Re: [CAD_CAM_EDM_DRO] Rolled Ballscrews / mapping errors
Posted by
Doug Fortune
on 2001-11-05 20:19:56 UTC
ccs@... wrote:
for each axis. But I recommend the other software vendors get that
functionality into their software too.
problem, since I doubt woodworkers care if their cnc router can hold
1/2 thou tolerance). But it is an interesting problem.
I plan to push a heavy block, incrementally along the table (along
both the X axes first (my routers have two X screws), then Y, in
both directions.
Assuming there is a digital indicator with SPC output, we can assume
we are calibrated to the current position. Then we move the block
(say) and inch forward, then as we move back to our calibrated zone,
the probe tip stays in contact with the block and extends.
Now we move forward (say .100") using our motor/encoder/pulley/screwlead
theoretical formula. But now we can compare to the real distance we
just moved (the difference between the last two SPC-emitted numbers).
And now we are calibrated to this point.
We go forward all the way down the table like this, and it might take
a while, but it is all automatic.
The particular guage I am looking at are the various Mitutoyo Series 543
versions.
These are available in 0-.5" and 0-1.0" in both .000 5" and .000 05"
resolutions. Of course the long range and high accuracy versions costs
more. The combo long range AND high accuracy version costs a lot more.
(this one is Mitutoyo Series 543 IDC PN#543-451B). The cables are
available in 1m and 2m lengths.
Doug Fortune
http://www.cncKITS.com
>I understand EMC already does it. There are two tables (forth and back)
> I've been thinking about rolled ballscrews and their potential lead error.
> It seems this would be a problem for two reasons:
>
> 1) inconsistent preload
>
> 2) cyclic and cumulative positioning errors
>
> For the cyclic and cumulative positioning errors, perhaps it's time
> that we recognize that software is cheap and iron is expensive.
> Programs like EMC could be modified to include a full screw correction
> formula.
for each axis. But I recommend the other software vendors get that
functionality into their software too.
> The problem with that approach is how to accurately map out theUsing acme screws, I also have the same concerns (probably an academic
> errors. What methods are available for a hobbyist to measure
> distances of one or more feet to an accuracy of .001"?
problem, since I doubt woodworkers care if their cnc router can hold
1/2 thou tolerance). But it is an interesting problem.
I plan to push a heavy block, incrementally along the table (along
both the X axes first (my routers have two X screws), then Y, in
both directions.
Assuming there is a digital indicator with SPC output, we can assume
we are calibrated to the current position. Then we move the block
(say) and inch forward, then as we move back to our calibrated zone,
the probe tip stays in contact with the block and extends.
Now we move forward (say .100") using our motor/encoder/pulley/screwlead
theoretical formula. But now we can compare to the real distance we
just moved (the difference between the last two SPC-emitted numbers).
And now we are calibrated to this point.
We go forward all the way down the table like this, and it might take
a while, but it is all automatic.
The particular guage I am looking at are the various Mitutoyo Series 543
versions.
These are available in 0-.5" and 0-1.0" in both .000 5" and .000 05"
resolutions. Of course the long range and high accuracy versions costs
more. The combo long range AND high accuracy version costs a lot more.
(this one is Mitutoyo Series 543 IDC PN#543-451B). The cables are
available in 1m and 2m lengths.
Doug Fortune
http://www.cncKITS.com