6
$\begingroup$

I was wondering how to mount a small rotary encoder to a platform in order to measure how far a small treaded robot has traveled. We're using this tread/wheel set: track set We've tried epoxying the encoder shaft directly to the end of an extra gear wheel, but it broke off quite easily. I'm thinking about grinding down the encoder shaft until it can just squeeze into the axle hole of the drive wheel and then using the epoxy inside the wheel. It would then be driven off of the tread. Like this:My Idea However I'm not sure this would be the best idea, and other suggestions would be appreciated. Especially with keeping any assembly either coupled to the drive shaft or to a tread.

Note: Its for an EE senior project, and none of us have much experience with construction skills. Its going to be the platform for an automated robot on Aruino.

$\endgroup$

3 Answers 3

4
$\begingroup$

The flat on the encoder shaft would typically be used for set screw. If you are able to get a gear or wheel with the right size shaft then adding a screw should be relatively straightforward. This will not only hold the gear onto the encoder, but also give it some increased tortional strength which should be better than just epoxy.

If you are unable to get a gear to exactly fit the encoder shaft size then I would suggest being careful about how you reduce the size of the shaft or increase the size of the hole. You need to ensure that the axis of rotation remains central. If it can be properly turned on a lathe (or even drill chuck if necessary) I think you'll get a more even result than going at it with a file by hand.

I would not expect your encoder to have a particularly robust bearing system within it, so it may not be suitable for use directly as a wheel axle, but the end decision really depends on your component specifications and the loading on your robot frame.

A final point, in case you've not considered it, is the resolution you require from your encoder measurement. This will be affected by the gear ratio you choose from your motor. Higher ratios will give you a more precise measurement, but also a higher count in your software, so it really depends on your expected move distance and sotware performance.

$\endgroup$
6
$\begingroup$

That model of encoder is not intended for mechanical measurements. It's meant to be used as a fancy knob with high friction, for turning by hand, say, to pick a position from menu on a display, and trying to couple it to a mechanism for readout is a misguided attempt.

What you need is an encoder based on slotted optocouplers or a reflective optocouplers.

enter image description here enter image description here

Note you can print the code wheel on transparency, to get the desired number of pulses per turn. Put the wheel on the axle, attach the optocoupler to the main plate - or even extend the tracks a little and use them as "code strip".

If you have trouble purchasing these, you can often find them in computer mice and cheapest inkjet printers.

$\endgroup$
1
  • $\begingroup$ Thank you for the advice against using an incorrect tool for the job. We hadn't gone with the optocoupler method due to the cost, but it turns out that trying to go cheap and "get away" with things is bad practice. Our force-fit and epoxied encoders kept going off-center and breaking. Would have saved a lot of effort and been more stable to do it right to begin with. $\endgroup$
    – rasz
    Jan 20, 2016 at 7:19
2
$\begingroup$

Another alternate solution would be to use a Hall Effect sensor. There are many ways to apply a hall effect sensor to measure distance. Most simple might be to attach the hall effect sensor to the motor and count the rotations. Then translated the number of rotation to distance. Below are few reference links that better explains how to use a hall effect sensor.

In some automotive application Hall Effect sensor is use to measure speed. Also hall effect sensor is used in clam-shell cell phone to detect open and close.

Hall Effect Sensor Gear Tooth Hall Effect Sensor


References:

$\endgroup$
1
  • $\begingroup$ Just a note: perform the measurement some distance from the motors - they could disturb the measurement. $\endgroup$
    – SF.
    Feb 17, 2015 at 12:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.