On Tuesday 28 July 2020 15:38:19 Chris Albertson wrote:

> I'm looking at this imported step file.    I see many details I'd
> change. For example the "bearing cap" has some M4 screws thread into
> holes with no threads and the blocks with the unthreaded holes look
> like they would break.  I'd fill that area with plastic or use some
> big fillets.  Lots of detailed design changes are needed.  Some are
> just personal preferences (like brass inserts vs nut pockets) and
> others are fixing shortcuts the designer took.

Your thoughts are correct, I just checked the hole size against a 4mm 
reamer, and its bigger, enough the screw will make its own threads. I 
don't care for that either unless the screw is installed hot. But all 
I'll do is ream the hole in the top plate below the head of the screw 
for sample #1..
>
> Also, it would be better if the unit was designed around 608 skate
> bearings just because those are so cheap and easy to get but I'm not
> going to de-dimension the entire unit for this.

608's would be bigger, which would expand it all.  And these aren't that 
much pricier.
>
> But I bet this would work as-is.      The idea is that you build one,
> see what needs changing then build a second version.

Precisely.

> After perfecting 
> it then I assume everyone reading this owns a CNC machine, you make
> some parts in metal from your perfected CAD files.     Certainly, the
> housings and output plates could be made on a hobby-grade CNC machine.
>  That is the best use of printing from CAD files, the same files can
> drive CNC or 3D prints so you prototype in plastic before cutting an
> expensive block of metal.

Not to mention the body alone is over 39 hours to print. Maybe even 40 
hrs. :(

> I bet I'd never find the exact motor the designer had and I'd have to
> adapt the design to whatever I wanted to use.

I'm salvaging old 5.25" drives, figure on spinning off some of the tape 
pulley, and double flatting it once I am inside the SQ nuts. And JBWeld 
it once it runs by itself.  That 5mm shaft is a huge weak point, nowhere 
near enough materiel in back of the nuts. JBWelding the top to the nut 
risers will also help considerably. I've not yet found if the drivers I 
have can be turned down enough to keep the motor cool enough to be 
bolted to the PLA.  Thats another consideration...

> If the goal is to drive a machine tool axis this would work as it
> can't be back driven like a belt of spur gear reductio can.

Ditto a well balanced Meade DS-10 scope, which may have a half pound 
eyepiece in the holder, or a 2+lb 35mm SLR film camera. Suitable 
balancing counterweights will be needed.
> On Tue, Jul 28, 2020 at 11:41 AM Gene Heskett <[email protected]> 
wrote:
> > On Tuesday 28 July 2020 12:52:09 Chris Albertson wrote:
> > > Don't scale the entire part unless caliper show you the entire
> > > part is scaled too small.  (this is very unlikely)  Better to
> > > simply adjust the one hole you need to be larger.   Or if it is
> > > less than about 0.1mm put the part on a lather and bore it out to
> > > size.   I did this to all my timing belt rings to make them fit
> > > perfectly to the 24mm hubs.   If you go slow so as not to melt the
> > > PLA the carbide boring bar works and just took off "fuzz"
> >
> > holes aren't bothering me, other than hard toremove support crap.
> > What is bothering me is the incomplete mesh I saw in the movies, and
> > that easily fixed by changing the printer x and y scales.
> >
> > I think its obvious now that whats published is not 100% optimized.
> >
> > > The hole seems small but usually, it is because the plastic was
> > > not laid down perfectly and there are bumps and ridges and lines
> > > inside the hole. Make the hole perfect rather then making another
> > > larger not-perfect hole. Also it is MUCH quicker to bore the hole
> > > then re-print the part.
> > >
> > > The other trick I do is to make "temporary" bearings with either
> > > aluminum or even PLA while waiting for the real bearing to arrive.
> > > You could make them on a manual lathe in a few minutes and then
> > > you could verify the parts work and there is not OTHER problem
> > > before doing another 16 hour print
> > >
> > > I'd be surprised if you did not need to clean up the fine pitch
> > > gears with a nail file.
> >
> > They are amazingly clean after removal of the brim. 10 minute job to
> > clean them up.
> >
> > > Most 3D prints require some post-processing with a
> > > tiny file and maybe #600 wet and dry paper.    A boring bar makes
> > > the inside of holes smooth as glass.
> > >
> > > A read a translation of the designer's blog page and it seems he
> > > wants to make a 6 axis arm with these.   He will probably need to
> > > make some of the parts near the base of the arm in metal
> >
> > I wouldn't be surprised. These are, if built as published, going to
> > have enough backlash to wreck a 6 joint arm. What I'm getting off
> > the printer, does not always match his pix.  Properly fine tuned it
> > definitely has commercial prospects to be exploited, and I can do a
> > bit of that fine tuning.
> >
> > Thanks Chris
> >
> > [...]
> >
> >
> > 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)
> > If we desire respect for the law, we must first make the law
> > respectable. - Louis D. Brandeis
> > Genes Web page <http://geneslinuxbox.net:6309/gene>
> >
> >
> > _______________________________________________
> > Emc-users mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/emc-users


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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
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