Torque Sensors

[Ajax]

Hi . I have an idea for a new torque based sensor, but first some background.

As most here already know there are PAS sensors where your cadence is used to activate the power to the motor. There are also torque sensors which act to provide varying leves of power to the motor by sensing the effort placed on pedals. The latest torque sensors are to be found in the rear wheel mount which requires the frame to be adapted.

The other kind of torque system are to be found in the botton bracket. These are either fitted in the crank botton bracket, or they form part of the chain ring assembly. My first ebike was the PowaCycle Salisbury electric bike which feature the chain ring torque assembly. At the time I didn't realise how special this was.

Now for my idea.

It seems to me a cheap retro fitted torque sensor could be designed to work off the tension in the bike's chain. What i have in mind is a device like a sprung chain tensioner. This would have a torque sensor like the TMM4 device in those rear wheel torque systems. This device would be mounted in the area of chain between the Chain ring and the main rear sprocket so that the derailer wasn't a factor on chain tension.

I did a basic search and couldn't find a patent for this, so this idea is now in the public domain.;0) If i could get hold of a torque senor i'd test it myself, but i don't have the means. So my question is are there any resourceful individuals in this group to test this idea? Can anyone see any obvious flaws with this idea?

Btw if someone here does develop a proto type, think of me.

 

[Woosh]

you can already buy very good bottom bracket torque sensors for around $50-$70.
Do you think your idea will lead to a better torque sensor?
Chain tensioners are not good sensors because they are exposed and can be moved by kinks in the chain. If your chain tensioner is falsely triggered, the bike will behave dangerously.

 

[cyclebuddy]

...If your chain tensioner [torque sensor] is falsely triggered, the bike will behave dangerously.

Oddly, I've experienced a similar problem with my mid-drive torque sensing Bafang MaxDrive…

Coming to a T-junction at the end of a road, on a slight incline upwards, with the bike totally stationary: One foot on the ground, and the other foot lightly held on the opposite pedal (just to hold the weight of the bike waiting for a clear road to turn onto), and the bike lunges forward!

There are no brake sensors (Shimano hydraulic), so even if applied firmly, that wouldn't help by cutting power to the motor whilst stationary (although that obviously does limit the lunge by holding the bike mechanically).

Does this "lunging" happen to all torque sensing e-bikes in such a situation, just mid/crank-drives generally, just the MaxDrive, or just MY MaxDrive I wonder?

It can be a bit scary at first, but now I'm aware of it, I'm very careful about applying any pedal pressure until I want to move!

 

[Nev]

Does this "lunging" happen to all torque sensing e-bikes in such a situation, just mid/crank-drives generally, just the MaxDrive, or just MY MaxDrive I wonder?

It can be a bit scary at first, but now I'm aware of it, I'm very careful about applying any pedal pressure until I want to move!

I have had similar experiences but not quite as bad as yours with the bosch CX motor, but only when using turbo mode. So now I either do what you do ie don't apply any pedal pressure until I want to move, or more usually make sure I am not in turbo mode when stopped at a junction.

 

[Woosh]

Coming to a T-junction at the end of a road, on a slight incline upwards, with the bike totally stationary: One foot on the ground, and the other foot lightly held on the opposite pedal (just to hold the weight of the bike waiting for a clear road to turn onto), and the bike lunges forward!

it shouldn't do that. Did you wobble on the saddle while resting your foot on the pedal?
You have to push firmly on a hill start to start the motor. If it happens again, get it checked out.

 

[Ajax]

you can already buy very good bottom bracket torque sensors for around $50-$70.
Do you think your idea will lead to a better torque sensor?
Chain tensioners are not good sensors because they are exposed and can be moved by kinks in the chain. If your chain tensioner is falsely triggered, the bike will behave dangerously.

It wouldn't be any more exposed than the TMM4 sensors used in those rear mounted torque sensors. It would be sensing the tension in the chain, with perhaps a smothing algorithmn to even out any fluctuations in readings.

I was thinking if this were coupled with a PAS, you could have it so its not triggered until you start to padel, as is the law , then its function would be to regulate the hub motor power against effort or chain tension. It should also be a lot cheaper to produce and install than the botton bracket kit, although i am looking into that.

Example of Bottom Bracket torque sensor.

 

[Woosh]

the cheapest solution is this one, made by XOFO and fitted into my Woosh Sport TS. It's built inside the motor.

 

[Ajax]

the cheapest solution is this one, made by XOFO and fitted into my Woosh Sport TS. It's built inside the motor.

http://wooshbikes.co.uk/?sport-ts

Quote -

" A throttle is convienient - but can strain your wrist and not encourage you to cycle.
The torque sensing technology in the Sport TS relies on a set of 4 coil springs built into the rear hub.
As you push down on the pedals, the springs are compressed, a set of Hall sensors measure the displacement and outputs a throttle voltage proprotional to the applied force.
You select the assist level to suit your physical condition and terrain, the electronics will try to maintain a constant human/motor ratio for you.
So if you want to keep cycling as much as possible, a torque sensor will retain that natural feel to cycling. "


I didn't think this was possible, ie to sense cycle 'effort' through the motor. I suppose you could tell something about resistance on the wheel, or how freely it was spinning, by comparing to the current being applied, but would this be the same as the effort by the cyclist.

After reading site [via image search], I'm not sure if by the rear hub, whoosh does this though the motor, or an additional feature of the hub - an inner baring, but its an interesting design for sure.

Ha ha I see it now. I suppose if there were space you could have a simiar device fittted to the freewheel, or come out with a special design of freewheel to incororporte this principle, where the load is transfered indirectly to the wheel via a built-in strain gauge. The problem then is routing the cable which would occupy the same area as the axel / cables used for the hub motor .

 

[Woosh]

the sport-TS uses the normal cadence sensor to stop false triggering. Doing so removes the most desirable feature which is hill start. You have to rotate the cranks 60 degrees before the motor powers up.
TS bikes need sophisticated software to make them natural sensing: guessing correctly hill start and gear shift. Without clever software, cadence sensor wins any day.

 

[wheeliepete]

I wouldn't mind having a play with a BB one, but it seems to me there is not a lot of choice when it comes to a suitable off the shelf controller for TS to work well.

 

[Woosh]

it seems to me there is not a lot of choice when it comes to a suitable off the shelf controller for TS to work well.

the controller is the Lishui one, same as used on most of my bikes, nothing special. The output of the torque sensor is connected to the throttle, the controller is programmed to activate the throttle / torque sensor after pedaling.
As I said before, you need clever software to accurately predict hill start and gear shift.

 

[wheeliepete]

OK, you use the same controller as on your PAS bikes, but the software has been modified to work with the TS, so to me that makes it a different controller. The point I was trying to make was, from the ones I've seen, if you were to buy a BB TS, you are limited to the controller that has been produced or modified to work with it, usually a choice of one.

 

[Woosh]

OK, you use the same controller as on your PAS bikes, but the software has been modified to work with the TS, so to me that makes it a different controller. The point I was trying to make was, from the ones I've seen, if you were to buy a BB TS, you are limited to the controller that has been produced or modified to work with it, usually a choice of one.

the choice depends on the torque sensor and yes, you usually have to ask your supplier to program the controller for you to suit the sensor. Nowadays you just ask for it, all the factories have ready made solutions. For a one off project, buy the XOFO motorr and use a KT controller, it can be programmed by end user for this. I still think that PAS before throttle is simple but not the good way to do it.

 

[Wisper Bikes]

Coming to a T-junction at the end of a road, on a slight incline upwards, with the bike totally stationary: One foot on the ground, and the other foot lightly held on the opposite pedal (just to hold the weight of the bike waiting for a clear road to turn onto), and the bike lunges forward!

I agree with Woosh, this shouldn't happen and can be adjusted with the controller software. The trick is to find the balance between having to push too hard to get going and the bike lurching forward. It took a while but we got there in the end!

 

[cyclebuddy]

I agree with Woosh, this shouldn't happen and can be adjusted with the controller software. The trick is to find the balance between having to push too hard to get going and the bike lurching forward. It took a while but we got there in the end!

TBH, holding a bike on an incline by the pedal is perhaps a riding habit I've carried over from the old days. As Woosh also suggest, I think it's just me being twitchy, and as Nev suggests, maybe I was in too high a power/wrong gear setting too. It makes sense that a torque-sensing e-bike doesn't know if you're just holding the weight steady, or trying to accelerate away. I suspect I just need to be more aware of the torque-sensing nature of the bike and adapt my riding style. Thanks for the input.

 

[Ajax]

TBH, holding a bike on an incline by the pedal is perhaps a riding habit I've carried over from the old days. As Woosh also suggest, I think it's just me being twitchy, and as Nev suggests, maybe I was in too high a power/wrong gear setting too. It makes sense that a torque-sensing e-bike doesn't know if you're just holding the weight steady, or trying to accelerate away. I suspect I just need to be more aware of the torque-sensing nature of the bike and adapt my riding style. Thanks for the input.

Or with your foot on the pedel, you could just hold the brakes until you need to power away.

 

[Woosh]

Or with your foot on the pedel, you could just hold the brakes until you need to power away.

he has no brake sensors on his bike.

 

[Ajax]

he has no brake sensors on his bike.

Geez, it a wonder he's able to stop with a torque sensor under foot,
not to mention the power he's very likely wasting having the motor running when he's also braking.