend my time on!

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 13 2019 11:46am

district9prawn wrote:
Mar 13 2019 3:05am
I had a play around with a small HK outrunner and a cheap UNI-T UT204 clamp meter. The meter is supposed to have true rms capability.

I tried pwm frequency from 5 to 30khz in foc. At 30khz phase current was 6a. Dropping the switching frequency resulted in higher measured motor current. At 5khz it was over 15a. In trap mode it was about 7a. Controller was a vesc.

Someone suggested earlier in this thread that ripple currents are likely the cause for the high readings. Seems likely as ripple will be a lot higher at low fsw.

EG I'm still a bit baffled why you still believe your measurements when many of us have pointed out they blatantly violate conservation of energy.
Back when Vedder was still developing the VESC, he made several videos of motors running at very slow commutation speeds with crazy high phase currents. He had one video where the motor would get super hot so he ran the tests with the motor immersed in a plastic bin full of water. I had to laugh, but the demonstration was pretty amazing too.

I believe your meters readings are probably accurate or at worst pretty close to true.

Phase currents and battery currents as you just saw can be wildly different. I don't think any conservation of energy laws are broken at all. It's just the nature of inductors and iron and magnets interacting that make it "look" like this is the case. Also, there are many examples to be found of real devices that collect the BEMF and reuse it in interesting ways that recovers that energy. The cycling effects of rapidly reversing magnetic fields, collapsing magnetic fields, BEMF, magnets passing by inductors, and other elements create these interesting effects. I agree it is interesting and somewhat mysterious, but there's not magical energy in the phases.

district9prawn   100 W

100 W
Posts: 252
Joined: Jul 20 2013 10:46pm
Location: Sydney

Re: Revolt RV-series motor review and comparisons

Post by district9prawn » Mar 13 2019 4:33pm

ElectricGod wrote:
Mar 13 2019 11:46am
Phase currents and battery currents as you just saw can be wildly different. I don't think any conservation of energy laws are broken at all. It's just the nature of inductors and iron and magnets interacting that make it "look" like this is the case. Also, there are many examples to be found of real devices that collect the BEMF and reuse it in interesting ways that recovers that energy. The cycling effects of rapidly reversing magnetic fields, collapsing magnetic fields, BEMF, magnets passing by inductors, and other elements create these interesting effects. I agree it is interesting and somewhat mysterious, but there's not magical energy in the phases.
Yes everyone has been trying to tell you that your results imply that there IS magical energy between phases. When you spin your motor at full speed, the voltage between the phase terminals is nearly battery voltage. So if your rms phase current at full speed is 5x higher than battery current, you violate conservation of energy. As Lars mentioned in his previous post you can multiply rms ac currents just like dc to calculate power. Myself, madin, lars and others keep saying this but something isn't clicking.
Neu 8057 6kW left hand drive hardtail with 18 fet Vesc: viewtopic.php?f=6&t=96754

User avatar
spinningmagnets   100 GW

100 GW
Posts: 11195
Joined: Dec 21 2007 10:27pm
Location: Ft Riley, NE Kansas

Re: Revolt RV-series motor review and comparisons

Post by spinningmagnets » Mar 13 2019 5:35pm

On the weekend of March 17, I deleted quite a few posts (on the last two pages, at the time) that were part of a discussion that devolved into an argument with a sprinkling of personal attacks.

Posts that only discussed motors and electrical phenomena were left intact. Posts that were personal attacks and insults were deleted. I regret that some posts were a mix, and I simply don't have the time or the expertise to determine where I should delete "half a post".

Few people have bought and tested a Revolt motor, so the general public can thank ElectricGod for the pics and info...If anyone disagrees with the results posted here, the accepted remedy is to buy a Revolt motor, and then test it. Then you will have the right to post what your results are.

If there are two Revolt motor threads, please trust that the reading public will determine which thread has the trustworthy information.

For future reference, if anyone here on ES is annoyed by someone, please take the first step of "blocking" them. That way, you won't see what they post.

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 17 2019 2:10pm

I'll be using 219 chain and sprockets. The shaft on the RV-120 is 15mm. I have several of these shaft adapters already made that have a 3/4" OD with a keyway. The ID's are 10mm or 12mm or solid.

Image

While not ideal to mill parts on a drill press, it's what I have. Carbide mills still work even on a drill press. Spinning the adapter around the mill hollows it out. I'm pretty sure even a cheap mill would do a better job, but this works too...more or less.

Image

This is more of a sleeve now. One of my limits is the length of the cutting surface of a mill. Lack of precision in the drill press is a problem too. As a result about all the deeper I can mill out and do it slowly is 1"

Image

I'll need to make a single rectangular key that catches the shaft keyway and the sprocket keyway. Probably it will want to roll over since it's taller than it is wide. Typical 219 sprockets have dual 5/16th set screws in them. I think I'll add a third one and then use longer screws so that 2 set screws can land in the shaft. I'll mill 2 flats on the shaft for the set screws to land on. Since I wont know in advance the exact position of the sprocket on the shaft, I'll make the keyway and flats the length of the shaft end.

Image

Image

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Revolt RV-100E

Post by ElectricGod » Mar 18 2019 12:45am

I got to mess with the RV-100E a little and made a paper template for the bolt pattern for a base.

Now that the base is made, I can pull the motor apart.

I made 3 blanks so one of those got used today to make a base for the E.

Since I didn't know for sure if the paper bolt pattern transfer would be accurate or not, I pulled the base off the table after I got the center hole and 6 bolt pattern done. As it turns out, it was a perfect fit. The M6 screw holes were not tapped very well and one of the screws bound up in the aluminum. I was concerned I wouldn't get the screw back out. Afterwards all 6 holes got retapped by me.

Image

The finished base. Countersunk screws are coming soon for it.

Image

Image

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Revolt RV-100E

Post by ElectricGod » Mar 18 2019 12:59am

With the motor base on it, I can now turn the motor. The much stronger magnets definitely make the cogging effect a lot stronger.

The bell bottom came off quite easily. Remove 6 screws and with very little prying, the bottom came right off the armature. The bell bottom and top are CNC milled quite nicely.

Image

There's a circlip on the bottom of the shaft. With the bell bottom and circlip removed, remounting the motor base and using a puller made quick work of pushing the bell off. I was hoping there was a radial fan in the bell top, but I guess not.

Image

Image

The hall board is interesting. Clearly it is by design, but 2 of the 41F halls face backwards. They each have a 1K resistor across pin 1 (+v) and pin 3 (signal). This is redundant since controllers have pull up resistors. I'm not sure how pull-ups at the halls will help.

Image

Image

Image

Image

Image

I'll take this apart later and see how good the connection is.

Image

Like all Revolt motors the stator looks pretty close to completely full. I've measured the lams and they are .3mm thick. Everything is coated in laquer. There's 6 strands per phase.

Image

Image

Image

Image

I'm not sure what happened here...

Image

User avatar
madin88   100 MW

100 MW
Posts: 2944
Joined: May 27 2013 2:02am
Location: Austria

Re: Revolt RV-series motor review and comparisons

Post by madin88 » Mar 18 2019 8:33am

From the looks the stator seems to be on another level in comparison to the previous revolt motors.
Thank you for the pics!
Why they decided to place the hall sensors beside the stator hanging free, and with a large gap to the magnets is kinda strange and also the thing with the large resistors looks unprofessional because they could have put SMD ones directly on the circuit board. I wonder if the magnets are wider as the stator?

Does the bell leak any magnetism (paper clip test)?
Did you do any measurements or no-load tests already?

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 18 2019 10:46am

madin88 wrote:
Mar 18 2019 8:33am
From the looks the stator seems to be on another level in comparison to the previous revolt motors.
Thank you for the pics!
Why they decided to place the hall sensors beside the stator hanging free, and with a large gap to the magnets is kinda strange and also the thing with the large resistors looks unprofessional because they could have put SMD ones directly on the circuit board. I wonder if the magnets are wider as the stator?

Does the bell leak any magnetism (paper clip test)?
Did you do any measurements or no-load tests already?
OR at least "next level" to the Regular motors...that's the only thing I have looked at so far. It is a nice stator.

The magnets and stator are the same length.
The hall board is held in place by 2 small hidden screws.
The resistors "hanging" by their legs is likely to get them broken off via vibration over time. A little thermal glue to secure them would resolve this.

Until yesterday, I didn't have a base made yet. No chance to do any live testing...yet.

Why place 2 of the 3 halls backwards?

I did a comparison with the C80100 and 12090. The Revolt Regular motors shield the magnets better than the AP motors. I used a wrench and a screw driver instead of a paper clip. The Pro motor shields better than the Regular and the E better than the Pro. This wall is really thick. I'm sure that adds a considerable amount of spinning mass to the motor. I bet these ridges creating an air gap to the magnets helps it look like there's less magnetic leakage too.

Image

Image

The Regular motors RV-110-reg and RV-120-reg) bell wall thickness.

Image

Image
Last edited by ElectricGod on Mar 18 2019 4:28pm, edited 4 times in total.

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 18 2019 11:17am

[off topic comments removed by moderator]

Last night I replaced the shielded bearings in the E with sealed ones. I'll be adding SMD halls between the stator teeth like I did to both Regular motors.

Hall placement isn't very consistent. As can be seen the halls don't overlap the magnets. It's not readily obvious in the images, but the halls are not placed dead center over the stator gaps.

Image

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Revolt RV-100 series motors

Post by ElectricGod » Mar 18 2019 8:47pm

This ought to make a few people happy...

I took apart all 3 RV-100 series motors so I could do comparison images.

Bell bottoms.

Image

Left to right: RV-100-Regular, RV-100-Pro, RV-100E. Same diameter (84mm), but 45mm tall for the Regular and Pro and 60mm for the E. The Pro and E both have .3mm laminations. Only the Regular has .5mm laminations. Looking under all 3 stators under a 35X magnifier, you can see the Pro and E lams are the same thickness. I also used my micrometer and measured both of them. The Pro stator looks a little less full than the E stator. I count 4 strands on the Regular phases, 5 on the Pro and 6 on the E. I guess the stator iron in the teeth of the Regular must be thicker than what's in the Pro and E. How else do 4 strands look like they fill up the stator as much as 6 strands? It's .8mm wire on all 3 motors.

Image

Image

Image

None of these motors have as much magnetic leakage as an Alien Power. In all 3 motors the leakage is very slight. I used a circlip off the Pro and in all 3 motors, the circlip is just barely attracted to the bell. Jar the bell even slightly and the circlip falls off. Testing again just now, I think the slightly thicker bell wall than what an AP motor has is the big reason for this. Just a "feel" measurement, but the magnets in the Regular are definitely weaker than in the Pro and those are weaker than the E magnets. The Pro and E both use curved magnets.

Image

Image

Image

Pro vs E: The stators are essentially shorter and longer versions of each other. One has 5 strands and the other 6 for a bit more current handling, but otherwise they look to be the same construction with the same laminations. Magnet strength and phase wire size are the other big differences.

Regular vs Pro: .5 or .3mm lams, otherwise the stators look identical. 4 or 5 strands per phase and magnet strength are the big differences.

In conversing with Revolt, they won't tell me anything other than what they post on their web site. I think they really dislike me scrutinizing their motors, but won't say anything. You'd think they would want to brag about their motors...at least the Pro and E...nothing. They get annoyed when I tell them what I have found in the motors. They think their motors are perfect or something as is.

This is a bit disappointing about the RV-100E. Seriously a single key in the shaft to bell union! Whose dumb idea was that? The bell keyway is 24mm and the key is 15mm. Another missed detail. Good thing I plan to reinforce this like I did to the 120-Regular! BTW...the RV-120-regular has 2 keys in it's shaft and supposedly this motor is stronger. How pray tell does that work?

Image

district9prawn   100 W

100 W
Posts: 252
Joined: Jul 20 2013 10:46pm
Location: Sydney

Re: Revolt RV-100 series motors

Post by district9prawn » Mar 21 2019 1:15am

ElectricGod wrote:
Mar 18 2019 8:47pm
In conversing with Revolt, they won't tell me anything other than what they post on their web site. I think they really dislike me scrutinizing their motors, but won't say anything. You'd think they would want to brag about their motors...at least the Pro and E...nothing. They get annoyed when I tell them what I have found in the motors. They think their motors are perfect or something as is.
I find their attitude just as concerning as the problems you've found in their motors.

In practice would the single key holding the shaft to the rotor be that big a problem.
Neu 8057 6kW left hand drive hardtail with 18 fet Vesc: viewtopic.php?f=6&t=96754

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-100 series motors

Post by ElectricGod » Mar 21 2019 1:59am

district9prawn wrote:
Mar 21 2019 1:15am
ElectricGod wrote:
Mar 18 2019 8:47pm
In conversing with Revolt, they won't tell me anything other than what they post on their web site. I think they really dislike me scrutinizing their motors, but won't say anything. You'd think they would want to brag about their motors...at least the Pro and E...nothing. They get annoyed when I tell them what I have found in the motors. They think their motors are perfect or something as is.
I find their attitude just as concerning as the problems you've found in their motors.

In practice would the single key holding the shaft to the rotor be that big a problem.
Aluminum is a soft metal. It deforms relatively easily. Even 6061 like these motor parts are made from is just a harder aluminum alloy,

1. 100% of the motors torque will be transferred from the aluminum bell top, through the key/keys, to the shaft and then to a sprocket.
2. T8f is all steel chain and capable of holding about 2kw reliably.
3. The amount of energy transfer area one 4mm key like found in the RV-100E is similar to that of t8F chain.
4. By comparison the key and key way in the RV-100E ought to hold about 2kw as well...maybe a bit more.
5. To make aluminum not deform under loading, you give it a lot of surface area to spread out the loading.

You have several mechanical things playing against each other in any motor:
1. 50-80kv motor RPM's translate to wheel RPM's fairly easily with a single stage of gearing.
2. The closer to 1:1 gearing you get, the more stress on the mechanical linkages (keys, keyways, chain, sprockets, etc) there is.
3. Further from 1:1 reduces the loading stresses on the mechanical linkages.
4. Using a 120-200 KV motor significantly reduces the stresses on the shaft to bell keys and keyways, but it means gearing is now 2 stages to get motor RPM's to a usable wheel RPM.
5. Simple, single stage, 3-4:1 gear reduction is compact, reliable, has low frictional losses and easy to maintain.
6. This means that probably most people will be using inrunners and outrunners in the 50-80kv range where the mechanical linkages deal with the highest stress levels and therefore need to be robust.

Revolt says the RV-100E is a 6-11kw outrunner. From Revolt, it has enough metal at the shaft to bell to reliably hold about 2kw. The keyway in the aluminum bell top will be the weak spot and over time will deform and fail. If Revolt used 20mm of the 24mm of length they have available in the bell top, they could have increased that measly 2kw to more like 2500 watts. That extra 5mm would have been zero cost to add. What's more there's a single key and keyway in a place that at a dead minimum needs 2 of them. IMHO, dual 20mm long by 4mm keys would hold up fairly well to 6kw.

I have 2 HLD inrunners good for about 6kw. They use a single 25mm by 4mm key. The shaft, key and sprocket adapter are all hardened steel parts. One of my HLD's has about 5000 miles on it and the other maybe 300 miles. The one with 5000 miles, they keyways are now about 4.5mm wide. The hardened steel has deformed and now there is about .5mm of slop in the keyways. This translates to about 1mm of rotational play between the shaft and sprocket in hardened steel parts. Imagine what similar wattage on a smaller key into soft aluminum will be like in 5000 miles. I bet that keyway in the aluminum bell top will have stretched considerably more than .5mm! I bet it will fail completely well before 5000 miles. As a rock bottom minimum Revolt should have made them 20mm long and used 2 keys.

In none of the Revolt or Alien Power motors I have looked at is the bell to shaft union adequate. In the Alien Power motors 2 4mm set screws on flats on the 12mm shafts will work very short term. I bet on the C80100, you'll get 50-100 miles before this mechanical linkage fails. In the 12090, if it lasts 5 miles, I'd be surprised! In the Revolt motors I have, I bet you will get a few hundred miles before the shaft to bell unions start giving way. In motors from either source, this mechanical union is insufficient for long term reliable use.

I still stand by what I have said many times before. Revolt motors are better than Alien Power motors.

The AP C80100 is a fairly close comparison with the RV-100-regular. Realistic wattage numbers are similar. The C80100 has 2 M4 set screws holding the bell to the shaft. The RV-100-regular has 2 3mm x 8mm keys. Both are inadequate for long term and reliable transfer of power. However, the dual keys in the Revolt motor will easily handle 2X the power of those 2 tiny set screws in the C80100 and do so for far longer time frames. The RV-100 has larger magnets and better copper fill too.

The AP 12090 is similar to the RV-120-regular. They both deliver similar power to their shafts. The Revolt motor has 2 4mm x 8mm keys between the bell and shaft. The 12090 has 2 4mm set screws just like the much smaller C80100. The Revolt motor has a 15mm shaft and the 12090 has a 12mm shaft. Every aspect of this is a win for Revolt over what AP has in their motors. Never mind better copper fill and bigger magnets.

Bigger bearings, larger shafts and other things make the Revolt motors mechanically better than similar AP motors. Better copper fill and bigger magnets help make the Revolt motors better too. The big advantage AP motors have is .3mm lams in all their motors. Revolt in the Regular motors use .5mm lams, but in the current Pro and E series use .3mm lams. In every way that I can see and measure so far, it looks like Revolt is the better option. Perfect...no not at all, but compared to Alien Power, definitely better products.

So despite the languid and complete inability for Revolt to fix anything in any of their motor product lines, I'll continue to use them instead of the Chinese made AP motors. The Chinese manufacturer of AP motors doesn't take feedback either!
Last edited by ElectricGod on Mar 21 2019 3:26am, edited 1 time in total.

larsb   100 kW

100 kW
Posts: 1143
Joined: Dec 10 2014 5:12am
Location: Gothenburg, Sweden

Re: Revolt RV-series motor review and comparisons

Post by larsb » Mar 21 2019 3:19am

The stall torque for the Rv100 motors is not known by me but checking the key size vs the material properties for alu 6061 does not show even close to the shear strength limit at 20 Nm:
key shear.JPG
key shear.JPG (92.67 KiB) Viewed 726 times
The shear limit is found around 75Nm IF only the key takes torque. The Rv120 motors i've used has an interference fit on axle and bell, enough so i had a hard time with a crowbar to adjust the rotation on the shaft to meet the key slot. This interface will share the load. Is this the case also for RV100?

-->I don't think there will be any problems due to the original key interface if only the key and axle fit is good.
Ride on :D

User avatar
fechter   100 GW

100 GW
Posts: 14069
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Revolt RV-series motor review and comparisons

Post by fechter » Mar 21 2019 9:04am

Any pictures of failed keyways?

One thing that would help is to glue the connection together with super Loctite or something that will prevent any movement between parts. Its when you get a little free play things start to hammer and you get deformation. Just high power isn't the real problem but shock loading and peak torque.
"One test is worth a thousand opinions"

User avatar
ElectricGod   100 MW

100 MW
Posts: 2670
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Revolt RV-series motor review and comparisons

Post by ElectricGod » Mar 21 2019 9:41am

fechter wrote:
Mar 21 2019 9:04am
Any pictures of failed keyways?

One thing that would help is to glue the connection together with super Loctite or something that will prevent any movement between parts. Its when you get a little free play things start to hammer and you get deformation. Just high power isn't the real problem but shock loading and peak torque.
That is true...a tiny bit of play...tends to grow...

I have tried blue and red locktite to help hold mechanical interlocking parts together and it didn't really work for long to stop movement. For threads they do a great job. For stopping movement between parts that already have slack and movement in them, typical blue and red didn't work so well. Maybe "super locktite" is something I don't know about? Are you talking about Locktite 277? It's the best of the "red" formulas and I've never tried it.

In the case of my HLD inrunner, the key was a snug fit back some 5000 miles ago between all the parts. I had to press it into place. The key didn't deform in any kind of appreciable way, but the motor shaft and the shaft to sprocket adapter both did. The shaft is hard enough that a file won't touch it. HSS drills will cut it, but they dull quickly.

If that will happen in hardened steel, the problem will be much more significant in aluminum.

That's why I do shaft to bell reinforcing. I don't trust the factory "solutions". IMHO, they are always inadequate.

Failed keyways...yeah...had that happen on that HLD inrunner. The original adapter failed at the keyway. I don't have a picture of the part. It's been a couple of years since that happened.

Post Reply