I want to measure the number of meters of running cloth off of a fabric roll. So that when I press a button a motor spins a roll of cloth and the fabric is rolled onto a separate roll. I want to be able to measure the number of meters of fabric that runs from one roll to the other. What kind of sensor would I need to measure the length of running cloth?

I was thinking maybe remove the laser sensor from a mouse and use it to look at the fabric. What other possibilities are there for this problem? This is to be used on a cloth inspection frame, so a fabric quality inspector will press a button, the motor will start running the fabric; if he sees a flaw in the running fabric, he presses the button to stop the fabric, a sensor detects at how many meters the roll stopped, and the inspector enters a code for a type of flaw and presses the button again to start moving the fabric again. This way, quality data is recorded for a roll of fabric in a database.

Eventually one could pull up the quality inspection data and know exactly at what meter in the roll a flaw exists.

  • 1
    While this is certainly an interesting question it doesn't appear to be specific to the raspberry pi. – goobering Apr 1 '16 at 7:41
  • umm..i'm going to be using a raspberry pi to generate this data and then bounce it out to a database server, through linux command line. – Shalimar Apr 1 '16 at 8:13

A couple of ideas which spring to mind - perhaps some will be more appropriate than others, given your setup.

  1. If you were to drive your roll of fabric with a stepper motor, you'd trivially know how many turns you'd done. Measure the diameter of the roll - perhaps with an ultrasonic distance sensor - and you can convert rotations into metres.

  2. Can you pinch the fabric between rollers or cause it to roll a wheel, and attach the wheel to a rotary encoder?

The optical mouse idea sounds quite nice also, provided it can reliably "see" the fabric. I'd be thinking along the lines of creating a new driver (heavily based on the existing mouse driver) so you can collect the data is a way that's convenient to you, not mixed in with real mouse data.

  • lovin option two...much simpler...a reed switch with a magnet will solve this...I already have a manual odometer like device running with wheels on top of the inspection frame.. very similar to the following video youtube.com/watch?v=hfjvwiKEW84 – Shalimar Apr 1 '16 at 8:24
  • It's easier to imagine having seen the machine! There are loads of places on that machine where you could put a rotary encoder - although if the accuracy of a reed switch (or hall effect sensor - more reliable long term, I'd imagine) is enough then fair enough. – Mark Smith Apr 1 '16 at 8:27
  • the main issue with the roller device is that there's slippage...which in the grand scheme of things can be lived with...since it probably only misinforms the actual length by maybe half a percent...but if ure inspecting a million meters a month, poopoo adds up. – Shalimar Apr 1 '16 at 8:30
  • the accuracy of the reed switch really isnt an issue, since i will only be listening for pulses. 1 pulse = circumference of the wheel. – Shalimar Apr 1 '16 at 8:32
  • Will the rollers ever need to roll backwards? You can't detect that with a (single) reed switch / hall sensor. – Mark Smith Apr 1 '16 at 16:32

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