LED Split Printing Enlarger Lamphouse

[mshchem]

This is great. I could never figure it out without some help. I can wire household stuff, but same with woodworking, I tell people I can get by with a bit of carpentry, but I'm no cabinet maker.

I use Beseler Universal RGB 3 lamp heads for color and VC black and white. These are temperamental, Beseler had a lot of issues, so far I've been able to keep them going.

Heiland units look amazing, maybe some time I will get the cash together to do more than look.

 

[Mal Paso]

Happy with the light distribution I did a final build of the LED array. As it turns out 20 of the Cree LEDs will fit in the 6.5 inch circle so I have 5 each, 18 degrees apart, of blue, green, red and white for focusing. I cut a new 1/4 inch thick aluminum disk, drilled and tapped it then attached the heatsink before mounting the LEDs. I'm waiting on one more Buckblock Driver so I can only light 3 of the strings at once. When it gets here red will be on the footswitch for positioning dodge/burn tools. Focus light is warm white, green and blue. The Cree XPE2 LEDs and Buckblock drivers were purchased from LED Supply.

 

[M Carter]

Very cool project. Keep in mind that the Beseler condenser head has a center hot spot that's well known, depending on height and f-stop. Wide open for a 16x20 print, mine is 3/4 stop off in the corners; someone used to manufacture a milled plastic disc that was thicker in the middle, but it ate about 2 stops. This is probably why you had to mess with LED placement and diffusers!

 

[Lachlan Young]

Wide open for a 16x20 print, mine is 3/4 stop off in the corners

Are you quite sure that isn't just the natural fall-off of a 150mm wide open?

 

[john_s]

Very cool project. Keep in mind that the Beseler condenser head has a center hot spot that's well known, depending on height and f-stop. Wide open for a 16x20 print, mine is 3/4 stop off in the corners; someone used to manufacture a milled plastic disc that was thicker in the middle, but it ate about 2 stops. This is probably why you had to mess with LED placement and diffusers!

The beauty of this design is that a hot spot can be avoided by LED placement, and even a "coldspot" can be arranged to account for the inevitable hotspot caused by the simple unavoidable light falloff of the enlarger lens. In other words, a "coldspot" could be superior to totally even illumination.

 

[M Carter]

Are you quite sure that isn't just the natural fall-off of a 150mm wide open?

Nope, I've fought with it for some time - this week, 135mm, 27" from the paper, F16, still some noticeable falloff. And I see it with 80 and 110mm lenses with MF. I do lots of lith printing, so wide open or close to it is nice, but tons of burning the corners in. It's still there when stopped down but not as bad. And the light actually isn't hottest in the center, but a tad hotter to one side of center. I've played with bulb placement, painting the inside of the lamphouse white (some people say that helps with the 23C).

I can't find the link right now, some guy that sells enlarger parts and manufactures a custom-milled diffuser for Beseler condenser heads, it's very thin at the edges and thicker in the middle. It's a lot of light loss though, been thinking of seeing if I can make something with litho film, like a soft mask that could go in the chain somewhere. But very hard to get gentle falloff with that stuff. I have a few sheets of 11x14 to play with it.

 

[Mal Paso]

Fascinating, I believe I found the diffuser fellow: http://www.glennview.com/index.htm $100 for the Beseler 45 isn't too bad but 2 f stops is a lot. I am at the other end of usable power now with 64.5 watts total. I've been printing 11x14s at 12-15 seconds f16 Ilford Classic at 20% on the power dial. And no infrared to heat negatives.

Light output is stabilized twice with a 24 volt switching power supply and individual drivers for each color. I can run the heater and print.

I remember Ansel showed a print and you could see the coldlight tube in the print. After reading your comments, I probably should have done the tests at f16 rather than f45 but at least the lens was part of the test.

Thank You.

 

[mrosenlof]

Very cool project. I've been trying to envision an LED array for my LPL4x5 enlarger.

A schematic, if only for one LED channel would be great to see, if you're willing to share. I'm an embedded software programmer, but not an electrical engineer. Even though, maybe I can imagine what it looks like. Are the LEDs (for one channel) in parallel or series?

 

[mrosenlof]

I'm assuming your circuit with the driver is something like xx mA across each LED. Are you dimming with resistance? You haven't mentioned some kind of microcontroller or something for Pulse Width Modulation?

thanks for any info!!

 

[Mal Paso]

I think a micro controller would be a good thing. The switching is done with the Buckblock driver. It can be standard 0-10 volt or a 20K pot as I have. Open circuit across the dimming leads runs the LEDs at full power, 1,000 MW in this case. Closed circuit turns the LEDs off. The Buckblocks have very fast switching and are left on.

 

[mrosenlof]

cool, thanks! I was just reading again, and guessing that if you were shorting the center LEDs they must be in series. But this looks well within even my circuit building capability. I have just found the datasheet for the driver, so I may have to give this a try...

 

[Lachlan Young]

Nope, I've fought with it for some time - this week, 135mm, 27" from the paper, F16, still some noticeable falloff. And I see it with 80 and 110mm lenses with MF. I do lots of lith printing, so wide open or close to it is nice, but tons of burning the corners in. It's still there when stopped down but not as bad. And the light actually isn't hottest in the center, but a tad hotter to one side of center. I've played with bulb placement, painting the inside of the lamphouse white (some people say that helps with the 23C).

I can't find the link right now, some guy that sells enlarger parts and manufactures a custom-milled diffuser for Beseler condenser heads, it's very thin at the edges and thicker in the middle. It's a lot of light loss though, been thinking of seeing if I can make something with litho film, like a soft mask that could go in the chain somewhere. But very hard to get gentle falloff with that stuff. I have a few sheets of 11x14 to play with it.

Glenn Evans of Glennview was making a diffusion disk - but to be used in place of the condensers I recall. I've been very slowly setting up a CB7 Beseler (too many other enlargers here!) - will have a look & see what the evenness of the head is like - the inside of my CB7 head is very definitely white. Bulb size and shape can be an issue too - I recall that a bulb with too long a neck can be a real problem on Focomats.

 

[john_s]

Are you quite sure that isn't just the natural fall-off of a 150mm wide open?

There's always some fall off. Much worse at full aperture, sure, but it's always there. Look at spec graphs.

 

[M Carter]

Glenn Evans of Glennview was making a diffusion disk - but to be used in place of the condensers I recall. I've been very slowly setting up a CB7 Beseler (too many other enlargers here!) - will have a look & see what the evenness of the head is like - the inside of my CB7 head is very definitely white. Bulb size and shape can be an issue too - I recall that a bulb with too long a neck can be a real problem on Focomats.

That's it, Glennview. I emailed him and it doesn't seem to replace the diffusers IIRC; you can put it above or below them.

Way back in the day, I shot a bunch of really pushed E6 (320T, my favorite color film ever) and asked the lab to dupe on Velvia. They said "that's impossible", so I got to work, taped a cheap flash unit to my crappy Beseler printmaker (for daylight exposure), grabbed a polaroid back and ended up making amazing dupes. I'm starting on a mural-sized project so I'm going to play with making a head with a powerful COB LED and cooling setup (for focus) and a strobe for exposure. Just for B&W though, should be interesting!

 

[Mal Paso]

The controller changed again. Blue and Green circuits got 10 turn potentiometers for repeatable settings. The Red pot (far left) made it onto the new panel. White is the regular pot to the right. Focus is now Blue Green and Warm White at full power. Really nice for focusing. The other toggle is the blue or green only or both at the same time as I only have 1 timer channel.

 

[albada]

I'd like to build one of these. But Digikey shows 239 models of XPE2 LEDs made by Cree. Could you post the manufacturer's part numbers of the R-G-B LEDs that you used?
Also, did you use the 1000 MA version of the BuckBlock drivers? (they come in 3 different current-ratings).
Thanks!
Mark Overton

 

[Mal Paso]

CREEXPE2-830-1 Warm White
CREEXTE-ROY-1 Royal Blue (1,500 MA) But CREEXPE2-ROY-1 is 1,000 MA which is a better power match
CREEXPE2-GRN-1 Green
CREEXPE2-RED-1 Red
Yes the BuckBlocks were 1,000 MA
I had good luck with LED Supply. Under $4 ea for LEDs and fast free shipping but I've used Digikey too, just not for this.

 

[albada]

Thanks for the LED part-numbers.
Another question: Are you using the timer to turn on/off the 24V supply? If so, when the LEDs are dim, have you noticed whether the LEDs remain on for a little time after the enlarger turns off? The power supply has some large capacitors in it that might contain enough stored energy to keep the LEDs on for a second or longer when they're dim. I've seen power supplies maintain output-voltage for several seconds under low load, so I'm wondering whether this could be a problem.

Thanks,
Mark Overton

 

[koraks]

That's why you should turn off the LEDs by simply setting a 0% PWM duty cycle.

 

[albada]

That's why you should turn off the LEDs by simply setting a 0% PWM duty cycle.

That's what I'm thinking, but if I understand Mal's circuit correctly, there is no PWM. Instead, I think it uses analog dimming via potentiometers connected to the Buckblocks. An easy way to get instant on/off would be to have a relay wired so that the DIM wires on the buckblocks are normally shorted (LEDs off), and when the enlarger timer is running, it activates the relay which opens the contacts, causing the LEDs to come on. I'm designing a circuit that does this with transistors controlled by an NE555 timer chip to generate the PWM. But I'm wondering if Mal has a simpler solution...
Mark Overton

 

[koraks]

Sorry, you're right. A simple solution is to short DIM to GND on the buck block using a relay or even cheaper a little FET or NPN.

 

[radiant]

Sorry, you're right. A simple solution is to short DIM to GND on the buck block using a relay or even cheaper a little FET or NPN.

+1

Also I would suggest using Arduino instead of 555 timer. It opens up lot of possibilities in addition to PWM generation such as exposure times and making few things automatic. I know that it sounds difficult to switch to but it is quite easy to write PWM control actually, you don't need to be "real" programmer.

The heiland head feature where it switches on red lights for a moment is brilliant. Easy to adjust burn/dodge tool before actual exposure!

 

[Mal Paso]

The LED on off switching is done with the BuckBlocks. The power supply stays on. A 3 pole relay runs off the timer's power, shorts Blue and Green and turns off the safelight. Print to print the exposure is dead on. Much more stable than a tungsten bulb.

Red is on a footswitch

I avoided PWM switching partly for aesthetic reasons partly because PWM will need compensation for actual system reaction times. Don't know how well it would work over the 4 stop range of the head.

 

[koraks]

If you use eg a mosfet for switching the led array, pwm is pretty much instantaneous, so no delays or warm up/cool down. In fact, the buck blocks likely use pwm internally for dimming. I used a pwm motor controlling driving mosfets for switching the leds at something like 1.5kHz frequency and LM350T's for current limiting. In a future version I'd probably use a mosfet/opamp current limiter but that's another story for another day.

 

[Mal Paso]

https://www.luxeonstar.com/assets/downloads/buckblock.pdf

No PWM inside the Buckblocks but it looks like they are fully capable of pulse or strobe.

I wanted a high power, low heat, even, stable, diffusion, split multigrade head and I have that now. Please continue if you have a better idea for control and I may copy you.