I am new to the forums, and I hope I've stumbled onto the right place to post this strange request. I'm wanting to build a device that will be part of an installation in a kid-friendly (young kids) hands-on part of an art museum, and it will be in use for a year. These are the characteristics of the device as I see them, but are open to change:

- Large, 8x8 to 10x10 feet square
- Drawing is achieved via a high-intensity LED flashlight shining down on a phosphorescent table.
- The framework would be built from found or off-the-shelf parts. I expect to use full sized, complete skateboards for the X-Y tracks, for example. We're talking low tolerance, simple drawing here...
- I don't mind spending a little money on the other hardware. It sounds like servos with encoder feedback would be better than steppers, because I want it to draw and move quickly.
- I only need it to draw simple polygons. I have tried Inkscape running on a Mac OSX via X11 and it seems to be a perfect starting point.
- User input: pushing one of four buttons makes it draw a shape. Ideally each button would correspond to a library, stars, for example. Push it once, a star is drawn. Push again, another different sized one in a different location would draw, etc. This could be randomly changing sizes and location of one star each time, or going through a library of a couple dozen stars.
- I initially envisioned this running off its own processor, like an Arduino, but it could have a dedicated Mac or PC if necessary. I work on a Mac, but I'm not opposed to using a PC...I'm well aware of the lack of CNC software for the Mac.

I know there's probably lots of details I'm forgetting or simply don't know about. BUT, thoughts on how this might come together? I'm hoping and assuming that there is no single right way to build something like this.

Thanks!

It sounds like the only forces you have to take care of it's the portal itself (if you're aiming for a portal design, that is). Check the DIY-router forum and its subforum or the Photo gallery to see different ideas how people have made their own designs.

If I were you, I would go for two "raised" rails to keep moving things away from children fingers etc in the table level, as they seem to find all the places were they shouldn't put their hands... Steppers or servos doesn't really matter I think, you have no cutting forces. Also, I would drive the gantry with a belt on both sides having an axle in the rear end across the table running the two belts. I'm an EMC-fan and EMC is a great controlling software. If you're not afraid of Linux and some manual tweaking it's awsome.
Something like this is what I have in mind:
http://www.cnczone.com/gallery/data/505/medium/mill31.jpg
http://www.cnczone.com/gallery/data/505/medium/DSC_3530.jpg

Sounds like a fun project! Hope you'll get it up and running.

I'm surious about this too. Would you have teh gantry pen holding part up high so that curious children fingers can't reach into it? And how fast would it have to draw to make the patterns on the phosphorous? would you use like.. a sheet of plywood with glow in teh dark paint on it? covered in plexiglass to keep it from getting on kids fingers?
I suppose in a slightly darkened room you might see the patterns drawn on it for a while...

maybe use an LED flashlight and replace all the LED's with UV led's to get a maximum glow effect out of 'em.. and use some kind of colimating lense so that they don't spread out and cover too large of an area?

as for the button thingy.. you might be able to do that and have a linux process that reads your buttons and then invokes EMC with a parameter string so that it will draw a pattern fed to it, then exit and wait for the next occurance... Dunno if that's possible or not, but it might be.

sounds kind of neat, actually. :) where are you located and where will this toy reside! I wanna play with it!

Oh, and you dont' need servos.. a big 1000 oz inch stepper motor can move things in a hurry.

Although I don't have motors or electronics yet, I am building the hardware side of it now to better get a sense of what is needed to power it, and to also show you all some real photos. BTW, this will be at the Wichita (KS) Art Museum.

This definitely will be raised to keep kids' fingers out of trouble, and will likely have plexiglas on all four sides just to make sure. I have both standard phosphorescent pigment as well as a high-persistence type on hand, so painted on plywood and in dim light this should work well. I already have one high-intensity LED flashlight (if anyone doubts the brightness of LEDs, try out one of these guys that puts out something like 150 lumens...whoa, they are BRIGHT). I actually considered one of the UV output LED models but figured people shouldn't be staring at it even if there is little light spill and is all 100% reflected light. My tests show that traveling at 1-3 inches per second leaves a very adequate image, and this is without any lenses that I plan to focus the light with. Fortunately, due to the size of this table, I can throw a beam an inch wide and it will work well drawing images 6-8 inches across. Precision is not going to be a factor.

I've read so much about steppers and servos and geckos and choppers, it's a lot to digest, and I'm not much further along in understanding what to buy. That's why I'm just building the table first, and hoping for suggestions after I post photos. I do know that I want it to move fast, if money allows it. Somewhere I ran across these, and while I like the idea of buying everything as a unit, the data says they can run 24 inches per minute, which sounds very slow to me. I was hoping it could travel 24 inches in a few seconds:
http://www.stepperworld.com/FET3dynamo.htm

I'll study up on EMC, since that seems to be a common suggestion. Thanks for all the input, and I'll post photos soon.

woof! WHY CHIT UH! I'll HAVE to go there to see this. I live near springfield MO so I'll be able to head up that way some weekend to check this out. :)

Okay, you want about 3 ips.. that's 180 ipm.. that's.. not VERY fast, but it IS fast. In order to figure out what kind of steppers or servos you would need, we need to figure out a few things.. First of all is the mass that has to be moved , in the worst case you would have a "fast reverse" where you have to decelerate the mass of the carraige and then accelerate it in the opposite direction. The reason we would need this information is to determine how much torque the steppers or servos would need in order to move at a rate sufficiently fast.. this allows us to figure out the maximum or minimum gearing necessary to acheive those rates for the given stepper.

So if you use a gantry type system, welll.. you 'll need bigger motors.. bigger motors means more money...

HOW ABOUT using something like.. a stepper at each corner that winds a cable around a drum and pulls on a central assembly that holds your led flashlight? Kind of like those stadium cameras that travel around over the football field.. only yours whould never 'leave the ground. The power to the led's would be a coiled cable (maybe even an airhose with small wires fed through it so that it acts like a spring and keeps the wires from dragging )

The advantage here is that you'd have a much lighter assembly to move around, and so could use smaller steppers.

The disadvantage is that it would require custom software... though MOSTLY that would be in the gcodes needed to move to a certain point. the code that figured out the gcode to feed to EMC would be.. rather custom. Not a HORRIBLE undertaking, but still, it'd take some time.

Another approach would be to hang the peice on the wall and let gravity take care of the x axis.. Though I don't think you'd get the kind of speeds you want out of it though.

Mach3 software
Gecko drives
Hybrid steppers
belt drive system
build the moving mechanism so it can be suspended above the table, assuming you can focus the light from a distance of more than a few inches.

Mach offers the means to drive the machine as well as import files for creating the code using LazyCam.

Mike

I have the table built, what you see here is about 8x8 feet square. Because skateboards use pretty good wheels and bearings, this rolls very well, but I've not put something like a fishing scale on it to see how much mass there is to move. It's a little hard to make out, but a center cord going up and over the top will be the drive mechanism for the entire assembly.

I'm thinking it will require some decent sized motors and drivers. At this point I'm completely open to suggestions. I haven't even thought about what the "end product" will be. It could draw shapes, or simply move to a locations and blast some light through a mask.

http://www.leeshiney.com/artmachine.jpg

THAT, is cool. :)
doesn't look like you have much mass there at all, mabye on the order of 50 pounds or less!

Just found your forum, kinda late now that I see you have a table built. ....I was going to suggest to build a big box with a painted plexiglass sheet on top. Install mechanics and light from underneath, no moving parts outside the box, probably cleaner looking and I think everyone would be amazed .....how does it work! It may also be brighter too since the light source could be very near glass.

That's a really good idea. Although everything out in the open works for this particular project, your idea for backlit plexiglas would work great for a wall-mount idea that I have been rolling around.

I have the table built,
It could draw shapes, or simply move to a locations and blast some light through a mask.

You seem to have captured the beginning of your "kid-friendly" installation quite well!
The brilliant design is sure to inspire creative artistic work as well as mechanical design ideas for those so inclined.:banana:

That is taking ACME 7 bearings to another level! :D
Nice work, keep the pictures coming.

--S

It might take a bit of specialized software to generate the code, but I would think it would be much easier and cheaper and maybe cooler if the light source were mounted on a turret with 2 rotational axes. Much like an army tank. The distance and angle would define the length of the lines. This design would have low low inertia and would be very clean as the light source could be mounted on the ceiling and pointed at a phosphorescent wall.

Matt

Hi,
It would be easier to use a 2-galvo style device, like those used on laser projection. You just need a couple of mirrors and steppers, or servos, and a bit of math. And it would be faster


http://www.zamisel.com/galvanometer.jpg

http://www.laserod.com/images/galvo.jpg

yes, that is what I was describing.

http://www.planetdj.com/images/DMX600.jpg

Matt

I recently saw a "Sand Plotter" at a local college Engineering building. It ran on stepper motors and one of the professors there said she had it moved away from the wall adjacent to her office due to the noise it generated. The accel / decel and step impulses would sing through the board and cover panels.
Here is a link to a video of a similar machine:
http://193.135.56.86/snowflake/technorama/sisyphus3eck_technorama.wmv
And another link: http://taomc.com/

Artzofacto, if you are not set on using the skateboards, I have a suggestion. Suppose the flashlight were suspended above the table with small steel cable in loops similar to the maner some copy machines move the mirror/lamp asembly. With proper isolation of the cable pullys you could run power to the led through the cable. You wouldn't even need to put the flashlight itself out there, just the LED. If you are interested in the cable idea let me know and I'll try a drawing of the idea. I'm in Tulsa and would like to see the result one day.

I recall many years ago having some phospoescent paint that was a light cream color under normal light and glowed light blue in the dark. A friend used some to put stars on his bedroom ceiling. Was really amazing when you turned off the room light at night.

Till next time
Steve

Couple of obsevations;

First, it would seem that with collimated LED's you would have roughly the same sized spot and watt-density delivered to the table whether it was a few inches or a few feet from the surface. There just isn't that much dispersion over that distance. Why not mount the light up off the table several feet?

Secondly, to get the speeds you are talking about, it would seem easier to use steppers to effect angular motion to the flashlight instead of moving it the whole length of the table were you to keep the flashlight perpendicular. If you were to simply tilt the flashlight, it would allow for VERY small motors to move the light as quickly as you wanted as an inch of angular movement of the end of the flashlight would effect a movement of the projected spot of light by several feet across the table surface. The incident angle of the light would not be relevant to what you are trying to do, correct?
Tilting isn't as trick or as fast as two mirrors of course, but way cheaper, simpler, and less technical with the math. Tilting probably is still easily capable of faster speeds than are needed here anyway.

Thirdly, with this approach you have no moving parts near childrens hands, and possibly no moving parts visible at all. Shapes just appear on the table - Magic!

Fourth, Size is no longer a limitation of physical axis length. Size is effectively unlimited. Also, as Keebler pointed out, with this setup a wall would be just as valid of a canvas as a table, which would just be cool.

I think you should give a quick search on google for both the "blu-ray phaser" as well as "blu-ray laser glow in the dark" the phaser has instructions on instructables and can be built for around $100 i think the effect in your machine would be amazing... just throwing it out there

actually.. if it was the route you were to look into... after reading a bit more about it i'd have to recomend going to laserpointerforums and checking it out... they have all the real info on how to hook up the lasers properly.... either way... blu-ray laser drawing in glow-in-the-dark paint on a skateboard cnc... that would be incredible

I only skimmed this thread, but the first thing I thought of when I read the description was this :

http://muonics.net/blog/index.php?postid=15

If nothing else it's a damned interesting read.

Enjoy.