re:welding linear shafts

Richard,

(snips, inserts below)

>Considering that linear shafting, even Thompson, is reasonably
>priced unless you want mounting holes drilled and tapped...

Two thoughts: Have you seen the supercam site re casting
epoxy "mounts" to the shaft? Second. It is CASE hardened. I have used
a grinder to get below the hard surface and drilled/tapped. Only
short shafts though, my time is worth more than the difference in
price! I have seen other machines which "glue" the plain shafts in
place.

>Any reason I can't weld mounts on the shaft? Assuming I take
>reasonable precautions to keep the shaft clean?
>I realize I'll be changing the hardness of the shaft with the
>heat... but how many cnc machines fail because the shaft isn't hard
>enough?

Quite a few, actually! One of the reasons we see less of
the "thomson" style shaft and bearing in high-load (read machine
tools) situations is that this system, by design, is point contact.
And this point contact wears "grooves in the shafting. Yes, even
hardened shafts!

Two circles will meet (theoretically) at a single point, tangent to
both circles. Actually, the metal used in the ball bearings of the
linear ball-bushing and the shafting, will "deform from roundness"
under the load and provide additional points of contact and load
carrying.

But this is still no match for the current generation of linear
bearings (thk, and others) which use precision grinding to achieve
contact over much more of the ball bearings' circumference. Equaling
higher loads/ longer life/ less deformity(which equals less heat),
etc.

Depending on the loads/life you calc for your proposed machine, none
of this may actually come into play. But do be aware that thomson
shafts in the woodworking industry are regularly replaced on CNC
routers and point-to-point machines (the woodworking equivalent of a
VMC). Also on autimatic saw stops, which have nearly no load; just
the pre-load of the bearings on the shaft is enough to wear the
grooves into the shafts.

Hope this helps.

Ballendo