Re: machine design was: Re: Re: re:Re: Lead screws for ever...(plastic)
Posted by
ballendo@y...
on 2000-10-28 18:23:51 UTC
Chris,
Thanks for mentioning what you will use the machine for.
Have you seen Dave rigotti's hobby CNC plans?
The luberth plotter site(s) I've mentioned?
Neil Gillies Seagull site? (R/C sailplanes and home made CNC
machines/SW)
One of the things I like about this list is its diversity. We've got
guys who hold "tenths" (while they're smilin' for the camera...)
And guys who are wondering what "tenths" are...
Which means we have a large VARIETY of machines to talk about.
You can make a machine from the hardware store parts bins which can
prob'ly do the sort of parts you may need. In models, it's RELATIVE
DIMENSIONS which usually matter, NOT ACCURACY! (I'm using strict
definitions here)
As long as the airfoil is proportionally correct, it will work!
As long as the WWII plastic model "tank" LOOKS right, it is.
As long as the parts FIT together, it all works.
There's NO WAY I'd pay anywhere NEAR $4K(don't know the exch. rate)
for 3 screws/nuts for a "hobby" machine. UNLESS I was going into
production; Then EVERYTHING changes!
Let's look at a "worst case" scenario. You build a machine out of
WOOD! Using 3/4 mild steel (cold rolled) rods for guides. And MAPLE
blocks with holes in them (soaked in oil) for linear bearings. Other
maple blocks to mount the leadscrews. And "hardware store" threaded
rod for leadscrews. You try "coupling nuts" on the screws, but decide
that there's too much "movement"(the word backlash is unknown to you
at this point), so you (big spender that you are) splurge and buy a
TAP!
And some DELRIN (from a plastics store). And make your own nuts. They
still bind, so you buy a DIE and "chase" the threads of
your "leadscrews". Then visit the DUMP and tear stepper motors out of
old IBM printers(and throw away those funny disks with the holes
in 'em attached to the motors,'cause ya don't know what good THEY
could be!).
You figure out some 4 transistor drives from an electronics magazine
article. And puzzle out some software that "spits bits" to
your "drivers"(you're learnin' the lingo, now) and makes the thing
go.
And you know what you find after all this? That it WORKS!!! You
make parts! And they work, TOO! and then...
You start to notice that parts made on one part of the machine table
don't ALWAYS fit parts made on another part of the table. So you
start to make parts that need to fit other parts on certain "parts"
of the table! And again, it WORKS!!
I did this. NOW, I've done a LOT more. With MUCH better machines. But
you know what? Machines don't hafta cost thousands and weigh hundreds
of pounds to WORK. That's STILL true.
The point of all this is to keep it in perspective. Most ANY cnc
machine you can contemplate will do a good job, within its'
boundaries. Good construction with poor parts beats poor construction
with "best" parts nearly EVERY time.
Ballendo
P.S. Good luck on the "nut" design. We'll be watching for it.
Thanks for mentioning what you will use the machine for.
Have you seen Dave rigotti's hobby CNC plans?
The luberth plotter site(s) I've mentioned?
Neil Gillies Seagull site? (R/C sailplanes and home made CNC
machines/SW)
One of the things I like about this list is its diversity. We've got
guys who hold "tenths" (while they're smilin' for the camera...)
And guys who are wondering what "tenths" are...
Which means we have a large VARIETY of machines to talk about.
You can make a machine from the hardware store parts bins which can
prob'ly do the sort of parts you may need. In models, it's RELATIVE
DIMENSIONS which usually matter, NOT ACCURACY! (I'm using strict
definitions here)
As long as the airfoil is proportionally correct, it will work!
As long as the WWII plastic model "tank" LOOKS right, it is.
As long as the parts FIT together, it all works.
There's NO WAY I'd pay anywhere NEAR $4K(don't know the exch. rate)
for 3 screws/nuts for a "hobby" machine. UNLESS I was going into
production; Then EVERYTHING changes!
Let's look at a "worst case" scenario. You build a machine out of
WOOD! Using 3/4 mild steel (cold rolled) rods for guides. And MAPLE
blocks with holes in them (soaked in oil) for linear bearings. Other
maple blocks to mount the leadscrews. And "hardware store" threaded
rod for leadscrews. You try "coupling nuts" on the screws, but decide
that there's too much "movement"(the word backlash is unknown to you
at this point), so you (big spender that you are) splurge and buy a
TAP!
And some DELRIN (from a plastics store). And make your own nuts. They
still bind, so you buy a DIE and "chase" the threads of
your "leadscrews". Then visit the DUMP and tear stepper motors out of
old IBM printers(and throw away those funny disks with the holes
in 'em attached to the motors,'cause ya don't know what good THEY
could be!).
You figure out some 4 transistor drives from an electronics magazine
article. And puzzle out some software that "spits bits" to
your "drivers"(you're learnin' the lingo, now) and makes the thing
go.
And you know what you find after all this? That it WORKS!!! You
make parts! And they work, TOO! and then...
You start to notice that parts made on one part of the machine table
don't ALWAYS fit parts made on another part of the table. So you
start to make parts that need to fit other parts on certain "parts"
of the table! And again, it WORKS!!
I did this. NOW, I've done a LOT more. With MUCH better machines. But
you know what? Machines don't hafta cost thousands and weigh hundreds
of pounds to WORK. That's STILL true.
The point of all this is to keep it in perspective. Most ANY cnc
machine you can contemplate will do a good job, within its'
boundaries. Good construction with poor parts beats poor construction
with "best" parts nearly EVERY time.
Ballendo
P.S. Good luck on the "nut" design. We'll be watching for it.
>Chris wrote:
>I was suprised, as they were more expensive... I was setup for the
>fall by a chap from the SKF dealership here telling me they were
>significantly cheaper..
>at the end of the day this machine is going to cut balsa and plastic
>for models, probably not a big deal..
>My problem is that I have a quote sitting on my desk for nearly
>$4000.00 NZ for the three screws/nuts I want. Considering this is a
>hobby machine after all
>I'm sure I could still get the end product for well under $1000
>for all three axis, time isn't really a problem,
>The other up side of doing it myself it that if it works I'll be
>able to pass on the nut design to the list.
>Cheers, Chris.
Discussion Thread
ballendo@y...
2000-10-28 18:23:51 UTC
Re: machine design was: Re: Re: re:Re: Lead screws for ever...(plastic)
Randy Gordon-Gilmore
2000-10-28 19:02:13 UTC
Re: machine design
Smoke
2000-10-28 19:44:08 UTC
Re: [CAD_CAM_EDM_DRO] Re: machine design
Chris Hellyar
2000-10-28 19:45:13 UTC
Re: [CAD_CAM_EDM_DRO] Re: machine design was: Re: Re: re:Re: Lead screws for ever...(plastic)