Re: [CAD_CAM_EDM_DRO] General machine discussion, was Bent Nook ballscrews ?????

On Wed, 2003-08-13 at 03:17, ballendo wrote:
> Kim,
>
> I'm confused... Here's what I understand so far:
>
> You have some motors which some members of the group told you were
> too small, saying you wouldn't get good performance using them...

Yes. We are using Camtronics Servos, Geckos, Nook 5 TPI ballscrews.
This setup has tons of power. We are running with 3 Amp fuses, but the
motors will take 10 Amps. 12 Amps is supposed to be the demagnetization
current limit.

Story: a bad Gcode command by our operator (not me) ran the tool into a
10 mm barstock about an inch out from the chuck and bent it about 1/8".
No lack of power here. The motors run cool all the time.

Story: We were roughing out CNC spindles yesterday with 0.020" cuts at
F16-20, depending upon how aggressive we were with the feed override.
This fully loads the spindle motor on a 13x40 lathe.

Story: we slowed down the spindle speed and were taking 0.100" cuts.
The motors are fine with that loading.

I highly recommend this setup over stepper motors. The tuning was no
problem at all.

At 900 RPM, this setup yields a G0 speed of 180 IPM, which is a bit on
the slow side. If I were building another machine, I'd aim for 300
IPM. However the software we are using (TurboCNC) currently limits the
stepping speed to 28ish KHz, which works out to be 140 IPM anyway.

> But
> you ARE getting good performance from them, and you attribute it to
> your anti-backlash nut design, which is of a non-compliant 2 nut
> spacer type.

I just know that this setup works. The non compliant backlash setup
works very well. No problems whatsoever and everything is tight. I
*suspect* that others have motor power problems because of a number of
reasons, one which may be backlash setup.


> And you have this design in operation (production
> setting) on both ground and rolled ballscrews. ?

Yep. We've got a lathe cranking parts as I speak using rolled screws.
The mill uses ground screws and is just becoming operational. There was
no real difference setting up the axes on either one.

> Also you post that the required torque from a given motor is FAR in
> excess of what many of us have experienced first hand in our own
> machines. This seems to be in conflict with the first part above???

I don't know what you are referring to here. The torque required to
move a load with a single ballscrew is given by equations, ie for 5 TPI,
force = 31ish x torque (in lbs). The only points I've made about motors
are:

a) the Camtronics motors work well when others told me they wouldn't.

b) others had problems with a similar setup to what we are using.

> So please help me understand how the motors which were too small in
> the first place (according to those who believe-perhaps erroneously-
> that the preload only adds a slight drag to the assembly) have great
> performance, since by your stated calcs they "should be" even worse?!?

I think you are confused here. I haven't stated anything should be
worse. By calcs, our motors should yield about 1000 pounds of force
driving the 5 TPI ballscrew and they do. Nothing revolutionary. It
just works.


> If you're onto something, I'll go ahead and listen... Maybe we can
> get back to something which will help us all build better machines.
> So how does this all fit together to give you great performance with
> small motors? or cheaper screws?

I think there is some advantage to using the non compliant backlash
adjustment over the spring loaded adjustment because there is no
constant drag on the ballscrew. Maybe that is why our setup works and
others don't.


> Thank you in advance,
>
> Ballendo
>
> P.S. FWIW, loading ballnuts with oversize balls to create a negative
> clearance IS something the ball screw and nut mfrs. HAVE been doing
> for decades...

If that were the case, then there would be no reason for a two nut zero
backlash ball system.


> --
> Kim Lux <lux@...>