Hello Guest it is March 29, 2024, 04:17:44 AM

Author Topic: Holes Not Round  (Read 2936 times)

0 Members and 1 Guest are viewing this topic.

Holes Not Round
« on: February 24, 2015, 06:19:46 PM »
I’m machining a plug panel that has assorted hole sizes for mounting plugs. Each of the holes that gets machined is eccentric (oblong). I thought at first that it was due to a backlash issue, but doesn’t seem to be the case. The holes are eccentric diagonally. If backlash, it seems they would be eccentric only in the x or y direction. This would indicate something to do with the software; a setting or such? Anyone seen this?
I’m running Mach 3/PMDX/Gecko/RF 45 rex

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Holes Not Round
« Reply #1 on: February 25, 2015, 05:40:05 AM »
First check that the machine is adjusted correctly ie; ways, load bearings, pulleys/screws etc.

Then make sure your steps per unit for each axis are correct based on the longest axis  travel.
Then check to see just how accurately you can move at some shorter equaly spaced distances along the axis ( after  refining steps per unit).
Then check the backlash for each axis and setup the backlash compensation.

The above will minimize inaccuracies in machining.

You will know just how accurate the machine is.

Then try some circles and see how accurately they are done. Makes sure that you use a finish cut when doing the cirlces.
They should be within what you machines linear axis move tolerances.

RICH
Re: Holes Not Round
« Reply #2 on: February 25, 2015, 11:53:44 AM »
Thanks Rich; sounds like a great approach to verify what's going on and to get the problem resolved.  What's the best way to measure table travel within machine tolerances?  I can see measuring short distances with a dial indicator setup.  Is it generally necessary to verify accurate table travel over varying distances, short to long?

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Holes Not Round
« Reply #3 on: February 25, 2015, 12:34:45 PM »
You use  the longest table travel to set the steps per unit. See ALT 6 page lower left hand corner and use the axis calibration feature of MAch. Any inaccuracy in the step setting is a very small amount in a short move as compared to the steps required for a long move.

Since the screw  lead may vary over it's total length, ie; ball screw specs sometimes are over some distance and thus the lead error may not be  not be linear over that distance ie; more in one  section as compared to another. For a short distance ie; 1" you can use a dial indicator, preferably one in 0.0001" resolution.

Now if you could, which I have done opticaly, you can actualy profile the screw and see how it varies in both directions.

You have what you have and then must deal with it.

As far as measuring something, well it all depends on what you have available to do the measurements with.

RICH
Re: Holes Not Round
« Reply #4 on: February 28, 2015, 08:26:36 AM »
Thanks Rich; I now have accurate round holes.  I was a little surprised to find that I have zero backlash in the y axis and 25 mils of it in the x axis.  Same lead screws and nuts.  Is that typical?  I using an RF 45.  I guess since the software seems to do a good job of tuning out the backlash, it doesn't matter.