Mini torque gage 9

[insideman]

I need to find or make a mini torque meter or gage. This will be used to measure the reaction torque of tiny motors driving propellers and ducted fans. Say 10-20 watts output at 10000-20000 rpm.

Very little space is available. About one inch diameter by one inch long to fir inside the fan tailcone. Smaller is better so as to avoid restricting cooling air flow past the motor.

 

[rb1957]

is this an electric fan ? ... why not measure the electric pwoer going into the fan and the fan shaft rpm ?

 

[insideman]

rb1957: This is a model ducted fan for propulsion. Testing will be done in a miniature wind tunnel. Measuring the electricals will not account for losses. I need to measure input volts, amperes, rpm, torque, thrust, air density, air velocity. The one that eludes me is torque. My objective is to improve the fan design, including outlet guide vanes, inlet configuration, exhaust cone diameter, blade configuration in general.

But thanks for the suggestion.

 

[rb1957]

but then measuring the torque probably won't tell you that either. i think you're intersted in the real work done by the fan, ie the change in airspeed across the fan.

or else you can say that with a smaller fan (less electrical power input) you're accomplishing the same cooling effect (and so are more efficient).

 

[CoryPad]

Try here:

http://www.interfaceforce.com/

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[BobM3]

I suppose mounting it in a cradle and letting the body rotate and be reacted by a moment arm (a dyno) would bring too much mechanical hardware into the airstream?

Probably not enough room to mount any sort of standard torque/load cell. Strain gaging the shaft isn't too hard but getting the signal out into the stationary world is tough. Any room for a rotary transformer or telemetry unit?

Something I've wondered about is using two encoders mounted some distance apart on the shaft. The torque would twist the shaft and change the relative phase between the two encoders. Again, probably a custom unit.

 

[sreid]

Here's a link to some folks who measured stress on a spinning gear. Silicon Strain Gages and RF data transmission. The same techniques could be used to measure strain due to torque.

http://www.sensorsmag.com/sensors/T...n-Gauge/ArticleStandard/Article/detail/361651

 

[Griz42]

Maybe you could measure the reaction forces on the motor mount. You could calibrate the output with known torque inputs.

 

[EngineerTex]

Potentially very cheap way of doing it:

1) Mount the fan on a couple of thin wires -- My first guess would be about half the size of a coat hanger. These will be your torque reaction springs.
2) Make sure that both ends of the wires are FIRMLY ATTACHED, i.e., one end is firmly attached to the motor and one end is firmly attached to the base.
3) Securely mount a very small mirror on the side of the motor.
4) Securely mount a laser pointer at some point off of the motor and aim the laser at the mirror.
5) Mark the point on the wall where the laser reflects at rest.

This can potentially be EXTREMELY accurate, but your problem may be calibrating it.

Good luck.

Historical side note: This is one of the ways that the gravitational constant was originally verified to exist between all objects. A barbell-shaped mass hanging off of a long string in a vacuum chamber had a small mirror affixed to the bar. A change in movement of the reflected point of light of a few millimeters at a distance of a hundred feet away was enough to measure the gravitational pull of an orange on a marble.

-T


Engineering is not the science behind building. It is the science behind not building.

 

[MikeHalloran]

The laser pointer and mirror is a nice idea.

For a small motor, instead of wire 'leaf springs', I'd try a piece of surgical tubing slipped over the motor and a fixed stub of similar diameter, say a diameter or two away. Make the tubing longer to get a lower spring rate. Secure with wrapped wire or ty-raps.



Mike Halloran
Pembroke Pines, FL, USA

 

[insideman]

Thanks guys!

Cory Pad: All way too huge!

BobM3: I cannot measure torque of a rotating shaft. I nust use reaction torque. Shaft is .094 inch diameter.

sreid: These strain gags offer some hope. I will study further.


Engineer Tex: The difficulty is that the mirror is inside the motor shroud AND the fan duct AND the wund tunnel wall. All this should destroy any pretense of accuracy.

Mike Halloran: Surgical tubing slipped over the motor would totally block cooling air.

 

[CoryPad]

You need to scroll down - the miniature one is around the size you need.

 

[MikeHalloran]

The presence of multiple windows does not affect the accuracy of the laser galvanometer at all. Clean the windows once in a while so the scatter doesn't mess up your photography.

You can eliminate half the windows by placing the mirror on the back of the motor at 45 deg to its axis, and the laser in the tailcone, right on the motor axis.



Mike Halloran
Pembroke Pines, FL, USA

 

[insideman]

Mike

CoryPad: I looked at the smallest one. Still too big.

MikeHalloran: Why does multiple windows not distort? I would need to contrive transparent parts of a 1.12 inch diameter motor shroud, a 2.50 fan shroud. The wind tunnel wall will be transparent anyway.

Laser being read from straight aft would be read through the rotating wind tunnel fan. Not good.

I have looked at a conceptually simple unit using a hairspring to restrain motor rotation and a simple pointer mounted on it reading against a scale on the housing. This can be read looking past te front edge of the fan housing through the wind tunnel duct window about 45 degrees forward. Might work.

 

[Martensyte]

When you say that electrical measurement will not account for losses, what do you mean? Are you talking about internal mechanical (bearing/windage) losses in the motor and/or electrical losses?

If so, the latter can be accounted for. I think mechanical losses probably could be too. If you know the motor's absorbed power at speed/no load, then doesn't that equal electrical + mechanical losses, and won't mechanical losses remain the same regardless of load?

I dunno, but I'd look into calibrating the motor.

 

[hydtools]

Here is a small torque sensor:

http://www.transducertechniques.com/SWS-Torque-Sensor.cfm

Ted

 

[MikeHalloran]

The laser is behind the motor. The mirror, attached to the motor, bends the laser beam 90 degrees, so it points RADIALLY away from the motor. You read the angular deflection by measuring the movement of the laser spot on the far wall. "Far" can be almost arbitrarily large.



Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

I Like Martensyste's idea - calibrate the motor elsewhere and then just monitor volts and amps (and preferably speed)in the wind tunnel.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[sreid]

Just a thought springing from others ideas. Wind tunnels usually have highly accurate aerodynamic force balances to measure lift, drag, etc. I would be surprised if the balance can't be configured to measure a rotation torque due to, say, aleron torque. This measurement would be no different than the motor reaction torque.

 

[insideman]

I'm the guy who must build the wind tunnel!