Build two wheel drive without cycle analyst

[Cyclomania]

So I watched Micah Tolls video here where he talks about making a two wheel drive but keeping only one battery. I would like to try this.

But I wonder if there is any cheaper alternative to that cycle analyst used in his video? And also what happens if I do this using only throttle.

Will it go on full power at all times then? Or is it possible to limit the power output also when doing it this way?

Looks like this when done with no cycle analyst: (See sreenshot)

 

[Cyclomania]

Also if I were to buy these two controllers and only one display. Is it possible to plug in the display but make it run both of the controllers. Looking at this one:

https://www.aliexpress.com/item/4000993964473.html?spm=a2g0o.store_pc_topSellerIng.8148356.16.59297915VOPedf&pdp_npi=3%40dis%21NOK%21NOK%20431.06%21NOK%20431.06%21%21%21%21%21%4021038edf16967523909602780e68da%2110000013671539437%21sh%21NO%21167266123

 

[glennb]

Sorry, I don’t want want to derail your thread into a pros vs cons of dual motor debate, but every time I I see someone proposing doing it I wonder what the compelling reason is.

It’s an unusual use case in which the redundancy, traction, or weight balance benefits counter the negatives (cost, complexity, weight, risk, ugly wiring, troublesome tube/tire repairs).

If you already had a more than capable battery running a rear wheel setup that was a little underpowered, and a motorised front whee fell into your lap for free, I could completely understand.

But as a starting point for a design … I have to wonder what the decision tree looked like for someone to arrive at such an outlying branch.

Like I said, the only practical things that could/should steer you in that direction are redundancy, traction, or weight balance.

And even then, weight balance is better solved by a mid motor, so that one’s tenuous, and most traction concerns can be addressed via geometry and tire selection. That leaves redundancy as the only real compelling practical reason. But it’s not compelling, because there’s already intrinsic redundancy in the form of legs and cranks.

I’m only stating the practical perspective, not denying the validity of other criteria. A bike with a motor is cool. Two motors is definitely more cool, and at least twice as impressive.

 

[scianiac]

I'm building a 2WD bike and I agree with all of that, I'm building it for the traction reasons and combined with all possible tire, wheel, etc traction improvements for tackling deep snow. And even then there are a lot of issues with 2WD for improved traction as the less weight on the front wheel means any traction limited situation the front slips first and you lose control. I've heard many describe such builds in low traction environments as "totally unrideable" with linked throttles and only dual throttles with manual control is passable. I've come up with a number of options to try and solve this but they require entirely new controllers.

 

[electric_nz]

You could use one half twist throttle and one thumb throttle to control the motors independently. You would need 2 displays ( or none and just use ones with Bluetooth and a basic cycle computer)

 

[Cyclomania]

Yeah it is all of the above. Also it would be fun to build and test drive such a bike. Especially since I live in an area with lots of snow in the winter.

So above all else, it is a fun experiment. That is why I want to know if there is a cheaper alternative to a cycle analyst and if I could just skip that step completely, and if so, what that would lead to. Full power all the time? Or could I limit power through a display to go in different speeds still?

 

[Chalo]

Like I said, the only practical things that could/should steer you in that direction are redundancy, traction, or weight balance.

There's also double the heat capacity, and double the thermal discharge surface area. You can get the benefits of a clutched geared hub while exceeding the limited continuous power available in geared hubs.

 

[Chalo]

So above all else, it is a fun experiment. That is why I want to know if there is a cheaper alternative to a cycle analyst and if I could just skip that step completely, and if so, what that would lead to. Full power all the time? Or could I limit power through a display to go in different speeds still?

One of my buddies built a 2WD cruiser bike with a Bafang G020 in front and a BPM in the rear. He started off with two throttles, but quickly changed to a conjoined throttle because it worked better that way. He found the bike underwhelming with one motor, but more than satisfactory with two (without pushing the cruiser out of its comfortable speed range).

 

[glennb]

There's also double the heat capacity, and double the thermal discharge surface area. You can get the benefits of a clutched geared hub while exceeding the limited continuous power available in geared hubs.

Assuming you’ve already installed the largest available geared hub motor. Otherwise, upsizing the hub is more practical than installing a second smaller one.

There’s duplicated (redundant) mass and complexity in a second motor and the associated controller and wiring. Mismatched gear reductions create other headaches. Avoiding front wheel spin is a major concern. Having both hands operating throttles without one free to operate a brake is dangerous. You know this.

I can concede there’s possibly situations in which for whatever reasons it might be more convenient to install two smaller motors on a two-wheeled vehicle, but it’s never going to be a better from a purely engineering standpoint. An example might be having one motor PAS controlled, with the other used for throttle boosts.

But if you’re looking clinically at what people are trying to achieve from dual hub motor setups it seems self-evident to me that a left drive mid motor setup would be more practical.

 

[Chalo]

But if you’re looking clinically at what people are trying to achieve from dual hub motor setups it seems self-evident to me that a left drive mid motor setup would be more practical.

So... you suggest reducing heat capacity and heat discharge area instead? Because that's what it sounds like to me.

I'm sure when AWD came to rally cars, there were folks making some of the same arguments you are. But history spoke.

 

[bengxe]

So... you suggest reducing heat capacity and heat discharge area instead? Because that's what it sounds like to me.

I'm sure when AWD came to rally cars, there were folks making some of the same arguments you are. But history spoke.

I think his point was that even if you add a second motor it's better for both to drive the rear wheel. 2wd 2 wheelers aren't a new concept, I think history has spoken here too.

 

[Chalo]

2wd 2 wheelers aren't a new concept, I think history has spoken here too.

I think it has not when we're talking about sub 2hp per wheel.

 

[glennb]

Yeah, sorry, I wasn’t suggesting adding a mid-motor in addition to a rear hub motor - although there’s less wrong with that, at least you don’t have to worry about balancing the power delivery to avoid front wheel spins - I was suggesting it as a better alternative to two hub motors.

Many of these proposed dual hub designs that pop up on ES are scratch builds. No sunk cost issues to contend with in these cases, so might as well objectively assess the options.

The dual hub setup gives you:
- resistance-free freewheeling and pedalling
- more top speed or acceleration than a 1000W geared hub can provide
- no additional drive train wear

So does a left-side mid motor. I just think people should weigh it up against dual hub motors. There’s pros and cons of each, obviously, but they’re directly comparable WRT the key criteria.

It mostly boils down to geared hubs’ inherent torque and thermal limitations.

Currie got a lot of things right with their eZip system. The concept was sound, and still is, more so now that we’ve got access to better motors and associated technology.

 

[Chalo]

The dual hub setup gives you:
- resistance-free freewheeling and pedalling
- more top speed or acceleration than a 1000W geared hub can provide
- no additional drive train wear

So does a left-side mid motor.

How do you propose that adding a chain, gear train, belt, or other method of transmitting power from a motor to a wheel elsewhere, gives you "no additional drive train wear"? That's the entire reason I left mid drives (left and right side) behind. My first e-bike was left side drive, and I ditched it because the chain was a constant nuisance. My third e-bike was a Bafang mid drive, and I ditched it because of its depressingly short mean time between failures. My second e-bike, and every one I've had after the third one, have been front hub driven and boringly reliable.

I've been tuning in for long enough to remember a ton of chatter about problems with the Currie roller clutch and chain. Those things haven't gotten better in the last twenty years like e-bike specific motors have. They're mature tech that won't give up much more after all this time.

 

[glennb]

You know what I meant - no wear in addition to what the riders power causes.

The “additional” drive train (that “wears”) goes in the cons column for the mid motor.

(I could point out that it’s cheap parts that wear, they wear less rapidly due to wider teeth and perfect chainline, and that it avoids the nonsense of having to match cadence to motor, but that’s a comparison to right-side mid motors, not dual hubs).

Anyway, if I ride past someone trying to simultaneously operate two throttles and poke at buttons to change power assistance levels, while staring at a screen (or two), I’m going to confiscate their bike on public safety grounds … as long as they’re not big or tough or scary or crazy looking, in which case I’ll probably compliment them on their bike.

 

[chuyskywalker]

That is why I want to know if there is a cheaper alternative to a cycle analyst and if I could just skip that step completely, and if so, what that would lead to. Full power all the time? Or could I limit power through a display to go in different speeds still?

(Effectively) every ebike controller on the market just needs a 0.8->4.2 (read: hall sensor) signal on the throttle input to make the motor go. You can very easily skip the entire CA and simply tie both controllers into the same throttle output. The CA is not necessary.

 

[Chalo]

(Effectively) every ebike controller on the market just needs a 0.8->4.2 (read: hall sensor) signal on the throttle input to make the motor go. You can very easily skip the entire CA and simply tie both controllers into the same throttle output.

Note that if you do this, use the +5V supply to the throttle from one controller only. The second controller will connect to throttle signal and GND only.

 

[Cyclomania]

Note that if you do this, use the +5V supply to the throttle from one controller only. The second controller will connect to throttle signal and GND only.

Nice. I think that is what he is proposing in the video as well.

By the way, do you guys know if I could convert one of these two motor back wheels(below images) so that one of them works like a front wheel instead of a back wheel?

For example, if I remove the cassette on one of them and then add a disc instead of the cassette, and then convert it into a front wheel?

Or are frontwheels designed to work out of a certain standard, making them impossible to convert into back wheels maybe? Because of the design.

Since these two are quite similar in motor power connections and rims I was thinking they might be fun to try out in a dual system.

I think they are about 250w each. So If I use a 36v battery and 17a controllers for both rear and front I could get something like 1200watts out of the bike, correct?

Or would you guys rather go down on controller amps and up on voltage? So for example use a 48 volt battery and a 14a controller? That would amount to approximately the same watt output.
But I am not sure what is harder on the motor, higher controller amps or higher voltage from a battery?

 

[Chalo]

By the way, do you guys know if I could convert one of these two motor back wheels(below images) so that one of them works like a front wheel instead of a back wheel?

For example, if I remove the cassette on one of them and then add a disc instead of the cassette, and then convert it into a front wheel?

There are a few problems you would have to address. First, rear hubs are typically 35mm wider than fronts, so you would have to use a fatbike fork or else heavily modify a normal fork to get the extra width (not just bend it wider, because then the legs would crowd the hub/rotor and tilt the brake mount).

Also, those are geared hubs that only drive in the forward direction. They can only be used with the freewheel side on the right. If you mount it that way, and weld a disc brake tab on the front side of the right fork leg, then a disc rotor installed on the freewheel thread would unscrew from the hub. If it's mounted on a cassette spline, the freehub body would ratchet without providing any braking.

It's there not a disc brake mount on the left side of those hubs? I really see a hub motor without one.

Rear hub disc brake spacing is 15mm from axle locknut to rotor mounting surface. Front hub spacing is 10mm. So you wouldn't be able to simply mount a caliper on a widened fork and have it work as intended, if the fork is configured for a front caliper.

I think they are about 250w each. So If I use a 36v battery and 17a controllers for both rear and front I could get something like 1200watts out of the bike, correct?

You could put in something like 1200W. You would get out something like 900W (peak) of mechanical power from that.

Or would you guys rather go down on controller amps and up on voltage? So for example use a 48 volt battery and a 14a controller? That would amount to approximately the same watt output.
But I am not sure what is harder on the motor, higher controller amps or higher voltage from a battery?

Amps make torque (and heat). Volts make RPMs. If you don't have enough amps available to push the bike along at an efficient percentage of the unloaded RPM (say 65-85%), then you'll trap the motor in an inefficient speed range and make extra heat. Controllers can trade off volts for more amps internally, but only up to a point. In this case it's probably best to go for either more amps, or a little more of each.

 

[AmericanElectricSolutions]

Sorry, I don’t want want to derail your thread into a pros vs cons of dual motor debate, but every time I I see someone proposing doing it I wonder what the compelling reason is.

Much better acceleration. MUCH better.

Pulling acceleration instead of pushing, if you need it. Affects your back differently, sometimes pulling acceleration is more comfortable on longer rides. The rear hub motor affects your lower back, while front hub doesn’t.

Speed. Cheaper than increasing motor size, then controller then battery.

Redundancy.

Weight balance.

Just more fun.

And for some reason I noticed that many people online and offline have strong feelings about 2wd for some reason. Like really strong feelings.

 

[Cyclomania]

There are a few problems you would have to address. First, rear hubs are typically 35mm wider than fronts, so you would have to use a fatbike fork or else heavily modify a normal fork to get the extra width (not just bend it wider, because then the legs would crowd the hub/rotor and tilt the brake mount).

Also, those are geared hubs that only drive in the forward direction. They can only be used with the freewheel side on the right. If you mount it that way, and weld a disc brake tab on the front side of the right fork leg, then a disc rotor installed on the freewheel thread would unscrew from the hub. If it's mounted on a cassette spline, the freehub body would ratchet without providing any braking.

Ah that was the reason! I can remember having that same issue once but had forgotten why. Of course the axle Thanks for clarifying that.

 

[Cyclomania]

You could put in something like 1200W. You would get out something like 900W (peak) of mechanical power from that.

Amps make torque (and heat). Volts make RPMs. If you don't have enough amps available to push the bike along at an efficient percentage of the unloaded RPM (say 65-85%), then you'll trap the motor in an inefficient speed range and make extra heat. Controllers can trade off volts for more amps internally, but only up to a point. In this case it's probably best to go for either more amps, or a little more of each.

How do you calculate that? The 900 mechanical peak I mean? I have learned something about multiplying peak amps from controller with volts but I am not really that good about understanding of how it all works. But mechanical power is more the kind of current I would get on a continuos basis, and more often on a typical bike ride perhaps? Do you calculate 0,75*1200? Is that how you came up with 900?

Is there a good online calculator or something where I could reliably see how much I could overvolt a 250w motor and how to do that in the most efficient way? Without frying up the motor. . Best combo of battery and controller.

 

[Chalo]

How do you calculate that? The 900 mechanical peak I mean?

Up to date, good quality hub motors have efficiencies in the 80-something percent range. Geared ones usually rate in the low 80s. The controller costs another few percent, up to 10 but not usually that high at full load. Add a little loss in your plugs and cables, and you're looking at 75% overall efficiency in a near-ideal case.

Chances are very good you won't even be getting 900W to the wheel. But since you can't easily measure that value, and you can easily measure input power, it's a decent rule of thumb to say 75% overall efficiency for a good quality system and maybe 65% for a cheap-cheap but recently made one.

 

[Cyclomania]

Up to date, good quality hub motors have efficiencies in the 80-something percent range. Geared ones usually rate in the low 80s. The controller costs another few percent, up to 10 but not usually that high at full load. Add a little loss in your plugs and cables, and you're looking at 75% overall efficiency in a near-ideal case.

Chances are very good you won't even be getting 900W to the wheel. But since you can't easily measure that value, and you can easily measure input power, it's a decent rule of thumb to say 75% overall efficiency for a good quality system and maybe 65% for a cheap-cheap but recently made one.

I think I am going to wait until I have a front wheel as well. I have a used one in mind that I am looking at. If I get that one I will go for it. Also waiting for a controller for this system.

 

[motomech]

I ran 2 WD's for almost a decade.When I started w/ it I was queer for mini motors, specificly the then wildly popular Q100's and I wanted to see how fast I could go using 2 of them. Generaly, when one runs a single high-speed range mini you might 25 mph, but with a very lazy mid-range. Worse than that, the "328" is so tall it never really gets into an efficient rpm range, the motor tries to get all the Amps it can and phase wires and connectors start melting. Using two, they tended to pull each other along a top speed and I could get 27 or 28 mph. Although the bike still slowed a bit in the mid range, it wasn't too bad. I could still melt wires, especially off-road where even with 2WD, the rear does the bulk of the work. What happens is the rear system, if it makes any power at all, lifts the frt. and that tire spins. Happens on the street too.
These days, I use better motors and if I deemed that my current single Bafang felt underpowered, I would go up a step to a mid running 35 to 40 Amps. More than that and traction becomes a problem and then maybe a 2WD would be useful.
The dual mini powered was good on wet grass and I suspose it would be good in snow.
After trying different combo.s of controllers, batteries, throttle and motors, I ended up with two of everything. I use LiPoly bricks which are modular, so no downside to 2 batt.s. Controllers are small and Cutes were light and cheap.
What made it work well was both throttles on the left side, a left-hand half-twist shoved up next the a left-hand thumb, very intutitive and easy to modulate. I would think, that even with todays better electronics trying to aportion the power in low-traction situations, the human brain can still do a better job.
Two minis weren't too bad in terms of weight, but two midi's were just too heavy for me to want to carry the bike up the stairs.
Although I never made use of having redundent systems in an emergency to get home, there is a sense of security that comes with having a back-up.
Most of the time, I just used the rear throttle and added frt. power as needed. As a result, the frt. batt. used about 1/3 rd of power reserve compared to the rear. The frt.. batt. was about half the size of the rear.
I did PAS on the rear and throttle for the frt. for a while and that worked fairly well. Just the controllers I was using back then didn't do PAS very well.