On Tuesday 25 September 2018 07:46:44 Les Newell wrote:

> > I am talking in terms of a
> > light cut on a stiff aluminium bar between centres.
>
> That assumes your tailstock is dead nuts on. If not you're going to be
> trying to compensate for a mix of tailstock + bed errors. If you use
> the bar unsupported there are flex issues. Having said that you always
> get some flex in the work so maybe it is a good thing that the test
> bar flexes slightly...
>
That too, Les. For long cuts, cut light and measure. That almost calls 
for a compensating file. Unforch I have not figured out how to use that 
as it seems to be intended as a per screw, screw compensator, with no 
way to cross couple the z location to the x offset that I could deduce.

Thinking out loud here:

Ideally, a pulldown menu item to select the file to reload a lincurve 
according to the material being worked would be handier than sliced 
bread or bottled beer.  Hint hint hint.

Or perhaps a slider to adjust the gain between the lincurve output and 
the offset module in could be used to compensate for machine flex under 
heavy cutting loads. Wear is consistent, where flex is dependant on 
cutting forces.

So that implies 2 lincurve modules whose outputs are summed together 
before being presented to the offset module. That way if we had the gain 
slider, one could even raise the comp on a dynamic basis to compensate 
for a dulling tool. Not useful for one offs like most of my stuff is, 
but production quantities where you are making thousands of an item, the 
out of tolerance scrap could be reduced. Heck, the gain slider could 
even be put over in the pyvcp portion of the screen, and that could be 
done right now.

I'll do that bit of gui stuff after I've actually matched the taper on 
the od of the 5c, with the id taper of the spindle. I am still 
increasing the taper per inch by about .0002 per pass right now. Getting 
closer, but not there yet. I should put a dial on the shoulder, and 
measure how far I can move in in the mouth of the spindle as it rocks on 
the small end, and add about 40% of what I measure to the ramp/inch in 
the gcode thats re-carving that taper.

Right now I need a way to turn off the comp so I am measuring the raw 
machine and this would be one way to do it without needing to edit the 
hal file and restart LCNC and rehome the machine.

> When I was playing with the head alignment on my Colchester I used a
> large chunk of Delrin bar and a very sharp cutter. The bar was pretty
> stiff and cutting forces were low.
>
> Les

That too sounds like a great idea, Les. Suggested OD of the Delrin to get 
that stiffness? At about 30" between centers...

Tailstock isn't deadnuts on. The last 4 or 5 thou seems very difficult to 
achieve as the tightening of the adjustment screws seems to move it in 
reverse. So I have been experimenting with less and less tension on 
those screws, depending on the central holddown to lock it. I think the 
adjusting screws are flexing the base, tipping it slightly as they are 
snugged up.  And I am not convinced the barrel is pointed dead at the 
center of the spindle, but perhaps a few thou above, so the rear of it 
needs some Reynolds Wrap shimming? Stuff Yet To Be Determined.

> On 25/09/2018 12:19, andy pugh wrote:
> > On Tue, 25 Sep 2018 at 11:15, Gene Heskett <[email protected]> 
wrote:
> >> This is also true. But that test bar costs money too.
> >
> > It can stay in stock as useful material. I am talking in terms of a
> > light cut on a stiff aluminium bar between centres.

Thanks everybody.

-- 
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>


_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to