Accelerometer vs. Gyroscope - measuring torque

[HarryLime]

Summary:: accelerometer Vs gyroscope - measuring torque

Torque can be measured with an accelerometer (tangential-acceleration):
t = F*r and F=m*a, so we get a from the accelerometer, giving:
t = m*a*r Nm

Torque can be measured with an gyroscope (angular-acceleration):
t = F*r
The relationship between tangential (a) to angular (omega) acceleration:
[OK gyroscope gives angular-velocity, so differentiate this once and you get angular-acceleration]
a = omega*r m / s^2
Giving:
t = m*(omega*r)*r = m*r^2 * omega
or if we include the moment of inertia:
t = I*omega

Questions:
[1] Is the assessment correct, Torque can be measured with an accelerometer OR with a gyroscope?
[2] what are the pros and cons of either method?

This seems to sum it up nicely:

 

[berkeman]

Welcome to PhysicsForums.

Summary:: accelerometer Vs gyroscope - measuring torque

Questions:
[2] what are the pros and cons of either method?

What are your thoughts? I can think of several tradeoffs, but we want to hear your thoughts first.

Is this a schoolwork question?

 

[HarryLime]

Welcome to PhysicsForums.

What are your thoughts? I can think of several tradeoffs, but we want to hear your thoughts first.

Is this a schoolwork question?

Thanks berkeman, Yes this is background for homework! statements, pros & cons -
1. Accelerometer give signal based on linear acceleration or tangential if at the edge of a spinning obj (like the drawing above) - So accelerometers react to linear/tangential forces, which also include Gravity. They suffer from drift (good in the short term but very noisy in the long term due to additive error), noise (any acceleration effects the sensors).

2. Gysroscope give a signal based on angular velocity, like the spinning obj above - So gyroscopes react to rotational motion. They seem to suffer from drift & lag, though I'm unclear as to why. They are also more prone to temperature dependencies (?)

3. Both gyro & acc are often fused to to offset the other's noise and drift errors, providing more accuracy. So they are complementary.

This could do with more - assistance welcomed. how would I tie/combine these sensors together?

 

[A.T.]

So accelerometers react to linear/tangential forces, which also include Gravity. They suffer from drift (good in the short term but very noisy in the long term due to additive error),
...
So gyroscopes react to rotational motion. They seem to suffer from drift

How is drift relevant, if you are trying to measure torque?

 

[HarryLime]

How is drift relevant, if you are trying to measure torque?

yes

 

[A.T.]

yes

That wasn't a yes/no question.

 

[hutchphd]

A linear accelerometer is not a good way to measure rotational motion because position relative to axis is a direct requirement. The rotational sensors have no such requirement and simply report any change in angular velocity regardless of linear velocity or acceleration. The MEMS sensors are quite remarkable.
What do you mean by a gyroscope?? Does anyone use them any more?

 

[hmmm27]

Does anyone use them any more?

They're a reasonably decent way of keeping smaller aircraft from landing ahead of schedule.