Leaf / leafmotor / leafbike high efficiency 1500w motor

If you're really keen, this is the original test I did on this back in 2015.
[youtube]1S0WSt_dZWk[/youtube]

Cheers

I intend to modify mine in a similar manner at some point in the future. How much extra weight and pedaling difficulty did the cooling fans, ferrofluid, heatsinks, and vents add while riding unpowered, if it was even noticable at all?

What voltage can these be run at before any arcing over occurs? If one could get away with running one of these at 130V, 10+ kW peak may be a possibility with some of the new controllers out.

john61ct said:

What does Grin rate the Phaserunner for continuous motor amps?

Found it:

The use of extra finned heatsinks thermally linked to the Phaserunner base can increase this continuous phase current to nearly 70 amps, while the current limit if placed inside an insulated cavity or bag with no airflow at all will be much lower, around 30A.

Click to expand...

Yeah i'm referring to battery current, not phase current.

70A phase current is very low for a powerful motor. On my 4T leaf run, i was running 180A+ phase current and 80A battery current.

Cowardlyduck said:

Yeah, if you're really keen the '1500W' Leafmotor can take a lot. I pushed over 8KW peak through mine and regularly held it at 4-5KW continuous for long periods up hills.

Click to expand...

I'm really impressed that you found the maximum of what the motor could take. I was limited by a short wheelbase that threatened to throw me off the bike at 6kw.. got video of your crazy power levels?

The Toecutter said:

I intend to modify mine in a similar manner at some point in the future. How much extra weight and pedaling difficulty did the cooling fans, ferrofluid, heatsinks, and vents add while riding unpowered, if it was even noticable at all?

Click to expand...

Only thing that really impacts pedalling is the FF. The more you add the worse it gets. I added around 10ml from memory and it was barely noticeable.

neptronix said:

I'm really impressed that you found the maximum of what the motor could take. I was limited by a short wheelbase that threatened to throw me off the bike at 6kw.. got video of your crazy power levels?

Click to expand...

Sure, any of my later ride videos on my Fighter show what it was like. I sold it since though.
[youtube]2td41WXB6xM[/youtube]
Cheers

Nice video Cowardlyduck, good to see people really using their bikes instead of only tinkering with it :wink:

I'm planning to make a new battery with max 213 cells. Based on your experience with te leaf what would you recommend for only offroad: 16S13P, 17S12P or 18S12P? For me torque is more important then speed...

raggertje said:

Nice video Cowardlyduck, good to see people really using their bikes instead of only tinkering with it :wink:

I'm planning to make a new battery with max 213 cells. Based on your experience with te leaf what would you recommend for only offroad: 16S13P, 17S12P or 18S12P? For me torque is more important then speed...

Click to expand...

Thanks!
If your after more torque, then decrease your wheel size and increase your phase amps. If you want less voltage sag, and/or more range go with the higher parallel cell count...so 16S13P.
If you get the standard 4T leaf and run it in a 26" rim you should get over 70kph at 16S. That is more than enough speed IMO. If you can use a 24" rim or smaller and you will get heaps more torque.
To give you an idea, my Fighter ran a 24" rim (with 3" tires) and my battery was 14S14P (~2.5Kwh) of Sanyo GA cells. Without field weakening top speed was around 60kph. With OVS 7 set on my Adaptto I could get over 100Kph, but that was insane and I only tried it a few times.
Whatever you do, don't get a 'high torque' motor and cheap controller. Better to just get the standard 4T and a better controller like the Nucular or similar.

Hope that helps.

Cheers

Cowardlyduck said:

If you get the standard 4T leaf and run it in a 26" rim you should get over 70kph at 16S. That is more than enough speed IMO.

Click to expand...

I'm getting 75 kph out of a 13S3P pack of Sanyo/Panasonic NCR18650GA cells. But mine has greatly reduced aero drag plus the aid of strong legs and a wide gearing range.

A 20S6P pack of the same cells is going to be fun...

Nice build, BTW. The amount of power AND longevity you've gotten out of that motor is nothing short of impressive. It looks like the two of you had a great time together.

Cowardlyduck said:

If your after more torque, then decrease your wheel size and increase your phase amps. If you want less voltage sag, and/or more range go with the higher parallel cell count...so 16S13P.
Whatever you do, don't get a 'high torque' motor and cheap controller. Better to just get the standard 4T and a better controller like the Nucular or similar.

Click to expand...

Thanks! That helps a lot I didn't know about the voltage sag vs. parallel cell count so that is clear and I better go for a 16S then... I did order the Nucular 12F but in the meantime I ditched the 22amp KT controller and setting up a Kelly 7218S to get the amps flowing

I lost an axle nut today. I have the cassette version and while executing regen, an axle nut on the drive side came loose along with the oblong shaped washer. I couldn't find them on the road and they may have rolled off into a ditch or gutter. Does anyone know where I can source replacements? I went to a hardware store and have a 12mm fine thread nut and normal circular washer in their place, but I think that is only a temporary solution.

Here is the build:

https://endless-sphere.com/forums/viewtopic.php?f=6&t=110298

If the axle nut came off during regen, then it was already very loose, ready to fall off.

This usually means, assuming it was tightened enough to start with, that the axle is rocking back and forth in the dropouts, even just a tiny bit, which eventually works the nut loose, and then vibration can unscrew it further depending on circumstances.

If it isn't the above, then something else is seriously wrong for a single regen use to eject the axle nut.


I would recommend first setting up either clamping dropouts / torque plates, or using sufficiently-precise-hole torque arms that they cant' move on the axle, and then snugging up the left side arm clockwise and the right side counterclockwise, so that any tiny slack they do have is now cancelled out, so the axle can't rotate at all.

Then use something like Nordlock ramped lockwashers under the nut to ensure the nut isn't going to back off from other circumstances (vibration, temperature changes, etc).


For the replacement nut, as long as the threads match your axle, and it's hard enough metal to not be deformed when tightened, then it's probably better quality than what came on the motor axle to start with. The threads on the axle itself are the limiting factor on tightness, usually, becuase they are discontinuous (because of the axle flats), and they may not be cut precisely (not really matching any threads on any nut, too shallow or too deep, etc.).


For the washer, a round washer works fine, unless you are looking for a torque washer with the little "tongue" that is bent to stick into the dropout...I've found TWs to be useless for much more than a couple hundred watt systems and then only if the bike dropouts are hard metal (cheap bikes can have their dropouts bent or spread over time even by those systems!).

If you are looking for the flat washer with the axle-shaped hole, I don't see a need for it on the outside--it is very useful on the inboard side of the dropout, as a way to help spread some of the load from the very narrow axle shoulder against the dropout face, but it only works (to keep everything tight) if the washer is hard enough metal to not be indented and deformed by the axle shoulder face pressing on it against the dropout.

A round-hole washer will do the same job of spreading the nut's force across the face of the dropout as one that has an axle-shaped hole, and keeping the nut from damaging the dropout face as it's turned.

by amberwolf » Feb 09 2021 12:38am

If the axle nut came off during regen, then it was already very loose, ready to fall off.

This usually means, assuming it was tightened enough to start with, that the axle is rocking back and forth in the dropouts, even just a tiny bit, which eventually works the nut loose, and then vibration can unscrew it further depending on circumstances.

Click to expand...

Had the same issue when enabled regen, torqued my axel nuts, run two grin's torque arms set up 90 deg apart but opposing directions, then lower my regen and ramp it up with a throttle. Have not had anymore problems in over 2000 miles. The smallest movement will back off the nuts. Hitting full regen from power their is some twist. I would like clamping dropouts that fix it.

amberwolf said:

If the axle nut came off during regen, then it was already very loose, ready to fall off.

Click to expand...

Likely.

This usually means, assuming it was tightened enough to start with, that the axle is rocking back and forth in the dropouts, even just a tiny bit, which eventually works the nut loose, and then vibration can unscrew it further depending on circumstances.

Click to expand...

I installed both sides as tight as I could get them.

I would recommend first setting up either clamping dropouts / torque plates, or using sufficiently-precise-hole torque arms that they cant' move on the axle, and then snugging up the left side arm clockwise and the right side counterclockwise, so that any tiny slack they do have is now cancelled out, so the axle can't rotate at all.

Click to expand...

I have a torque plate installed on the non-drive side that was not yet modified for clamping but has the potential to be drilled for such. I will have to do that when I get the chance. On the drive side I have a Grin TorqueArm Kit v4.

Then use something like Nordlock ramped lockwashers under the nut to ensure the nut isn't going to back off from other circumstances (vibration, temperature changes, etc).

Click to expand...

Good recommendation.

For the replacement nut, as long as the threads match your axle, and it's hard enough metal to not be deformed when tightened, then it's probably better quality than what came on the motor axle to start with. The threads on the axle itself are the limiting factor on tightness, usually, becuase they are discontinuous (because of the axle flats), and they may not be cut precisely (not really matching any threads on any nut, too shallow or too deep, etc.).

Click to expand...

That may in part explain why it came loose...

For the washer, a round washer works fine, unless you are looking for a torque washer with the little "tongue" that is bent to stick into the dropout...I've found TWs to be useless for much more than a couple hundred watt systems and then only if the bike dropouts are hard metal (cheap bikes can have their dropouts bent or spread over time even by those systems!).

Click to expand...

My frame is carbon steel, including the dropouts.

The Toecutter said:

I lost an axle nut today. I have the cassette version and while executing regen, an axle nut on the drive side came loose

Click to expand...

Yes, that's regenerative braking all right.

There's an analogous problem with crank arm bolts on fixed gear bikes. Because torque can be transmitted both ways through a fixed gear, the left arm budges ever so slightly back and forth on its taper, working the bolt loose. Once the bolt is fully loose or missing, the same tiny movements gradually walk the crank arm off its taper until the hole becomes wallowed out and the arm ruined.

In both cases, I find the best remedy in practice is not to do that. If your reaction torque is only in one direction, then it doesn't matter so much whether the torque arm fit has zero clearance. It simply takes up however much play it has-- once-- and stays there.

And that's why we use tight fitting, meaty torque arms on a motor like this.

Here are the torque arms i made for the bike i ran 6kw + 1kw regen on. I used ebikes.ca torque arms as a starter, then added hardware:





The cassette version of this motor has a smaller axle on one side, correct? if so, it's even more critical to use tight fitting torque arms!

I would make your torque arm even stronger than this. You can see i had a long aluminum bar in one pic. A shorter distance would have been preferable. But the thing held up to ~100ft-lbs of torque.. the bolt holes were all drilled and then hand filed to ensure as close to zero play as possible before the assembly was bolted together.

neptronix said:

The cassette version of this motor has a smaller axle on one side, correct?

Click to expand...

Correct.

if so, it's even more critical to use tight fitting torque arms!

I would make your torque arm even stronger than this. You can see i had a long aluminum bar in one pic. A shorter distance would have been preferable. But the thing held up to ~100ft-lbs of torque.. the bolt holes were all drilled and then hand filed to ensure as close to zero play as possible before the assembly was bolted together.

Click to expand...

Part of my problem is that before I installed the motor, the dropouts were not perfectly aligned. This caused my rear wheel to be installed at a slight angle. The non-drive side torquearm is this:

https://endless-sphere.com/forums/viewtopic.php?f=31&t=69073

Another part of the problem is that on the drive side axle, I can barely fit what I have. It is extremely difficult to tighten a nut onto a washer because there is almost insufficient axle threading to get the nut on. It is only when it wants to cooperate and tighten down that I have enough axle length to fit the nut, washer, and torquearm on the drive/cassette side.

I'm definitely going to have to change my rear tire out ASAP too. The sidewall is down to the blue from it rubbing the wheelwell after it came loose. It's a shame too, because the tire still has good tread, even at 22,000 miles. Schwalbe Marathon Plus Tour 26x1.5".

The Toecutter said:

Another part of the problem is that on the drive side axle, I can barely fit what I have. It is extremely difficult to tighten a nut onto a washer because there is almost insufficient axle threading to get the nut on. It is only when it wants to cooperate and tighten down that I have enough axle length to fit the nut, washer, and torquearm on the drive/cassette side.

Click to expand...

Ifyou mean that the dropouts are spread far enough that they have to be "squished together" in order to put the hardware on...my first recomendation is to, with the wheel out, squeeze them together with an old threaded axle and it's regular hardware, until it reaches the point that it's "permanently" bent them inward as far as you need them to. That will take experimentation and repeated applications, most likely, releasing tension and checking distance, retensioning, etc.

A bar clamp could also be used for this, with the bar inserted in the dropout gap to keep it lined up as well as possible during the procedure, and use pairs of blocks of wood screwed to each other on either side of the stays/dropouts on both sides, to give a place to clamp to.


If you don't want to or can't do that, you could still use the barclamp and blocks to squeeze the stays together and force the dropouts against the motor axle shoulders, so that it would be then easy to install the axle hardware.

Leaf motors dont have long enough axles, its the same on my Leaf 1500W from Leafbike. I have used a cut off wheel on a grinder to shorten the width of the axle nut. My old MXUS 45H 3kw had a very long axle, had like half an inch past the nut.

That's not necessarily true...I ran a Leaf on my Fighter with it's 165mm shoulders. How....with a custom made axle I got Leaf motor's to put on mine before I ordered it. At the time (2015) they told me they did it by ordering a 'sample' from their axle manufacturer.
Total width was 190mm from memory.
So it is possible....but no idea if they would still do that these days.

Cheers

I think the replacement nut stripped my threading on the cassette side of the axle. The odd thing is that it did not show any indication of cross threading when I installed it with a small crescent wrench and took very little force to turn. I was mostly able to turn it by hand. The nut was still tight when I attempted to loosen it, but did not require much force to loosen.

Does anyone know the precise type of nut that the cassette version of this motor uses on the cassette side? Or how to replace the axle or restore axle threading? I'd like to not destroy it beyond repair. The threading is sort of still usable as it is, but there's not much left.

The Toecutter said:

I think the replacement nut stripped my threading on the cassette side of the axle. The odd thing is that it did not show any indication of cross threading when I installed it with a small crescent wrench and took very little force to turn. I was mostly able to turn it by hand. The nut was still tight when I attempted to loosen it, but did not require much force to loosen.

Does anyone know the precise type of nut that the cassette version of this motor uses on the cassette side? Or how to replace the axle or restore axle threading? I'd like to not destroy it beyond repair. The threading is sort of still usable as it is, but there's not much left.

Click to expand...

I had a similar thing happen to me years ago with a Crystalyte HS4080 motor. It used 1/2" UNC threads and I tried my damdest to make them work...till the day my rear wheel FELL OFF during a ride while regen braking. Managed to get home using duct tape, zip ties and a tow strap with a mate. That was a funny day.
Anyway, I 'fixed' it by using an M12 die with close as I could manage thread pitch. This worked better than I could have hoped and I ran that motor on and off for several years on my Fighter like that before recently selling it with the bike.
Obviously M12 is not ideal for high power/torque, so in my case I had plenty of other mitigations to prevent axle issues after doing that.

In your case, assuming you have a Leaf with the standard M14 threads, maybe M13 or 1/2" UNC would work? Failing that M12 would definitely work, but may be a challenge to cut depending on how worn out your existing threads are.

Wish you luck.

Cheers

I think the cassette side already has an M12 compatible pattern.

I'm using an M12 nut on the cassette side, but I'm going to upgrade it to an automotive grade M12-1.50 wheel nut.

I received the following email from Leafbike:

if your motor from our company.
We have washer and nuts to replace. (Put 2 nuts per side,will not loose)

Our 1500w rear cassette version is fit for min 135mm ~ 155mm rear dropout. (Very good fit for 140mm dropout.)
Total length of this shaft is 220mm

One shaft is M14 nut
Cassette side shaft is M12 nut.

Click to expand...

A big part of my issue is insufficient axle length to fit everything. Now I can't even get a washer to fit underneath the replacement nut I have because the torquearm is taking up too much available space on the axle.

While I reprogrammed the controller and computer to make the regen more gradual and with less peak power, I'm currently avoiding the use of regen for the time being. I'd really like to be able to use it. With all of the hills where I live, it extends range an extra 5-10%.

When money permits, I may just have to order a new motor with a longer axle length, probably the freewheel version. I ordered the cassette version because at the time I could not readily find 7 speed freewheels with an 11T small cog, and it appeared as if the only one on the market, the DNP Epoch, was being discontinued and was commonly cited as "unavailable" across various retail websites that normally carried them. But now it appears the DNP Epoch are widely available once again.

If I go for a 1,000W freewheel version, which is available in a 3T wind, I could continue using 46.8V nominal without overvolting my torque sensor or having to run the torque sensor off of a separate 12V system, while getting a top speed of about 55-60 mph in a 26" rear wheel. I do not know when money will permit to do this. I had to pass on Neptronix' sale of a nice 20" version of this motor for cheap because there's been a death in my family and funeral costs were a worry...

I'm on the verge of scoring a decent paying but temporary engineering job, but I need this velomobile to stay operational as it is my transportation, and I will be using it on the job to go to sites and collect data and getting reimbursed for mileage at the IRS rate. The motor is key to allowing car-like average speeds in the city which will enable me to do this, but without the motor, it won't be doable. *If* this works out, I'll be set...

Toecutter, I meant to say that I use the DNP Epoch 7-speed 11-28 freewheel linked here on all three of my bikes, and they've been good to me. My legs aren't as strong as yours, though.

The Toecutter said:

I ordered the cassette version because at the time I could not readily find 7 speed freewheels with an 11T small cog, and it appeared as if the only one on the market, the DNP Epoch, was being discontinued

Click to expand...

DNP freewheels are pretty junky and failure prone.

There's another concern when setting up gearing on a rear hub motor, which is the tendency of the freewheel thread to break out of the sidecover when pedaled too exuberantly. There best way to mitigate this problem is to use the largest sprockets that will give you the ratios you want, thereby lowering the chain's operating tension. So rather than using an 11 rear sprocket to get a tall gear, you'd bump up the chainring size. 66/14 is the same ratio as 52/11, but 79% as much chain tension for the same power.

Torque at the freewheel thread would be the same given the same gearing ratio, even if the chain tension is reduced. Using the lowest gear up a steep hill would be hardest on those threads of course.

I've heard the DNP freewheels are failure prone, so I try to be mindful of how hard I push 'em