Measur bending of carbon tubes


strange title i know :) sorry about that.

so, what i would like to do, is a garden installation.

the idea would be like this: several carbon tubes, diameter 5mm, length between 1 and 2 meter, and on top a LED. the tubes are fixed in the ground. when the tubes are bent, from the wind, or by hand, the LEDs on top would flash or blink or shine with different strenght (PWM like)

any ideas how i can measure if the are bent?

You could have a very tight string inside the tube connected to a switch in such a way that when the tube is bend the string is tigthend a little more and the switch closes.


BUT carbon tubes are very stiff so it would take a lot of wind to bend them, thats why they are used for building kites.

That tensioned string idea sounds like a good possibility. What about connecting the string to a piezo sensor to get a proportional response?


If you get enough bend the string and piezo sounds like a good thing to try.

  • You could also try a little anemometer at the end of the tube.
  • Tilt sensors mounted at different angles (assuming you get enough bend).
  • Accelerometers would work but would be expensive.

(* jcl *)

How about sending light also through the tubes? When they bend, less light will fall directly onto the receiver.


Not too sure of the tubes you are using but, in general, a carbon composite will exhibit a change in resistance when bent, stretched, or compressed. Methinks if you were to measure the resistance from one end to the other, you will find a significant difference when bent and unbent.

The question is, is that difference large enough to be sensed by the Ardy, in a simple manner.

Without more information about the tubes and their characteristics, that's all that I can say.

If you want to measure bending, then you are measuring strain. So…Strain gauges come to mind!

You glue on these little [<5mm wide, <20mm long) strips and start measuring their resistance.

Now, now you’ll want to glue at least two on to either at the bottom of the rod, or possibly the rod support. That way you can measure longitudinal strain in both the x and y.

(Cross-sectional rod view:
o x

With some solid mechanics, we can calculate horizontal direct and magnitude of the force applied to the tip of the rod. Or, with a bit of fluid mechanics we could estimate wind speed/loading.
But it’s probably easier to just do it experimentally rather than analytically.

I presume that you could use something like these flex sensors from sparkfun:

If you could attach one of these on the tube then it would be able to measure the flex.

I'm not sure how accurate they are or how accurate you want the readings.

Are you just wanting to know when they move or by how much?

Are you wanting to measure wind speed or similar?