Using LED bulbs in an enlarger.... GREAT!

[trondareo]

Thanks for the link!
I see that I should be looking for a 110mm diameter bulb and 180mm high. In Norway where I am, it seems the standard diameters are 95mm or 135mm. I´'ll look into fitting one of those, and look forward to hearing the results of any LED tests you do.
I have come across an option here: https://lampa.lv/led-bulb-sky-light...-g125-2720f.html?___store=en&___from_store=en
A 135mm might be a tight fit though.

 

[Ron789]

Trondareo, I bought one one these (in your link) a couple of weeks ago to fit in my Durst Laborator 138s. It works fine for contrast and for even light coverage across the entire image. The weird thing is that it gives less light, thus longer exposure times, compared to a smaller 7w or 10w LED bulb, contrary to the official Durst recommendations. I suggest you go for a smaller, more commonly available bulb and find out how that works for you. My experience is that this huge (125mm diameter) bulb does not perform any better than a standard size LED bulb in the 138s.

 

[trondareo]

Ron, I have a traditional size Paterson opal bulb in there now, and can't achieve even coverage with any combination of bulb centering, condensers and lenses. What kind of exposure times do you typically end up with, using the 135mm bulb?

 

[Ron789]

Trondareo, my setup is:
Durst Laborator 138s
125mm diameter 20W 6400K LED bulb
Latico 240 and Latico 200 condensors (the proper combination for 4x5")
Rodagon 150mm F/5.6 lens
I recently printed a 20x24" on Ilford MG FB WT from a 4x5" negative. At F/11 with a Grade 2 filter the exposure time was 43 sec.
With a light meter I measure 15-30% light fall-off from the center to the outer corners. When printing that is insignificant, especially since the negative will have some vignetting as well which is then nicely compensated by the vignetting of the enlarger light.
Using a smaller LED bulb the light fall-off is 20-35%; more vignetting but even that I have never seen causing problems in real printing.

 

[Danlo]

Very interesting! I have tried with LEDs but 2700K ones, and that was not good Going to try much stronger ones, 806 lumen and 1000 lumen at 4000K now. Thanks!

 

[mnemosyne]

Very interesting! I have tried with LEDs but 2700K ones, and that was not good Going to try much stronger ones, 806 lumen and 1000 lumen at 4000K now. Thanks!

the standard size opal bulb is good for negatives up to 6x9cm, this is according to the L138 manual and my own tests confirmed this. Also a standard size E27 LED bulb showed too much vignetting with 4x5" for my taste. In practice you might get away with it depending on the angle of illumination of the lamp. For the larger formats I would consider a 120mm diameter lamp (G120) which should be good for all formats up to 5x7", especially when mounted horizontally (depends on the type of lamp mounting bracket in your lamp house). You might consider looking for a stronger bulb like something around 20 watts / 2000 lumen, which should be available in the G120 form factor. In my smaller enlarger (up to 6x9cm) I use a 1500 lumen standard LED bulb. Of course in practice the effectiveness of the bulb will depend on other factors as well (the spectral distribution of the light in relation to the spectral sensitivity of paper or the angle of illumination of the bulb).

 

[ic-racer]

Those of you experimenting with different lamps in condenser enlargers, consider the Philips 150. That enlarger uses the lamp to project its light on a uniform disc. The disk is focused by the condensers, rather than the lamp. (mirror not shown, condensers are square).

 

[naaldvoerder]

Do LED bulbs exist that can replace 12v/100w halogeen lightbulbs, with a similar or higher light output?

 

[locutus]

Yes those exist, but to know more specifically if theres one that fits your purpose you need to know exactly what mount, GU10 MR16 etc

 

[Craig]

I have a DeVere 10x10 and it takes 8 bulbs with the MR16 base. I could not find an LED replacement for them, so I am staying with traditional bulbs. I also plan on doing colour printing, so I wanted to stay with the light spectrum the dichroic filters were designed for.

 

[spijker]

Do LED bulbs exist that can replace 12v/100w halogeen lightbulbs, with a similar or higher light output?

No, I don't think these exit yet. A halogen bulb is about 20 lm/W so a 100W bulb is about 2000 lm. The brightest MR16 bulb I've seen sofar in stores is a Philips MR16 bulb at 620 lm, 7.5 W, 3000 K. I bought one to try it out. The normal bulb in my LPL670/7700 diffuser enlarger is the exact one shown in the picture. When measured with a incident light meter on the base board, the 100 W halogen bulb measures 1.5 stop brighter than the 7.5 W / 620 lm LED bulb. That puts the 100W halogen bulb at 620 * 2^1.5 = 1754 lm. If you don't print large and/or use a fast paper, the 620 lm LED could work fine. I haven't printed with it yet.

 

[phil61628]

I just switched from a GE incandescent 1500w 3000k to a generic Ecosmart bright white 150w 3000k 2200 lumens bulb from Home Depot.

Using a Beseler 45m, the LED stopped my negatives from popping. Virtually no heat, much sharper and longer workflow. My ilford multigrade filters work well, but some contrast tuning is needed.

Much brighter, could probably do with a 100w, but I like the extra light. Easier to subtract than add.

 

[koraks]

Brilliant! Thanks for sharing your experience @phil61628 !
So it basically was a snap-in replacement?

 

[titrisol]

I just switched from a GE incandescent 1500w 3000k to a generic Ecosmart bright white 150w 3000k 2200 lumens bulb from Home Depot.

Using a Beseler 45m, the LED stopped my negatives from popping. Virtually no heat, much sharper and longer workflow. My ilford multigrade filters work well, but some contrast tuning is needed.

Much brighter, could probably do with a 100w, but I like the extra light. Easier to subtract than add.

I had issues with my timer (Gralab) making the LED flicker when they were supposed to be off, I replaced all the timers capacitors and didnt help
Then I went to dimmable ones (I bought 100W GE), and since then its been smooth sailing

 

[phil61628]

Brilliant! Thanks for sharing your experience @phil61628 !
So it basically was a snap-in replacement?

Yep! I did a quick test with grades 00, 2, and 5. Got expected results.

LEDs are weird tho, they work off oscillation to give the illusion of constant light, rather than the incandescent filament. Dimmable LEDs might be the way to go. They oscillate at a variable rate, so they can handle a wider range of voltages without freaking out, but it might freak the enlarger out. Hard to know. If there’s flickering issues, I’d try both a dimmable and non-dimmable bulb

 

[koraks]

LEDs are weird tho, they work off oscillation to give the illusion of constant light

That's how most of the interior lighting 'bulb' style LEDs do it, yes. They generally simply flicker at 50/60Hz grid frequency and rely (if you're lucky) on the brief afterglow of the phosphors that turn the native blue light into wide-spectrum white for reduction of the flicker effect. B&W paper won't mind the flicker, btw, neither will the enlarger. The enlarger is just a chassis; it basically doesn't care what's going on electrically.

Dimmable LEDs might be the way to go. They oscillate at a variable rate

The rate (frequency) is generally the same, but the pulse width varies. However, I assume most dimmable LEDs work similarly to LEDs that are run off of a DC supply; i.e. they dim through PWM (pulse width modulation) but this PWM driver is fed itself by DC (rectified from the AC line supply). This PWM frequency is generally much higher than grid line frequency (e.g. upwards of 1kHz instead of 50/60Hz), which will make any flicker much less noticeable - unless you're recording video (esp. on high frame rates), which may result in some oscillating light intensities.

For the paper and the enlarger, all this doesn't really matter. Paper doesn't care whether it's exposed with a 50Hz modulated light source, one modulated at 10kHz (much higher may result in odd non-linearities) or one that isn't modulated at all.

 

[albada]

The rate (frequency) is generally the same, but the pulse width varies. However, I assume most dimmable LEDs work similarly to LEDs that are run off of a DC supply; i.e. they dim through PWM (pulse width modulation) but this PWM driver is fed itself by DC (rectified from the AC line supply). This PWM frequency is generally much higher than grid line frequency (e.g. upwards of 1kHz instead of 50/60Hz), which will make any flicker much less noticeable - unless you're recording video (esp. on high frame rates), which may result in some oscillating light intensities.

If LEDs contain full-wave rectifiers, then frequency would double, making it 100/120Hz. But is that high enough to avoid visible lines when dodging/burning for a short time and moving the tool quickly? As @koraks mentioned, hopefully the phosphors' afterglow lasts long enough to smooth out that frequency, yielding nearly-constant light.

Mark

 

[phil61628]

If LEDs contain full-wave rectifiers, then frequency would double, making it 100/120Hz. But is that high enough to avoid visible lines when dodging/burning for a short time and moving the tool quickly? As @koraks mentioned, hopefully the phosphors' afterglow lasts long enough to smooth out that frequency, yielding nearly-constant light.

Mark

30fps is awfully fast for the human hand.
Results check out fine for me.

 

[koraks]

But is that high enough to avoid visible lines when dodging/burning for a short time and moving the tool quickly?

Nope, I don't think so.
LED armatures differ quite wildly but most low-end bulb make do with the bare minimum of components - so half wave rectifiers instead of full wave.

 

[Donald Qualls]

Quick reality check. All of us are old enough to remember CRT television; most of us have probably seen the effect of rapidly moving a hand in front of a CRT screen (which refreshes at 29.97 Hz -- how that figure arose is a very long story and I don't know all of it). Short version: you need to move your had very rapidly to generated zones from 30 Hz flicker, never mind 4x that (as you'd get from a cold cathode fluorescent light source), and the exposure difference between the strips so generated would be more convenient to measure in milliseconds -- which isn't enough to see in a typical enlarging exposure of ten to thirty seconds.

 

[albada]

Nope, I don't think so.
LED armatures differ quite wildly but most low-end bulb make do with the bare minimum of components - so half wave rectifiers instead of full wave.

To me, an armature is a coil on the rotating part of a motor or generator.
What is a LED armature?

Anyway, half wave would mean the downstream components would need to be rated at twice the current, boosting their costs. Maybe I should disassemble a few LED bulbs to see what's in them...

 

[koraks]

What is a LED armature?

Read it as "fixture", or "entire bulb assembly".

half wave would mean the downstream components would need to be rated at twice the current

Nope.