Welcome! Log In Create A New Profile

Advanced

optical xyz positioning?

Posted by Anthong Redbeard 
optical xyz positioning?
July 18, 2010 10:22PM
I've been building my repstrap for a while, loving every bit of it, adding my touches as I go. I've been fanatically reading everything I can about hobby 3d printers.

I ran across the Red Blue CNC [redbluecnc.blogspot.com] and loved it. It got me wondering about some new confiugrations of the machines, I loved the way this one could work below its own surface.

I imagine a cnc that has powered wheels, can sense it's placement optically and print anything as long as you have space for it. I'm guessing that it would require a printed pattern of some sort that runs the length of the print area, and it could keep track of itself like an optical mouse does. This could end up being cheaper than a stepper motor, or at the very least open up some new ways of printing and print dimension limitations.

I'm interested to hear what people think. My goal with this is to separate the dimensions of the device from the dimensions of its products, if you can think of a better way to accomplish that please share.

thanks,
-Anthong
Re: optical xyz positioning?
July 18, 2010 11:40PM
Something I have meant to share since last spring. When I attended the embedded conference, I got to talking with one of the ATMEL reps about the reprap projects and stepper and servo control.

One suggestion that he has was to use several of the inexpensive usb enabled digital calipers, which could be used to supply real time feedback from the table position. Clamp one side of the caliper to the table and the other to the bed.

The downside of this, would be the requirement for at least a chip that supports USB on-the-go. This would pretty much require an ARM processor, although ATMEL does make some versions like the AT90S128USB which is similar to the mega128. Not sure if there is an Arduino Library for on the go. There is an open source project by Dean Cameron in Australia called LUFA which could provide this.

-julie
Re: optical xyz positioning?
July 19, 2010 04:40AM
Quote

I imagine a cnc that has powered wheels, can sense it's placement optically and print anything as long as you have space for it.

How about the positioning part of an optical mouse? This is cheap (mice start somewhere at $ 6,-), has high resolution (800 dpi or more) and works on almost any surface. I'm not sure how reliable this is, however. Mouse-users don't care if a tenth of a millimeter is dropped, a 3D robot will collect such errors over time.


Generation 7 Electronics Teacup Firmware RepRap DIY
     
Re: optical xyz positioning?
July 19, 2010 10:46AM
You will not be able to verify the internal Firmware without access to it, which is quite a drawback.

So there is no way to be sure what the mouse really does and how. You are left to experimentation, deduction, reverse engineering.

That's the typical problem of non-open Hardware/Firmware.
VDX
Re: optical xyz positioning?
July 19, 2010 03:16PM
... go through the linked posts in this search - in a german robot-forum we programmed the image-transfer from a mouse-cam to a micro ... the program is for free too ...


Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]
Re: optical xyz positioning?
September 26, 2010 08:57AM
sheep Wrote:
-------------------------------------------------------
> The downside of this, would be the requirement for
> at least a chip that supports USB on-the-go.
> This would pretty much require an ARM processor,
> although ATMEL does make some versions like the
> AT90S128USB which is similar to the mega128. Not
> sure if there is an Arduino Library for on the go.
> There is an open source project by Dean Cameron
> in Australia called LUFA which could provide
> this.
>
> -julie


The new Aruino Uno & Mega have a much better USB support that is using LUFA [twitter.com]
Re: optical xyz positioning?
September 27, 2010 03:13AM
[teamcyberdyne.blogspot.com]
[reprap.org]

This is about as far as I've gotten with using calipers as an XY table. I think the idea has promise, and still intend to develop it further, but I've been busy lately. One improvement I already have in mind is shifting to a gantry design and using the 12" calipers for the X axis.
Re: optical xyz positioning?
October 01, 2010 06:37PM
I was trying to track down USB output digital calipers. From US sources, they are expensive, especially if you want a build platform bigger than 6". Then I ran into a site that mentioned DRO digital scales from China on eBay for cheap. The interface is not USB but more like the synchronous serial that the microcontrollers a used to. Some send multiple nibbles in BCD, others 24 bit binary. Some report on demand and others every 300ms. But the prices and sizes are good, and I think I will base my cubist, gantry crane like Open Air design on these digital scales, so that I can used motors or steppers, and in either case know that a missed step will not affect the piece being built.

Mike
Re: optical xyz positioning?
October 01, 2010 11:41PM
You got me thinking. I just pulled apart a bubblejet printer and it has a printed fence (a mylar sheet with vertical bars) that is read optically which can also be slavaged. I would expect most printers would have these, so that could be a cheap source to determine the location of the x & y axis (and z for that matter).
Sorry, only registered users may post in this forum.

Click here to login