Re: [CAD_CAM_EDM_DRO] Re: Optical mouse for DRO anyone? (maybe, But how about...?)

If you look back in the archives of this list or the Mach list, this was attempted by a few different folks and it was found to be unreliable. It is tempting to build a DRO out of a trackball or mouse as all the hardware is there presumably. Unfortunately, it has no precision or repeatability, neither of which are needed to move a cursor around a screen. If I recall correctly it was in the software interface where the issues were and not so much the hardware. Windows priority thing...

Ron

ballendo <ballendo@...> wrote:

>>Has anyone had success using an optical mouse or it's
>>components to make a DRO or an accurately reading scale? If
>>so, can you share any tips, links, or other info?
>>TIA, Chris

>In CCED, "Tony Smith" <ajsmith@...> wrote:
>Do you want the good news or the bad news? Well, it's mostly bad
>news. I do recall people dismantling the old ball mice (usually
>serial), and hacking up something to turn the wheels in those.
>This fact that everyone isn't hacking up their old mice, coupled by
>the fact that digital calipers are dirt cheap probably means their
>success rate isn't high, or it's just not worth the effort.<snip>

Having read all that...

Since this IS a DIY group...

Let's have a look at what might be possible. You've got a ready-made
serial rs232 interface circuit on a relatively small form factor
PCB, and a couple of decent optical sender receiver pairs. And a
really crappy slotted wheel. A bios that reads the pulses and switch
states... And a really crappy slotted wheel.

Seems to me if you;re going to pursue this--and for a DRO, the cheap
caliper scales ARE going to work better; you start by replacing that
crappy slotted wheel...

You could probably get some improvement by "Gearing" the existing
wheel such that it spins many more times per unit of travel. But I
think you're gonna still have poor accuracy due to molding flash on
the slots and inconsistencies in their placement going that route.

Becasue the mouse is designed to detect movement, but with no real
need for repeatability or fine resolution. But IF you pursued an
OLDER mouse (Which as I recall WERE more industrial-encoder-like,
before the engineers realised they could dumb the rodents down quite
a bit and shave costs dramatically) and IF you chemically etched or
optically made a new encoder wheel (using Laser printer or PCB
imaging techniques); you just might get a good enough result to be
worth the effort. ORrrrr...

You might ALSO consider if there are any OTHER places in DIY-CNC
where a low resolution encoder might have some use? MPG's for
example? Index pulses for use with steppers? Tachometer sensors?

And if you just use the parts of an old serial mouse separately...
3 or 4 Switch detection over a ready made serial interface?

Or by using the emitter/detector pairs, how about some CHEAP optical
homes or limit switches for your desktop DIY CNC machine?

Hope this helps,

Ballendo

P.S. He wrote:

>The real downside is that mice are USB these days, and good luck
>building an interface for that. You're stuck with having to plug
>your mouse into a PC (although you might be happy with that).

Yes, you might be happy with that! Instead of seeing the USB as
a "problem"; look at it again as a ready-made plug-n-play(hardware
at least) solution for getting inputs into a CNC application via
USB...

>On top of that you get to deal with everyones favorite, dirt.

Anyone building a CNC machine which cuts has to deal with this!<G>






[Non-text portions of this message have been removed]