On Tue, Nov 18, 2014 at 10:32 AM, dave <[email protected]> wrote:

>
> On 11/18/2014 06:45 AM, Mark Wendt wrote:
> > Stuart,
> >
> > I have no tooling plates in use now.  ;-)
> >
> > Mark
> >
> >
> Hi Mark,
>
> I don't know if the site is still active but Les Watts used an
> interesting compensation
> scheme for his router. Something about a cam on a tensioned wire IIRC.
>
> Being an old curmudgeon ( is that redundant?) I just have to ask what
> the precision
> of the process is if the bed were perfect?
> Maybe I envision the tooling plate incorrectly but why wouldn't  one
> that was simply
> flat do. After all you do have a controllable Z axis.
>
> Despite claims by various people about resolution to .01 mm or so it
> still comes
> down to the accuracy and precision of the process.
>
> On another tangent I wonder about using a polynomial (curve fit) but
> with the number of inflections points you have it may be pretty high
> order. Linear interp instead of a continuous function may still be best.
> This is the kind of thing that starts email wars. ;-)
>
> Best wishes however you decide to cure the problem.
>
> Dave
>


Dave,

If the bed were perfect, I wouldn't need a correction factor for the Z
axis.  ;-)  I need to keep the finished cross-section to + or - .001" at
each 1" X station.

I'd have to rebuild the machine to use some kind of tooling plate.  The
"table" is actually my vacuum hold-down.  I have no way to "cut" the table
flat using the machine itself, unlike a router.  The cutting head is made
up of two saw blades held at an included angle of 60 degrees, with the
"point" of the "V" at the bottom.

I've tried to flatten the bed by filing, sanding, etc, but the problem is,
I really have no reference to the cutting head.  That's how I end up with
being 2 or 3 thou low in one area, 3 or 4 thou high in other areas and so
on.  I've shimmed out the table as best as I can, and this is what I'm left
with, so I'd like to correct for the Z using some kind of compensation read
into the machine.  Be a lot easier to do it once, rather than having to do
it each time I create a different G Code file for each different rod and
each different section.

I like Andy's lincurve idea.  I just have to go through all the reading of
the suggestions given this morning, and work out how I can apply it to my
machine.  I think lincurve will probably be the simplest approach.
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to