Enlarger conversion to LED (WS2812 & Fujimoto G70)

[radiant]

I have now test printed with Fujimoto G70 enlarger converted to WS2812 led strips.

The conversion was successful; I could increase the light output by 3-4 stops compared to halogen bulb with even illumination and the conversion was pretty easy to do.

I purchased (gladly) two 1 meter 144 led / m strips. I was about to use only one meter because the eBay sellers promised 42 watt power from single meter. I think the sellers have messed up with readings from 5 meter strip. So I soldered two meters of strip and glued the strip on the walls of the enlargers light "chamber", above condenser lens.

The power consumption was really a surprise: single color channel consumes about 2.2 amps on 5V so the power is about 11 watts. That is enough to make 5x7" print at 2.2 stop exposure while lens is stopped down 4 stops! My halogen setup required 3-4 stops exposure at lens stopped down 2 stops. That is 3-4 stops difference, only at 11 watts of light. I have always thought the halogen light path is pretty stupid

The illumination seems to be really even, I made a "grey" print without any film inside and it looks quite even. I need to dry the print and scan to make final conclusion. So far I would say no problems at printing.

Also the exposure time seems to be pretty even with green and blue light so adjusting contrast should be really easy.

Attached is LED printed papers in fixer. Upper is printed with green only and below blue only.

 

[ic-racer]

Nice, it works!

I know how to program Arduino, but it might be easy, both physically and conceptually, to just hook each LED driver to a separate timer. Use 'time' to dial in the shadows and highlights. The timers could be started together or sequentially for selective burning.

 

[radiant]

Nice, it works!

I know how to program Arduino, but it might be easy, both physically and conceptually, to just hook each LED driver to a separate timer. Use 'time' to dial in the shadows and highlights. The timers could be started together or sequentially for selective burning.

Yes, this opens a lot of possibilities. For example pre-exposure red light to get dode/burn tools in correct positions etc.

Looked at the dried prints this morning and those look great. The gray print looks really even to my eyes, cannot find any hotspots. There is no return to halogen light enlargers after this

 

[radiant]

.. the discovery in this is that it is possible to replace enlarger light with WS2812 led strips and using those is really easy. No need for PCBs or PWM based led drivers, really!

 

[UnderwoodPhoto]

I love my G70 enlargers. One has the analyzer built in thr baseboard the other without. I found the JCR 75 watt lamp is 2 stops brighter, less heat than the 100 watt lamp so I stocked up on a dozen. Most likely due to less voltage drop and loading on the power supply. Most printing is on LOW mode now.

 

[radiant]

I love my G70 enlargers. One has the analyzer built in thr baseboard the other without. I found the JCR 75 watt lamp is 2 stops brighter, less heat than the 100 watt lamp so I stocked up on a dozen. Most likely due to less voltage drop and loading on the power supply. Most printing is on LOW mode now.

Probably my enlarger has just non-efficient lamp then! However I didn't have any issues with exposure times etc. so I was pretty happy with my enlarger in it's halogen setup. I just wanted the nice features from computer controlled colors

I have already replaced Fujimoto's own control panel with my own timer system so after this I'm basically using only the mechanics of Fujimoto

 

[radiant]

I've now done few prints with LED conversion and I've found few things:

- Overall sharpness (microcontrast?) seems to be better with LEDs. This might be placebo - but I really think the prints have a bit extra "punch" or clarity with LEDs.
- LEDs don't have linear response, what a surprise If I get nice print at 50% "contrast" (both colors at 50% of full time) then the single color prints are a bit darker. I mean if I print only with blue color for 100% time, the highlights start to get too much exposure / overall print is a bit dark. I fixed this using curve matching with bezier curve (because I'm looser at matching logarithmic curve).
- Printing with this system is stupidly easy. No need for swap contrast filters, no need to do two exposures for different filters. My stopclock exposes both grades at the same time.

 

[radiant]

Here is the curve plotted what I was previously writing about. The target exposure is here set to 16 seconds. As you can see at 50% the combined time is 16 seconds, but either color itself produces only 14.4 seconds.