Picking high-speed commuter hub motor: Motor Sim. results

Background:
--I'm an experienced road cyclists and have been commuting for a couple decades. I'll be starting a 50 mile daily round trip commute this fall and want to build a high speed, pedaled (+200-250W human power), lightweight road ebike. I want to average 25 mph and so cruise at 30 mph in flats. (I realize this speeds may make some folks nervous, but on my short commute they're already commonplace to me.) Calculators show that I need around 700W to achieve this, or about 500W of additional power.
--chas58 turned me onto geared hub motors (https://endless-sphere.com/forums/viewtopic.php?f=28&t=89455), and I'd like to build something like his high-speed commuter (https://endless-sphere.com/forums/viewtopic.php?f=3&t=49691), except about 5mph faster.
--I understand that DD motors can operate more efficiently, but I'm having trouble a) giving up the ability to freewheel, which also affords efficiency, and b) putting on a rear wheel that weighs as much as the rest of the bike (!).

I've simulated three geared hub motors with 52V/14S batteries: the Q100 CST 328, the MAC 8T, and the Bafang CST. To try to make the comparison "apples-to-apples," I've compared them with the motor output / effective speed. Here are three simulations (broken into three posts):

30 mph, ~725W This is the speed that I'd like to cruise at (by adding 200-250W of peddling).

400W, ~24.2mph This is the 400W that I'd like out of the motor in the flats (that I'd be adding 200-250W of peddling to).

500W, 16.2mph on a 5% grade A scenario for a couple short climbs. (I would add 300W for the climbs to get back up to 20 mph.)

So here are my questions:
1. I'm concerned with efficiency because of size/weight/cost of batteries. Do I understand correctly that the Bafang CST wins hands down in terms of efficiency, or am I missing something in my analysis of the simulations?
2. The Bafang CST and MAC 8T both weigh double that of the Q100 CST. Is the efficiency worth the weight hit?
3. How can I interpret the graphs to tell what happens when I pedal? This one really has me stumped!!
--Me adding 200W to the 400W of the motor puts the speed at about 28 mph.
--If I look at the graph of the Bafang CST at 400W, for example, then 28 mph is extremely close to the no-load speed/0 amps expended, so that doesn't seem to make sense.
--Is there a way to accurately simulate what happens when I add 200W of pedal power?
4. Aside from the difference in cost and weight, why pick of these hubs over another? Or why not pick the Bafang CST?
5. I'm planning on pedaling up to the teens before engaging the motor, and I can treat the system gently (I just changed the original front brakes pads on the Odyssey at 106k.) However with commuting 12,000 miles annually reliability is a huge concern. Thoughts about reliability given these simulation results?

Thanks!

mystryda said:

So here are my questions:
1. I'm concerned with efficiency because of size/weight/cost of batteries. Do I understand correctly that the Bafang CST wins hands down in terms of efficiency, or am I missing something in my analysis of the simulations?
2. The Bafang CST and MAC 8T both weigh double that of the Q100 CST. Is the efficiency worth the weight hit?
3. How can I interpret the graphs to tell what happens when I pedal? This one really has me stumped!!
--Me adding 200W to the 400W of the motor puts the speed at about 28 mph.
--If I look at the graph of the Bafang CST at 400W, for example, then 28 mph is extremely close to the no-load speed/0 amps expended, so that doesn't seem to make sense.
--Is there a way to accurately simulate what happens when I add 200W of pedal power?
4. Aside from the difference in cost and weight, why pick of these hubs over another? Or why not pick the Bafang CST?
5. I'm planning on pedaling up to the teens before engaging the motor, and I can treat the system gently (I just changed the original front brakes pads on the Odyssey at 106k.) However with commuting 12,000 miles annually reliability is a huge concern. Thoughts about reliability given these simulation results?

Thanks!

Click to expand...

I'm concerned with efficiency because of size/weight/cost of batteries. Do I understand correctly that the Bafang CST wins hands down in terms of efficiency, or am I missing something in my analysis of the simulations?

Click to expand...

You are placing too much importance on efficiency numbers. In the real world, there will not be that much difference in range. What you had for lunch or did you put the Lycras on will have more effect. What it does show is, the 328 Q100CST is not happy at cruise and what is not evident on the sim. is that it is Current limited and will be trying to suck the controller for everything it's worth, making for a very HOT controller.
My 328 Q100CST in a 26" whl. on 46 V will not go over 23 mph, it just runs out of power. I can only add 100 Watts w/ my legs, but I can add plenty w/ my frt. motor. w/ 2 328's I can hit 27 mph because they help pull each other into their "zone of higher Efficiency". I'm just a normal rider and chas8 seems to be a Superman w/ performance for a single high-speed mini that few can equal. If you are not a "Superman", I would rethink trying to use a high-speed mini at those power levels.

Aside from the difference in cost and weight, why pick of these hubs over another? Or why not pick the Bafang CST?

Click to expand...

The BPM is a mid-range motor and is a better match for your projections. The mid speed provides good climbing performance(efficiency), while the 52 Volts raises the top speed a bit. The BPM is a good motor up to 30 mph, above that, look at the more expensive MAC.

. I'm planning on pedaling up to the teens before engaging the motor...

Click to expand...

It really doesn't work that way and I think you would find that both tiresome and tiring. Focus on getting the best PAS experience you can afford(the KT series sine wave controllers w/ 5 level "torque imitation) from elifebike or BMS Battery)and stay on the PAS all the time, is the way you want to go.
Personally, if I was going to scratch build a higher speed assist bike, I would use the newly refined Q128CST. Especially if the Current was limited to 20 Amps. Those larger geared motors want 25 to 35 Amp controllers to be in their "happy zone". The 48 Volt Q128CST 328, if measured on 36 Volts would be in the same speed range of the BPMCST(270 rpms @ 36 V).
Whether or not the KT sine wave controllers mate up w'/ a 52 V pack is the question. You could research that. I would just run it w/ 48 Volts, it would still do 30 mph and would allow the use of one of BMS Battery's intergrated Li-Ion bottle batteries.
P.S.- Dave(D8veh)recently did a Q128 build w/. a sine wave controller and is happy w/ the results and he is not an easy guy to impress. I suggest you look that build up(just reading his posts is a great way to learn).

If you want a reliable commuter get a DD motor and use regen braking to save your brake pads. And since you're in the US, might as well go with a 1000W motor kit that can cost as little as $160 shipped to your door in a week or less.

No, because for him weight is important.

You are on the right track, but I agree that efficiency numbers wont be what matters. What will matter, is how aero is the coat you are wearing today, are you packing panniers that cause wind drag, etc.

You are correct, that you will need about 1000w ( including electric watt losses) to go 30 mph. this is why one guy leading the peloton at 30 mph can't do it forever. They need to rotate so they can pedal that much wattage when leading.

Any of the motors you looked at should work fine. Choose the one with the speed wind that suits your cruising speed. If the motor won't cruise 30 mph without you pedaling, then it won't help you at all at 30 mph. It will push till 25 mph or whatever, and then stop pushing. This is how motors work, they spin to a certain rpm depending on the voltage. So you need a faster wind, such as the Mac 8t. The motor the company I work for sells will be no good for you with 48v, ( 54.6v 13 cell) because its a 10 turn. It would only get you about 25 mph cruise.

Go for the fast motor, because as you ride, voltage drops. halfway there, you will be slowed down to 25 mph.

If this makes no sense to you now, that's normal, this is weird stuff. But KNOW THIS, what you can do is pedal hard and add efficiency. You WILL NOT, add much more speed. So when you need 1000w, and pedal hard, your 250w will only subtract what the motor pulls from the battery, not so much add speed. Sure,, you can sprint and add 5 mph, but in cruise mode, 2 mph more speed is typical. Even for a very fit cyclist, that 5 mph more speed simply adds too much drag, and then if the motor stops helping at 25 mph, well, there you go.

So you need to buy a motor that can haul ass. And you need to feed it volts. the 13 cell 48v may not do. go for a 14cell, usually called a 52v battery.

What you need to look for on the graphs is not efficiency, but max speed with no load. get the fast one.

As for reliability, any of those motors will be very reliable. 12,000 miles a year is a lot, but most of your wear and tear will be brakes, tires, wheel build, and especially, the battery. You are going to need to carry a much bigger, much heavier battery than you think.

You will need to carry 25 watt hours per mile, plus,, you will need to carry more. Why more? because one fine day, it will be 25 miles all the way home into a 40 mph wind. So a 48v 20 ah battery would be the minimum for your commute.

Lastly, if you choose to cruise at 25 mph, then you will have a lot better efficiency, and less need to pick the very fastest motors.

If you wonder where my comments come from, I commuted 30 miles a day for 5 years by e bike. Not quite your distance, but close enough to know how it goes. It's just not the same thing if when you get there matters less. On a commute, you need a lot of reliability, and you need more speed, and more battery, for that day you have to stop and change a tube. Then haul ass to get there on time still.

I found max reliability meant not having to wait for any motor kit parts I might need unexpectedly. It could be something simple as a wire broken off the plug, but you are stopped till you fix it. So then,, having two motor kits, it made sense to mount it on a second, not so good, bike.

So your spare bike, it could be a 50 buck bike, with that 160 buck direct drive kit on it.

The Bafang CST and the MAC are both relatively heavy motors. the Q100 is nice and light, but You'll kill it running it at 52v and 20A. Even 48v and 20A will wear the gears rapidly with a light rider and bike.

I'd say that the two motors you should consider are the Q128C from BMSBattery and the Bafang SWX02C. Both weigh just over 3kg and will handle 48v at 20A.

The 36v 201 rpm Q128c does about 25mph on the flat at 48v with hard pedalling using 26" wheels, so 52v and a 700c wheel should take it pretty close to 30 mph. I'm not sure about the speed versions of the SWX02C. I know there's one 36v version that does 20 mph at 36v and another that does 25 mph.

d8veh said:

The Q100 is nice and light, but You'll kill it running it at 52v and 20A. Even 48v and 20A will wear the gears rapidly with a light rider and bike.

Click to expand...

Regarding the Q100, could it perform at a higher speed by running a higher voltage/lower amperage? Or can the mechanism simply not handle 750W?

(For reference, I'm about 170-175 lbs, and in the spec that I'm working up so far the bike comes in at 35-40 lbs.)

d8veh said:

I'd say that the two motors you should consider are the Q128C from BMSBattery and the Bafang SWX02C. Both weigh just over 3kg and will handle 48v at 20A.

Click to expand...

Thanks, I'll try to look those up. Neither are in the simulator, so I'm going to have to do some "real math," I suppose.

MAC is the motor you want if light weight and 30mph is desired at the same time.

The mac on the simulator has the old 0.5mm lams. The newer eZee motors are similar and will give you a good approximation of what the performance of a new MAC is like, although i believe they are less powerful than the MAC.

eZees are cool because they're really 8.5lbs instead of ~10lbs like the MAC.

Why not a mid drive like BBSHD? Gearing it for the amount of power & speed he wants is super simple, just put a big 48+ tooth front gear. The speedo PAS would be more intuitive to his pedaling. A hubbie motor can sometimes lead to a bad behavior of hand throttling for power.

I'd go with a larger capacity 48v battery vs. a 50v. Hell, one could put a even larger capacity 36v battery and gear it appropriately. With a fit rider, battery consumption would be minimal. It's ohso easy to pedal assist @25-30mph. The picture changes when he decides he wants a faster ride. Then go direct drive and high voltage.

Just bring an extra chain if your doing long mileage trips with a middrive as it will wear chains faster. 8)

melodious said:

Why not a mid drive like BBSHD? Gearing it for the amount of power & speed he wants is super simple, just put a big 48+ tooth front gear. The speedo PAS would be more intuitive to his pedaling.

Click to expand...

I appreciate the suggestion, and I had originally been leaning towards a BBS02, but given the high mileage (50 miles/day, 12k+/year) I've been advised that mid drives come with considerable concerns with regards to reliability and maintenance (time and part replacement costs).

neptronix said:

The mac on the simulator has the old 0.5mm lams.

Click to expand...

Thanks for letting me know. I've read all the documentation the main page for the simulator, but there's no substitute for insights like this.

neptronix said:

The newer eZee motors are similar and will give you a good approximation of what the performance of a new MAC is like, although i believe they are less powerful than the MAC.

Click to expand...

I had looked at the "eZee 350 rpm" initially, but it didn't seem to do as well as the others in simulation in terms of efficiency as well as headroom. Here's what I put in:


I'll do some more reading on the MAC 8T, though. Do you know when the new version came out or how it's different?

mystryda said:

I appreciate the suggestion, and I had originally been leaning towards a BBS02, but given the high mileage (50 miles/day, 12k+/year) I've been advised that mid drives come with considerable concerns with regards to reliability and maintenance (time and part replacement costs).

Click to expand...

I agree with that.

Damn, that's a hell of a commute. If it's mostly flat.. many hub motors have higher overall efficiency. leafmotor peaks at 90.5%.. MAC peaks at 84%ish.. BBSHD/BBS02 are said to be in the low 80's, but i've never seen a dyno chart for one.

mystryda said:

I had looked at the "eZee 350 rpm" initially, but it didn't seem to do as well as the others in simulation in terms of efficiency as well as headroom. Here's what I put in:

I'll do some more reading on the MAC 8T, though. Do you know when the new version came out or how it's different?

Click to expand...

The reason is that you're simulating that motor wrong. That's a high speed per volt motor.. it will do better at 48V and 30A.. use the IRFB/infineon type controller.

But for a 700C bike, i'd go with a 300RPM model.. the 350RPM winding is too fast for that big wheel. In my 24" wheel, i get 30mph.. on your 70C, you're looking at 35mph, which is a bit too much for the motor to handle continuously.

I've used all those motors in various builds. I'd agree that they're be good for what you want if you were considering an MTB type of builds, but for a light-weight road bike type, they're too heavy and agricultural. The Q128C is a lot better. it's exceptionally smooth and totally silent, yet still it gives a lot of power because of its high reduction ratio.

I wouldn't get too hung up on the simulator. The curves are not massively different between the motors. If you substitute a geared motor with the same maximum speed (kV), you'll get a good approximation. It's the voltage and current that you give to it that makes the most difference.

The ezee would work. notice that you have only 76% throttle for 30 mph. I'm not sure how to simulate your massive pedal input. subtract 75 pounds?

Its sort of coming down to does he want it fast and really light, or fast and more reliable.

He pedals hard enough to take a lot of the strain off a smaller motor, but if he wants to cruise 30 mph the whole ride,, then he simply needs to have something that will go 35 mph without pedaling, leaving the garage.

Lets not forget, the second half of the commute, he will have a lot less battery voltage. So he needs a faster wind, like the 8t. Or, more volts into something else.

Returning to the math, he pedals at 250w!!!! That means he can run at 1000w speed using 750w. So a smaller motor able to really handle 750w continuous will do.

I'd still pick the mac though,, because it can handle 1500w fine, and has more mass to dissipate the heat when it does get run hard. And it comes in a good fast wind.

But it is true, with the 8t he will sacrifice a lot of the efficiency on any starts and stops. Which is where his massive pedal wattage comes in. He pedals hard enough to be quite efficient on starts, especially if he just cranks out three pedal strokes before he hits the throttle. Or sets his PAS for a big delay in start up.

One thing to bear in mind,, It's almost never going to be the motor went out that stops you getting to work on time. Its going to be shit like the spokes broke, the chain broke, or you got two flats in one day. Or with the motor, a broken wire, janky plug, or rarely, a controller just goes zzzzt.

You can try the Ezee, and if it does not satisfy, then move on just replacing the motor. Just the motor wheel is not that expensive, compared to a complete kit with wiring and display.
In any case, 350 rpm at 36v is the winding ballpark you need, to have the extra speed you want to cruise 30 mph. Oh,, and a 60t front chain ring.

The load line remains the same for all the above sims, as it is determined by the initially selected bicycle type (wind resistance, tire friction) and rider weight. This should make calculating contribution of pedalling straight forward. Where the red watt line + 200W intersects the load line (indicated by horizontal red line) should be a fairly reliable indicator of the speed gains you should achieve (solid black vertical line). Note that motors fed more amps reach top speed nearer to where the red line falls off a cliff (approaching no load speed), in which case, adding 200W would achieve little gains in speed.



In my case, my ride is primarily for exercise, but I need to cover the 15 km quickly to arrive at work on time after taking the kids to school. I currently use the Q128c slower wind, but plan on implementing the 328RPM wind and limiting amps to the point where I need to contribute 200W on level ground to achieve my final desired speed.

Now compare this with a motor provided more amps:



The speed is a bit higher, but an input of 200W pedal power yields a paltry 1 mph, as the motor is approaching its unloaded speed.

Apart from the motor, you need to consider the control systems. If you want to pedal, you need a good PAS system. The Kunteng controllers are really good, but they don't seem to work with Ezee motors. If you use an Infineon controller, you need a Cycle Analyst, so you end up with a lot of expense, complicated wiring and complicated setup to get it working. A KT controller with LCD is plug and play and about 1/3 the cost, and better IMHO.

An eZee and MAC motor are pretty much the same. Best controller is an infineon clone from em3ev, ebikes.ca, etc. These controllers will handle the high electrical RPM.

My initial thoughts:
1) You need to change the amps for the Q100, as 52 volts and 20amps isn’t going to work. For the simulation, I use 15 amps.
2) Realistically for us, the maximum current isn’t too important. It only provides power to get you going, and you are going to supply most of that yourself (I’m out of the saddle when I accelerate). I’m happy with a low current (12amp) controller, as it forces me to provide the initial acceleration torque myself, and won’t get too hot. Remember, your cruising speed is to the right of the inflection point, so you (as a strong rider on a road bike) should not be using maximum current very often.
3) You can move your “speed” line to the right to account for your 200+ watts. You want the distance between the black line and the red line to be 200+ watts (that is the difference between the load and the motor’s power). (Question 3)
4) With those graphs, your cruising speed is dangerously near the max current of the controller – meaning that your controller will get very hot and likely overheat (unless you have a good heat management plan). (in other words, the cruising speed is near the inflection point of the red line on the graph – you want to be well to the right of this)

The power output of your legs seems to be similar to mine. Keep in mind that your leg power for an endurance ride is about 50 watts less than a shorter faster ride – 200 watts is pretty strong for riding 2 hours a day.
The MAC and the Bafang are pretty similar motors as I recall – it’s a tossup between the two.

And thoughts after reading the thread:

(Q2) The larger size of the larger motors is going to allow you to handle more power (more current) and give you faster speeds. You can decide if that is worth the weight.

I do agree, that 25mph is a lot easier, lighter, and more reliable than 30mph. However, you are talking a 10-minute difference in a 25-mile commute.

Personally, I would use the Q100 for speeds up to 25mph (with you pedaling 200+ watts), or the Q128 for speeds of 30mph.

And for the endless big wheel debate: 700c and 26” wheels are roughly the same size. (compare a 622-25mm vs 2.125”x26”). A road bike and road tire is 300-350 watts more efficient than a mountain bike at 30mph. Thus, with a Q100, I use a slower speed winding on my mountain bike (26” wheel) and a higher speed winding on my “big wheel” road bike.


A mid drive would be easier to design, but isn’t going to be faster on a flat commute. Where they excel is being able to downshift and climb.

Reliability (Q5): my biggest wear item is going to be the battery and tires. Plan on charging at work. Ideally you would not charge above 80% and not deplete below 30%. Most people just charge to 100% and leave themselves minimum ~30% unused capacity. That gives you a little head room for occasional detours, headwinds, and the battery capacity degrading over time.

mystryda said:

So here are my questions:
1. I'm concerned with efficiency because of size/weight/cost of batteries. Do I understand correctly that the Bafang CST wins hands down in terms of efficiency, or am I missing something in my analysis of the simulations?
2. The Bafang CST and MAC 8T both weigh double that of the Q100 CST. Is the efficiency worth the weight hit?
3. How can I interpret the graphs to tell what happens when I pedal? This one really has me stumped!!
--Me adding 200W to the 400W of the motor puts the speed at about 28 mph.
--If I look at the graph of the Bafang CST at 400W, for example, then 28 mph is extremely close to the no-load speed/0 amps expended, so that doesn't seem to make sense.
--Is there a way to accurately simulate what happens when I add 200W of pedal power?
4. Aside from the difference in cost and weight, why pick of these hubs over another? Or why not pick the Bafang CST?
5. I'm planning on pedaling up to the teens before engaging the motor, and I can treat the system gently (I just changed the original front brakes pads on the Odyssey at 106k.) However with commuting 12,000 miles annually reliability is a huge concern. Thoughts about reliability given these simulation results?

Thanks!

Click to expand...

For the simulator, I use a ' 26" wheel ' for my road bike.

A 700c road tire on a road bike has approximately the same diameter as a 26” mountain bike tire.

I just ran some calculations, and it looks like the simulator uses these tires:
700c = 56mm tire
26” = 2.0” tire (or ~23mm 700c tire)

Unless your bike uses the so called 29” tire, you are probably better off modeling with the 26” tire diameter.