h a l f b a k e r y
See website for details.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
Water graphics is the art form where drops falling in adjacent rows form a pattern. Currently, this is achieved by valves sitting side by side, triggered electronically. The resolution is limited due to the cost of that many valves, the problems of triggering a sufficient number of valves, and spatial
limitations due to the size of the valves. The drops being gravity-accelerated on the way down means a natural boundary to the resolution in height.
My idea is to supplant the valves by small nozzles, that will hold the water if no pressure is applied. The nozzles sit on the underside of a tube holding the water, being resupplied. At the ends of the tube sit underwater loudspeakers with a broad frequency range. For every column that is to be 'printed', the loudspeakers give a noise designed to lead to defined pressures at the nozzles at a defined time. At this time, water is protruding from the nozzles according to the local pressure, and a long bar swings in from the side, swishing the drops from the nozzles simultaneously. (Alternately, the nozzles might be given a shake by piezo elements).
As the graphics are known in advance, the necessary wave computations (Fourier transformation of the desired waveform) can be done in advance too.
As an upgrade to this it might be possible have water streaming from the nozzles at the end-velocity of water (~30m/s), and having a pin-board installed horizontally along the length of this curtain. The pins being movable and ending in a tube of hydralic oil, that is subjected to the loudspeakers. In this case the pins will punch holes in the curtain, and the resolution in heighth is not limited by the drops accelerating.
'smearing' due to acceleration, low resolution in width (96pix/m at 125rows/sec) [loonquawl, Jun 04 2009]
||Is this part of a wider project to display graphics using the
most inappropriate medium? Has anyone considered a low-
definition cheese-ripening based system?
||[bigsleep]: the Y-junction approach would still require every nozzle to have their own adress - my idea strove to cut down on the number of individually adressed devices (two loudspeakers, one global shaker).
||[MaxwellBuchanan]: Watergraphics usually aim to deliver stunning effects for a water curtain, they are not meant to be the most efficient way to project an image, but merely an upgrade to a normal water curtain; This is comparable to cereals in the form of, well, anything. The main reason there is a cereal with a face in your bowl is not that you wanted a face in your bowl, but a cereal; yet what distinguished this cereal from others is that is has a face. - I changed the category from display to art.