Re: long lead screws

Hi Elliot,

Very interesting analysis.

I an not overly concerned about cumulative accuracy, if I can get
within about 0.015" over 5 feet, I'll be really happy. If I get
0.005" I'll be laughing for a few days !

Also, since the drive is in only one direction, backlash will not be
a concern either.

Gottal love real world spec's.

I have 'lost' some of the length. There is a part holding table and
an extention that cut the actual table travel distance to 9 feet.
Of course when I make the long machine, (19 feet travel) I will have
more problems and tighter tolerance. I will be shooting for 0.005"
over 120 inches. Hey, we all have to set goals.

regarding sagging of the screw, well that is another matter.
Vermicelli comes to mind, al dante. The dip is closer to an inch,
but since it rests on the bench at that point, I can't measure it any
better. : )

I've come up with a way to support it in sections with a hinged
support that will be pushed out of the way as the table travels, but
have not implimented that fix as yet. I'm putting together the power
supply for higher voltage and cleaning up the wiring.

I was thinking of going to 1 inch ACME in the next iteration. That's
1 inch, 9 feet long. Prices I've gotten so far, shipping is about
the same price as the screw.

Dave

--- In CAD_CAM_EDM_DRO@y..., "Elliot Burke" <elliot@h...> wrote:
> A recent email about a 1/2"-13 12' long leadscrew caught my
attention. I
> modeled it in SolidWorks, imported into Algor (FEA program) and
analyzed 1 g
> case and the lowest 5 modes of vibrations.
> Results are as follows:
> for a .407" diameter (minor diameter 1/2-13) CRS rod, 12 feet long,
> rigid support at one end and sliding fit at other end (axial motion
allowed)
> deflection 1 g is .512 inch at center
> 1st modal frequency is 4 Hz
> 2nd 5 Hz
> 3rd 11.2 Hz
> 4th 13.8 Hz
> 5th 22 Hz
>
> Even adding a few supports won't help much, as there are two nodes
between
> the ends for the 5th mode.
>
> I suppose you could tension the screw, like a big guitar string.
That might
> be fun to analyze too.
>
> I understand this is a rotating nut design, so the screw won't be
twisting.
> But the force applied into the screw by the nut will make it bounce
around
> quite a bit. Sorry I can't analyze what happens with axial
impulse, yet.
>
> And when the nut is at one end of the screw it will sag by .5 inch.
>
> Let us know how the long screw works.
>
> Elliot