On Saturday 30 May 2020 14:12:28 Chris Albertson wrote:

> [image: image.jpeg]
>
> On Sat, May 30, 2020 at 7:31 AM cogoman via Emc-users <
>
> [email protected]> wrote:
> > On 5/29/20 8:44 PM, Gene Heskett wrote:
> > > In this case, openscad gives me the tools I want right now, and I
> > > could even do a set of gears for stage 2 of the reduction since
> > > that spacing is almost too close for a belt coupling. So by using
> > > the correct tooth profiles it could make a very usefull gearset,
> > > no belt at all, and a higher reduction at the same time. But would
> > > those teeth be strong enough. Worth a try IMO. And a swarf cover
> > > from the printer should be doable.
>
> Two points:
>
> 1) can you (you, not someone else) actually modify a downloaded gear
> design using openscad?  Try it Go get any random gear from SPD/SI or
> Grainger and change the hub and face width and save as STL.     I bet
> this is not easy. It is nearly trivial in Fusion or Onshape or any
> that work like them. Seriously, go through the steps using each
> system.   Downloading and installing 3 or 4 CAD systems takes days but
> doing the same test project in each is the only way to see what each
> can do.    Downloading a gear and making a trivial modification like
> (boring out the hub) then saving as STL is a good test case.
>
> 2) Yes, the TEETH would be strong enough to carry the force from any
> reasonable size motor.   Remember that a 1NM motor can only generate
> 1NM of torque.  What is the radius of the gear?  lets guess at 0.1
> meter.  If so then the force n the tooth is 10 Newtons (or in
> barbarian units about 2.5 pounds) and this is in the worst case where
> the motor is stalled at full current against a hard stop.
>
>   When you slice the gear you can specify the wall thickness and the
> infill percent and style.  For gears of this size in PLA I'd go with
> 1.6mm walls and skin and 50% infill density.    Cutting those numbers
> in half to 0.8 and 25% would still make a serviceable gear for a small
> < 2NM motor.   PLA is the most rigid plastic.  It is not the toughest
> against impacts but for driving a CNC machine rigidity matters more
> than impact resistance
>
> I've been doing experiments with plastic gears.  I can print small
> ones down to module 0.5 but the percent error in tooth shape is very
> high for small gears.   For my self-drive car project I am using
> module 2.0 spur gears with 24mm face width.  These gears are huge but
> the front tire will be hitting rocks and the computer will be making
> fast steering corrections. These gears are very strong so that
> "something else" will break first.
>
> But notice I said the TEETH are very strong.   In a gear the stres is
> greatest near the hub and minimal on the teeth.  Larger diameter gears
> have less force on the teeth then do smaller diameter gears.   So you
> always want to use the largest gears possible.      If the gear fails,
> it fails at the hub.  It slips on the shaft after the keywhy shears
> off of the threads holding the setscrew strip out.      With PLA gear
> slipping on the hub is a "death spiral" because slipping make heat and
> heat softens the PLA and makes it slip more.   And the hub to shaft
> interfce is the highest stressed place.  It fails if not
> over-engineered.
>
> My solution is to make my plastic gears with a 24mm (or larger)
> diameter borehole.   Then turn a 24mm steel bushing that fits the
> shaft and has set screws.  he gear is press fit with epoxy to the
> bushing.    This way the hub is steel and will never fail and the
> teeth never do fail.    (Later I might try making hexagonal hub.)
>
> That said.   When you design the gear, just figure that PLA s about
> 1/3rd as strong as metal and give the plastic gear a wider face and
> try to use module 1.0 size at least if you can.
>
> Finally one more thing:  With plastic gears, you really can benefit
> from using *helical* gears as more tooth area is in contact and they
> slide better.  It costs no more to print a helix than a spur gear.
> Next, you think you don't want to deal with axial loads (side force)
> so you use TWO helix gears to make herringbone gears.  These have zero
> side force, are self-aligning, and cost no more than spur gears.  
> MANY people doing gears with printers use this kind of gear.  They are
> expensive to cut in steel but cost no more in plastic.
>
> If you have a CAD system herringbone gears are easy.  (1) find a helix
> gear some place like Grainger or McMaster Carr,  Download both the
> left and right-hand versions.  (2) in the CAD system stick them
> together, side by side, (3) cut the face width as required, (3) bore
> the hub to 24mm, (4) slice and print.    Try doing the full exercise
> using OnShape then Oopenscad then with fusion and see which works best
> for you.  (If you are a hard-core Linux user I bet onshape works best)
>
> As a further exercise try getting you CAD system to make g-code for a
> 4-axis mill to make the same herringbone gears.  Try using a tiny
> ball-end cutter.
>
> Here is a photo.
>
> [image: images.jpeg]
>
> > Strength of 3d printed parts is a deep subject.  PLA if done right
> > can be pretty strong, but other filaments done right are stronger.
> >
> > I like this video showing how a veteran at 3d printing goes about
> > refining his process for printing in nylon.
> >
> > https://www.youtube.com/watch?v=5MFX1whwjGg
> >
> > Here is a video where some guys test out different filaments on a 15
> > horsepower outboard motor.
> >
> > https://www.youtube.com/watch?v=scUEEQyC0GI
> >
> >
> > Keep in mind, the odor will be much worse using any filament besides
> > PLA, and most will need a hot end that can do the higher
> > temperatures.
> >
> > Also with higher temperatures you might want to put a box around the
> > printer to keep drafts from messing up a print. One guy uses grow
> > tents around his printers.
> >

All of which are very good points. 

Funny you should mention Onshape though, formerly Solidworks. I made the 
mistake years ago of downloading their so called freebie that had a 30 
day free trial, then found it had zero help unless you had paid for it, 
and I was totally lost, never having dealt with any CAD like program 
before.  So I deleted it, then spent the next year feeding their legal 
threats to spamassassin and replying to them by denying that a copy of 
it existed on any system I owned.  To say they were insistent assholes 
about it is being way nicer than they deserved.  They did every threat 
in the book short of sending it to a bill collector.  I half expected to 
get a registered letter at anytime, demanding I cough up the $600 they 
wanted, but eventually they gave up.  So you can guess how fat the 
chance is that I'll ever deal with those people again.

> >
> >
> > _______________________________________________
> > 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