Re: [CAD_CAM_EDM_DRO] robotic workcell design

My opinions are in line below. By the way I believe that Servo is
getting better but it has a ways to go before I will bother with a
subscription. That is a wholly different discusion though.

skykotech wrote:

>I was trying to come up with something interesting to base a few
>articles in Servo around. The thing that keeps popping into my head
>is a robotic xyz workcell. Most of the construction articles are
>about the Nth sumo robot, line follower, etc. etc. A workcell would
>be a worthy project that could be followed up with many articles on
>things to do with it. Something around 18" x 18" by 6" travel (xyz)
>

It sounds like you are attempting to build a machine not an automated
work cell. There is a differnce. To that end a gantry type machine
could be of great interest to the majority of your readers. It could
also be the basis of a work cell but not likely to be one in and of itself.

>
>Some of my ideas:
>
>Add machine vision and a vacuum pickup tool to create a pick and
>place machine, or even a puzzle solving robot.
>

Pick and place (P&P) are just mechanisms that bridge real machine
operations. In industrial machinery these are often as simple as air
cylinder mechanisms. I do not see great value in trying to build such
mechanism unless you have the other tools to hook them up to. In
other words there are more important thing for the average robotics
hobby shop to be developed or purchased. Maybe after a series of
articles, but do realize that such mechanisms imply something to be mass
produced otherwise they are not worthwhile. There are applicaitons
beyond mass production but I don't see the average hobbiest involved in
these.

>
>Add machine vision and a electro mechanical syringe to create a
>solder paste applying machine.
>

There are actually two projects here both worth while. The solder
paste machine could simply be a adaptation or attachement to a
unerversal gantry type machine. Acurracy would be important and
fabircation of the syringe mechanism would be non-trivial but it seen as
part of a larger project it could have big appeal. To get people to
invest all of this money though I think you need a pay off that is
bigger than one machine. This is why I believe that the device should
be an attachment to another more versatile machine. By the way you
should be able to do this without machine vision in an initial introduction.

Machine vision is in itself very interesting but not cheap. This
really should be presented as an independant project that can find use
on a number of platforms.

Both of these projects could be presented in the context of GPL'ed
software running on Linux. This is where one can leverage the cost
issue of software. There are CNC and vision system projects going on
in the Linux world so there is a base to build from. Frankly being
involved in robbotics and ignoring Linux is rather stupid unless of
course you have the $$$ to pay for all sorts of software.

>
>Add sensors to create a 3D digitizing machine.
>

Nice idea. Again I wonder about a special purpose head for a gantry
type machine. Actually a well done head would likely result in
application by a large number of people for referencing positons in CNC
machinery. This is not a trivial project either so don't
underestimate the electircal and mechanical requirements.

>
>Add a small sealed CO2 laser to create an engraving machine.
>

Same as above with one over riding consideration. That is
SAFETY!!!!!!!. These types of machines are often of the gantry type
machines, so we are back to the idea of an unirversal platform. The
problem is that installation of a head like this, introduces very
serious safety issues that can not be ignored. Your machine would
literally require a cover and interlocks not see on the other machines.

>
>
>The problem is coming up with a suitable gantry design that can be
>built for a price. I would like to have a design that could be built
>with minimal hand tools for under $2000.00 using new non-surplus
>sources. I have spent a few days looking into the mechanical side of
>things and I do not think it can be done at that price. Opinions?
>

I'm Assuming that the cost of a PC is taken out of the price equation.

Well it certainly can be done or gotten close to that price.
Generally people involved in these boards may take a different approach
such as buying surplus/used or making their own. The problem becomes
to what level do you want the machine to perform and how big should it
be. If you search around the net you will find all sorts of example
machines built by guys for vastly different applications. Some are
simple gantry machines for routing out circuit boards others handle
large sheets of wood or metal.

The biggest problem I see is your requirement that parts be fabricated
with minimal hand tools. One saves the most buy fabricating as much
as possible with his own tools. The other issue is the parts
themselves will in many cases have to be straight, this requires a well
equiped shop or a skilled craftsmen. So you may bet into a situation
where you have to offer a kit of parts. This has it's own
problems. Thus I think it is better assume certian capabilities that
the reader either has or can get by enlisting friends and neighbors.
There are so many possiblities with respect to materials and
construction that it would take me a very long time here to list them
all, best to search the net for info on homemade machines. Just in
materials you have the following possibilites: wood, plywood, MDF,
Aluminum, steel, concrete, stone, fiberglass, castings (aluminum, zinc,
brass, etc) and more plastics than you can imagine. Do not be
surpirsed if you use all of the mentioned materials.

So is it possible to hit you design price point. Yep it most
certainly is though you will likely have an extremely limited
machine. If that machine is seen as a base for many possible
attachements the price doesn't look all that bad in the end. Though I
would resist going this route, you could build such a machine out of MDF
and steel found in most hardware stores. There is the issue of
fabricating the stuff with hand tools but that can be overcomed.
Again a huge number of options here.

Dave

>
>
>
>
>
>
>