This is one of those cheap "laser kaleidoscope" heads that uses two
layers of dual-axis holographic diffraction material to project a big
spray of dots that move around.
Back in the day our local laser-show
crew used to build similar setups and our term for it was "D-squared device",
as it was two diffraction axes times two layers and the rotation of
one relative to the other made all the magic movement. We even
took it to the level of "D-cubed" with three layers, and all with
independent rotation either way -- *that* was lots of dots, and
especially pretty with a multiline argon source firing through it.
Now the concept is a common commercial product for less than a hundred
bucks from China, snapped up by countless DJs and small clubs.
But their construction is flat-out garbage. This one still had a working laser, but the rotation motor wouldn't spin anymore.
The innards. The 532 nm green DPSS modules are a little less efficient
than the red diodes and throw a little more heat, so green ones get more
of a heatsink and a tiny fan added. The little gear-motor turns a ring
that sits around the front of the brass laser module and holds the second
diffraction piece, and that's all there is to it. A 3V regulator
feeds the laser diode driver circuit itself, which is floating in a wrap
of heatshrink instead of attached to the back of the diode.
The wiring for the various parts is just sloppily tack-soldered to the control board; there aren't even holes to accomodate the connection points. Some brands of these things do use small connectors to make disassembly easier, but this manufacturer chose to cut corners wherever possible. "Brands" is relative because these things probably all come out of the same factory town with the design changing every month anyways.
Wiggling things around made the gear-motor temporarily try to run,
so it was clearly a loose connection of some sort. I finally tracked
it to the commutator brushes themselves, which involved un-swaging where the
back plate of the motor was attached to its can and carefully popping it
out. It's obvious how flimsy the brush springs are, and they don't
bear very firmly on the commutator surface. And no carbon pieces
are involved -- these *are* the brushes, in direct metal-to-metal
contact. A little bit of crap-up from running in dusty environments
later, and they weren't contacting anything at all anymore.
What are the humans who design stuff like this actually thinking?? Where do they derive *any* pride in workmanship, or are they in the business of simply not giving a soggy rats ass about durability?
|The temporary fix was to give it all a little shot of DeOxit, and !carefully! wipe the relevant surfaces. Reassembly was done slowly while working the brushes out around the commutator flange by poking a pin through the slots in the back plate. What a pain the tuchus. I probably mangled them slightly in the process -- the factory must have a jig that holds them out in a loading position for this, but I was damned if I was going to construct one just for this. Eventually it was together, and a little more prodding with power applied got the motor going and it seemed to run fairly reliably thereafter.|
A similar unit in red was also on the bench with the same nonrotation
problem, and as it turned out from the same cause. In that case I was
able to shoot some DeOxit into the back of the motor and prod the brushes
and motor around just enough to get the motor to run a little, after which
it pretty much self-cleaned and continued running. With two units
at least temporarily working again, I aimed them both at the ceiling.
White light through the grating material makes little rainbow spreads, but because the laser output is a single wavelength, we get dots. The matrix comes from the fact that each first-order split gets split again in two axes and those get split again, and so on, with a certain intensity loss every time. Then *that* whole pattern hits the second holo grating. With the axes in relatively close alignment, the pattern sort of resembles square pillows, or chiclets, or something.
|The overall effect is rather amusingly complex. Clicking this image gets you a short video [8.5 Mb] of the red and green pair of "kaleidoscope" toys running together. For the moment the drive motors in both of these units are working, but with the flimsiness of the parts I don't expect that to last long. As a substitute for when they break again, you can always download this video and play it in a loop projected on a wall. Whoopie.|