Re: shafting question
Posted by
ballendo@y...
on 2001-11-12 00:41:50 UTC
Rick,
I'd agree with Jan about using the largest diameter rails that you
can if they are to be end mount only.
For your original question, Thomson publishes the formulas to calc
the load for their products, including bending due to load. Both as
s/w and in "print". Techno-isel will also send you a catalog which
has many useful cnc type calcs.
IMO, crossed roller bearing are overkill for most "home made" cnc
machines. Do become familiar with what the thk type linear bearing
are though. THK is a manufacturer, and will prob'ly be found with a
web search. Thomson also sells similar competing products.
Back to the thomson type round rail stuff:
I agree, the pre-drilled shafting and supports are pricey.
First step is to look beyond thomson. The "round rail" technology is
no longer "state-of-the-art", and others have joined the market. I
think "Hiwin" is one "same size" inexpensive competitor to thomsons
rails and ball-bushings.
Second, the cost of plain thomson shaft is not too bad. Epoxy
mounting this rail to the supports will save significant dollars. The
supports are also expensive, but ARE ground; so the cost is
justified, IMO. But,... IF you can hold the shaft accurately
straight (and in proper relation to the support) while the epoxy
sets, you CAN get away with using an inexpensive angle for the
support. And this epoxy mounted rails technique IS used by some
commercial mfrs, tho not all are necessarily for cnc. For some other
ideas, look at the super tech(superCAM) website.
Another way you can save is by noticing the cost of the linear ball-
bushings WITHOUT housings. Not too bad. If you have the time and
skill to fabricate your own housing(s) you can come out ahead DOLLARS
wise (but you WILL NOT get away for free; your time spent will likely
be significant if you go this route).
Next, notice that the newest generation of super smart ball-bushings
have much greater load carrying capabilities than
their "forefathers". And although they often don't "look right"
to "design by eye" types, the specs ARE there, and the money saved is
again significant. You just have to keep convincing yerself that
those skinny rails "really can" carry the load (of course these WILL
BE the fully supported type).
Note: these newer types DO depend MORE upon the housing than some of
the older ball bushings for their load carrying specs, so factor the
housings into your "figgering"...
Hope this helps.
Ballendo
P.S. The pacific bearings are a good product with some "caveats"
(buyer beware): They require looser fits to the shaft (no pre
loading here) than linear balls. This will mean that a bridge design
is better than a gantry if these are going to be used. They are high
speed/light load by design, so you may want to go the opposite way of
the above "thomson info" (use bigger than what "looks right" for best
results). And DO pay attention to the "too smooth shafting" problems
mentioned by pacific. these bearings definitely "work in" and NEED
a "rougher" shaft for good results.
I'd agree with Jan about using the largest diameter rails that you
can if they are to be end mount only.
For your original question, Thomson publishes the formulas to calc
the load for their products, including bending due to load. Both as
s/w and in "print". Techno-isel will also send you a catalog which
has many useful cnc type calcs.
IMO, crossed roller bearing are overkill for most "home made" cnc
machines. Do become familiar with what the thk type linear bearing
are though. THK is a manufacturer, and will prob'ly be found with a
web search. Thomson also sells similar competing products.
Back to the thomson type round rail stuff:
I agree, the pre-drilled shafting and supports are pricey.
First step is to look beyond thomson. The "round rail" technology is
no longer "state-of-the-art", and others have joined the market. I
think "Hiwin" is one "same size" inexpensive competitor to thomsons
rails and ball-bushings.
Second, the cost of plain thomson shaft is not too bad. Epoxy
mounting this rail to the supports will save significant dollars. The
supports are also expensive, but ARE ground; so the cost is
justified, IMO. But,... IF you can hold the shaft accurately
straight (and in proper relation to the support) while the epoxy
sets, you CAN get away with using an inexpensive angle for the
support. And this epoxy mounted rails technique IS used by some
commercial mfrs, tho not all are necessarily for cnc. For some other
ideas, look at the super tech(superCAM) website.
Another way you can save is by noticing the cost of the linear ball-
bushings WITHOUT housings. Not too bad. If you have the time and
skill to fabricate your own housing(s) you can come out ahead DOLLARS
wise (but you WILL NOT get away for free; your time spent will likely
be significant if you go this route).
Next, notice that the newest generation of super smart ball-bushings
have much greater load carrying capabilities than
their "forefathers". And although they often don't "look right"
to "design by eye" types, the specs ARE there, and the money saved is
again significant. You just have to keep convincing yerself that
those skinny rails "really can" carry the load (of course these WILL
BE the fully supported type).
Note: these newer types DO depend MORE upon the housing than some of
the older ball bushings for their load carrying specs, so factor the
housings into your "figgering"...
Hope this helps.
Ballendo
P.S. The pacific bearings are a good product with some "caveats"
(buyer beware): They require looser fits to the shaft (no pre
loading here) than linear balls. This will mean that a bridge design
is better than a gantry if these are going to be used. They are high
speed/light load by design, so you may want to go the opposite way of
the above "thomson info" (use bigger than what "looks right" for best
results). And DO pay attention to the "too smooth shafting" problems
mentioned by pacific. these bearings definitely "work in" and NEED
a "rougher" shaft for good results.
--- In CAD_CAM_EDM_DRO@y..., "Rick Miller" <apollo@a...> wrote:
<snip> I'm not familiar with the THK grooved rails, but I do have a
catalog next to
> me with Del-Tron Crossed-Roller rails... $420 for one 11.8" long
piece -
> ouch! WAY out of my price range. That would wind up more than $2400
in rails
> for a 3-axis machine.
> For continuous support, I was thinking of using frelon open
bearings w/ the
> ceramic coated linear shaft assembly from Pacific Bearing. The
shaft &
> support are one piece of 6061T6. That was still more $ than I had
in mind,
> but it looks to be the most cost effective solution I've been able
to find
> so far. It's still cheaper than Thomson pre-drilled rail + the
supports.
>
> Thanks for your help!
Discussion Thread
Rick Miller
2001-11-09 15:18:31 UTC
shafting question
machines@n...
2001-11-09 15:37:55 UTC
Re: shafting question
ccs@m...
2001-11-09 15:41:13 UTC
Re: [CAD_CAM_EDM_DRO] shafting question
JanRwl@A...
2001-11-09 16:02:05 UTC
Re: [CAD_CAM_EDM_DRO] shafting question
Rick Miller
2001-11-09 16:44:45 UTC
Re: [CAD_CAM_EDM_DRO] shafting question
Jon Elson
2001-11-09 21:44:13 UTC
Re: [CAD_CAM_EDM_DRO] shafting question
andesign2@y...
2001-11-09 23:40:33 UTC
Re: shafting question
ballendo@y...
2001-11-12 00:41:50 UTC
Re: shafting question
Rick Miller
2001-11-12 15:03:15 UTC
Re: [CAD_CAM_EDM_DRO] Re: shafting question
ballendo@y...
2001-11-13 15:24:13 UTC
follw-up on pacific bearings was Re: shafting question
Rick Miller
2001-11-13 17:05:31 UTC
Re: [CAD_CAM_EDM_DRO] follw-up on pacific bearings was Re: shafting question
thscarince@h...
2001-11-14 06:20:26 UTC
follw-up on pacific bearings was Re: shafting question
Eric Keller
2001-11-14 15:56:38 UTC
Re: [CAD_CAM_EDM_DRO] follw-up on pacific bearings was Re: shafting question