Re: [CAD_CAM_EDM_DRO] re:Re: Lead screw twist... some musings

Smoke wrote:

>
> Yes BUT....how many inventions have been created by individuals who
> were told by the "Experts" that it was impossible?

Not disputing that at all, in fact that sort of what I was trying to
say. I did state that amateurs have a good record in the field of
innovation.
But, to solve a problem, you have to know what you are up against, or
determine the problems by emperical methods.
It's easy for the layman to overlook factors that are obvious to others.
When new data shows up, one has to account for that and see if the
original goal is still attainable, feasible, or practical.

Referring back to the original post. If one's lead screw is twisting
enough to have a tangible effect on accuracy, then one has too small a
lead screw. Fact. If someone wants to try to compensate for that, it's
their choice. I merely wanted to point out that in the real world, the
effects are probably minimal enough that they ought to at least consider
devoting their energies elsewhere. Easiest solution is to slow feedrates
or acceleration parameters below the problem point. The postulated fix
is significantly more difficult, does not fix the underlying problem of
too small a screw, and -may- open a can of worms related to the systems
stability. Putting the encoder at the opposite end of the screw may
result in slightly better positional accuracy, but introduces a lag
between a servo move and a response from the encoder.
I'd think this would result in an immediate increase in power to the
servo motor to try to catch up with the lag. I think this is what Jon
Elson was talking about. BTW, this lag will vary slightly depending upon
how close the nut is to the encoder, regardless of which end it's
mounted to.
Should the original poster pursue the problem, prove it -is- a problem,
and come up with a workable solution, I'll be among the first to admire
and congratulate them on their work.

Jon