Re: plasma accuracy

--- In CAD_CAM_EDM_DRO@yahoogroups.com, "sargossa_99" <rmcbeath@i...>
wrote:

>
> I've been looking at some plasma machines, and some of the accuracy
> numbers they throw about seem ridiculous. I mean, the cutter

itself

> isn't going to be too accurate, is it?
>
> So, realistically, what sort of accuracy level would you want to
> build a table to? Do you need precision rails, or would cold

rolled

> with care do?
>
> Just seems to me to be a case of why build it more accurate than it
> can possibly cut anyway (therefore much more expensive), so I

thought

> I would ask.
>
> Thanks
>
> Rod

total error is the square root of the sum of the square of all the
errors.

If your table is only good to 0.05 postion and your plasma is also
good to 0.05, then you can be as far as +0.1 one way, or -0.1 the
other, or +/- 0.2. that is nearly a quarter inch off.

There are what I call windows.

Getting a POC machine to hold 1/16 postitioning is easy with stone
knives and bear skin coats.

Getting 0.02 is easy with general home shop novice abilities.

Getting 0.005" is where the wall effect comes in. beyond that, the
costs start to rise geometrically.

The truth is that a steel plate, 1/2 thick will flex a couple thou
when putting things together.

And, there is a a not-often discussed phenonomen (sp?) your machine
may be within 0.00005 for short distances so a small part will mate
up with another small part. but over a foot the machine could be out
0.01" but... the parts still fit easily becasue both parts are out
that much. Not sure if it is a false sence of accomplishment or if
it is the nature of the beast.

A bicycle chain will cog to some degree, but for plasma, that is not
a horrible error.

Dave