Re: [CAD_CAM_EDM_DRO] Newbie Questions

On Mon, 21 Oct 2002, C.S.Mo wrote:

> At any rate, I've discovered that I really can't stand using the dials

I find them functional but annoying.

> 3) I don't expect to have the machine converted to CNC anytime in the
> near future, in the meantime I'd really like a DRO for at least the X and
> Y axis.
> The least expensive method (once you factor in the cost of the
> PC/Monitor..etc) and the quickest to get setup seems to be to use
> digimatic scales (the ones that are kind of like digital calipers without
> the caliper part) with SPC output to an external 3-axis readout. WTtool
> sells the readout for $249 and the individual scales are available for
> roughly $200 for three, so call it $500 once I make the

You can probably get the scales for a bit cheaper but not a lot. Search
for "JTS machinery" on ebay.

> brackets/shipping..etc. The disadvantage to this is, as far as I know, I
> can't get the SPC outputs into a PC.

It's possible to do this but you will probably have to build the device
yourself. I think it's been done before but I can't remember where
off hand. Anyway I don't think the SPC output is fast enough (50
samples/sec.) to use for machine control. It's fine for a DRO but that's
probably about it.

> I'm planning on using servo motors (already have two) with Gecko drives
> for my CNC upgrade. The servo motors have encoders attached so in a sense
> the external DRO would be redundant. So, I *could* get the X/Y motors
> hooked up (which has a lot of other advantages, but will be very
> expensive because all of a suddenly I will have my credit card out and
> will be calling Gecko for drives, and getting a power supply
> and...and..and..) and use the encoders on the motors with one of the free
> DRO software utilities. One disadvantage to this is the servo encoders
> will have to compensate for backlash which may or may not be a problem. I
> assume it isn't a huge problem because it appears to be common practice.

If the encoders are attached to the motors they won't help with backlash
at all. The encoders need to reflect movement of the *table* and not the
motors. The advantage to servos is that the control system will turn the
motors until the table gets to where it's supposed to be, whether it's X
revolutions or X + a little more because of backlash. Where as steppers
just turn a set amount and assume the table got to where it's supposed to
be.

> The other method would be to build my own rotary-to-linear encoder system
> which certainly has appeal and could have better resolution than either
> of the two previous methods. However I don't know that it is necessary...

This might be harder than you think. I investigated this for awhile but
decided it wasn't really worth the effort. Even if you do get a system
that can read to 0.0001" is it really going to be accurate? At that level
all kinds of things start to become an issue; flex in various places like
the column and table, play in the bearings of the encoder mechanism, even
thermal expansion.

> So, bottom line, is there an advantage to having two DRO systems on a
> single mill?

Depends on how long you want to run with just the DRO. If it's going to
be awhile before you switch to CNC then it might be worth the effort.

> If I put on a digimatic scale system will I end up junking
> it once I get my CNC stuff completed?

When you switch to CNC you will most likely be able to sell the DRO setup
for not much loss.

C|