Measuring Chain Wear

Technical Details
21

No, what I was saying that in order to do this fancy measurement process that lets us get readings to solve 1000+ simultaneous equations and get a wear value for each link, we ideally need to keep both sprockets at 12 o’clock and “pull” the motors apart to tension it. That’s not practical to do, obviously, so doing this (i.e., getting individual chain link wears) might not be practical. But that’s ideal situation and I don’t know if it matters. It’ll introduce error but I don’t know by how much.

For the time being, I’m just going to try to use my calipers and see if the wear is uniform while the chain is extended. shouldn’t even need to remove chains from sled.

22

possibly, but that requires a differnet mechanism

there are B commands:

  1. to turn a motor a given distance
  2. apply power to a motor
  3. return the encoder position of a motor
  4. reset the encoder position of a motor to 0

so the programming for this can all be in g-code (plus looking at the log)

  1. attach the chain to the left sprocket

  2. zero the encoder positions of both motors)

  3. feed out chain from the left motor until it goes two links past the right
    motor

  4. attach the links to the right sprocket

  5. turn both motors 1/4 turn clockwise

  6. pull tight by applying power to the motors (low power first for a short time,
    then full power to minimize sag)

  7. get motor positions, add them (the right motor position will be negative),
    this is the distance for the base chain slop

  8. zero the encoders

  9. rotate right motor a bit to the left to release tension

  10. rotate both motors to the right a fixed distance

  11. apply tension (see 6)

  12. get motor positions, add them, they should be 0

  13. go to 8

I’d have to look up what the b-codes are for this to turn this into a g-code
program, but it should be pretty trivial to do.

David Lang

1 Like
23

both sprocket should move, but you know how much they have moved, so you can
return them to a 12 o’clock position (you don’t need the same tooth at 12
o’clock, just A tooth at 12 o’clock, so you should only need to rotate a small
amount)

David Lang

24

I used my calipers and found the chain had somewhat variable wear… the part that seems to ride over the sprocket the most seemed to wear the most… I’ll have to check when I get back in the shed, but I don’t think I’m measuring anywhere near 127.4… I think more like 127.1 to maybe 127.2 at worse… I need to check though (got called in for dinner)

1 Like
25

it should be possible to calculate chain wear, by measuring the actual chain length by hand, and having the machine match that to the number of links in the chain… the hard part is accurately measuring the chain without getting it covered in dirt.

26

I verified my wear measurements were between 127.1 and 127.2 mm (0.1 mm and 0.2 mm wear over 127 mm). That’s 0.15% and no where near the 0.34% maslow measured.

1 Like
27

put the chain under tension in the maslow and re-measure, the different tension
can make a big difference.

David Lang

28

Those meausrements we’re done with the chains attached to the sled. The part between sprocket and sled was under high tension, but there was still decent tension on the slack takeup side as well