Leaf / leafmotor / leafbike high efficiency 1500w motor

Kodin said:
Do you think a direct-plug CAv3 can run as-is or will I need an external shunt for the power levels I'm running?
The whole point of the CA-DP model is to use the shunt in the controller. By definition shunt capacity is a non-issue - unless you have a controller without a shunt...

neptronix said:
...make sure to get a kV figure on your motor if you can. A cheap ebay laser digital tachometer is something like $10 on eBay.
If you have a CA, follow the adjustment in the Guide (section "6.9 Displaying Wheel RPM") to set it up as a tach.
(The setup works with both V2's and V3's...).
 
Nep, if you were able to sustain 50mph on a 3% grade, and faster than that on a flat, with no observed heat problem at the time on 72V, since you were having trouble sustaining 45mph in regular use at 48V, why didn't you just run 72V full-time? I'm wanting to know if it was more the convenience factor for you with batteries and charging (as you mentioned 72V is inconvenient for you), or was it more that you feared you'd hurt the motor?

Part of the reason I'm asking is to determine the feasibility of running the 7-turn/361RPM version at 72V for myself, normally at speeds considered slow for this group, and was wondering if that's alright for the motor. 72V might actually work better for me with how much space I have in my frame, and the multistars I already own. I crunched the numbers last night, and at 72V, the 361rpm motor would give me all the speed I'd ever want (~31mph), plus all the torque I'd ever need: no downsides for me personally. And when I run 48V (~25mph tops), I'd be a lot closer to that 100%/peak efficiency speed you mentioned, with that 7T wind. So the only concern is for the motor's sake.. any idea if the motor may not like 72V, or what might be the downside for it? All I've read indicates that slower winds deal better with higher voltages anyway. Considering your torture test, and my slower wind and speeds, I figure it'd be fine, but so many of yous have so much experience, I know I should try to nail this down before buying. Thanks, man. I'd invite anyone else to chime in, too.. esp SpinningMagnets and TommyL.
 
Yeah, amps matter more than volts in this application, and since this motor has the thinner stator laminations, it's gonna have a higher saturation level than a lot of Chinese hubmotors. 72V should be just fine. Nep was running his motor at about 4KW, which caused it to overheat even when vented. You'll have very high torque and very low current draw at 25 MPH with a 7T, so the motor should pretty much just sip on your batteries. With multistars, I'd expect really really good range with that wind. :) Downside is I suspect it'll generate more torque than you can possibly use without a very long/front-heavy/low-to-the-ground bike frame.
 
Nep was actually pushing 6000W through on his 72V tests :shock:
 
...Wait, I thought thinner laminations meant you could push more through a motor before it reaches saturation and starts generating massive amounts of heat... :eek:
 
Kodin said:
...Wait, I thought thinner laminations meant you could push more through a motor before it reaches saturation and starts generating massive amounts of heat... :eek:
Thinner laminations will give lower eddy current losses but, if you're talking about flux saturation in the core, thinner laminations will give a lower stacking factor so, slightly lower saturation, everything else being equal. Does that make sense?
 
Indeed it does! Thanks!
 
Thinner laminations = less eddy current = less heat generated per watt inputted because the motor is spending less of the inputted power fighting itself.
The higher the RPM, the more eddy current.. so you can spin up this motor without taking as big of a penalty as you do with a 0.5mm lamination motor. But just think of the efficiency of motors like these. 80% vs. 90% means that the 90% efficient motor is generating half of the heat as the 80% efficient motor, literally. Your saturation point and continuous wattage handling for the same design changes in your favor.

As for my 72v test, the motor wanted to do 60mph and would sip 5,500w-6,000w or so while achieving 50mph up a hill, but the hills i tested were short compared to my route. On a much higher turn count on the same voltage, it will do.. what.. 35mph? and only sip maybe 3,000w going up the same hill.
The heat generated by doing a low turn count motor on the same voltage doing the same thing is gonna be MUCH lower. So is the speed.

And my tests were done at about 65F, and i went up relatively short, but very steep hills. When it hit 80F-90F here in Utah, the situation with the motor staying cool changed very quickly. Also, i started using 120% mode almost everywhere. End result is that i average 3000W+ for long periods of time, climbing long hills repeatedly on the way home. These are 20+ mile rides where i am totally flogging the motor about 75% of the time. Of course it's reaching the critical heat point.

But where i live is very demanding. I ran a MAC 8T on 36v for 4 years everywhere else i lived, and got away with it.. nowadays, i live about 500ft. higher than downtown and about 15 miles away. Most of my route has almost no bike lanes and speed limits of 40-45mph on the road. You should only worry about what i'm saying if you have my conditions.

Let's revisit the dyno sheet, shall we? because we know that this is indicative of what happens in real life, so it's useful data:

leafmotor.jpg


30mph = 1000W input point ( ~90.5% efficient ).
40mph = 2000W input point ( ~88% efficient ).
45mph = 3000W input point ( ~81.3% efficient ).

Now, add a big, long hill to any of these figures and your efficiency is gonna go way down. Add high levels of ambient heat, and your air-cooling is also going to be less effective. That heat that i build up while flogging the motor is dissipated at a slower rate.

A crystalyte 35mm motor would be producing basically double the heat, just because of the laminations. It could not survive this torture in the slightest. That's why anyone who buys one today is kind of a fool to be honest :mrgreen:
 
neptronix said:
Your saturation point and continuous wattage handling for the same design changes in your favor.
Well, I don't see how the saturation point is affected. Why complicate matters? Thinner laminations mean lower eddy current losses.... that's it.......
 
Mine is a 5T, with temp sensor, cassette freehub covers, and 3mm^2 phase wires.

Just ordered myself a Sabvoton SSC48150 off Aliexpress. I might be able to spin the motor in a couple weeks! :D

Judging by that the controller is "150A max" but rated for 4KW, I assume constant current would max out at ~80A. If not, oh well, I over-specced my controller. :p
 
Nep, some time ago you said that if you didn't need to go over 30mph, you'd still be running a MAC. However, over time, it seemed your affection for this new motor progressed. I know you are usually pretty enthusiastic when you've got a new motor, but to me this one seemed more than before. (I remember you saying in a video, "Everyone should have one".) In light of this, I wanted to ask you if that's changed or, if, for whatever reason, you just couldn't or didn't want to ride over 30, but still had the same steep hills, which would you buy today?


neptronix said:
You'd probably want a more expensive controller that has torque throttle functionality instead of the speed control that infineon-like controllers and many others use. Go drop $300 on the experiment and buy a controller that's capable of 100v max so that you could potentially adjust your voltage up or down 2 cells to make it hit the correct 100% speed you're looking for.
Can someone give me a tip on name brands/models to research? I've never heard of a controller like this. I don't see really obvious or clearly searchable terms here. Pricey, but I'd at least like to know.

Lately, I've been thinking maybe I should just get the 4110 40A controller, not 60, if I go with the 7T wind. If there will be as much torque as people predict, I don't see what the point of the extra 1000W would be? AFAIK, it wouldn't help me go faster. And if the motor's so torquey, I don't see how it would help me get up hills better, either. I once watched a trail video of a guy with a BBS02. He was in his lowest gear, and towards the top of the fairly long, sloping hill, the bike was still bogging down a bit. Not wanting a BBS02 anyway, this motor seems like the best way to avoid that scenario. I also extrapolated from the chart Neptronix posted, that if 30mph is peak efficiency for a 4T on a 26-inch, peak efficiency on the 361rpm/7T on a 29-inch would be around 19mph. Which, if I could choose, I'd drop the sweet spot even a bit lower, but it's a hell of a lot closer to what I'll typically be doing.
 
You can actually get "torque control" (a.k.a. "Current throttle") on an infineon controller by using a CA V3. You would just need to set up the CA for the appropriate closed-loop throttle mode.

But I think what Nep was referring to are the more advanced sine wave controllers, like the Sabvoton and adaptto controllers, which have this type of throttle control built-in.
 
Kelly controllers also support torque-based throttle. Cheap/basic controllers react to throttle input by changing speed. The "torque" throttle style instead tells the controller how much torque to push the motor at, much like a motorcycle reacts to throttle input.
 
SprocketLocket said:
Nep, some time ago you said that if you didn't need to go over 30mph, you'd still be running a MAC. However, over time, it seemed your affection for this new motor progressed. I know you are usually pretty enthusiastic when you've got a new motor, but to me this one seemed more than before. (I remember you saying in a video, "Everyone should have one".) In light of this, I wanted to ask you if that's changed or, if, for whatever reason, you just couldn't or didn't want to ride over 30, but still had the same steep hills, which would you buy today?

If i didn't have the same steep hills and only speeds up to 35 to deal with ( bike lane, or low speed limits ), my #1 choice would be a MAC due to it's lightness. I'd take a hit in efficiency, but having a 10lb hub on a dual suspension bike would rock. ( 16lb. of unsprung weight does hurt ya there ).

If i had the same steep hills but didn't need >30mph, i might actually go with the newer 9C, MXUS or leafmotor 30mm 0.35mm lamination models and save a bit of weight. Or just keep the 1500W leafie if i still had it.
 
Question for those of you who have ordered the Cassette spline version

How many cogs does the cassette take in a 135mm(standard) frame?

What are your thoughts on the quailty of the cassette?

What torque arms(commercial) are people using on thier frames?

thanks
 
I think I saw someone say here that the axle on the motor was 17mm with the 3mm phase wires (which now appear to be standard?). Is that right? While it's great, I only have 12mm and 14mm torque arms available. I'm not seeing the axle dimension on their diagram.
 
12mm & 14mm are the current bicycle standard width. I have a hard time believing that a Chinese motor company is trying to change the norm. It would mean many many folks would have to modify their existing bicycle's dropout to fit.

Maybe a 17mm axle are the for moped & light motorcycle category?
 
neptronix said:
The axle is 14mm wide, so you can use a 14mm car oil pan drain plug washer to gain that additional space, no problem. Easily found on eBay:

Found this on page 2 of this thread.

This obviously refers to the standard axle. I think re reading there are actually 2 other axles.
One comes with 3mm wires and the other comes with 3mm wires and cassette spline. No idea what the diameter of these are though.

I hope they are 14mm because the only grin tech options are 12 or 14mm.
 
Kodin said:
Technically almost 17mm; they had to change to a larger shaft, left side cover, and bearings to admit the 3mm phase wires. ...Which come sheathed in a nice bundle other than the temp sensor wire. I'm assuming the other end of the temp sensor goes through ground or something... Not sure honestly as I haven't bothered to test anything yet. Probably will hook it up to my 24-36V 20A Kelly to make sure it spins when I get back home this weekend. Covers have plenty of space for appropriate seals to be installed too. I'll have to see what I can source once I receive the replacement shaft and freehub cover I'm trying to order. Here's hoping I can easily seal the freehub side so it doesn't spew ATF everywhere... Otherwise I may have to choose between cooling and a proper cassette.

Did you just use a cheapo crappy freewheel to remove the freewheel side?

I think this is where I got the 17mm axle width from.
 
I'll confirm both shaft sizes with thread diameters tonight guys. Also the cassette should take a 9 speed cassette... I'll check tonight. When I convert the motor to cassette I'll post pics of my conversion as well. And maybe start my build thread finally. :roll:

Also to clarify the 17mm section is only through the center. It necks down smaller at about the 125-130mm mark.
 
neptronix said:
If i had the same steep hills but didn't need >30mph, i might actually go with the newer 9C, MXUS or leafmotor 30mm 0.35mm lamination models and save a bit of weight. Or just keep the 1500W leafie if i still had it.
Thanks for the answer. I didn't know about the new 9C. Looks like you're referring to the M3007RC and similar, as sold by ebikes.ca?

The MXUS I've seen referred-to the most is the "mini cro" ("V3"?), which you must not be referring to. MXUS's website only has one DD hub motor listed, it's an 83% 1500W, so that must not be it either. So, curious which you meant.

Interesting that you'd consider the 1000W LeafBike, as as you pointed out, it's only about 2 pounds lighter. That's kind of why I kept just looking at the 1500W. I imagined the wide diameter of the 1500W, though sucky, does create more torque, kind of like I learned vicariously with your Magic Pie 2. I figured the 1000W is a "little" stealthier (more like "less-un-stealthy"), but for 2 pounds the overall drop in power might not be worth it. Maybe I should reconsider.

The bottom line for me is: I wonder if any of these above motors you mentioned ship with a wide axle, or have one available. RideTheLine has a nice writeupabout the M3007 from ebikes.ca, and it's definitely standard width
http://endless-sphere.com/forums/viewtopic.php?f=2&t=68922
Don't know about 1000W LeafBike axle, probably standard, but they seems pretty willing to work w/people to customize either way, so that could be a possibility, not sure it's worth it.
 
Okay, so I've been studying the graphs for the 1000W and 1500W. I think I am missing something when I read the graphical charts, because what I interpret from it, is seems different than what I read in their tables below the graphs.

For the 1000W motor, when I look at the graphic, it looks to me like the motor begins to approach saturation at around 19 amps or about 900W. You can see on the bottom graph how the black line [power out] starts to flatline at around 1000W when power in hits around 1150w. When I read this graph my impression is that if people give it more than 1200W, they're really just dumping the extra electricity as heat, but I don't think this is the case. On the graph version, efficiency (to my reading) appears to start to get to 80% efficiency somewhere around 280W. This 80+% band seems to me to last through the rest of its curve. On the table though, 80% efficiency doesn't start till (it's hard to tell, but I feel more confident in this) around 525W, then lasts on through the 1200 listed watts at above 80%. With the 1500W, looking at the table, I interpret 80% being reached around 466W, and lasting through about 3100W, and doesn't drop below 75% efficiency until nearly 4000W(!).

Just sticking to the tables now, where I'm more confident, both motors appear to hit 80% efficiency basically around the same point, but the 1500W gets there sooner (~466W vs ~525W). Both reach peak/90% effiency around 1000W, but the 1500W keeps it in this high-80s range for far more watts--till about 2000W. It's disappointing the 1000W test stops at 1200W, while it still is in the 80% range. Of slight note, the 1000W is done at 50V, the 1500W @48V.

This doesn't make sense to me: when I look at the Torque columns of both motors' tables, the 1000W version shows around 12 Newton-meters (I think that's their unit, remember I'm new to this), while the 1500W is shown to be producing about 16--of what I assume to be Nm, doesn't say, at 86%/90% efficiency, respectively. So if 16 is Nm, how can 1000W on one be producing 12, and 1000W on another be producing 16, and yet they're still close to the same efficiency at that point? Doesn't make sense to me.

So a lot of you guys are way better at interpreting these charts. I know I've biffed here, so anyone feel free to clean up after my attempt. That didn't feel very fruitful to me. The only thing I feel confident about my reading is that the larger motor does seem to come out stronger in almost every sense, but in some places more than others, most clearly in power capacity, but there is still a tradeoff in other ways. Thoughts welcome.
 
Back
Top