Homebrew PCBs

javaguy11111 wrote:
> Definitely very peculiar. I would expect the exposure time to go up
> with the square of the distance. So a four times increase in distance
> would result in a 16 times exposure time. In your case it should have
> taken 8 minutes. 
>   

I've since been browsing the web and it appears manufactures of 
photoresist go to great efforts to try make
their exposure a non-linear function of intensity. The purpose is to 
improve the "contrast" of photoresists.
If the resist photochemical reaction occurs abruptly above a particular 
intensity then it will improve contrast
of the image. If the photomask lets a little light through then it won't 
cause any problem because the intensity
is below the threshold. Also any scattering or light detraction will 
less likely cause the the resist to be exposed.

So its expected that low light levels won't cause photoresist to expose. 
At low light intensity, the exposure time is not
proportional to total light energy as I originally expected.

> Do you have any kind of material that fluouresces when you shine the
> LEDs to see if there is uniform emission. I wonder if the UV could be
> getting emitted in a ring pattern. If that were the case, when you get
> too far away, the UV would not be hitting the resist.
>   

I shined the LED on some gloss inkjet photo paper and the light looked 
relatively uniform across a 1cm x 1cm exposure test
area.
>
> I bought the Best HongKong LEDs as well. I received a notice from the
> post office that I have to pick them up. When I get them I will try
> the same kind of experiment to see what happens

As noted in my other post the BestHongKong 390-395nm LEDs were about 3 
times more efficient at exposure than the cheap LEDs.