Large diameter direct drive hub motors

[pwd]

Is there a direct drive hub motor that has a larger diameter than the Magic Pie 2?

My Pie is working well but I've been trying to find out if there is a worthy upgrade when the time presents itself. What I am looking for is a similar dd hub motor but with greater efficiency and torque around the same weight. I also need it to fit 135MM dropouts with a 7 speed freewheel. I love the low-end toque of the Magic Pie, but I'm wondering if there is anything modern out there that is even better. My system is limited to 52V 40A and that is about the max current I can draw from my battery. I understand that the large diameter of the motor is a major plus for low end torque but I can't seem to find motors with a similar diameter. Also, with the MP stator being only 18MM; is there a similar diameter motor with wider magnets (18-35)? I do realize that a wider motor will have weight costs.

I have been reading through the leafbike 1500W thread, but it seems that motor doesn't produce as much torque when given the same power. I wish Grin's ebike simulator has the Magic Pie on there I've also considered the Crystalyte 35XX series.

Does anyone have any hub motor recommendations?

 

[kneedeep]

I also run the Pie but there are plenty of motors out there and in its range of power the Pie is probably the heaviest. Check out the QS motor site they run fine at 48volts. I think that will be my next build staying at 48volts.

 

[pwd]

I just had a look at the QS motor:
http://www.cnqsmotor.com/en/article_read/1000W%20205%2035H%20V1%20135mm%20dropout%207%20speed%20freewheel%20electric%20bike%20bicycle%20brushelss%20dc%20spoke%20hub%20motor%20for%20bicycle/613.html

it looks like a pretty solid motor; however I'm not convinced it will have the same low end torque as the Magic Pie. I'm looking at the QS datasheet @ 48V 50A (second last one), but to be honest, I'm sure sure how to compare it with the Magic Pie's datasheet:

https://www.goldenmotor.com/magicpie/MP-performance%20data%2048V.pdf
https://www.goldenmotor.com/magicpie/MP-performance%20curve%2048V.pdf

Here is what I am comparing (I would appreciate some advice on this comparison):

Magic Pie
Nm 27.66 RPM 279 WattsOut: 808.16 Voltage: 47.85 Amps: 23.45 Watts In: 1122.09 Efficiency:72.0

QS Motor
Nm 84.7 RPM 56.4 WattsOut: 500.7 Voltage: 47.84 Amps: 37.94 Watts In: 1815 Efficiency: 27.5

Those are the max toque figures, but the rpms and watts in are quite different. How do I compare these for low end toque when the rpm/power parameters are different?

 

[kneedeep]

If you email them from the site tell them what you have and what you would like and they will give you there best setup. They can give different setups of the same motor to match your needs. I want the same and emailed them 3 weeks ago.

Qsmotor "I think what you need is low speed type but with high torque, so we can make the motor with high torque winding.

If you have any other questions, please inform."

The only issue is I inquired about 150mm drop out. But if you ask them they will tell you strait up what they can do for you.

 

[neptronix]

I have been reading through the leafbike 1500W thread, but it seems that motor doesn't produce as much torque when given the same power. I wish Grin's ebike simulator has the Magic Pie on there I've also considered the Crystalyte 35XX series.

I've owned both motors and pushed both to their limits. The magic pie doesn't compare to the leaf.. if you are seeing higher torque with the magic pie, you might have your parameters set wrong, or you're looking at peak torque instead of overall.

The leafmotor in the ebikes.ca simulator is the 5T winding and requires way more amps to produce torque than volts. You need to set the custom controller to like 90A if you haven't tried that already.

BTW, for efficiency, you don't wanna go tall, you wanna go wide. The magic pie's achille's heel is the end turn losses. It's something like a 19mm stator, so there is a higher ratio of the copper at the end of the motor doing less work, than say, a wide motor with the same amount of copper at the end being semi-useless.

Per the leafmotor dyno, that motor peaks at 90.5%. I've found the dyno to be true. A MXUS 3kW, cromotor, or MXUS 5kW might peak higher, if the windings are as tight on those motors.. but we haven't got proper dyno sheets for those motors, so who knows.

That QS motor you're looking at is basically a heavier and less efficient leaf.

 

[neptronix]

I also run the Pie but there are plenty of motors out there and in its range of power the Pie is probably the heaviest. Check out the QS motor site they run fine at 48volts. I think that will be my next build staying at 48volts.

The pie, crystalyte, and leaf are about the same weight, give or take 0.2kg.

The QS? now that's a pig.. actually, all their motors are much heavier than they should be. I wonder if they are using higher amounts of lower quality materials somewhere. Haven't found a single product in their lineup that enticed me to buy it.

 

[neptronix]

One last thing.. in a lot of these dynos, they do not push the motor as hard as they should. You never get to see the saturation point on high voltage, which is the actual maximum torque the motor can output.

Also, with motors of differing KVs, you're gonna be doing a skewed comparison. The leaf motor dyno on leafbike's website is the 4T winding, and does 40mph on 48v. The magic pie does 20mph on 36V. Golden motor does not dyno their motor at an equivalent voltage. Consider that the torque of the leafmotor is happening at about double the RPM. This also skews the results.

The newer magic pie with 0.35mm lams will probably have a torque advantage over the leaf at low RPMs because of the higher pole count, however, as the speed starts climbing, the high pole count starts to become an efficiency ( and thus torque per watt inputted ) disadvantage. The continuous power will be lower.

I've been looking for something better than the leaf motor since i discovered it in 2014. Nothing interesting since.. manufacturers aren't willing to play with 0.27mm or 0.3mm lams yet, which is a shame because we could have a motor in the same weight and form factor that produces a bit over 2000 watts continuous and peaks at ~92% efficiency, actually.

 

[pwd]

Thanks a bunch neptronix !

The leafmotor in the ebikes.ca simulator is the 5T winding and requires way more amps to produce torque than volts. You need to set the custom controller to like 90A if you haven't tried that already.

My system can only supply around 2000W (52v X 40A), in this case, do you think the Leaf will still have more torque than the Pie at this level?

Is this the correct dyno for the leaf you are referring to ?
http://www.leafbike.com/u_file/images/15_01_05/0a32577ead.jpg

The newer magic pie with 0.35mm lams will probably have a torque advantage over the leaf at low RPMs because of the higher pole count, however, as the speed starts climbing, the high pole count starts to become an efficiency ( and thus torque per watt inputted ) disadvantage. The continuous power will be lower.

I didn't realize the newer Pies were using 0.35mm laminations vs 0.5mm - I'll have to keep my eye out for one of them if they will supply more low end torque than other dd hubs at my 2000W level.

As a general question (I realize I have some homework to do), will increasing the max phase amps increase the the power draw from the battery?

 

[neptronix]

Is this the correct dyno for the leaf you are referring to ?
http://www.leafbike.com/u_file/images/15_01_05/0a32577ead.jpg

Yessir. That's the 4T.

I didn't realize the newer Pies were using 0.35mm laminations vs 0.5mm - I'll have to keep my eye out for one of them if they will supply more low end torque than other dd hubs at my 2000W level.

As a general question (I realize I have some homework to do), will increasing the max phase amps increase the the power draw from the battery?

It all depends on what you want to do. If you want a large hub that goes slow, the new pie might output more initial torque in the first hundred RPM. However, even with the fast wound 4T, which many would consider 'low torque', i had great troubles with keeping the front wheel on the ground at 80 amps and 48v. I had to progressively tune the phase amps down to a ratio of 2.15:1 phase:battery on my infineon clone controller in order to avoid doing a power wheelie.

More than enough!

On 72 volts, i had to tune the battery amps down to 65 amps and just barely nurse the throttle while leaning all the body weight i can on the front of the bike in order to avoid instantly flipping over. I have handlebars that go very far forward.

You can see me slowly accelerating on 72v here.. only because physics were the limiting factor..

[youtube]TmoRNMSzHjY[/youtube]

So unless you have a particularly long wheelbase bike, or you weigh above 250 pounds, i think the leaf will produce enough torque for you. The advantage of the leaf's overall higher efficiency is that it will actually make more torque across the powerband than the magic pie.. i can almost guarantee it.. but i don't have a dyno graph or experience with the 0.35mm leaf to make a definitive statement. But i really doubt that, with the extra end turn losses from the thin stator design, that the new pie is anywhere near as efficient across the power band.

As for the phase amp question, it will not increase the total peak amps drawn, but it will give you peak battery amps for a longer duration through the powerband. A cycle analyst or other watt meter will show you the result in real time.

PS, if you can find a more efficient DD hub motor and you can prove it, i'll put my leaf up for sale, buy one, and stop talking about the leaf. Efficiency is everything to me since my trip distances are very long, and every watt hour i can stuff in my triangle is a precious commodity.

 

[pwd]

Awesome. Your firsthand experience and testing is quite valuable. 8) Thanks for the input

PS, if you can find a more efficient DD hub motor and you can prove it, i'll put my leaf up for sale, buy one, and stop talking about the leaf. Efficiency is everything to me since my trip distances are very long, and every watt hour i can stuff in my triangle is a precious commodity.

I will let you know for sure haha. The "leaf" thread is actually what inspired me to start looking for a more efficient motor. In turn (pun intended), I'm beginning to understand the differences between different hub motors. I thought my ebike journery was starting to cool down, but this just opened up a can of whoop ass.

 

[rumme]

Neptronix , is there a rear brake rotor on your ebike ? Are you only using a front brake that is disc brake but non hydraulic ?

 

[pwd]

I believe there is no rear brake on his ebike. He uses regen to slow down.

I pulled this from the famous leaf thread:

If i had the opportunity to install a disc brake, i wouldn't have. It's up to you guys to find out if everything is okay with the disc brake. I understand that if you're offroading a lot, a modulatable disc brake in the rear could help out, but 99% of my riding is on the street, so having strong regen is no issue. I've done just a tiny bit of offroading and not had a problem with rear tire slippage though.. but not up/down any serious offroad hills.

 

[bingosunnoon]

Why do you guys, and I mean those of you with the larger dia. idea; Why is the heavier of two motors considered to be less desirable? I prefer heavier motors because they are heavier. It's hard to explain.

 

[amberwolf]

For a hubmotor in a wheel, suspension or not, heavier generally means a harsher ride.

Heavier doesn't mean better (either better suited or better quality), either, or even more powerful. It might be either of those, but no certainty just because it's heavier.

Because of quality / manufacturing issues, I'm actually going down to lighter "lower power" DD hubmotors on my heavy-cargo SB Cruiser trike, from the MXUS 450x ("3k" 45h) motors to the Ultramotors used in the A2B Metro and at least one version of the Stromer (though I'm not using their itnernal controllers; I'm still using the same controllers that ran the MXUS motors).

(If I were to use heavier ones again, it'd be the QSMotors hubs, like the QS205 50H series, because they are at least significantly better-built than the MXUS.