Light power #
Prototype 0 proves that LEDs are providing more than enough power to expose multigrade photographic paper.
4x1W green LEDs at 50% duty cycle during 30 seconds was enough to give some acceptable exposure with the head at the top of the column (resulting image was above 50x60cm).
Intensity evenness #
This one might be a bit tricky but as the LEDs provide a lot of light, I think some "brute force" (i.e. a lot of diffusion layers, baffles, ...) will tackle any issue on that side.
Wavelength evenness #
I still have to verify. It is possible that the (cheap, not binned) LEDs I bought have slightly different wavelength. The result would be slightly different grades on the print. Again, the brute force diffusion may even out the spread of each LED.
Other possibility is to use only one LED (5W) per channel. This moves the issues of having different wavelength to the previous point, i.e. intensity evenness across the negative surface.
Intensity stability #
Output of a LED will change with temperature, which might be a problem. Two solutions:
- temperature stabilization
- make the exposure independent of temperature
Temperature stabilization would involve keeping the LEDs at a constant operating temperature (possily heating and / or cooling). Also all LEDs don't have the same response to temperature variation.
The other solution seems more interesting to me. Photographic paper is a light integrator, i.e. the final result depends on the total amount of light it receives. By placing a sensor that receives a portion of the light going through the negative and the lens (i.e. on the base board of the enlarger and close to the paper), it is possible to measure the quantity of light received and stop the exposure once the set value is reached. This solution also point in direction of a single LED per channel because there is no temperature stabilization involved. The sensor used to measure the contrast range of the negative can also be used to measure the amount of light during the exposure.
Possible implementation: A small hole through the negative holder will led some light pass. Place the sensor in the light projected through that hole on the base board before starting the exposure.
Wavelength stability #
Seems to be a non-issue (very small variation for blue and green relative to the temperature).
Length of pulses of PWM - Reciprocity failure #
If the pulses of light are to short and the gap between two pulses is too long, we might hit the reciprocity failure. I don't have figures about that but that's something to keep in mind.