Leaf / leafmotor / leafbike high efficiency 1500w motor

[spinningmagnets]

Humidity has been found to "breathe" through the strands of copper phase wires. Sounds bizarre, but trusted members have documented this.

It involves hot and cold expansion/contraction of the air volumes inside the hubmotor.

 

[neptronix]

So what is the max voltage one can run this Leafbike motor at without arcing over?

I hit 60mph on a 26" wheeled upright bike and it took a number of multi mile 7-10% hill climbs for it to get hot.
If you're going with a smaller wheel, expect your power density to go up versus my big wheel. Expect total average aerodynamic load on the flat to be a fraction of what it was on my bike.

Honestly, if i had a velomobile, i'd give 96V a try. If we're talkin' about a 4T anyway. 3T? probably pushing it..

Add ferrofluid and a hubsink and watch the heat go down.. there's your headroom to blast full throttle on 96v for decent periods of time.

I would have ran >100V on the leafmotor 1.5kw but my problem was that it was hard to keep the front wheel on the ground even at 4kw peak.. even more of a challenge at 72v/6kw, lol.

 

[The Toecutter]

I'm wanting to do this to the 3T in my possession, which will be built into a 16x1.5" light-duty motorcycle wheel with a Mitas MC2 16x2.25" tire.

96V would get me about the same mph in top speed. I have both hubsink and ferrofluid in a box waiting to be added to the motor. I eventually plan to do about 8-10 kW peak, 250A max phase current, and with a 10k NTC thermistor feeding a signal into the CA3 to roll the current back if the motor starts to overheat.

I'm going to need a roll cage. I'm building a car that can pass as a "bicycle" as long as I don't do anything stupid, and is fully functional as such(only faster) when I disable the motor thanks to the pedal drivetrain and the slippery aerodynamics.

My first build met or came close to meeting all of the goals I outlined when I first posted here back in 2015, things members here were saying was not possible, said goals which were 30+ mph cruising speed, 200 mile range @ 30 mph, retains functionality as a velomobile with motor disabled, 50 mph top speed. It is pretty much there right now in all categories, or close to it. I have Chalo to thank for introducing me to parts and concepts that I was not aware of. The vehicle linked in my signature is the first "bicycle" I put together completely on my own starting from a frame. Before building it, I found it much easier to work on cars than to work on bicycles and had difficulty with maintaining my road bike.

I think a 300+ mile range @ 35 mph, 80+ mile range @ 70 mph and car-like acceleration/cornering/top speed is possible in a sub 100 lb vehicle that remains fully pedalable to velomobile-appropriate speeds on a dead battery. The only limitation to performance with the off the shelf parts available is the motor technology we have access to, which the current Leafbike 3T may still allow 0-60 mph ~8 seconds with the setup I'm planning for it. 96V would yield 90+ mph top speed if I get the drag nice and slippery like my Milan velomobile. Lots of things would need to be upgraded before I'd dare push such a thing to its limits; do keep in mind a spindle failed at 40 mph from a pothole, and my cable-pulled Avid BB7 brakes aren't exactly trustworthy in any panic stop above 35 mph. Working on rectifying all the issues I expect to pop up.

Even with my current 4T wind in a 26" wheel, with the humble Phaserunner set to 2,500W and 96A phase current, wheelspin is already an issue.

I hit 60mph on a 26" wheeled upright bike and it took a number of multi mile 7-10% hill climbs for it to get hot.
If you're going with a smaller wheel, expect your power density to go up versus my big wheel. Expect total average aerodynamic load on the flat to be a fraction of what it was on my bike.

Honestly, if i had a velomobile, i'd give 96V a try. If we're talkin' about a 4T anyway. 3T? probably pushing it..

Add ferrofluid and a hubsink and watch the heat go down.. there's your headroom to blast full throttle on 96v for decent periods of time.

I would have ran >100V on the leafmotor 1.5kw but my problem was that it was hard to keep the front wheel on the ground even at 4kw peak.. even more of a challenge at 72v/6kw, lol.

 

[neptronix]

Your biggest problem with that bike is going to be weight if you're adding a rollcage and all sorts of stuff.

I'm imagining something in the realm of 150lbs or more.. am i right?

+100lbs over a regular ebike = you're gonna need a lot more power while accelerating. Although you have a velomobile, so your energy use during cruise is very low.

What i imagine is that the motor is saturated ( therefore makes a lot of heat ) from a start until you are rolling, but then we have hyper efficient aerodynamics getting that motor to cool down during cruise since it's at a super low load point..

Who knows how the heat-cool stasis works out.. i'd say bring a laser temp gun at the very least with you.. but.. better is a temperature display on the screen. ( cycle analyst, etc )

I think the amount of motor you will need will be determined by the topography of your area due to the amount of weight we're talking about moving with an ebike motor. If you were in my area ( rollercoaster-like topography, hills ranging from 1-30% ), i'd say go with a moto rear wheel!

 

[The Toecutter]

Your biggest problem with that bike is going to be weight if you're adding a rollcage and all sorts of stuff.

I'm imagining something in the realm of 150lbs or more.. am i right?

Closer to 100-120 lbs.

Current weight is 91 lbs, including trash/debris, heavy battery pack, unoptomized steel and aluminum mounting and reinforcement for my coroshell, 5+ lbs of nuts and bolts on the coroshell, and the unoptimized/heavy multi-layered coroshell itself. There's a lot of room for improvement regarding weight, which will free up weight in the budget for all the heavy duty items I need for higher speeds. This said, the vehicle is very much a death trap. It has been hit by a truck already, but I got lucky(I would not have fared well if broadsided or hit head on), and was able to fix the tail section that was damaged. My body design protected the rest of the trike.

Keeping the bike as it is today with the exception of adding just the hydraulic disk brakes with slightly thicker rotors, reinforced spindles, rear gas shock, and DOT rims/rubber would put it right at 100-105 lbs.

Thicker spindles, heavier duty DOT rims/tires, hydraulic disk brakes, thicker wiring, each will add a few lbs of weight. Some of that will be offset by going to some NCR21700 cells to shave a few pounds off the weight of my battery pack while improving power and capacity.

I can easily remove about 10-15 lbs from my current setup. I've probably got about 1-2 lbs of random debris inside the shell by itself from riding it around in bad weather and accumulating leaves and mud plus random trash. The roll cage is not going to be very heavy. A friend of mine has built roll cages for dragsters at his shop and they are about 100-150 lbs. My vehicle is much smaller, won't be subject to race car speeds, and doesn't need nearly as much material, and I only need the roll cage to cover me and not the rest of the vehicle. I could probably get away with one weighing ~15-20 lbs and be slightly stronger than the ones the Pedal Prix HPV racers use. I can also offset some of the weight of the rear suspension by installing an aluminum rear frame section that I purchased, and use that to mount my hub motor and derailleur after I hack the steel frame apart. If I build a fiberglass or carbon fiber body, I may end up not even needing a roll cage if I get the safety cell design right, and if I build a monocoque, I won't need the heavy KMX frame anymore either, freeing up space for a much larger battery pack. Making a new body shell out of carbon fiber or fiberglass may end up weighing even less than my current coroplast one if the Milan SL I purchased is any indication.

+100lbs over a regular ebike = you're gonna need a lot more power while accelerating. Although you have a velomobile, so your energy use during cruise is very low.

What i imagine is that the motor is saturated ( therefore makes a lot of heat ) from a start until you are rolling, but then we have hyper efficient aerodynamics getting that motor to cool down during cruise since it's at a super low load point..

Who knows how the heat-cool stasis works out.. i'd say bring a laser temp gun at the very least with you.. but.. better is a temperature display on the screen. ( cycle analyst, etc )

I think the amount of motor you will need will be determined by the topography of your area due to the amount of weight we're talking about moving with an ebike motor. If you were in my area ( rollercoaster-like topography, hills ranging from 1-30% ), I'd say go with a moto rear wheel!

I live in a hilly area. Flat ground is rare here, and there are a lot of 3-8% hills, some 20%+ hills. I can turn the motor off and do 4 mph up a 20% grade near where I live without issue, as I have the gearing for it.

Me, plus the bike, plus any tools/luggage, I'm budgeting for a laden weight of 300 lbs.

 

[999zip999]

Does the leaf Electric Motor Company sell a good 72 volt 60 amp controller with a cycle analyst plug ?

 

[The Toecutter]

Leafbike's products were not intended to be Cycle Analyst compatible. I had to modify my Leafbike motor's plugs to be able to use the CA3. Get a Frankenrunner from Grin. You won't regret it. 96A phase current, and up to 72Vnom, as long as you're happy with 1,500W or less. It's a good, reliable, well-built, fully featured controller and is perfectly suited for practical everyday use, even if it may not be the most powerful thing around.

My experience with a Phaserunner(the predecessor to the Frankenrunner): I run my Phaserunner at 2,500W peak, 96A phase current, on a 46.8V system, but my vehicle dramatically reduces power requirements once it has accelerated up to speed, so the controller rarely goes into thermal rollback mode and usually only does so when it's 90+ degrees outside. I have no heatsink, and climb steep hills. The XT60 plug used is the weak link for my application. If I would have restricted my max phase current to 60A, I may not have ever had a problem with the connector and plug. The Leafbike 4T wind motor in a 26" wheel in my heavy trike performs quite acceptably and I can accelerate like a slow car. If you're going to run this in a normal bike, you will probably want a heatsink, and you'll probably have zero issues regarding overheating the controller or motor. I've ridden my build in snow, thunderstorms, extreme cold or heat, and other than the connector/plug issue, the controller has been reliable in those conditions for many thousands of miles on it so far.

Grin has configured a system that is generally waterproof or close to it, so it's hard to go wrong there, even if it is a bit expensive. The value you get for the money is greatly more than the increased cost over its cheaper competitors.

Does the leaf Electric Motor Company sell a good 72 volt 60 amp controller with a cycle analyst plug ?

 

[999zip999]

I have an asi bac855 and can't get it running I think that is the base to Frankenjustin.

 

[The Toecutter]

I have a bac855 and can't get it running.

Contact the vendor you bought it from. They have to assist in programming it. I have a BAC4000 and I'm going to eventually run into this issue when the time comes to install it.

 

[999zip999]

Ok he changed the secret password twice and maybe again.

 

[silentbike]

Yeah, true, silicone is also too thick. Thanks for suggestion :thumb:

Take plenty of pics when you do the upgrade. I'm running in the lower end of the ranges you mentioned, but haven't had any issues with the phase wires, so less motivated to upgrade. But, seeing some pics of a relatively easy installation could be just the motivator. :thumb:

It wasn't easy at all, took me two attempts! In the end, I used round file and dremel to widen axle hole a little, and smoothen sharp corners of cable exits. 11 awg(inside)/8awg(outside) phase wires, 24awg ptfe hall wires and thin red heat shrink
heatshrink wouldn't fit through but it doesn't have to. far enough just pull the wires with nose pliers
I used 100 watt flat spade tip iron and 60/40 lead solder, and high temp RTV silicone from china
cover copper fill with tape when soldering

I tried bunch of different things and wire gauges but in the end, this is what worked for me.
stock phase wire+10 or 8 awg outside is much simpler , i took some photos if you are interested






dont mind the silicone oops

 

[Cowardlyduck]

Do those wires just have clear insulation or is there not any?

I did a similar thing years ago using heat shrink instead of the thick silicone insulation:


I also made the axle hole larger and used very thin teflon coated hall wires so I could fit 10AWG wire through.

That and the cooling mods done were how I was able to hit 8KW+ on my Leaf.

Cheers

 

[silentbike]

Do those wires just have clear insulation or is there not any?

I did a similar thing years ago using heat shrink instead of the thick silicone insulation:
DSC_2398.jpg

I also made the axle hole larger and used very thin teflon coated hall wires so I could fit 10AWG wire through.

That and the cooling mods done were how I was able to hit 8KW+ on my Leaf.

Cheers

Looks awesome. might do something similar in future, I'll see how temperature foldback/hubsink combo goes.
(I also live in colder climate, might not be needed at all).


Phases have clear teflon insulation. had them colored but failed first attempt...then was too short in length .
At least I was smart enough to order more just in case :lol: from different seller but same product. just transparent
controller can auto detect phases and halls so only had to mark three small wires. Gnd, 5v and temp


After all this work i saw temp sensor didnt work swapped factory 10k ohm for ntc 100k used in 3d printers
now all works and motor spins. first tests tomorrow

 

[silentbike]

Had some fun with 18 fet monster vesc :lol: 85A battery 200 phase seems like a good everyday setup.
72v nominal 84v max. 1500wh copper battery. tried 150/300A but things start getting hot after while.
but this power levels are total overkill anyway, for fun test rides

 

[john61ct]

Link to that controller please?

 

[silentbike]

https://trampaboards.com/1x-vesc-100v-250a--p-28115.html
for cheaper alternative -> https://www.makerspev.com/products/little-focer-rev3-standard
cheapest -> https://flipsky.net/products/flipsky-75100-with-aluminum-case-based-on-vesc-for-electric-skateboard-scooter-ebike-speed-controller
(flipsky is rather shady business , and first batches of 75100 sucked, but this new one is promising)

just be aware, this is an efoil-eskate focused platform.
it does not come with display, PAS or wiring harness.
controllers are also typically non enclosed or sealed so 3D printer is useful. it's a tinkerers box :wink:
being FOC tuning needs more work than simple trapezoid or simulated sine wave bricks
this controllers are rarely used on ebikes. probably for good reason :lol:

plus sides are it's 100% open source hardware and software on github, no shitty logins like ASI
super small in size and hardly gets warm
built-in app can do advanced programming on the fly, unmatched to any controller on market
true FOC (dead silent and efficient) smooth as butter acceleration
MTPA support (this is an advanced topic but Ill post link below if anyone is interested)
https://hackaday.io/project/164932-axiom-100kw-motor-controller/log/179447-mtpa-merged-into-vesc
four different methods of sensorless startup, should your halls fry! it can be made to work pretty good!
accurate current and voltage measurement on all motor phases (3x shunts)
auto detection of phase wires and halls
easy to program compared to ASI controllers
great community support online https://vesc-project.com/

 

[neptronix]

Had some fun with 18 fet monster vesc :lol: 85A battery 200 phase seems like a good everyday setup.
72v nominal 84v max. 1500wh copper battery. tried 150/300A but things start getting hot after while.
but this power levels are total overkill anyway, for fun test rides

Jesus that's a cool bike you've got there!!!

 

[Ownzies]

Hey, this will be my first ebike build and I'm considering this motor. Someone on Reddit recommended that instead I get the Golden Motors Blackpie 5 Vector. This is only 1000w and has an integrated controller.

Which do you think is better? Do these LeafBike kits have any big issues regarding build quality? How long will the controller etc last me?

Is the Leafbike 26" Wide Rim 1500w motor the same as the one you're talking about here? I'd assume so, but just checking

Thank you

 

[Chalo]

Hey, this will be my first ebike build and I'm considering this motor. Someone on Reddit recommended that instead I get the Golden Motors Blackpie 5 Vector. This is only 1000w and has an integrated controller.

Which do you think is better? Do these LeafBike kits have any big issues regarding build quality? How long will the controller etc last me?

I'd choose the Leaf 1500W over any hub with an integrated controller. It's just not worth it to commit to a weird proprietary controller that is hard to access and may or may not be available when you need one.

Manufacturing quality of the Leaf motor is, well, adequate. No glaring deficiencies but nothing outstandingly good, either. The underlying engineering seems to be better than average, at least from an electrical performance standpoint.

The controller that comes with a Leaf kit is a generic KT trapezoidal wave model. I like those, but like the motor they're best described as adequate. Mine is working fine after about three years, with its only minor troubles being related to plugs and cables.

 

[neptronix]

AFAIK the golden motor hubs, and pretty much every hub with an integrated controller, all have a history of reliability problems with the internal controller.

It seems like an ultimately bad idea for a motor and controller to be sharing the same heat space.

In the case of the magic pie, more power is realized always by using an external controller.. the reason is that there's less heat in the case.

As for the leaf motor controller, i personally wouldn't touch it and run this nice motor with a nice controller

 

[Ellou]

Hi. I haven't been on this forum in years but it seems like the best place for the info I need.

I have 3 20 inch fat ebikes. I'm planning on making at least 2 of them reach 50ish mph but typically will be riding between 30 and 40mph. I stumbled across this thread and seems like this would be a good option for what I want to do.

I plan to buy a kit to replace the 500w/750w kit currently on one of them and upgrade controller and display later.

Is this the motor this thread is referring to?

https://www.leafbike.com/products/beach-snow-fat-bike-kit/20-inch-48v-52v-1500w-rear-hub-motor-beach-snow-fat-tire-bike-conversion-kit-1233.html?VariantsId=10019

I would rather not open up the motor to replace phase wires so if this is the motor this thread is referencing are the phase wires large enough to handle 50 to 60 amps?

I would like decent amount of torque and speed. I don't need to freeway speed, I'm just looking for enough ooomph to get out of danger and move at a reasonable speed while riding on the street.

I plan to upgrade the 52v controller in the kit to a grin tech when I'm ready to bump up to 72v.

I did search and read a lot of this thread but it is massive and was information overload.

Thanks

 

[The Toecutter]

I have 3 20 inch fat ebikes. I'm planning on making at least 2 of them reach 50ish mph but typically will be riding between 30 and 40mph. I stumbled across this thread and seems like this would be a good option for what I want to do.

I plan to buy a kit to replace the 500w/750w kit currently on one of them and upgrade controller and display later.

Is this the motor this thread is referring to?

https://www.leafbike.com/products/beach-snow-fat-bike-kit/20-inch-48v-52v-1500w-rear-hub-motor-beach-snow-fat-tire-bike-conversion-kit-1233.html?VariantsId=10019

That is indeed one of the variations of the Leafbike 1500W motor referred to in this thread.

I would rather not open up the motor to replace phase wires so if this is the motor this thread is referencing are the phase wires large enough to handle 50 to 60 amps?

Yes, but not continuously. Continuous rating is about 35A. But they might do 50-60A for the duration of your battery's discharge without the motor overheating, depending upon your battery pack size.

You'll want to open up the motor to add a 10k NTC thermistor for insurance purposes once you upgrade to a CycleAnalyst computer.

I would like decent amount of torque and speed. I don't need to freeway speed, I'm just looking for enough ooomph to get out of danger and move at a reasonable speed while riding on the street.

I plan to upgrade the 52v controller in the kit to a grin tech when I'm ready to bump up to 72v.

I did search and read a lot of this thread but it is massive and was information overload.

Thanks

If you go with 72V, a 4T wind motor will get you to ~50 mph on a fatbike w/20" drive wheel, accounting for aerodynamic drag. If you retain 48V, a 3T wind motor would allow you to top out ~40 mph on a 20" drive wheel fatbike, accounting for aerodynamic drag.

The 3T version definitely lends itself well to massive power upgrades in the future. To maximize the power density of the motor, you want to use a winding that can fit all 66 coil spaces in the motor. Meaning, you want your number of turns in the winding to be a number that 66 can be neatly divided by, without a remainder. THEN there's an advantage from having a shorter path for the current to flow through, which favors a lower turn count winding. 3T is the lowest turn count Leafbike will make for you, and you will have to specify the winding you want in the comments, as the standard they will give you is 5T.

No matter which winding you choose, the motor will always have the same continuous torque without overheating, but the current vs. torque graph will vary. However, the 3T wind allows the best power density of all the possible configurations.

How much power are you planning to run? A Grin Phaserunner or Frakenrunner is good for ~1,500W continuous, which on a fatbike, suggests you'd be good for ~35 mph at that wattage. You can get more power out of a Phaserunner with a heat sink. In which case, a 48V system with the 3T wind is probably a good choice, especially if you intend to upgrade the controller for even more power in the future.

Also, I hope you have full suspension and nothing less, a good hydraulic braking system, and at minimum Schwalbe Marathon Plus tires(ideally you'd have DOT-rated rubber, but the Marathon Plus are rated by the ECE for 50 km/h continuous). 30-40 mph on a fatbike is asking for death without these things.

 

[Ellou]

Thanks for the response. I have hydraulic brakes on one and one of the others will be upgraded when I choose which to upgrade.

2 of the 3 are full suspension.

I already have a set of shinko 241 for the first bike I convert and will definitely use motorcycle tires on the other.

I will be starting off with 52v batteries 16a or higher. I make my own batteries so I can cannibalize them to make more volts or amps.

How do I know the turns of the motor? I don't see that listed.

Will a 3t be a nice mixture of torque and speed? On 52v and/or 72v?

 

[The Toecutter]

How do I know the turns of the motor? I don't see that listed.

If you do not request a specific turn count, Leafbike will send you the default 5T wind motor.

Will a 3t be a nice mixture of torque and speed? On 52v and/or 72v?

3t is the fastest rpm wind available. It makes the lowest torque per amp of any of the winds available, in exchange for offering the most rpm per volt. 6T would be a more balanced wind IMO, but even at 72V it won't provide the 50 mph speed you are after, albeit it will still do somewhere slightly below 40 mph top speed @ 72V in a 20" drive wheel. In contrast, at 52V, the 3T wind would get you somewhere around 45 mph top speed, but you won't get there nearly as quickly given the current limit imposed by your controller.

You could go to a 4T or 5T wind, but you will take a slight penalty regarding the motor's maximum continuous power capability versus a 3T or 6T wind because you are sacrificing some copper fill(of the space available for 66 windings in a Leafbike motor, the 4T can only use 64 of them, and the 5T can use 65 of them, while the 3T, 6T, and 11T fill all 66). If you're only going to use a Phaserunner or Frankenrunner controller from Grin and never upgrade, then this isn't an issue, as you'll never push the motor to its limits with that controller, and a 4T or 5T will work just fine. I've used a 4T wind in a 26" drive wheel for my build with a Phaserunner controller without major issue and the motor never so much as got warm to the touch running 2,500W peak. I live in a hilly area, but my vehicle is also more slippery to the air than the average ebike, so by the time I'm doing 45 mph on flat ground, I'm well below 1 kW demand from my controller giving it time to cool. I don't even have a heatsink for the controller. Accelerating with 2,500W, I can out-accelerate almost any moped I come across and keep up with cars.

IMO, I wouldn't bother doing more than 40 mph with a Grin controller on a fat bike. The aerodynamic drag imposed on the vehicle is just too much for the controller to maintain the required power output without eventually going into thermal rollback and/or destroying the XT60 connector. If you decide to go with a 72V system, a 6T wind would give you acceleration roughly twice as fast as a 3T wind, given the 96A phase current limitation of the Phaserunner/Frankenrunner. Do note that 96A on a 6T wind will risk burning the motor out faster than 96A on a 3T wind, because continuous torque remains roughly the same between the two(the Phaserunner will fail before ever damaging a Leafbike motor). As the turn count decreases, continuous current capability of the motor itself increases(this is separate from the continuous current capability of the phase wires, which is dependent upon the size of the wire. Leafbike 1500W uses 12AWG wire).

The advantage of the 3T wind is that if you decide you later want to go to a high-end controller that can do 250+ phase amps and want to turn your fatbike into a decently quick motorcycle, the motor will be able to handle it with minimal modification. The other winds are not as suited to that task. By selecting a 3T wind, you won't need to upgrade to a 72V battery to get close to your desired top speed either. But since the Grin controllers only go to 96A phase current, your acceleration will not be as good as with a higher turn-count motor should you choose to limit yourself to a Grin controller. If you haven't ordered a Grin controller yet, and like the idea of a 3T wind motor, consider other controller options like the ASI BAC2000, Nucular, Flipsky, PowerVelocity, Sabvoton, Lyon, Lewbowski, and similar. Do your research before you buy.

 

[ZeroEm]

Think the default is a 4T now used to be a 5T.

by The Toecutter » Jun 04 2022 11:01pm

Ellou wrote: ↑Jun 04 2022 10:28pm
How do I know the turns of the motor? I don't see that listed.
If you do not request a specific turn count, Leafbike will send you the default 5T wind motor.
Will a 3t be a nice mixture of torque and speed? On 52v and/or 72v?
3t is the fastest rpm wind available. It makes the lowest torque per amp of any of the winds available, in exchange for offering the most rpm per volt. 6T would be a more balanced wind IMO, but even at 72V it won't provide the 50 mph speed you are after, albeit it will still do somewhere slightly below 40 mph top speed @ 72V in a 20" drive wheel. In contrast, at 52V, the 3T wind would get you somewhere around 45 mph top speed, but you won't get there nearly as quickly given the current limit imposed by your controller.

You could go to a 4T or 5T wind, but you will take a slight penalty regarding the motor's maximum continuous power capability versus a 3T or 6T wind because you are sacrificing some copper fill(of the space available for 66 windings in a Leafbike motor, the 4T can only use 64 of them, and the 5T can use 65 of them, while the 3T, 6T, and 11T fill all 66). If you're only going to use a Phaserunner or Frankenrunner controller from Grin and never upgrade, then this isn't an issue, as you'll never push the motor to its limits with that controller, and a 4T or 5T will work just fine. I've used a 4T wind in a 26" drive wheel for my build with a Phaserunner controller without major issue and the motor never so much as got warm to the touch running 2,500W peak. I live in a hilly area, but my vehicle is also more slippery to the air than the average ebike, so by the time I'm doing 45 mph on flat ground, I'm well below 1 kW demand from my controller giving it time to cool. I don't even have a heatsink for the controller. Accelerating with 2,500W, I can out-accelerate almost any moped I come across and keep up with cars.

IMO, I wouldn't bother doing more than 40 mph with a Grin controller on a fat bike. The aerodynamic drag imposed on the vehicle is just too much for the controller to maintain the required power output without eventually going into thermal rollback and/or destroying the XT60 connector. If you decide to go with a 72V system, a 6T wind would give you acceleration roughly twice as fast as a 3T wind, given the 96A phase current limitation of the Phaserunner/Frankenrunner. Do note that 96A on a 6T wind will risk burning the motor out faster than 96A on a 3T wind, because continuous torque remains roughly the same between the two(the Phaserunner will fail before ever damaging a Leafbike motor). As the turn count decreases, continuous current capability of the motor itself increases(this is separate from the continuous current capability of the phase wires, which is dependent upon the size of the wire. Leafbike 1500W uses 12AWG wire).

The advantage of the 3T wind is that if you decide you later want to go to a high-end controller that can do 250+ phase amps and want to turn your fatbike into a decently quick motorcycle, the motor will be able to handle it with minimal modification. The other winds are not as suited to that task. By selecting a 3T wind, you won't need to upgrade to a 72V battery to get close to your desired top speed either. But since the Grin controllers only go to 96A phase current, your acceleration will not be as good as with a higher turn-count motor should you choose to limit yourself to a Grin controller. If you haven't ordered a Grin controller yet, and like the idea of a 3T wind motor, consider other controller options like the ASI BAC2000, Nucular, Flipsky, PowerVelocity, Sabvoton, Lyon, Lewbowski, and similar. Do your research before you buy.