Re: [CAD_CAM_EDM_DRO] Conversion Of Mill to CNC (LONG)

camerashy44 wrote:

>Although I have been monitoring this list for some time this is my
>first post. I am a complete novice at both milling and CNC so need the
>help and advice of the very experienced and knowledgeable members of
>this list.
>
>I have recently purchased a GEM500G mill/drill from Busy Bee. Yes I am
>in Canada. The mill has a 32 x 9.5" table and is a gear head mill with
>dovetail column. It has a x axis travel of 18", y axis travel of 14"
>and a z axis travel of 18.25". It is my intention to convert this mill
>to CNC. There are pictures of this mill in the photos section. It is
>the mill of Allan Reinhard who is also a member of this group. He has
>also converted his mill to CAC using 720 oz/in steppers and Geko drives.
>>From what I have been able to pick up, servos seem to be a better
>alternative than steppers. Is this true and what are the advantages
>disadvantages of each?
>

Servos with encoder feedback to the computer, allow you to switch back
and forth between
manual with DRO and CNC. It also allows the computer to constantly
monitor actual
position, and cause an emergency stop anytime an axis lags behind the
commanded position.
They also allow finer control, since there are no discrete positions to
the motor. Finally,
most servos, even without encoder feedback to the computer (such as
Gecko 320) should allow
better smoothness at low speeds. True velocity-mode servos with
tachometer feedback to
the servo amplifiers can achieve incredible smoothness at very slow
speeds, important
for very smooth walls when milling large holes.

The major difference from a machine motion standpoint is that servos
retain full
torque up to the maximum speed, while steppers lose torque as soon as
they start
moving. That's why steppers are rated at HOLDING torque, while servos just
have torque -at all speeds. Also, servos have a continuous torque and a
peak torque.
The peak torque is often 4 or more times greater than continuous.

> I want to have enough power to make sure that I
>never have a problem and am inclined to ere on the higher power side.
>I was looking at Sanyo-Denki P1 series servos, the 1kW size, 2000 rpm.
>

That is roughly 1.25 Hp! Are you prepared to drive such a big motor?

>This servo has a rated torque of 665.5 oz/in and a continuous stall
>torque of 835.5 oz/in. I was told that this was overkill and that a
>425 peak oz/in 2500 rpm servo would work fine with a 3 to 1 gear-belt
>drive. I am sure that this is correct at least for the x and y axes.
>However the z axis(Column)takes more manual power to move especially
>up and I like the idea of all axes motors being the same. I want the
>simplest cleanest(neat) and most trouble free system possible. Yes I
>want a lot!
>What are others opinions on this? Is gear-belt drive superior to a
>direct drive?
>

Direct drive requires motors with great torque at very low speeds, and
that means a big, expensive
motor. Belt drive allows a reduction ratio, allowing a much smaller
(and cheaper) motor to
be used. Also, direct drive requires very expensive high-torque
couplings to be used, much
more expensive and harder to align than belts.

>I like the Camtronics kits and the idea of getting everything from one
>place like Dan offers. However I have to consider that if I ever need
>to replace or repair any part of the system, items that are available
>here in Canada would be preferred.
>I am not going to go into ball screws yet but I know that these will
>depend on the motors that I choose (pitch) so comments on these would
>also be appreciated.
>

Ball screws make the machine precise and repeatable. With Acme screws,
there is no guarantee
the table is where you think it is. I really don't think a full-size
machine should be converted
to CNC unless it has ballscrews or at the least, compensating nuts.