backlash comp was Re: I need some help with a shopping list. . .

Hello,

Note that Jon's excellent answer about backlash need not apply to
DRILLING-only machines. In this case, large amounts of backlash may
be corrected-for and accomodated by S/W compensation as the moves to
compensate are made while there is no cutter in the part.

Also, machines with smaller amounts of backlash (good enough for
milling) which need to hold tighter tolerance can use the above info
by using b/l comp for/during any drilling type operation. There is
also single direction positioning (G61, usually) for the same
reasons/ purpose.

Hope this helps.

Ballendo

P.S. to control writers: Might be worthwhile to have a g or m code to
turn b/l comp on and off from within the program. I don't recall
any "standard" code for this, but will look into it. For now, maybe a
G61.1 or M61 ??? (I'd favor the M code)

--- In CAD_CAM_EDM_DRO@y..., Jon Elson <elson@p...> wrote:
> allenstepper wrote:
>
> > I am a new member here and I am wanting to convert my Cinncinatti
1-D
> > toolmaster (1967) mill into a CNC machine. I want to be able to
> > control x,y, and z axis from my PC. Is there any easy way to do
> > this? Is it possible to just take off the handwheels of the three
> > axis and bolt on some drive motors hooked into the PC? Am I
being to
> > simplistic in my thinking? What kind of motors will I need to
move
> > the table? Right now it has a 1HP three phase. There is a fair
> > ammount of backlash in the table as well ( about 25thous), will
this
> > mess everything up or does the backlash compensation from the
> > software compensate that much?
>
> Well, the original Acme screws can be made to work, but the results
will
> not be very good. .025" backlash cannot be compensated by software,
> as the time it takes the motors to go from one side of the backlash
to the
> other is considerable. During that time, the table is not
constrained by
> the computer, it is free-floating. Also, the risk of an accident
caused by
> the table (and workpiece) being pulled into the cutting tool is a
major
> problem in certain operations. Climb milling would be out of the
question
> except for the lightest finish passes. Any attempt to make large
circular
> holes or pockets would come out with bumps at the points where each
> axis reverses. I have .0015" backlash (some in the ballscrews,
some in
> flex in the machine, rocking of the table, etc.) and it causes my
circles
> to be out of round. With 20 x the backlash, ouch! Also, with this
much
> wear, the screws are also worn, and so there will be varying
backlash
> at different positions.
>
> You really need to tell us what you intend to do with this
machine. Some
> people want a fancy machine to drill holes, and +/- .010" accuracy
and
> .005" repeatability is fine for them. others want to make
miniature Merlin
> engines or jet engines, and they really are serious about holding
tolerances
>
> to .002" or so on the finished part. So, without further detail,
it is hard
>
> to advise.
>
> I don't know the Cinci 1D, is it about the size of the standard
Bridgeport,
> and a knee mill?
>
> Jon