Sensors

Sir Vival

Is there any sensor available that can measure torque?

Say for example i wanted to measure the torque on a wheel of a wheelchair as the person is applying force to it.

DublinDilbert

Sounds like you want to measure force at a fix radius from the centre of the wheel, hence you should be able to calculate the torque. A load cell will do this.

Does the wheel have to rotate freely?

If you know the weight of the person, wheel diameter ect, you should be able to calculate the torque required to move the chair

Sir Vival

DublinDilbert wrote: »

Sounds like you want to measure force at a fix radius from the centre of the wheel, hence you should be able to calculate the torque. A load cell will do this.

Does the wheel have to rotate freely?

If you know the weight of the person, wheel diameter ect, you should be able to calculate the torque required to move the chair

I haven't an engineering backround so excuse my answer if it's a bit uneducated.
I would like to be able to calculate/measure instantaneous torque as the person moves the wheel chair so i guess, yes the wheel has to rotate freely.

DublinDilbert

So you want to measure the force being applied to the rail by the persons hand. If you could set the rail up on a pivot at the wheel centre, a load cell at one point in the rail would work.

You will probably need some wireless data logger...

Sir Vival

DublinDilbert wrote: »

So you want to measure the force being applied to the rail by the persons hand. If you could set the rail up on a pivot at the wheel centre, a load cell at one point in the rail would work.

You will probably need some wireless data logger...

Yes that's what i would like to measure. I'm trying to work out a system that would output instantaneous power as the wheelchair is being moved so i need torque and angular velocity.

I'll definitely research the load cell and some kind of a data logger and see what i can come up with.

Thanks.

DublinDilbert

So you want to measure the force being applied to the rail by the persons hand. If you could set the rail up on a pivot at the wheel centre, a load cell at one point in the rail would work.

You will probably need some wireless data logger as it has to rotate..

bnt

Just a thought: if you could measure the acceleration of the wheelchair somehow - e.g. by analysing footage from a high speed camera - you could then work out the force being applied to accelerate it (F=m a), and then the torque (T = F r), where r is the radius of a wheel. This would give you a total net figure after friction losses etc.

Turbulent Bill

bnt wrote: »

Just a thought: if you could measure the acceleration of the wheelchair somehow - e.g. by analysing footage from a high speed camera - you could then work out the force being applied to accelerate it (F=m a), and then the torque (T = F r), where r is the radius of a wheel. This would give you a total net figure after friction losses etc.

I'd suggest a mash-up of this and DublinDilbert's idea. Put two wireless accelerometers on fixed points of the wheelchair (near to the wheel hubs if possible) - these should be pretty cheap. Interrogate these at a reasonably high sample rate and you get an acceleration profile of the person+chair, and measuring the total mass gives the force. Torque is then calculated as above.

Sir Vival

Thanks for all the input guys. All of yer ideas are really giving me food for thought.

After some thinking, I would prefer to be able to measure the torque on the wheel using a sensor preferably (rather than calculating it using the force). I understand a dynamometer can be used to measure torque so am wondering if there is any sensor that basically does the job of a dynamometer and outputs an electrical quantity in either analog or digital form?

Turbulent Bill

Sir Vival wrote: »

Thanks for all the input guys. All of yer ideas are really giving me food for thought.

After some thinking, I would prefer to be able to measure the torque on the wheel using a sensor preferably (rather than calculating it using the force). I understand a dynamometer can be used to measure torque so am wondering if there is any sensor that basically does the job of a dynamometer and outputs an electrical quantity in either analog or digital form?

The difference here though is that the wheelchair wheel can rotate freely on its axle, rather than being rigidly connected to a driveshaft (as in a car). From what I remember a dyno loads the engine to measure the torque - fine for a steady state engine, not great for the person in the chair!

DublinDilbert

Ummm you have to ask yourself where the torque applied to the wheel comes from. This comes from the persons hand applying a force to the circular 'hand rail'.

It is not normally possible to measure the torque applied from the hand rail to the wheel as they are fixed/screwed together. You need to change how the hand rail is fixed to the wheel. You might be able to get an axial torque sensor, attach one side to the
Wheel centre and other side to hand rail. The other option is to use a bearing as I've explained in my earlier post.

Get it to work with wires first before making it wireless! Also you can test the system by hanging known weights off the rail.

Sir Vival

Turbulent Bill wrote: »

The difference here though is that the wheelchair wheel can rotate freely on its axle, rather than being rigidly connected to a driveshaft (as in a car). From what I remember a dyno loads the engine to measure the torque - fine for a steady state engine, not great for the person in the chair!

Ok I think the picture is getting a bit clearer in my head. Thanks again

Ummm you have to ask yourself where the torque applied to the wheel comes from. This comes from the persons hand applying a force to the circular 'hand rail'.

It is not normally possible to measure the torque applied from the hand rail to the wheel as they are fixed/screwed together. You need to change how the hand rail is fixed to the wheel. You might be able to get an axial torque sensor, attach one side to the
Wheel centre and other side to hand rail. The other option is to use a bearing as I've explained in my earlier post.

Get it to work with wires first before making it wireless! Also you can test the system by hanging known weights off the rail.

Wires aren't a problem for the moment anyways.

Why i'm so intent on wanting to measure the torque directly is that I would like to display the torque and power output of the wheelchair in real time as the person is moving in it. (power = torque * ang. velocity). Would it be possible to measure the torque on the axel?

Again i'm not from an engineering backround so if i'm not making much sense, i apologise.

DublinDilbert

There is no torque on the axel, it's simply just a rigid piece of metal.

Just to confirm we are talking about a manual wheel chair??? If it's electric the motor current will be directly proportional to the torque.

I think you'll need to do some reading up on what torque is and how it's transferred. We'll do our best to help too...

Sir Vival

There is no torque on the axel, it's simply just a rigid piece of metal.

Just to confirm we are talking about a manual wheel chair??? If it's electric the motor current will be directly proportional to the torque.

I think you'll need to do some reading up on what torque is and how it's transferred. We'll do our best to help too...

Yep manual wheel chair.

Yeah i think your spot on about the reading up thing. I never really grasped the concept of torque. From wiki's definition: "Torque is the tendency of a force to rotate an object about an axis" therefore i thought it was the torque that was propelling the wheelchair forward (and backwards) once the person applied the force to the hand rail and therefore was a measurable quantity. Obviously my line of thinking is wrong.

I'll come back here once i've had a bit of a think and done a bit more research.

Once again, thanks, i know you've probably better (and more interesting) stuff to be at.

DublinDilbert

Torque is a turning force, but it's typically generated from a linear force.

Think of a spanner tightening a bolt, your hand applys a linear force to the end of the spanner, this is converted to a torque at the bolt. In this case the bolt is the pivot.

Understanding the pivot point is key to understanding torque. Torque is the force applied x distance from pivot, hence the units newton metres, foot pounds ect...

FesterBeatty

To answer your question - yes there is such a machine. Any good motorbike mechanic/workshop will have a dyno and I'm sure they'd let you throw a wheelchair on to undertake your experiment (for a fee no doubt!). At the end you'll get a graph which will indicate torque and horsepower.

http://www.howstuffworks.com/horsepower1.htm

Check out biker.ie - might be a few lads willing to provide you with a dyno in the interests of science!

Alternatively (cheaper!) have a look at an electric wheelchair supplier's technical specification. This should give the motor's peak torque which will give you a ballpark starting point for whatever it is you're designing

Turbulent Bill

Sir Vival wrote: »

Why i'm so intent on wanting to measure the torque directly is that I would like to display the torque and power output of the wheelchair in real time as the person is moving in it. (power = torque * ang. velocity).

This is probably a better way of looking at the problem, i.e., what you want to achieve rather than how to do it. Torque measurement on the wheel itself is likely to be tricky, but I don't think you need it to get power. It boils down to getting the wheelchair's linear acceleration and calculating force, power etc. from this.

Even something like an off-the-shelf bicycle computer would give velocity and distance for each wheel, and with some hacking might also give linear acceleration (i.e., dv/dt). Presuming the ground is flat and rolling resistance is fairly small you could assign almost all the acceleration to the user's applied force.

mawk

I've never heard of a wheelchair being put on a rolling road, but I'd like to see it..

FesterBeatty

[QUOTE
Why i'm so intent on wanting to measure the torque directly is that I would like to display the torque and power output of the wheelchair in real time as the person is moving in it. [/QUOTE]


To me, as an engineer, I believe that necessity is the mother of all invention. But why is it necessary for a wheelchair user to know the magnitude of the torque being produced (if you dont mind me asking)? I dont understand..