Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Igaging scales hysteresis
Hi everyone, 

I have a set of igaging scales (not absolute) on my mill. I have no problems with the x and y axes. On the knee, the scale/reader is accurate to within 0.001" over 3", which is plenty good for me. But, I'm seeing 0.003" of hysteresis. That is, there's  3 thousandths of backlash or lost motion when changing direction of feed, before the reader head begins to show a change. I've tightened all mounti g bolts, but with no improvement.  

I've seen quite a few mentions of this problem with regard to the igaging scales, but no solutions have been proposed. I'm wondering if perhaps the guts of the read head are loose within the plastic housing? Is it reasonable to dismantle it,  or should i be looking elsewhere?

Thanks in advance, 

Are you sure it is not backlash in your table?

I would test with an indicator to confirm where the problem comes from.

I would also make sure one end of your dro is floating, i.e. not connected solid use foam or even no screw at all.

I have over .0500 backlash on my Clausing 8520 table (X & Y).

So I turn the handle a lot before getting any actual motion of the table, I have made inquiries about adopting ball screws but never got any response.

I think I will go with the ball screws.

Hi Ralph,
Thanks for the suggestions. The table has a bunch of backlash, but that's not the problem. I have an interapid DTI in the spindle, measuring movement of the table. The table moves a few thousandths before the dro registers any movement.

I took the read head off and looked inside. I adjusted the tightness of the retaining screws (first tighter, then looser). I don't think that was the problem. The amount of hysteresis did change each time, but not predictably. I'm thinking it's maybe related to the scale mount and parallelism with the machine axis. I'm intrigued by your comment about leaving one end free floating - i thought that was only acceptable for short (6") scales? The scale on my knee axis is 24" long.

I have some more indicating to do - first the scale's parallelism in both planes, then on the read head itself to see if it's moving in those few thousandths that the dro isn't seeing...


Old issue, but, I had a thought.

I've noticed with the scale unmounted that there is some apparent "elasticity" in the reader head case. In other words, I can press on it in one direction, without the scale actually sliding and the numbers can climb a thou or two until I release pressure. The problem is that these scales have a tight felt wiper at either end of the case slot. These apply enough pressure to the scale that sliding is prevented until enough pressure is applied. Together, these two characteristics mean that the head may feel "pressure" without sliding, and that this pressure will affect the reading by a few thou.

When mounted, a head stops sliding in any direction with residual additional pressure built up at that position. As you reverse the travel the pressure is first reduced in the primary direction, then increased in the secondary direction, before the scale actually slides. This will show up as non-linear positioning at the reversal -- something akin to backlash.

So, as a guess, it might help to lubricate the slide, so that the head slides more easily with the carriage movement.

Forum Jump:

Users browsing this thread: 1 Guest(s)