An imagined LED enlarger light source for 4x5 -- NEW INFO FOR TECHNICIANS

[Ulophot]

Hi, folks. I just spoke with someone at SuperBright LEDs, who responded to my long email, including watching the video at aspenimages on building that fancy, allegedly DIY model. In case you missed my earlier post:

Our cold light sources are not going to last forever. My hypothesis is that a plain vanilla, white LED (i.e., non-VC) enlarger source should be inexpensive to make, and that it could either fit the round Beseler or Omega D-series collars or, perhaps better, be made as a rectangle. If made in two parts in the latter configuration, the bottom could be the light source, the top could be one of several modules made to connect the bottom to various 4x5 enlarger models.

Here's where the technical part comes in, regarding feasibility, at least as far as my poor understanding allows: SB's 3000K strip lights come spaced about 2/3" apart, in 3-LED segments, 18 LEDs per foot, sold in 5-meter lengths, if I read the specs chart correctly. Each foot produces 115 or 220 lumens -- I can't understand how they are figuring this between 18 or 36 "density." The strip is less than 1/3" wide, so even at two strips per inch, just 9 strips, i.e., 4.5 feet, would fit a 4.5 x 6" enclosure, so three heads could be made from either $30 or $50 worth of strip lights, plus one of their $20 transformers (12V to 120AC) for each. Would this lumen output be any where near comparable to the output of a cold light head? That's really the potential deal-breaker as I see it, unless for some reason they would not go on and off pretty much instantaneously.

How the enclosure might be constructed, I must also leave to those who do such things. It just doesn't seem as if it would be a terrible challenge, since heat is not a major factor.

Okay, full disclaimer: I work for a huge IT company now, because I couldn't run my own. I'm the world's worst businessman, full stop.

That said, my guess is that, provided the lights are sufficient, someone with know-how and a shop could, after planning, prototype, and set-up, make 20 of these in at least one configuration, maybe three hours a piece, for a cost of maybe $90 (~$40 for the light and transformer, $30? for plywood or 3-D printing material, $20 for white plexi, additional hardware, paint, and other materials) and sell them for $250 a piece for a gross profit of $3400, about $50 an hour. Heck, a good one would be a good long-term investment at $300. The fellow at Modern Enlarger was selling his for $400, but he stopped long ago.

Can someone qualified chime in on the light output or other considerations? Do we have any retired builders with experience in this sort of thing?

[Peter De Smidt]

Personally, having a bit of experience, I think you'd be better off with one of the 50-100w white COB LED video lights for this purpose. Or, if you must absolutely keep costs down, a COB LED worklight.

[jose angel]

Philip, your idea is good, but in my experience, the main problem you will find is lack of brightness.
With larger formats, a lot of light is needed, so strip LEDs use to be a bit poor on the lumen account.
But they work. Getting on and off is not an issue, they are fine.

[koh303]

Why not just buy a Heiland LED light source, readily available, made to fit, with a multitude of controls?

[sperdynamite]

Why not just buy a Heiland LED light source, readily available, made to fit, with a multitude of controls?

The Heiland LEDs are quite expensive, however I love mine! I'm guessing it will last as long as I will and I'm only 36. No more bulbs, no more weird power supplies. No aging filters.

[Eric Woodbury]

I know somebody printing with the white LEDs on tape. He used his old coldlight housing less the burned out tube and added as many strips as he could cram in there. Then he uses filters for contrast control; as before. It works fine, but as mentioned, it is a stop or two dimmer than the coldlight was. Not a problem for some printing combos, but he is printing 8x10 negs, so it takes a little longer or smaller f/#. Good news is that the LED brightness doesn't drift much.

[Eric Woodbury]

Here's the LEDs I'm using. These are the 1 amp GREEN super brights pre-mounted on star heatsinks. I've attached the wires before gluing to the heatsink. Once mounted, I'll connect the standoff wires in series to build groups of 9 LEDs. Two groups of green and two groups of blues. Blues and greens are controlled separately.

(Stars break apart before gluing to a 10" x 12" x 3/16" aluminum sheet.)

[Peter De Smidt]

LEDs in practice do not necessarily have long lifespans, especially if heat is not well-controlled. For example, I've had to replace LED light bulbs just as often as tungsten. I expect that this is especially true of cheap ribbon lights.

[Drew Wiley]

Labor-wise, all kinds of things can be done affordably if you do them yourself. Trying to contract it out is another story. No competent shop technician with decent equipment would even bother talking to you for that kind of unrealistic pittance. Some kind of task-sharing arrangement might work. I personally view a lot of these LED experiments analogously to the Wright Brothers and others climbing up beach dunes and gliding off, and often crashing. A necessary step; but this whole kind of technology still seems rather adolescent compared to tried and true previous options. Making a VC printing unit is one thing, serious color enlargement still off in the distance. And ANYTHING involving wiring and lights, no matter how seemingly cool, deserves a nonflammable housing! So forget the plywood. Don't kid yourself. What is a house worth? - just another reason why nobody in their right mind is going to produce something like this for resale without doing some serious homework first and factoring liability. As far as Heiland goes, without even factoring R&D, manufacture, and profit, there would have been considerable expenses and hassles just getting the thing legalized in the EU and then through UL here.

[Eric Woodbury]

True enough with LED light bulbs, but it is not the LED dying, it's the power supply inside the bulb either dying or killing the LED. A good article about this in EDN.

The other reason to keep them cool and uniform temperature is the emissive output is a function of temperature. Not a great dependence, but great enough to notice in your highlights. If the junction temperatures are held under 80C, they will last a very long time. Longer than I will.