Designing an instrument with strain gauges

[EngineeringDaN]

I am busy studying engineering and I have a subject that requires us to design and build a working instrument using strain gauges. We have a workshop at our disposal and all necessary materials and electronics.

At this moment my group and I are stuck with coming up with an idea for an instrument we can make using strain gauges that will not be too simple or generic.

So far, we have come up with several ideas, but only two seemed to be viable with our limited time and knowledge.

Idea 1: We would like to measure the torque produced by different propellers by mounting an electric motor to a fixed shaft. The strain gauges would then be placed on this shaft at 45 degree angles to measure the shear caused by the motor trying to twist the shaft. But, since we will only be using relatively small electric motors (with +- 3kg pulling force) we may have a problem with the sensitivity of the system. One idea was to use a tube of some type of aluminium alloy to try get decent strain. However, I have no idea if a tube can be used instead of a solid shaft. Another thing I am unsure of is what type of torque would we be measuring, dynamic or static ? (if say for instance we measure the strain produced at max rpm for each propeller). I know that in industry the strain gauges are mounted on the rotating shaft, and not on a fixed shaft attached to the motor firewall, to measure the strain. Also what else would we need to take into consideration with this design ?

Design 2 of idea 1: The second design involves mounting the motor between two cantilever beams.
The torque caused by the rotating motor and propeller at max rpm would result in the one beam deflecting upwards, and the other deflecting downwards. Is this idea even viable for measuring max torque of a rotating object ?

Idea 2: The second idea is to make an accelerometer to measure the acceleration of different propellers on a motor. The idea is to make a vertical cantilever with a certain mass at the top together with a mounted motor and propeller and to measure acceleration from zero to max rpm.The strain gauges would then be placed on the beam near to the fixed end. I was told that I may need to consider practical aspects of the design such as damping and the systems natural frequency if I am to design an accelerometer. However, I have no idea how to design it to take these into consideration, or if I even need to even worry about it for the stated purpose.

The golden question is, will I be able to get either of these two ideas to work ? We have a limited time to design and build a strain gauge instrument, we have other subjects that also require a lot of attention. I would like to get a final viable design asap. I would highly appreciate any advice or any additional information. If you think any of these two ideas can work, please help me to fully understand what I need to take into consideration to make a precise and accurate final instrument.

 

[Baluncore]

It is much easier to measure the torque of a motor against fixed engine mounts than on the rotating output shaft. If you also measure the RPM you can compute shaft power.

 

[EngineeringDaN]

It is much easier to measure the torque of a motor against fixed engine mounts than on the rotating output shaft. If you also measure the RPM you can compute shaft power.

Alright, since the motor will be spinning, I am guessing the torsion developed in the fixed shaft will give me the dynamic torque value of the motor and propeller ? Also, as far as I know, it should be possible to use a fixed tube, instead of a solid shaft, to measure the strain since the max shear develops at maximum radial distance from the center. I only wonder now why would they mount the strain gauges on the rotating shaft to measure dynamic torque ?. Would it not be easier to moun the motor/engine to a fixed shaft, then measure the deformation in the fixed shaft ? It feels like I am missing something here.

 

[Baluncore]

Torque measurement on a rotating shaft has a signalling problem. There is a very significant twist in a “torque tube”. If another concentric rod or tube is used as a reference between the ends of the rotating tube, the twist along the tube can be measured mechanically as a phase angle without a strain gauge.

It takes more strain gauges to monitor all the engine mounts and eliminate the engine mass and vibration noise. Alternatively, a torque tube that fixes/controls the engine block rotation can be isolated in alignment with a couple of universal joints. The engine mounts can be soft rubber with retaining chains to prevent the engine escaping if the torque tube mounting fails.

I do not know if a tube is better or worse than a solid shaft. The neutral axis is not needed.

 

[EngineeringDaN]

Torque measurement on a rotating shaft has a signalling problem. There is a very significant twist in a “torque tube”. If another concentric rod or tube is used as a reference between the ends of the rotating tube, the twist along the tube can be measured mechanically as a phase angle without a strain gauge.

It takes more strain gauges to monitor all the engine mounts and eliminate the engine mass and vibration noise. Alternatively, a torque tube that fixes/controls the engine block rotation can be isolated in alignment with a couple of universal joints. The engine mounts can be soft rubber with retaining chains to prevent the engine escaping if the torque tube mounting fails.

I do not know if a tube is better or worse than a solid shaft. The neutral axis is not needed.

The reason why I wanted to go ahead with a tube is to obtain more strain, and therefore better sensitivity. Remember, I am trying to measure the torque produced by a relatively small motor and propeller. That is the trick with this instrument design. The max specified mass of the instrument is 5 kg , so we cannot use a large motor to produce a large strain. Therefore I need to use a material with a lower modulus of elasticity and a lower polar moment of inertia to obtain a large enough strain.

However, my group seems to want to go ahead with the idea of creating an accelerometer as described. There are many more variables with the design that can be manipulated to suit our needs as compared to an instrument measuring torque.

 

[EngineeringDaN]

I just realized that using a tube to measure the torque of a motor with propeller will result in an axial tension in the tube as well as shear. The strain gauges aligned at 45 degrees on the fixed shaft to obtain the strain due to principal stresses will no longer be just due to the torque of the motor, as I originally assumed. Is there a way to position the strain gauges to "cancel out" the axial stress caused by the pulling force of the propeller ?

 

[Baluncore]

Use two perpendicular gauges, both at 45° to the tube axis.
That is usually sold as a two gauge rosette, (90°).
Then subtract the signals, which will cancel only the axial component.
Put the torque tube in an axial press to set the gain of the two channels and so zero the axial component.