Homebrew Waterfall Display HOWTO @ phoenixgarage.org
All - after thinking about it a bit, I thought I would repost my article here as well; for posterity, or as a backup; whatever. It will also allow people to more easily discuss it with it being on the forum rather than another website. If you have any questions, or want/need clarification, I will try to do what I can to clear up any confusion (and if drawings would help, I can try to supply those - but I'm no artist!):
**broken link removed**
After reading this thread today...and the responses, and watching the videos; it got me to thinking how it could be implemented in as cheap a manner as possible. After thinking about it for a while (a few hours), I think I have found an idea on how to implement the valves, very cheaply (but there's going to be a bit of fabrication involved, TANSTAAFL); if you implement this, please at least throw a bit of credit my way, OK? Thanks. I'd build this myself, but I don't have the time or inclination. I honestly don't know if this system would work, but it seems to me that the idea is what is most important; I am sure someone out there could get this to work with some experimentation.
This device (well, one valve) could be easily built in an afternoon to test out a single instance of it; for a decent home-brew waterfall printer system, you'd probably want 100 of these spaced about 1/2 inch or so apart.
First, buy a whole mess of cheapo BIC "round stic" pens; a 10-pack is a $1.98 USD on Amazon right now, I'm sure you can find more for cheaper. Take the cone-shaped tip off each pen; discard the rest of the pen.
Then, buy an equal number of 1/8 inch chrome-plated steel ball bearings from smallparts.com (or another vendor).
Then get a piece of PVC or ABS pipe (length will depend on how many valves you want and their spacing - I think 100 is easily doable, if time consuming to fabricate). It should be about 1.5-2.0 inches in diameter; why will be explained later. This will form an "upper trough" for water; you'll need to fabricate a stand or something for it to hold it level. You could easily do this with more pieces of PVC. Just remember to leave things in such a state as to allow the disassembly of the display for cleaning and/or maintenance.
Along the length of the pipe, using a drill press, drill a line of "spotter" holes, perhaps 1/8 inch in diameter, spaced equally (approximately 1/2-3/4 inches apart).
Then, pick one side or the other's line of holes, and drill these holes the diameter of the inset section of the pen tips (there's an inset that fits into the tube body of the pen, about a 1/2 inch long; when you take one of the pens apart, you'll see what I mean); you want these holes a tad smaller than the pen tip (maybe 5mm - yeah, I mix it up - sue me) for a tight friction fit. You'll fit the pen tips into these holes later. For the opposite set of holes, drill them out slightly larger (you'll need to play with the diameter - you'll see why in a bit).
Notice that one open end of the pen tip is wider (~1/4 inch) than the other end (< 1/8 inch), while our steel ball bearings are 1/8 inch in diameter. This is going to form the valve.
Take a ordinary steel nail, one which has a diameter of the larger "top" hole(s) on the pipe trough that was drilled. This needs to be measured, so that from the head to the end is the length of approximately 1 inch, plus enough to span the pipe and fit down inside the BIC pen tip, minus the diameter of the ball, plus about 1/8 of an inch. You will likely need to cut the nail so that it is this length; after cutting it, chuck it in a drill and wind a coil using magnet wire from the head down an inch, and form a coil about 1/2 inch in diameter (you will need to add fiber or rubber washers at each end to form the coil properly). Count the number of turns of wire needed (or build a winder of some sort); you want every valve to be the same. Once you have the coil wound, check its resistance - verify that there's continuity, and that its resistance seems reasonable for the gauge of wire being used. Tape the wire down so it doesn't come unwound, then remove the nail from the drill, and use polyurethane, epoxy, or something to coat the coil, and let it dry. Once dry, file the cut end of the nail it flat. Then give the entire nail and coil another couple of coats of polyurethane, etc (this is ultimately to prevent corrosion from happening). Each nail you fabricate should be as identical in fabrication as possible. This is why the upper trough can't be too large in diameter, because you are constructing an electromagnet attractor, and the coil is at one end - if the diameter of the pipe were larger, then these homemade coils would have to be made beefier to work, if they could be made to work at all.
Once the nail is dry, it is stuck through the upper hole (and this is where you may need to play with the diameter, because of the polyurethane coating, mainly) and down into the BIC pen tip, where the end is just slightly above the steel ball bearing. Now - let us suppose we have things set up so that below the valve there is a catch basin for water, and a pump which has enough power to raise the water to the height of the trough. If things were set up right, with water added to the system, the pump would pump the water up and "flood" the upper trough pipe with water, above the level of the ends of the BIC pen tips, so that water flowed into the tips, but met the blockage of the 1/8 inch steel bearings. Now, hopefully, the balls will seal the tips well enough so that water didn't leak, but I doubt that it will be perfect, since this is homebrew and such. Now, all that would need to be done to activate the valve would be to apply voltage/current to its respective electromagnet; this would cause the ball to rise and stick to the end of the electromagnet, which is positioned a very short distance away, but enough to allow water to flow around the ball and out the tip, hopefully in a droplet form. When the electromagnet is shut off, the gravity and the flow of the water should move the ball back down into the tip, plugging the end again. Due to the small distances and sizes involved, it should be possible to achieve fairly high open/close oscillation speeds to create small droplets.
Implementing the electronic hardware to actuate the electromagnets is fairly straightforward; a BJT or FET drive to handle the current, perhaps driven via a bank of shift registers or such, with the data clocked in via a microcontroller.