hill climbing? 750W hub versus 250W mid drive

[spokewrench]

you sure it was 22% thats a hard climb for anyone. and for a bike really hard. here are pics of 20% and the last one at the top is 22%

I calculated 23.6%. I was mystified! Then I realized that if the camera had been level, the corners of the barn would have looked plumb. Chain link fence posts on a slope? Who knows! (That bike has a seat position that would work for me.)

 

[fooferdoggie]

I calculated 23.6%. I was mystified! Then I realized that if the camera had been level, the corners of the barn would have looked plumb. Chain link fence posts on a slope? Who knows! (That bike has a seat position that would work for me.)

ya I realized it was not level it was hard to stand sideways on that slope and get it accurate.

 

[fooferdoggie]

here fixed it. fist time.

 

[AvalancheRun]

I mentioned earlier that Bafang motor power is overstated. Here's what I mean: A BBSHD is rated for 160 Nm. Sounds enormous compared, for instance, to a Bosch CX's 85 Nm. But Area 13 put a BBSHD on its dyno and found that 160 Nm at the axle translated to 58 Nm to the ground. Its still a lot, but nothing like what you would think by reading the specs. If a 48v/28a BBSHD is only putting down 58Nm, whats a 20a / 36v BBS01 putting out? Not much. What I see very often in EU cargo bike conversions are 48v BBS02's, which from what I hear creates really good middle ground insofar as power is concerned.

160 vs 58 nm is worlds apart. My two bikes are 75 and 50 nm and they are night and day in terms of climbing. Another 90 nm bike I had was again significantly better than the 75. Do you know how manufacturers typically measure torque for spec sheets?

Another example. The Priority Current is claimed to have 140 nm, but owner reviews liken it to the 85 nm CX in terms of climbing ability.

IMO the bicycle manufacturer must know this. If that were my product I would reevaluate those numbers and come up with something equivalent to how Bosch/Brose, etc. measure torque.

 

[spokewrench]

I mentioned earlier that Bafang motor power is overstated. Here's what I mean: A BBSHD is rated for 160 Nm. Sounds enormous compared, for instance, to a Bosch CX's 85 Nm. But Area 13 put a BBSHD on its dyno and found that 160 Nm at the axle translated to 58 Nm to the ground. Its still a lot, but nothing like what you would think by reading the specs. If a 48v/28a BBSHD is only putting down 58Nm, whats a 20a / 36v BBS01 putting out? Not much. What I see very often in EU cargo bike conversions are 48v BBS02's, which from what I hear creates really good middle ground insofar as power is concerned.

If the Foxbat's 26 x 4 tires have a diameter of 28.5 inches, 58 Nm on the axle would be 160 N on the ground. I think "Nm" was an error.

 

[mschwett]

Ah, that makes a lot of sense. I do everything I can to keep a cadence of 85±5 RPM just to keep my knees happy. Sounds like a mid-drive would be happy with that, too.

yes, this is the issue. all of these motors are geared way down - they spin internally much much faster. on the mid drive, you can gear them down internally with that roughly 80rpm cadence in mind regardless of the speed of the bike, because the 4-5x range of th e-bike‘s drivetrain takes care of the wheel speed. on a hub drive, you need to do the internal gearing in a way that produces usable torque all the way up to 300 rpm or so. so while there is very rarely a torque advantage (except on mountain bikes) between chainring and rear cog, the internal gearing of the motor can be designed to greater advantage on a mid drive. small mid drive motors produce much more torque than small hub drives - compare the mahle x20 to the specialized SL - 23nM vs 35 or 50nM for motors of similar weight and power rating. mahle could have geared it down another 2x Internally to make it 46nM, but then it probably wouldn’t work well at 300rpm!

 

[mschwett]

Because most bikes sold are wildly geared too high for normal people. And electric bikes make this even worse.

I think most bike manufacturers think all of their riders have thighs and quads that are bigger than most tree trunks.

nah, the real reason is that most people don’t ride up steep hills, and the people who make bikes know this.

 

[Stefan Mikes]

I was riding my Vado with the 36-46T gearing (250 W nominal, 520 W mechanical peak power 85 Nm mid-drive...)

...could stop at the road-sign then restart the ride... and was actually singing while climbing!

I was on a 58 mile ride with 5,000 ft elevation gain that involved six steep climbs...

...was taking pics of Slovak kids practicing mountain riding...

...and was racing against a Slovak roadie

All in high mountains of Slovakia.

 

[kevinmccune]

ah,vados! give it up Stef! just admit you are one those double strong east bloc superlative humans,dont believe me watch ufc?

 

[spokewrench]

Another example. The Priority Current is claimed to have 140 nm, but owner reviews liken it to the 85 nm CX in terms of climbing ability.

The front belt wheel looks twice as big as the rear belt wheel, so 140 Nm from the motor would be about 70 Nm on the rear belt wheel. From there, the torque on the rear tire would depend on the gear range in the hub transmission. They call it a commuter bike, so I guess it would be geared for speed, not torque.

I found Radpower egregious. A 500 W direct drive bike of another brand would take me up a 6.1% a slope as fast as a 750 Watt geared hub motor Radrunner. I've seen a published disclaimer that their power and torque claims are what Bafang says a motor is good for, not what it produces with a Radpower controller.

 

[AvalancheRun]

The front belt wheel looks twice as big as the rear belt wheel, so 140 Nm from the motor would be about 70 Nm on the rear belt wheel. From there, the torque on the rear tire would depend on the gear range in the hub transmission. They call it a commuter bike, so I guess it would be geared for speed, not torque.

I found Radpower egregious. A 500 W direct drive bike of another brand would take me up a 6.1% a slope as fast as a 750 Watt geared hub motor Radrunner. I've seen a published disclaimer that their power and torque claims are what Bafang says a motor is good for, not what it produces with a Radpower controller.

Are the Bosch and Brose mid drives measured this way?

FWIW my butt dyno found perfect agreement with the way all of the Specialized, Trek, Gazelle, etc. bikes are with regards to advertised torque spec.

 

[spokewrench]

Are the Bosch and Brose mid drives measured this way?

FWIW my butt dyno found perfect agreement with the way all of the Specialized, Trek, Gazelle, etc. bikes are with regards to advertised torque spec.

In conjunction with the Radpower founder, EBR tested the Radrunner's power. The reviewer pointed to a sign saying hills in that direction were as high as 13 degrees (23%). He said he would use throttle only, because his knees were wrecked. On a hill he passed two staff riders using PAS, going about 2 mph faster than them. By their cadence I calculated their speed at 13 mph and his at 15.

By a camera shot of a house along the street, I calculated the grade at 17%. He given his weight as 140 pounds, and the bike weighed 60. Climbing alone at that speed required 1070 watts against the pavement. That was more than twice what I was getting. In addition, he'd said he'd reduced tire pressure for comfort. I knew from experience that you had to reduce it to half, 15 psi, to notice an improvement, and at that pressure, pedaling was difficult. I'll call it 150 watts at 15 mph. There didn't seem to be a breeze. With no tailwind, air drag would have meant another 100 watts at that speed. That's 1320 watts on the pavement, almost 3 times what I was getting. The motor may have been 75 to 80% efficient at that speed, using 1650 to 1760 watts. At 48 volts, that would be 34 to 37 amps.

It appears that Radpower had used a 16 amp controller so they could claim it was a 750 watt motor because it used 750 watts, and they demonstrated one with a 35 amp controller. If I wanted a Radrunner like the one they'd tested, I'd have to pay a third party, and I'd lose my warranty.

I don't think Bosch and Brose test their mid drives with aftermarket controllers. The combination of sprocket wheels chosen by the bike manufacturer would determine what your butt dyno feels.

 

[m@Robertson]

@Jeremy McCreary says 34 gear inches is too high for some grades he pedals up. He must be working at speeds below 10 mph. A 250 W mid drive might do better than a 750 W hub if, as @Stefan Mikes says, the chainwheel is suitable. When I encountered a steep grade on soft ground with a heavily loaded Abound, I needed torque on the wheel, not speed. In that case, my direct drive hub motor was adequate.

I dropped my car (what we call a station wagon here in the States... an Estate everywhere else) to the shop to get some work done. I had that little red bike I pictured in the earlier post in the back so I could ride home. A single BBSHD motor with a middling 12ah 52v battery pack.

My normal ride home includes a 16% grade, and since I don't ride the red bike so often, I paid attention to my speed/gearing. On pedal assist level 5 (of 9) and the fifth cog up from the bottom (a 21) which gives me dead straight chain line, I traveled between 6 and 7 mph pedaling, which was easy and relaxed (probably 40-50) on that 16% segment. To see what I was leaving on the table, I hit the throttle halfway up and jumped right up to 16 mph. It had more in it - I was still accelerating - but I ran out of hill.

Gears were 40 in the front and 21 in back. Bikecalc tells me I was running 51 gear inches, and I still had 4 bigger cogs that I didn't need to try and tap into (24, 30, 37 and 46, all of which I can get to with the Box 2 derailleur that can go up to 51T). So I had loads of surplus power untapped. My cadence up that hill was fast and while I don't ghost pedal, this was my first decent ride in over a week after a light injury so I kept the workload light.

At a stoplight further on I forgot I wasn't riding a big cargo bike today and used throttle when the light turned green... On that 21T rear cog I still popped a wheelie right up off the ground and had to let off or lose it. And I weigh 235 lbs. If instead I was on say the 37T cog ...

 

[m@Robertson]

Because most bikes sold are wildly geared too high for normal people. And electric bikes make this even worse.

ABSOLUTELY this. Storebought bikes flat out suck when it comes to gearing. I think this is to cater to people's perception that if the bike is not fast on flat ground, it sucks somehow. So they gear it to go fast on flat ground. I saw a lot of first time mid drive riders in the Sondors world take the already overgeared Rock Star and want bigger chainrings on it. To make it go fast. Which also has the effect of creating chainring tacos, bogged motors etc.

When doing a bike build, fussing to get the right gearing is a huge deal. And I seldom got it right the first time on paper. Usually I had to actually ride the thing to shake out what the lab results said I should experience versus what actually riding the bike in the real world showed me.

The BBSHD: Musical Chainrings

It seems inevitable. When I build a bike, I go through front chainrings trying to get the gearing just to my liking. My Mongoose Envoy build has pretty much set the world record for tweaks in thi…

talesontwowheels.com

I wrote that back in 2020 and am still only marginally better at gearing a bike right on the first try without riding it. One thing I learned though is how to not have any of the alignment wear/tear issues I noted there.

What's too high in gear inches?

I'll argue you should ignore gear inches. Its a unit of measure everyone swore by... decades ago. In the modern world we have much better tools that give much more detailed performance answers. In particular when building a bike I would spend a lot of time with this table

Bicycle Gear Speed Cadence Table | BikeCalc

Calculate cycling speeds at a specified cadence given chainring, sprocket, tire and wheel sizes.

www.bikecalc.com

and this one.

Bicycle Gear Cadence Speed Table | BikeCalc

Calculate cycling cadences at a specified speed given chainring, sprocket, tire and wheel sizes.

www.bikecalc.com

Using these you can give yourself a very near idea of how the bike will behave across a range of its gears, taking into account wheel and tire size. It is still not perfect, but its going to get you a whole lot closer to a correct result than trying to picture gear inches, which even if you master them which is not hard to do, will still not tell you speed at cadence, or cadence at speed which I would argue are actual end results and what you really want to know.

ya I realized it was not level it was hard to stand sideways on that slope and get it accurate.

I have a clinometer app for my phone that does a pretty good job. You just lay it on the ground and read what it says. But such a short sample of a hill is bound to have a pretty decent margin of error.

 

[m@Robertson]

160 vs 58 nm is worlds apart. My two bikes are 75 and 50 nm and they are night and day in terms of climbing. Another 90 nm bike I had was again significantly better than the 75. Do you know how manufacturers typically measure torque for spec sheets?

Its all over the place and nobody is giving a straight answer. Your two bikes tell you they are 75 and 50, but is that torque at the motor axle or to the ground at the rear wheel, after its been run thru the drivetrain? This is exactly like the horsepower numbers you see on automobiles. The manufacturer tells you one hp and torque number ... at the crank. But get onto a dyno and you will get RWHP and RWTQ, which any hot rodder knows are the real numbers that matter (and need to be measured rather than taking the old-standby rule of thumb of 17% loss thru the drivetrain).

My Bafang fat motors are rated at 80 Nm and I don't believe that, either. Neither should anyone else for ANY motor measurement that hasn't been independently tested. You could argue Area 13 is a seller and thats not independent, but I know the owner well enough to be confident he is doing everything he can to be as consistent as possible in obtaining his results.

Dyno Testing Ebikes and more!

Electric Bike Dyno Testing

area13ebikes.com

Another example. The Priority Current is claimed to have 140 nm, but owner reviews liken it to the 85 nm CX in terms of climbing ability.

See above I think this is exactly like the ebike range numbers you see... if you've been around long enough you know that every one of them is a lie.

IMO the bicycle manufacturer must know this. If that were my product I would reevaluate those numbers and come up with something equivalent to how Bosch/Brose, etc. measure torque.

Who is to say Bosch/Brose is any better than all the rest of the gang of thieves/cutthroats? The engineers know, but the marketing department writes the ad copy.

 

[Stefan Mikes]

Who is to say Bosch/Brose is any better than all the rest of the gang of thieves/cutthroats? The engineers know, but the marketing department writes the ad copy.

All the big mid-drive motor manufacturers first take their mechanical motor peak power, and then divide it by 6.28 rad/s (it is 60 rpm) to get the torque figure. If the number is not satisfying for the e-bike manufacturer marketing department, the latter is advertising any number to make better sales.

For instance, Specialized gives 50, 70, and 90 Nm for their three motor models. In fact, the two weaker models have far higher torque than advertised (68 and 75 Nm). It is done to attract the buyers to choose the most expensive e-bike model. (The strongest motor is indeed 90 Nm).

 

[tomjasz]

(It looks @Jeremy McCreary has Ignored me).

Just one of many...

 

[Mr. Coffee]

ABSOLUTELY this. Storebought bikes flat out suck when it comes to gearing. I think this is to cater to people's perception that if the bike is not fast on flat ground, it sucks somehow. So they gear it to go fast on flat ground. I saw a lot of first time mid drive riders in the Sondors world take the already overgeared Rock Star and want bigger chainrings on it. To make it go fast. Which also has the effect of creating chainring tacos, bogged motors etc.

I think a lot of it is because the bicycle industry is really bad at listening to customers who aren't racers. And bluntly in a lot of cases they don't care about your opinion unless you are a super athlete.

The near-extinction of 2x (and to a lesser extent 3x) drivetrains is a symptom of this problem. A strong, fit, and experienced cyclist can make a narrower range 1x drivetrain work for them. A beginner or weekend warrior will struggle on all but the gentlest hills. Pushing a bike up a hill is not very fun and if your introduction to cycling includes a lot of pushing your bike uphill you aren't very likely to stick with it.

E-bike manufacturers, in a lot of cases, basically have engineered systems where you'll have to push your 50lb bike up the longest, steepest hills where with an acoustic bike at least you're only pushing 25lbs (or maybe quite a bit less).

One reason I go with alternate drivetrains like Pinion and Rohloff is at least there you can get a decent gear range. I was working with a LBS on a modified Surly Bridge Club build for a friend and we got totally stuck on finding a suitable 2x drivetrain, and it was a deal breaker for her to not have one. Which is a shame because that is an otherwise excellent all-around bike at a reasonable price.

Another bike I really kind of like is the Otso Fenrir, but it suffers from a similar issue with respect to drive trains and gearing.

 

[m@Robertson]

Another bike I really kind of like is the Otso Fenrir, but it suffers from a similar issue with respect to drive trains and gearing.

Interestingly, bikes like the Fenrir are oddly suited to a mid drive conversion oftentimes. Look at the straight downtube and thats ideal for an external motor to tuck up against. But what sticks out much more to me is the solid drive-side root on the chainstay, which lets an ultra-small front chainring tuck in close. Its also incidentally perfect to let a secondary gear housing snuggle up close, too. I have one bike - my ti fat Lynskey/Chumba - where I was able to work a miracle thanks to that chainstay root. I know from helping someone build one that the Marin Pine Mountain benefits similarly.

So while the death of 2x/3x is something I hear a lot of complaints about as well, it can often benefit those of us who have given in to the dark side.

 

[spokewrench]

All the big mid-drive motor manufacturers first take their mechanical motor peak power, and then divide it by 6.28 rad/s (it is 60 rpm) to get the torque figure. If the number is not satisfying for the e-bike manufacturer marketing department, the latter is advertising any number to make better sales.

For instance, Specialized gives 50, 70, and 90 Nm for their three motor models. In fact, the two weaker models have far higher torque than advertised (68 and 75 Nm). It is done to attract the buyers to choose the most expensive e-bike model. (The strongest motor is indeed 90 Nm).

I don't understand. I've never used a dynamometer, but I've read that the old ones measured torque with an adjustable brake with a spring attached to show how hard it was being pulled. Watching a tachometer, an engineer would increase throttle and braking until the brake was holding the full throttle engine at a certain speed. He'd plot that, then repeat the process for a different speed. I've read that nowadays, a computer can simply track the speed of a heavy wheel, calculating torque by its acceleration. If a dynamometer chart has torque and power, I don't know why engineers would have to calculate torque after looking at the power.

I don't understand why they would give the torque at peak power. For that matter, I don't understand Area 13's charts. Their computer dynamometer was made for gasoline engines, and that's what the charts look like. A DC motor's maximum torque coincides with maximum current, and that's at locked rotor, where there is no counter emf. It's a little different with an ebike motor because the controller limits current to a certain maximum, so maximum torque would be flat at low rpms, where the motor produces little power but, because it accepts the most current, consumes the most watts.

Do mid drive controllers make motors perform like gasoline engines? It's easy to estimate the torque of a hub motor. Find a grade it will climb at a crawl. Find your gross newtons (pounds x 4.54) multiply by the sine of the elevation angle. ( I like to call it the percent grade, but the percent is its cousin the tangent.) Multiply that by the axle height as a precent of a meter, and that's your Nm.

On lesser grades, it will speed up until the torque comes down to what the hill requires. In a low-speed situation with my loaded Abound, I needed 86 Nm. It accelerated so fast that I knew I probably had more than 100. On another grade, it accelerated to 17 mph. Multiplying my gross newtons by the grade, I found that I needed 96 Newtons of thrust to climb. Seventeen mph x 0.45 is 7.65 m/s. 96 N x 7.65 m/s = 734 Watts.

That's what the motor was putting on the tire for climbing, but at that speed it also had to fight significant air drag. On page 275, "Bicycling Science, Fourth Edition" says an upright rider needs 345 watts against air drag at 22 mph. It varies as the cube of speed. (17/22)^3 is .46. That would be 159 watts of air drag. Climbing watts and air drag watts would total 893 watts. If the motor was turning fast enough to be 80% efficient, it would have been consuming 1116 watts, with only 223 watts heating the motor.

893 / 7.65 = 117 newtons of thrust. My axle height, or wheel radius, is .27 meter. 117 newtons x .27 meter is 32 Nm of torque coming from the motor, perhaps a third of what it had produced at 5 mph, but 3 times more torque meant 3 times more current, squared to 9 times more heating.