Torque rating on E-bikes - marketing gimmick?

[Ravi Kempaiah]

I have been studying E-bike drive systems for sometime now. I find it amusing that so many people just go by what they read in the brochures. Especially, the Torque ratings!
Yet the subjective experience is nowhere near to the corresponding values. To put in crude, vulgar terms, It is like you're courting a girl and you decided to pursue her because this girl has 32B, 36DD.. etc

Torque rating is just one of the parameters and how the overall system is designed is much more important than one single parameter. The algorithm, controller design, power draw etc matter more than just "Torque". A system could have more torque but at what expense? power? less speed?

Bosch - 60Nm and now upped to 70Nm.
Derby Impulse 2.0 - 80Nm (newer RS versions)
TranzX mid drive - 73Nm
BBS-02 - 120Nm
M1 Pin Drive - 120Nm
Dapu - 40Nm
Currie electro drive - 42Nm
Stromer SYNO drive - 35Nm

I could go on...

Why is it misleading?

I have called tech people and even those who design the drive systems to find out how they measure these Torque ratings and the response was appalling. They have no clue!!

Is this measured at the wheels?
at the derailleur? (if so, is it amplified because of gears?)
at the hub?

Out of all mid-drives I have tested only two systems stand out in terms of Torque ( BBS-02 and M1 Pin dirve), rest have been very subdued experience but the torque sensor on some of these mid-drives suck!

You run any of the mid-drives at 23mph on a slight incline (3-4% grade) and see the range! and compare it to efficient hub drives. You won't notice any difference and in fact, given the specs, you'll see far less range than advertised.

Mid-drives supposedly get more range because they make you work like donkey. The mid-drives don't over exert and are extremely conservative in terms of power output (exception of few mid-drives I mentioned above).

So, all I can say is -

Beware of bombastic sounding blowhard people (if it's me, so be it). Is the claim backed up with some real data that can be verified? Is that a credible source?
Most importantly , test drive several bikes. Take it out for extended 24 hour test drives and put it through some vigorous testing. It's worth it. You'll learn a lot and not yield to marketing gimmicks.

Anyhow, the point is to educate yourself. At the end of the day, all you want is reliable E-bike to do 30 mile rides? most bikes would suffice. If you're a geek like me, you'll want to tear it apart and keep learning

Enjoy

 

[George S.]

@Ravi Kempaiah said:

Most importantly , test drive several bikes. Take it out for extended 24 hour test drives and put it through some vigorous testing. It's worth it. You'll learn a lot and not yield to marketing gimmicks.

Alas, you probably mean ride it for 24 hours. Not a test I will be doing. Honestly, it seem to be easier to find a good bike and then put a good kit motor on it.

I think reviewers could get much more solid numbers about things like hill climbing ability. Maybe a table showing the grade and rider weight. Pete has a course, more or less, with hills. You could probably figure out some number for a given bike and then compute some kind of graph for range at various speeds, boiled down to some constant watt hours. I don't see any reason people can't get a pretty complete profile of bikes that get reviewed. Maybe reviewers need a set of standards and then let the chips fall where they fall.

All people talk about is speed, which invariably kills range. The industry wanted mid-drive and now it wants speed pedelec. New riders seem much more motivated by the speed ebikes. A bike should competently climb any hill you normally encounter. That seems obvious.

 

[Deleted member 803]

Torque application can be supplied in so many ways. Mid-drives have strong torque at low speeds when used with lower gears for climbing. Rear hubs have torque applied from a stand still and typically can outblast a mid drive to speed (M1 bikes being the exception). Mid-drives, as you all know, are very gear specific to get the most out of the motor. For example, motorcycle cruisers do not typically have very high torque ratings but ALL of the available torque is applied from 0-40mph. Cruisers yank the hell out of you at a stop sign. Whack the throttle at 70-80 mph on the freeway and most cruisers will be very anemic. Sport bikes have much higher overall torque and the torque curve is typically flatter. A powerful sport bike will wheelie from a dead stop but will accelerate form 70mph to 130mph like a bat out of hell.

So, if you like to get up to speed quickly, a rear hub will (in most instances) do a better job. If you do lots of climbing at slow speeds, a mid-drive will serve you better. And, of course if you significantly alter the output or toque you can get a better responding bike no matter where the motor is placed.

My Neo Carbon had a small 350W geared rear hub and delivered the most amount of dead stop yank out of all the bikes I tested and rode. This is partly due to the fact that geared rear hubs are more efficient at delivering power. How the motor is designed also affects the efficiency of power delivery.

 

[bluecat]

Thanks for raising up this topic. I agree with my previous speakers, robust data aren’t available. The chart from my Stromer AG is outdated, but ST1 is the successor of P48. So, it gives an impression on the values:

(Link Removed - No Longer Exists)

I'm convinced, all measurement is done by the motor manufacturers on a test bench using an unlimited external power supply. But at what voltage? The ST2 runs with 48V, but the fully charged battery has 54.6V But how about the amperage? Under load, the ST2 might deal with 20 to 25A. And the rpm vs. efficiency. How will the curve look like, at what rpm in NM max?

Maxon claims for its hub motor 50NM peak and 25NM typical. Brose claims for its mid-drive 90NM; limited to 50 NM to protect cassette and freewheel. Both companies have large track record on building electric motors long before entering e-bike market. So, I believe them.

The drivetrain acts as a pulley tackle. In the low gears, the torque you (and a mid-drive!) gives on the pedal is amplified. If you stand on the pedal, 150NM might be possible. Just sitting and pedaling, it might be a fraction.
But all tech specs are nothing against experience. On the foot of my test-hill: ST2 35NM and ST1/M33 40NM. Additionally, I had removed the speed limit on the M33. The result was clear and consistent over several days: With the ST2 it's far easier.

As Adrian mentioned, Bafang claims 120NM for its hub drive. I couldn't find any data, but from Crystalyte - a competitor of Bafang - I found this chart:

(Link Removed - No Longer Exists)

 

[Deleted member 803]

The most scientific analysis of torque for 99% of all concerned is to simply ride an ebike and determine whether the real world performance provides the performance you desire. I pity the poor folks who are so consumer by the analytical side of their purchase that they fail to sense a very complex and seldom understood emotion: Joy!

 

[Bike_On]

I have been studying E-bike drive systems for sometime now. I find it amusing that so many people just go by what they read in the brochures. Especially, the Torque ratings!
Yet the subjective experience is nowhere near to the corresponding values. To put in crude, vulgar terms, It is like you're courting a girl and you decided to pursue her because this girl has 32B, 36DD.. etc

Torque rating is just one of the parameters and how the overall system is designed is much more important than one single parameter. The algorithm, controller design, power draw etc matter more than just "Torque". A system could have more torque but at what expense? power? less speed?

Bosch - 60Nm and now upped to 70Nm.
Derby Impulse 2.0 - 80Nm (newer RS versions)
TranzX mid drive - 73Nm
BBS-02 - 120Nm
M1 Pin Drive - 120Nm
Dapu - 40Nm
Currie electro drive - 42Nm
Stromer SYNO drive - 35Nm

I could go on...

Why is it misleading?

I have called tech people and even those who design the drive systems to find out how they measure these Torque ratings and the response was appalling. They have no clue!!

Is this measured at the wheels?
at the derailleur? (if so, is it amplified because of gears?)
at the hub?

Out of all mid-drives I have tested only two systems stand out in terms of Torque ( BBS-02 and M1 Pin dirve), rest have been very subdued experience.

You run any of the mid-drives at 23mph on a slight incline (3-4% grade) and see the range! and compare it to efficient hub drives. You won't notice any difference and in fact, given the specs, you'll see far less range than advertised.

Mid-drives supposedly get more range because they make you work like donkey. The mid-drives don't over exert and are extremely conservative in terms of power output (exception of few mid-drives I mentioned above).

So, all I can say is -

Beware of bombastic sounding blowhard people (if it's me, so be it). Is the claim backed up with some real data that can be verified? Is that a credible source?
Most importantly , test drive several bikes. Take it out for extended 24 hour test drives and put it through some vigorous testing. It's worth it. You'll learn a lot and not yield to marketing gimmicks.

Anyhow, the point is to educate yourself. At the end of the day, all you want is reliable E-bike to do 30 mile rides? most bikes would suffice. If you're a geek like me, you'll want to tear it apart and keep learning

Enjoy

Ravi,

I don't think torque ratings are a gimmick, but they can be misunderstood and overplayed.

If you check out ebikes.ca simulator for hub motors, most dd hubs give 40-50nm at 0 rpm, and decay to 20nm nominal. Geared hubs can start at 100+ nm.
Mid-drives get more because they have more gear reduction than even the geared hubs, thus Optibike advertises 135-165nm.

These numbers are all about INITITAL torque at 0 mph/rpm speeds. Looks at the ebikes.ca curves, they all decay with speed.

 

[bluecat]

Looks at the ebikes.ca curves, they all decay with speed.

Yes, as seen in the chart above.

But how do you explain the complete different values my Stromer AG published?

 

[Cameron Newland]

Shouldn't mid-drives have flatter torque curves when measured at the crank instead of steeply declining ones, and declining torque curves when measured at the wheels (due to the effects of gearing)? Are all these torque curve graphs measuring torque at the wheels?

The Tesla Model S's torque curve doesn't decline nearly as quickly as ebike motors' do:

Obviously, comparing 350W/500W ebike motors to a 300+kW Tesla is comparing apples to oranges, but I'm curious to know why they differ so much.

I'm no engineer, obviously.

 

[bluecat]

I'm no engineer, obviously.

Me too.

The amperage is limited to 1200A until the point where the declining begins. Otherwise, the curve might look similar.

I'm convinced, all torque measurement is done by the motor manufacturers on a test bench using an unlimited external power supply (except the example above).

 

[Bike_On]

The Geek Skinny:
from: http://electronics.stackexchange.co...ated-to-torque-and-speed-of-a-brushless-motor

The relationship between a motor's electrical characteristics and mechanical performance can be calculated as such (note: this is the analysis for an ideal brushed DC motor, but some of it should still apply to a non-ideal brushless DC motor).
A DC motor can be approximated as a circuit with a resistor, and voltage back-emf source. The resistor models the intrinsic resistance of the motor windings. The back-emf models the voltage generated by the moving electric current in the magnetic field (basically a DC electric motor can function as a generator). It's also possible to model the inherent inductance of the motor by adding an inductor in series, however for the most part I've ignored this and assumed the motor is at quasi steady state electrically, or the motor's time response is dominated by the time response of the mechanical systems instead of the time response of the electrical systems. This is usually true, but not necessarily always true.
The generator produces a back EMF proportional to speed of the motor:
Vemf=k i ∗ω
Where:
k i is a constant.
ω is the motor speed in rad/sec
Ideally at stall speed there is no back emf, and at no the no-load speed the back emf is equal to the driving source voltage.
The current flowing through the motor can then be calculated:
I=(VS−Vemf)/R=(VS−k i ∗ω)/R
VS=source voltage
R is the motor electrical resistance
Now let's consider the mechanical side of the motor. The torque generated by the motor is proportional to the amount of current flowing through the motor:
τ=k t ∗I
k t is a constant
τ is the torque
Using the above electrical model you can verify that at the stall speed the motor has the maximum current flowing through it, and thus the maximum torque. Also, at the no load speed the motor has no torque and no current flowing through it.
When does the motor produce the most power? Well, power can be calculated one of two ways:
Electrical Power:
Pe =VS∗I
Mechanical Power:
Pm =τ∗ω
If you plot these, you'll find that for an ideal DC motor the maximum power comes at half the no-load speed.
So all things considered, how does the motor voltage stack up?
For the same motor, ideally if you apply double the voltage you'll double the no-load speed, double the torque, and quadruple the power. This is assuming of course the DC motor doesn't burn up, reach a state which violates this simplistic ideal motor model, etc.
However, between different motors it's impossible to tell how two motors will perform compared to each other based only on the voltage rating. So what do you need to compare two different motors?
Ideally you'd want to know the voltage rating and stall current so you can design your electronics appropriately and you'd want to know the no-load speed and stall torque so you can calculate the mechanical performance of your motor. You may also want to see the current rating of the motor (some motors can be damaged if you stall them for too long!). This analysis also somewhat neglects the efficiency aspect of the motor. For a perfectly efficient motor, ki = kt, or rather Pe = Pm. This would cause the power calculations using the two equations to be equal (i.e. electrical power equals mechanical power). However, real motors aren't perfectly efficient. Some are close, some aren't.
p.s. In my calculations I used motor speed as rad/sec. This can be converted to Hz or rev/sec by dividing by 2*pi.

 

[Ravi Kempaiah]

But at what voltage? The ST2 runs with 48V, but the fully charged battery has 54.6V But how about the amperage? Under load, the ST2 might deal with 20 to 25A. And the rpm vs. efficiency. How will the curve look like, at what rpm in NM max?

I think THIS here is the key! You nailed it.
Understanding the "Dyno curves" gives you nice insight into how the system works. One may have all the right ingredients (motor, battery, controller) but the dish may still turn out to be different and this largely depends on the how engineers set up the power delivery via controller software.
Tesla has a whole power electronics division to handle this. The AC motor used is the same 100+yr old motor designed by Nikola Tesla but very evolved of course.

So, in short, how will the curve look at different RPM's, loads and of course, how's the controller designed to handle all this + shifting?

 

[bluecat]

I expect, the chart for the ST2 is similar to the old one published above. The controller limits the amperage all the time. This makes the torque curve more flat. There is no motor support from 0km/h to protect the system. Furthermore, the ST2 motor is a complete new construction (amount of poles), so I expect maximum efficiency around 40km/h. Additionally, the engineers stated, the construction is “sharp”, so efficiency falls dramatically above 46km/h.

Note: With the tech. explanations from Bike_On (which I don't understand ;-) it becomes clear, that the statement from Adrian was full ironic. He accuse Stromer advertising gives non-real promises, but for his Bafang he claims about 4 times the values of ST1…

 

[Bike_On]

Short answer:
Yes, torque depends on the current, and thus the limit of the controller.

IF the design allows 30A peak from a stall, you will get a lot of torque and pound battery life and use up Ahrs.

Torque is also is limited by motor winding resistance and battery voltage. I = V/R. 48V is > 36V, thus more juice and more torque @ 48V for a given motor, unless limited by controls.

 

[Bike_On]

Shouldn't mid-drives have flatter torque curves when measured at the crank instead of steeply declining ones, and declining torque curves when measured at the wheels (due to the effects of gearing)? Are all these torque curve graphs measuring torque at the wheels?

The Tesla Model S's torque curve doesn't decline nearly as quickly as ebike motors' do:

Obviously, comparing 350W/500W ebike motors to a 300+kW Tesla is comparing apples to oranges, but I'm curious to know why they differ so much.

I'm no engineer, obviously.

The original Tidalforce and E+ torque response was like the curve above. 85nm flat, out to a given rpm, then decay. It is a function of the controller, and it uses losts of Ahrs to do it with a full throttle response.

 

[kanakaraju]

For brush less dc motors motor manufacturers generally specify
motor torque constant NM/amp
motor voltage constant volts/rpm
nominal speed
nominal torque

 

[Trail Cruiser]

For brush less dc motors motor manufacturers generally specify
motor torque constant NM/amp
motor voltage constant volts/rpm
nominal speed
nominal torque

It's confusing to compare hub drives vs the mid drives.

On hub drives, it's a given that torque is measured directly at the wheel, whereas, on mid drives, torque is measured at the crank arm. So how do you compare the two?

I suppose, you can up gear the mid drive at the wheel to have an equal resultant wheel torque and then compare the two at that specific gear.

Let's take for example a 20 mph Dapu hub motor putting out 40 NM torque and compare it to, let's say, Impulse 2.0, putting out 80Nm (Focus Aventura). In order for the Impulse to have comparable torque at the wheel to the Dapu, it has to be geared up to 1:0.5, or 38T chainring to 19T sprocket. That would be roughly the 5th gear of the 9 speed Focus Aventura (with a Shimano 11-34T, 9 speed Cassette).

The starting torque at the wheel for both Dapu and Focus Aventura (in 5th gear position) are now equal (40 NM) but the Dapu can easily power up to 20 MPH regardless of what gear you are in. However, the Focus in 5th gear can only attain 13-14 MPH (at 85 crank RPM). The Focus has to shift up to at least 8th gear to reach 20 MPH. But then the wheel torque at 8th gear also drops inversely to 24 NM.

On the other hand, the Focus has higher wheel torque (better climber) than Dapu if it is geared below 5th gear (1st gear to 4th gear) in exchange for slower road speed.

The hub motor has the advantage of simplicity, ease of development, ease of operation, raw power, and speed.

The mid drive allows for the use of smaller lighter and energy efficient motor, has the potential to have better battery range, better climber but at lower speed, requires more engineering R&D to make it operate properly to match and supplement the rider's input.

 

[McSpiffy]

I have been studying E-bike drive systems for sometime now. I find it amusing that so many people just go by what they read in the brochures. Especially, the Torque ratings!
Yet the subjective experience is nowhere near to the corresponding values. To put in crude, vulgar terms, It is like you're courting a girl and you decided to pursue her because this girl has 32B, 36DD.. etc

Torque rating is just one of the parameters and how the overall system is designed is much more important than one single parameter. The algorithm, controller design, power draw etc matter more than just "Torque". A system could have more torque but at what expense? power? less speed?

Bosch - 60Nm and now upped to 70Nm.
Derby Impulse 2.0 - 80Nm (newer RS versions)
TranzX mid drive - 73Nm
BBS-02 - 120Nm
M1 Pin Drive - 120Nm
Dapu - 40Nm
Currie electro drive - 42Nm
Stromer SYNO drive - 35Nm

I could go on...

Why is it misleading?

I have called tech people and even those who design the drive systems to find out how they measure these Torque ratings and the response was appalling. They have no clue!!

Is this measured at the wheels?
at the derailleur? (if so, is it amplified because of gears?)
at the hub?

Out of all mid-drives I have tested only two systems stand out in terms of Torque ( BBS-02 and M1 Pin dirve), rest have been very subdued experience.

You run any of the mid-drives at 23mph on a slight incline (3-4% grade) and see the range! and compare it to efficient hub drives. You won't notice any difference and in fact, given the specs, you'll see far less range than advertised.

Mid-drives supposedly get more range because they make you work like donkey. The mid-drives don't over exert and are extremely conservative in terms of power output (exception of few mid-drives I mentioned above).

So, all I can say is -

Beware of bombastic sounding blowhard people (if it's me, so be it). Is the claim backed up with some real data that can be verified? Is that a credible source?
Most importantly , test drive several bikes. Take it out for extended 24 hour test drives and put it through some vigorous testing. It's worth it. You'll learn a lot and not yield to marketing gimmicks.

Anyhow, the point is to educate yourself. At the end of the day, all you want is reliable E-bike to do 30 mile rides? most bikes would suffice. If you're a geek like me, you'll want to tear it apart and keep learning

Enjoy

I have been studying E-bike drive systems for sometime now. I find it amusing that so many people just go by what they read in the brochures. Especially, the Torque ratings!
Yet the subjective experience is nowhere near to the corresponding values. To put in crude, vulgar terms, It is like you're courting a girl and you decided to pursue her because this girl has 32B, 36DD.. etc

Torque rating is just one of the parameters and how the overall system is designed is much more important than one single parameter. The algorithm, controller design, power draw etc matter more than just "Torque". A system could have more torque but at what expense? power? less speed?

Bosch - 60Nm and now upped to 70Nm.
Derby Impulse 2.0 - 80Nm (newer RS versions)
TranzX mid drive - 73Nm
BBS-02 - 120Nm
M1 Pin Drive - 120Nm
Dapu - 40Nm
Currie electro drive - 42Nm
Stromer SYNO drive - 35Nm

I could go on...

Why is it misleading?

I have called tech people and even those who design the drive systems to find out how they measure these Torque ratings and the response was appalling. They have no clue!!

Is this measured at the wheels?
at the derailleur? (if so, is it amplified because of gears?)
at the hub?

Out of all mid-drives I have tested only two systems stand out in terms of Torque ( BBS-02 and M1 Pin dirve), rest have been very subdued experience.

You run any of the mid-drives at 23mph on a slight incline (3-4% grade) and see the range! and compare it to efficient hub drives. You won't notice any difference and in fact, given the specs, you'll see far less range than advertised.

Mid-drives supposedly get more range because they make you work like donkey. The mid-drives don't over exert and are extremely conservative in terms of power output (exception of few mid-drives I mentioned above).

So, all I can say is -

Beware of bombastic sounding blowhard people (if it's me, so be it). Is the claim backed up with some real data that can be verified? Is that a credible source?
Most importantly , test drive several bikes. Take it out for extended 24 hour test drives and put it through some vigorous testing. It's worth it. You'll learn a lot and not yield to marketing gimmicks.

Anyhow, the point is to educate yourself. At the end of the day, all you want is reliable E-bike to do 30 mile rides? most bikes would suffice. If you're a geek like me, you'll want to tear it apart and keep learning

Enjoy

Who is this girl with 36DD? (JK)

 

[smitty]

The most scientific analysis of torque for 99% of all concerned is to simply ride an ebike and determine whether the real world performance provides the performance you desire. I pity the poor folks who are so consumer by the analytical side of their purchase that they fail to sense a very complex and seldom understood emotion: Joy!

Right on 86...enjoy learning all about torque but much more so applying whatever about is available...

 

[Deleted member 4210]

Ravi. Great Points !!!

As he says, you have to ride the ebikes to compare and notice exactly what the differences are. The problem with going by torque alone, is that it varies by RPM, and NO ONE tells you what RPM they measured it at. Do you look at the torque rating on a car engine, when you go to buy it ? Of course not. The specs are there for geeks, but at least car engine manufacturers can provide charts to show how torque varies over the RPM curve. So as a stand alone data point, just freaking ignore it. Its meaningless drivel, and provides no basis whatsoever for comparison, again bc e-Bike OEM's don't tell you what RPM its being measured at. I'd also be willing to bet that 99% of these motors coming out of China don't meet any QA or consistency specs, and that they probably did a one time measurement, or even calculated theoretical torque rating when they first designed the motor 1 or 2 decades ago. Someone like Bosch might actually do bench testing. The others from China ? I wouldn't count on it.

P.S. Degreed mechanical engineer. BSME Purdue University, 1985.

 

[smitty]

spot on...