I have been fooling around with these UV LED's for several months by
fits and starts. I bought a bunch from BestHongKong, and started out
to use those, but then I found some suspiciously similar devices
assembled into arrays at Sure Electronics, another vendor hailing from
China as well.
I was pretty happy with BestHongKong since they put up more technical
information than I have seen from anyone similar on eBay, and service
was as good as it can be from that distance. Sure Electronics has
proved to be an equally good source for different stuff - quite a
range of things. The information on LEDs is not quite so complete,
but the service has been very good.
The arrays I purchased are assembled on an aluminum strip with thermal
compound and include some sort of power conditioning that allows them
to work direct from up to 13.5 v DC or 9 v AC. They just seem to
work, and don't get particularly warm as far as I can tell. I later
purchased larger modules that have 24 UV LEDs per strip.
Sure claims (I have no means to verify) the strips produce a narrow
emission peak between 385 and 395nm. This seems to match the optimal
sensitivity of commercial (and home brew) organo-bichromate emulsions
pretty well. I thought that something closer to the actinic UV-B peak
at 254 nm would be required, but several sources indicate otherwise.
Polymeric activation of UV sensitive plastics may be another matter.
There is wide variation depending on chemistry. But these are the
shortest wavelength LEDs I have been able to find at these kind of
prices. If you can pay, Roithner-Lasertechnik and others have devices
almost to the soft x-ray region (at > $US1k-3K ∗per∗ LED).
Anyway, being somewhat conflicted about spending a lot of time on
this, and more than a little lazy, I opted to put together an array of
of the initial 8 LED arrays on a piece of heat sink. I put this inside
a box constructed from 0.5 inch foamcore that I had glued a thin mylar
'spaceblanket' material to as a reflector. With a generous
application of duct tape this made a pretty good base for my unit.
For the upper half I used the upper portion of an old HP Greyscale
scanner that had given its life for science sometime before (they sure
don't make them like they used to). For simplicity I power the whole
works with a couple of 12v Gel Cells that I trickle charge with an old
car charger.
I have not characterized this contrivance rigorously as you seem to be
trying to do, and even I see this as a largely inadequate design, but
I have worked with it enough to be surprised at the total output. I
keep overexposing things. I began with my own coating that I
concocted from purified fish glue (gelatin made from cold water
species of fish), ammonium bichromate, and a bit of pigment to make it
easier to see.
I began by trying to measure an H & D type exposure curve using a
stepped gradient exposure strip leftover from my father's darkroom
equipment. The biggest problem I had was getting into the correct
range. Particularly because, as I believe Adam observed, most of the
formulations used for this sort of thing are tuned for high contrast.
This is just as it has been in the graphic arts for many years where
materials like Kodalith Ortho #3 were used for producing halftone
images and line art for arcing printing plates.
What I did find was that uniformity of illumination was not ∗as∗
critical as I believe you are assuming it is (the high contrast
corrects for that in large part) and that these things put out a lot
more light than I would have guessed.
I hope to get back to these researches soon ("wife be willing and the
creek don't rise"). I have acquired a substantial stash of the dry
emulsion to play with and I 'm impressed with the potential for the
chemical machining of steel and other metals in addition to the
fabrication of PCBs.
Some links to (kind of BIG) pictures that may work:
http://206.152.116.84/Pictures/Projects/PCB/PCB_073.jpghttp://206.152.116.84/Pictures/Projects/PCB/PCB_074.jpghttp://206.152.116.84/Pictures/Projects/PCB/PCB_075.jpghttp://206.152.116.84/Pictures/Projects/PCB/PCB_076.jpghttp://206.152.116.84/Pictures/Projects/PCB/PCB_077.jpghttp://206.152.116.84/Pictures/Projects/PCB/PCB_078.jpg-- Dave W.