Re: More on Encoders&Index

"Arne Chr. Jorgensen" wrote:

> Back to index & encoders:
>
> Jon, you would like to control the EDM current. I don't know what
> you mean with " sinker EDM ",
> - I have seen your "wire EDM". I have tried to understand this EDM
> thing, and I found something that
> looks like you could use a shaped tip. Say you used a circular end
> of a pipe, - could you "burn" a hole with
> this shape ?

Nope, my 'wire' EMD is truly a sinker EDM, using a wire as the 'sinking'
electrode. Wire EDM uses a wire moving between two rollers, and each
roller can be independantly moved in X, Y and Z axes. If you think of
an abrasive-coated wire sawing through material, that is about what a
wire EDM does.

Sinker EDM is a common term (in the US, at least) denoting a fixed shape
electrode cutting a cavity of a shape that is complementary to the electrode's
shape.

Wire EDM denotes a moving wire that cuts out a complex shape that may
have almost any arbitrary contours.

> Okay, - I use the "wire EDM" in the following: You would like to
> control the height above the surface, - by using
> the current. If you are too high, then no current will flow. If you
> are too low, then you have a short. I would believe the
> distance you are talking about is in the micro area, - why would you
> control it with the EMC ?
> Well, I do understand that you would need to or want to follow some
> path, hence use the EMC, but I would think that the
> electrode would need a independent control ?

For sinker EDM, like burning out a tap, or boring a shaped hole straight
into material, you just have to worry about plunging smoothly at a very slow
rate into the material, until the required depth is reached. But, due to the
variations in the materials and the removal of the 'chips', the rate has to be
varied to keep the process going properly.

For making special shapes, etc, you might want to move in some other axis
once the depth has been obtained.

>
> The way I imagine it, would be something like this: use a power
> op-amp in a control loop with a voice coil actuator. You could read
> in the "position" with an A/D converter, - and run the Z axis, - but
> this would just move it when the value came above some threshold,
> the fine movements was up to the voice coil.

But, I already have Z axis control, with resolution of .000025" (roughly
half a micro-meter). I have already demonstrated that is enough! The
backlash of almost .001" is a problem, but also can be handled.

Jon