Re: [CAD_CAM_EDM_DRO] Re: measuring scales and thermal compensation

Ian Wright wrote:

> But this won't be the case. The problem here is that if the workroom
> isn't
> at 20C - and mine, for instance, can often be 10 or less when I start
> working in it - then the machine itself will have contracted more than
> the
> glass scale. So, if you work to the scale, the part will come out big.
>
> Conversely, of course, if the shop is above 20C or the machine has
> warmed up
> through use, the machine will be bigger than the scale and the parts
> will
> come out small. The workpiece may be the same temperature as the
> machine or
> it may be different due to the coolant temperature which, in a cold
> shop,
> may increase the problem. So, make the scale out of the same stuff as
> the
> bulk of the machine, and they both change length by the same amount.
>
> Ian
>
> --

Well, you seem to have missed my point. If you use steel scales, then
you will have some
measure of compensation when machining steel, but not necessarily when
machining
anything else. Even with steel scales, the workpiece temperature is
rarely constant, or
controllable, or the same as the machine's bulk temperature.

What I was trying to point out is that assuming that all errors will be
somehow
magically compensated by making the scale out of the same material as
the
MACHINE is a fallacy. It might, possibly, compensate for machine
temperature changes, but
does NOTHING to compensate for WORKPIECE temperature changes, which
are the worst of all, because that's where the heat is.

The difference in temperature between the machine and the scale will NOT
affect
the size of the part! (Admittedly, if it is 10 C in the shop, the
machine AND the
part start off cold, and smaller than at 20 C.) But, I'd rather have a
known accurate
scale over a range of temperatures, and have to worry about workpiece
heating,
raher than blindly assume that all thermal changes are magically
compensated, and
find out that they AREN'T. This compensation will work fine for ruling
engines,
but just doesn't make sense for milling, turning and grinding. I have
had to stop
and wait for a workpiece to cool before taking final cuts on it, so that
measurements
would be accurate.

Jon