Re: [CAD_CAM_EDM_DRO] MSA 665 linear Encoder

Steve Haushahn wrote:

>----- Original Message -----
>From: "Thomas" <teecee@...>
>To: <CAD_CAM_EDM_DRO@yahoogroups.com>
>Sent: Thursday, November 24, 2005 11:08 PM
>Subject: [CAD_CAM_EDM_DRO] MSA 665 linear Encoder
>
>
>
>
>>Hi Tim
>>I figured it out, that part is working nicely !
>>
>>What I want to do is plot a profile from a small Camshaft into Solidworks
>>
>>
>either Directly from X & Y Coordinates using a Linear and Rotary Encoder and
>just left click to create the point in CAD then join the Dots with a spline.
>
>
>>Taking this a step further I took a cheapo Serial PS2 PC mouse apart and
>>
>>
>started messing around inside with the X & Y axis and connected the 2
>Encoders and the Cursor moves nicely.
>
>
>>Does any one now of any open source Drivers or a suitable program for the
>>
>>
>Humble ol Mouse or this type of interface ?
>
>
>>I will need to set this up so the Encoders have a Scaled or known movement
>>
>>
>on the screen the Rotary Encoder is 360 PPR and the Linear Encoder has a
>Grating of 0.04mm ?
>
>
>>Would this be accurate in Windows ?
>>
>>any thoughts
>>
>>Thankyou
>>Thomas
>>
>>
>>
>Thomas
> I certainly have some thoughts on the subject.....I think you are doing
>what I want to do..Actually , I think what you are doing is very important
>work that somebody needs to do and make available to us types who havent the
>interest in developing our own electronic circuits and software....Lemmee
>tell you my application and idea conception - see if this dont parrallel
>your own ideas..
>.
> Just about all my machineshop work involves either copying an existing
>part or making a part to fit an existing part or welding up and recutting an
>existing part....
>.
>.For sake of explanation I'll use a lever from a lever action rifle as an
>example..We want to make a quick replacement and we have one to
>copy....Digitizing is slow.. Mach 3 Wizard 'Teach and Learn' does work -
>takes me about 20 minutes to 30 minutes to jog around and map the
>coordinates and check for errors...............This I believe is best
>currently available method for quick coordinates mapping.
>.
>...What would be far better would be to have a coordinate measuring machine
>of sorts sitting to the side dedicated to mapping coordinates of lines
>,radius and hole patterns..This CMM would be in shape of small vertical mill
>without having any lead screws.Table moves freely with human force
>alone......... Clamp sample workpiece to tabletop , lower carbide stylus in
>quill down to edge of sample and then push table in X and Y directions so
>that stylus follows outside the lever and inside the lever loop --- taking
>mere moments. Change stylus to 60 deg. V point - drag table over till pivot
>pin of lever is under V point and pressing down the quill - the V point
>enters hole and pulls the table into perfect alignment with the
>hole.....All coordinates being recorded , you then in good shape to step
>over to big mill and replicate the sample part or use the coordinates to
>work up a mold or dieset.
>.
> If anything is wrong with this idea it would be that such a unit would not
>be capable of extreme accuracy so would not do for some applications..That
>being because of compromize in setting gibs loose enough to move with just
>human power...But in the real world of hobby shop or maintenance/repair
>shops and a lot of mold shop work , the accuracy that machine would be
>capable of would be well adequate.
>.
> Dunno if anybody else would spring for one but I want one.
> Steven
>
>
>
>

Actaully such a machine would be pretty easy to build. all roller
slides, a counterweighted Z would work for the 3D aspects. An easy
encoder interface for positioning might be just some encoders with a
rubber wheel against a flat surface. The harder part would be the
stylus offset to the work to get surface coordinates, One way around
that is to use a cutting tool with the same radius as the stylus and
work with no tool offset in the final path.