Guaka at 18:09, 19 April 2011

2011-04-19T18:09:53Z

← Older revision		Revision as of 18:09, 19 April 2011
Line 89:		Line 89:
	-Ed Howdershelt (Author)		-Ed Howdershelt (Author)
	It is easier to get forgiveness than permission.		It is easier to get forgiveness than permission.

			[[Category:People]]

WikiSysop: New page: Re: torch table followup From: Gene Heskett (gene.heskett@verizon.net) Medium risk You may not know this sender. Mark as safe | Mark as unsafe Sent: Sat 2/09/08 7:17 PM To: marcin_ose (mar...

2008-05-24T22:15:34Z

New page: Re: torch table followup From: Gene Heskett (gene.heskett@verizon.net) Medium risk You may not know this sender. Mark as safe | Mark as unsafe Sent: Sat 2/09/08 7:17 PM To: marcin_ose (mar...

New page

Re: torch table followup
From: Gene Heskett (gene.heskett@verizon.net)
Medium risk You may not know this sender. Mark as safe | Mark as unsafe
Sent:
Sat 2/09/08 7:17 PM
To:
marcin_ose (marcin_ose@hotmail.com)

On Saturday 09 February 2008, marcin_ose wrote:
>Gene,
>
>Thanks for the hints.
>
>On the drive system, then, the 3/8" 'resolution' of the belt is
>sufficient to get accuracy about 10 times better than 3/8"? That was
>my main concern. I would want to use microstepping. I was concerned
>that the 3/8" pitch makes the belt too sloppy for 3/100 accuracy.

I think it would do that, I'm using a similar but finer toothed belt on the z
axis drive for my mill, and as long as I have the thing tensioned up so the
belt can't throw some slack on the push side, the backlash is zip. Its the
finest pitched 3/8 belt McMaster-Carr had. Belt tooth pitch is relative, the
finest resolution is the motors step size, with Jeffs drivers, that is 1600
steps per turn of the motor. I suppose one could find a cyclic error with a
fine enough resolver, caused by the shape of the teeth, but for driving a
cutting torch, I'd think any pattern you could see would be well swamped by
the turbulence of the cutting flame. As a friend of mine would say, good
enough for the girls I go with. :-)

>Regarding rack and pinion type drive - I was planning to use not a
>belt loop, but a belt attached firmly to a surface. The pinion would
>drive on top of that. Do you think that is a feasible solution, or I
>wouldn't get the traction that I want? Could gravity make a sufficient
>hold, or would I need extra clamping down onto the track?

Fastening the belt to a solid surface is the first problem, I wouldn't trust a
superglue like substance to do that since the belt has sufficient flexibility
that all the forces must be absorbed by the glue joint within a few thou of
the actual tooth contact. Then the gravity or helper springs implied to
maintain contact would, in the bearings of the swing mount, represent another
source of backlash unless on very good swing bearings in order for it to
follow the wibbles of the rack holder. I'd much rather use a continuous
loop, over an adjustable for tension idler pulley on the far end, with the
end of the carriage attached to the belt by clamping a short section of rack
to the belt, no holes punched. Short pieces of rack can be had at
McMaster-Carr I believe. As long as the clamp is the same elevation as a
sight line from pulley to pulley, then the major precision worry is the sag
of the carriage guide rods and the sag and whip of the belt over that length
unless suspended in a track so it cannot sag or whip. The track will need a
few holes to allow the dirt to be swept on through though.

Here also I think I'd take a lesson from the panel saw at your local Lowes and
use a larger pipe with skate wheel bearings, maybe arranged on jack screws
above the main frame to allow the pipe to be jacked as true as you can
measure. Spot weld (resistance weld if you can, less warpage of the pipe)
the bolt to the side of the pipe and drop it into the main frame with a nut
above and below the frame. 4 or so per 8 foot length as long as we're
playing "what if" here should do it nicely. A 1" solid rod is fairly rigid,
but a 2" piece of 4340 pipe with .032 walls is even more rigid, and weighs a
heck of a lot less. Not to mention its a bunch cheaper than the same length
of drill rod and more than adequate dimensional tolerance.

Like I said, I wish I had the room for such a tool. OTOH, then I'd have to
hit the road and find work for it too but I'm too old to undertake another
career. :( So I'm semi-happy with my modified micromill I've tried to make
an X3 out of, should have bought one in the first place. Hindsight as
always, is 20-10. See my mess at <http://gene.homelinux.net:85/gene/emc> (vz
blocks port 80) 8-)

I do hope you find my comments useful.

>Sincerely,

>Marcin

>PS. I like your boxes quote.

Its been there for quite a few years now Marcin, and there are times when I
think its time for the last box to come off the shelf, the jury box isn't
working that well anymore. And then they do manage to stuff somebody that
needs it, putting off the inevitable. For a little while.

--
Cheers, Gene
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
It is easier to get forgiveness than permission.

Gene Haskett - Revision history

Guaka at 18:09, 19 April 2011

WikiSysop: New page: Re: torch table followup From: Gene Heskett (gene.heskett@verizon.net) Medium risk You may not know this sender. Mark as safe | Mark as unsafe Sent: Sat 2/09/08 7:17 PM To: marcin_ose (mar...