On Friday 29 July 2016 11:20:57 andy pugh wrote:

> On 29 July 2016 at 16:06, Gene Heskett <[email protected]> wrote:
> > For those interested, it actually looks easy if a mill is available.
> >
> > <http://warhammer.mcc.virginia.edu/ty/7x10/vault/Lathes/7x10-7x12-Pr
> >ojects/TaperedSaddleGibs/>
>
> Or my way, which was even simpler:
> https://picasaweb.google.com/108164504656404380542/Gibbs
>
> First 14 pics.
>
Picassa is in a shutdown mode, and nothing displays there now.

There is a message link there now.

> In brief: machine an angle. Set straight. Clamp some brass in there,
> machine both at the same setting. Fit.
>
> But it seems crazy to go to so much trouble to improve one lathe to
> make a part for the other. Why not make a not-so-good part on the
> little lathe, then use the Sheldon to make a better part for itself?

Because the bronze crossfeed nut, has been bored out and cobbled up with 
a brass helicoil insert, and is about 5 thou from stripped again?  Its a 
10 tpi screw, and the backlash at the handle is about 90 thou, so there 
is very little brass left to engage the square threads of that screw.

And the compound gibs are some sort of home brew that are too short to 
allow proper adjustment?  Ed Nisley was kind enough to supply some shim 
stock, so I can tighten that up but they'll still be an inch shorter 
than factory on the load bearing end

And I've not a clue how much bed wear there there may be.  That looks 
good, but until I have  made some longer turnings, no clue.

But this is precisely why we need a cross-coupled screw comp that LCNC 
has to cobble up with lincurve. What it really needs is a 3 var format 
for the zcomp file at least, where the 1st var is the current zpos, the 
next one is the real zpos, and the third one is a corrected xpos at this 
zpos to be used to correct the xpos at this zpos. The only thing that 
can't correct is the tool tip contact height as its traverseing a worn 
bed.

Using lincurve, it may be possible to arrive at something that is useable 
because it can interpolate between points, but it seems it would more 
properly be done by sending the xpos to the file input of the x driver 
so it can be corrected at 50-100 points of the z range as it fixes cheap 
screws where the pitch may even have a cyclic variant per rev so you'd 
have multiple points per turn.

Because there is so much to fix just to make it useable before it grows 
motors, I may as well bypass that by tightening up the little monster.  
It would be weeks faster to 1st swarf on the Sheldon at my speed. :(

Does this help explain it?

Thanks andy.

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