Homebrew PCBs

You might want to find a way to lose the glass layer.
It may be sucking UV energy or converting it.
How about vacuum suction or really thin glass ( 1mm ) ?

Glass blocks UV
http://www.newton.dep.anl.gov/askasci/chem00/chem00539.htm
Plastic versus UV
http://www.newton.dep.anl.gov/askasci/eng99/eng99272.htm   see below**
UV transparent materials
http://www.coatings.de/radcure/reading/skinner.htm
http://oh-dog.com/UV

** But I could list specific plastics.
Pure Plexiglass ("PMMA", poly-methyl-meth-acrylate) transmits most of  
the UV that will give you a suntan.
Clear poly-styrene plastic is chemically simple, just C's and H's  
(Carbon and Hydrogen), and no big electron clouds. So it transmits UV  
better.
Poly-ethylene is even simpler, and will transmit even farther into  
the UV.  But it always has scattering, always looks cloudy or milky  
(translucent).
DuPont's Teflon (TM) has only Carbon and Fluorine atoms, and  
transmits so far into the UV that scientists have difficulty getting  
UV lasers to cut it.  But you have seen that it is really white  
("PTFE" type), not just cloudy like poly-ethylene, and this kind of  
DuPont's Teflon (TM)
has very strong scattering.  If it is too white to see through, you  
will get a low percentage transmission in the UV too.  Then there are  
clearer types ("PFA" type) which are only a little cloudy.  They  
transmit well too. Good thing is, almost never does
someone bother to put UV-blocking dye in Teflons.


Gus


>>
>> On Feb 27, 2008, at 8:16 PM, Adam Seychell wrote:
>>
>>> Russell Shaw wrote:
>>> >
>>>
>>> > Have you tried using the same higher current and just  
>>> increasing the
>>> > distance? Maybe the wavelength is increased at lower current,  
>>> which
>>> > would reduce the photon energy.
>>> >
>>>
>>> Yes, I tried 20mA at 120mm distance and still no exposure. I also  
>>> tested
>>> 15mA at just 20mm distance and got complete polymerisation in  
>>> under 2
>>> minutes. So, I don't think its anything to do with wavelength  
>>> change,
>>> even if an LED's wavelength could change significantly with current.
>>>
>>> LED arrays solve the problem because intensity is increased  
>>> greatly, and
>>> is also less dependent on spacing between the exposed surface and
>>> LEDs. Increasing the spacing also improves uniformity because the
>>> increased overlap of adjacent LEDs.
>>>
>>> For anyone interested, here's some more results.
>>>
>>> Here are approximate times for complete polymerisation of  
>>> negative dry
>>> film photoresist using a 3x3 LED UV array of 15mm pitch. The  
>>> light path
>>> was travelling through 3mm window glass plus one sheet of inkjet
>>> transparency film to represent actual PCB exposure conditions.
>>>
>>> LED	Current	Spacing	Increment	Polymerisation Time
>>> Type (mA)	(mm)	(seconds)	(seconds)
>>> -----------------------------------------------------
>>> A	10	150	10	 40
>>> A	20	150	5	 20
>>> B	20	150	10	 60
>>> B	20	60	10	 50
>>>
>>> LED types are:
>>> A "UV 390-395nm 5mm NR F/R" ebay: besthongkong, price: AUD$0.21/ea
>>> B "UV 3000mcd 5mm" ebay: winsome_house, price: AUD$0.11/ea
>>>
>>> Its clear LED "A" is a winner here. But then again it might be  
>>> better to
>>> use the cheap "B" LED and then double the density in the array so
>>> uniformity is improved. Both LED "A" and "B" when assembled in a  
>>> 15mm
>>> pitch square array at 150mm distance seems to give pretty uniform
>>> looking light when shined on some white paper.
>>>
>>>
> 


[Non-text portions of this message have been removed]