LED enlarger head for B&W -- question in passing

[Ulophot]

The LED head that a friend is constructing in a custom 3D-printed housing for my Omega D2 enlarger to replace my clearly aging Zone VI fluorescent, finally reached first trial stage yesterday. Yes!

The light source is an RGB LED panel of 16x16 LEDs. I will use it only for B&W (all I do) and with Ilford contrast filters. So, my first test was to arrive at a preliminary Grade 2 color balance so that the filters would work normally. Once set, it will stay there, unless I find special cases in which altering it will be advantageous.

Realizing that Red would have little or no effect on constrast, I simply dialed it down to zero and adjusted the Blue and Green to render approximately the same contrast with or without the Gr. 2 filter in the light path. (I will borrow a Stouffer step wedge from a friend to hone the settings later this week.)

Here’s my question, just to make sure I’m not missing something. The light coming from the head is intensely blue, though it gives me Gr. 2 at about 2550 Blue, 140 Green. I infer that this is so, simply because I have removed the red component, which, in a tungsten bulb such as 212 would be making the light much more yellow. Theoretically, I could add back red with no effect on the paper other than to change the light color. The actual spectrum of the red LEDs could affect the contrast.

Have I got this right?

[Michael R]

The extent to which the red LED emitted spectrum extends into the sensitivity range of the emulsion (or vice versa, same thing) determines the extent to which exposure from the red LEDs will affect both exposure times and contrast.

A common issue with most consumer-grade red LEDs is the existence of "anomalous" emission spikes in the blue/green spectrum, which is why people sometimes get paper fogging with red LED safelights. You can test for that fairly easily and then determine whether or not using the red LEDs in your head will affect anything.

[Duolab123]

Beyond my knowledge.

[neil poulsen]

I use Ilford Warmtone VC paper, which I really like, with a Beseler 45s color head. My grade "2" contrast is no filtration. To increase contrast, I add magenta filtration and leave both Cyan and Yellow filtration dialed to zero. Correspondingly, to decrease contrast from my no filtration grade, I increase the yellow contrast to filter blue with both magenta and cyan contrast set to zero.

Any paper testing in which I engage is without filtration.

Who cares about a "grade 2" contrast? For me, it's irrelevant. I begin with one contrast (no filtration), and if I need greater contrast for that particular image, I add magenta. If less, I add yellow. If neither, then I begin my dodging and burning.

[koraks]

The light coming from the head is intensely blue, though it gives me Gr. 2 at about 2550 Blue, 140 Green. I infer that this is so, simply because I have removed the red component, which, in a tungsten bulb such as 212 would be making the light much more yellow. Theoretically, I could add back red with no effect on the paper other than to change the light color. The actual spectrum of the red LEDs could affect the contrast.

Have I got this right?

1: You don't need the red for B&W printing. It's plain and simple like that, so don't worry about it.

2: There's something very strange with your setup if you need more than an order of magnitude more blue light than green to hit grade 2. In my experience with LED sources, it's the other way around usually. This has to do with many factors, including spectral sensitivity of the paper (it's inherently more sensitive to blue than to green, because that's how VC works in the first place) and the so-called 'green gap'; the low efficiency of green LEDs compared to blue ones. So while this may work, I would be somewhat concerned if this were my project if the drive electronics are working as they should. Assuming a led panel with roughly the same amount of blue and green LEDs, in terms of nominal power, I would expect grade to to be at something like 10%-25% blue and the rest green (again speaking in terms of power output).

What kind of driver are you using and how is the power level modulated for each channel?

PS: also what neil poulson says above - who cares in the end if grade 2 is grade 2, or 2.2, or 2.5 or maybe 1.8. What matter is that you can somehow use the full contrast range that you need from the paper and that it has to offer. Don't lose too much sleep over the question if your grade 2 is the 'correct' ISO-R.

[Ulophot]

When will I learn to scrupulously proofread each post/email before sending? koraks, that 2250 should have been 250. This demonstrates that a mere 2 in the wrong place at the wrong time between friends can be more than it seems. Like all RGB systems with which I'm familiar, the scale for each hue component in my LED panel ranges from 0-255.

I can't answer your question about driver and modulation; that's why someone else is building this for me.

As for caring about Gr. 2, my approach is simple. The Ilford filters, as you may know, are based on tungsten illumination and designed so that exposure time remains constant for whatever tone they selected (I don't recall, but it's some mid-scale value) while changing between filters from 0-3 1/2. The 00 and 4-5 filters require double the exposure. This change -- made sometime in the mid-'80s, if memory serves -- represented a significant improvement over the original filters, whose factors were all over the place.

I have been using a coldlight with a Gr. 2 filter as my base for the past several years, which is more like a 2 1/2. It used to be standard practice (mine as well) to put a CC40Y in the path first to mimic the tungsten balance. I got lazy and have made do, but now that I have the opportunity to simplify my life and have the filters getting the light for which they were designed, I'm happy to align my set up with this.

Fine printing is, of course, an art, not an application of fixed "recipes." To each his or her own when it comes to what fits a chosen approach. For me it's lining up with Ilford's design. I am hoping that this new source will also solve my previously posted problem with respect to the lacking high-contrast end with Foma Classic; I haven; gotten around to testing that yet. Ilford WTF is my standard paper, and initial tests show it behaving very nicely with my new head.

[Michael R]

Philip, as you note, the grade spacing and speed matching of the filters are based on a specific incandescent light source. The issue you will run into in trying to replicate all of this with your LED head is that as opposed to an incandescent source, your LED head is not putting out a continuous spectrum, but rather some relatively narrow bands/peaks of wavelengths which you see as blue and green. This issue applies to some extent with any light source that differs from the light source for which the filters have been "tuned".

So, even if you get everything dialed in on the LED head to produce the equivalent of whatever you had been accustomed to getting with an Ilford grade 2 filter, all bets are off (to some degree) once you change filters. You might be close, but you might also end up with some different grade spacing and/or paper speeds at different grades.

As others have said what's most important at the outset is to make sure you are able to get the full contrast range you anticipate needing. Then you can "calibrate" (if required) as you go.

[Doremus Scudder]

Philip, as you note, the grade spacing and speed matching of the filters are based on a specific incandescent light source. The issue you will run into in trying to replicate all of this with your LED head is that as opposed to an incandescent source, your LED head is not putting out a continuous spectrum, but rather some relatively narrow bands/peaks of wavelengths which you see as blue and green. This issue applies to some extent with any light source that differs from the light source for which the filters have been "tuned".

So, even if you get everything dialed in on the LED head to produce the equivalent of whatever you had been accustomed to getting with an Ilford grade 2 filter, all bets are off (to some degree) once you change filters. You might be close, but you might also end up with some different grade spacing and/or paper speeds at different grades.

As others have said what's most important at the outset is to make sure you are able to get the full contrast range you anticipate needing. Then you can "calibrate" (if required) as you go.

Philip,

A couple of things:

First, Michael brings up a good point; all bets on grade spacing are likely off when you switch to LEDs. Still, I think it's a good idea to find a "middle value" that allows you the most flexibility in either direction, which is what I believe you are trying to do. So, use your arrived-at grade 2 setting, get that step wedge, and then make quick comparisons with different filters in place. Probably you don't need to use all of them, just, say, #2, #3.5 and #5 on the contrasty side and then #1.5 and #00 on the soft side. The object being to see if the contrast grades space fairly evenly and if you can get maximum contrast and maximum "softness" from your LEDs (it would be helpful to have a comparison for the extremes, so try an exposure with Wratten #47 and #58 filters respectively if you have them to establish those extremes).

Second, while your red LEDs might not be exposing the paper, they may come in handy for focusing and for viewing the image when dodging and burning. If you just use the blue and green components of your head, the image may be rather too dark to work with when you need a lot of contrast (think #5 filter). I'd dial the red up to max and do my tests with them always on.

If you want to see if you have any blue or green spikes in the output of your red LEDs, you can use a CD as a makeshift diffraction grating. Turn on just the red LEDs, hold the CD at an oblique angle to the light and observe the reflection, which should be separated into rainbow colors like a prism does to white light. If you see only red, then you're good to go. If there are some spikes of green or blue, they could, potentially, have an effect on the contrast when exposing. However, unless they are fairly intense/substantial, the effect is likely to be negligible. The test would be to make the same test at extreme settings with and without the red LEDs on and see if there is a difference in contrast (i.e., more or fewer stripes visible on the step wedge).

Keep us posted on your progress.

Best,

Doremus

[Pieter]

Just curious if a color temperature meter would be of any use gauging spikes or dips in LED light output?

[Duolab123]

Philip,

A couple of things:

First, Michael brings up a good point; all bets on grade spacing are likely off when you switch to LEDs. Still, I think it's a good idea to find a "middle value" that allows you the most flexibility in either direction, which is what I believe you are trying to do. So, use your arrived-at grade 2 setting, get that step wedge, and then make quick comparisons with different filters in place. Probably you don't need to use all of them, just, say, #2, #3.5 and #5 on the contrasty side and then #1.5 and #00 on the soft side. The object being to see if the contrast grades space fairly evenly and if you can get maximum contrast and maximum "softness" from your LEDs (it would be helpful to have a comparison for the extremes, so try an exposure with Wratten #47 and #58 filters respectively if you have them to establish those extremes).

Second, while your red LEDs might not be exposing the paper, they may come in handy for focusing and for viewing the image when dodging and burning. If you just use the blue and green components of your head, the image may be rather too dark to work with when you need a lot of contrast (think #5 filter). I'd dial the red up to max and do my tests with them always on.

If you want to see if you have any blue or green spikes in the output of your red LEDs, you can use a CD as a makeshift diffraction grating. Turn on just the red LEDs, hold the CD at an oblique angle to the light and observe the reflection, which should be separated into rainbow colors like a prism does to white light. If you see only red, then you're good to go. If there are some spikes of green or blue, they could, potentially, have an effect on the contrast when exposing. However, unless they are fairly intense/substantial, the effect is likely to be negligible. The test would be to make the same test at extreme settings with and without the red LEDs on and see if there is a difference in contrast (i.e., more or fewer stripes visible on the step wedge).

Keep us posted on your progress.

Best,

Doremus

Compact disc diffraction grating. Brilliant idea.

Wouldn't using a combination of green and/or blue LEDs be easier.???

Best Mike

[Duolab123]

For 4x5 and smaller films I use a Beseler 45 Universal head. Closed loop, 3 halogen lamps on the VC version, green and blue dichro filters, and a clear filter. Blue and green are for VC paper, clear for graded papers. The computer controller figures out "grades".

For larger negatives I was lucky to have found a 5x7 and 8x10 Zone VI enlarger. Both have the last version of the green tube/blue tube cold light head. These are closed loop through a Metrolux II compensating timer.

The Zone VI heads were somewhat intimidating at first. The box that controls the ratio of blue and green, has 2 dimmers, one for each tube, instead of contrast grades it has letters, ABC etc. I of course, found it necessary to mark the combination of settings that produced "grades".
I usually don't do much fancy, but I have done split printing using separate blue and green exposures.

[koraks]

@Ulophot, I understand about the typo and the fact someone else is building this for you, ok! The blue/green ratio still seems way off to me for a mid-range (ca 2.) contrast. You might want to look into this and have a chat with your designer to see what's up. However, it's not an issue as long as you can hit grade 0 and grade 5 with the head.

As to calibration: you can actually create a calibration curve for the light source that approximates equal exposure times across most grade settings. In my experience not all grades because you run out of green light at some point, but it's usually possible to keep exposure constant across grades 1 through 4 or so. As far as I recall from working with Ilford filters that's similar to those.

Calibrating your led grades to ilford grades is also possible. It's a little additional work and math, but certainly doable. I didn't find it important enough to go there, but I did do the equal exposure calibration thing on both led heads I built. It sure is convenient.

Don't fuss about what people are saying about continuous spectrum etc. It's inconsequential. Calibration is needed anyway, and it still would be if the led source would have been continuous. Anyway, it's something you work around "automatically" if you calibrate your leds to match ilford grades.

Regarding the use of red led for focusing etc: yes, that can be done (in fact, I do it as well), but be careful using only red as a focusing tool used when the paper is already on the easel. It WILL fog. The question is by how much and if it'll turn up as fog or as a subtle contrast reduction. You don't want either. If you want to use pure red for focusing on the paper itself, be sure to add a good red filter.

[Ulophot]

Thanks to all.

Duolab, what I am attempting to create is something about which I had started a thread or two some time ago (2020?), which I have since “shopped” around to various photo gadget makers to see if they had any interest: a plain vanilla, budget-model, white light diffusion LED head to replace our aging coldlight heads. There are DIY models for those who can build, and then there’s the “Rolls”, the Heiland, for those who can afford it. Nothing in-between, to my knowledge.
The one my friend is building is already more complex than I had suggested. I had envisioned white-only LEDs; he opted for the RBG panel. I had offered a blue &green LED alternative, but he may well have gulped hard when I sent him this link:

http://www.jbhphoto.com/wp-content/uploads/2015/01/LEDsVCPrinting03.pdf

I am paying for materials, he is doing the work, based loosely on Marco’s (elgatosuizo’s) design, with a 3D-printed a head enclosure of his own invention that takes my suggestions into account. As indicated, we’re still ironing out issues. The photo realm is new to him, and it’s taking much longer than he expected, due to various problems encountered along the way. There may eventually be wider benefits for others.

The Gr. 2 calibration, again, is just to provide a useful standard. If the other contrast grades don’t space evenly, due to the differences between incandescent and LED with its spikey nature, testing will at least show that. If needed, I will at some point create a new exposure chart, based on some tone -- possibly a Zone VI, maybe VIII -- for maintaining that value when switching between filters. My preliminary tests suggest that I won’t need one. I’m not a great printer, but I am a good one and experienced.

Doremus, I’ll keep in mind adding the red back in and see if it makes a difference. At this point, I am not having difficulty with the projected image, but my testing has just gotten started with a couple of negs.

[Duolab123]

If I outlive my dichro/cold light heads I will take up golf :o

Anything else that requires computer etc. I'm running away. Always can make contact prints

I hope this works out as the beautiful Heiland made heads look confusing.

[Ulophot]

I'm happy to report that testing with a borrowed Stouffer 21-step wedge put my Gr. 2 just about on the mark at ISO(R) 112 with Ilford WTF (my standard paper), and that both #1 and #5 filters produced their proper contrast levels as well.

My old contrast issue with Foma Classic remains; the #5, even after eliminating Green entirely, gives only a Gr. 3. I have tested for safelight fogging exhaustively and eliminated that as a factor. Whether or not I just got a box that sat in a hot delivery truck or something, or the paper just doesn't produce that contrast (as at least one person here reported), or the blue of my LED is just not blue enough (added spikes or whatever that the paper is reacting to) will have to wait till I run out. Meanwhile, 3 is not so bad most of the time, and it still gives me an alternative of color and surface for the great majority of negatives.

[interneg]

Blue filtration tends to max out around G4.25 - at least that's Ilford's verdict both on their MG500 head & the Heiland LED. The magenta G5 seems to pass deeper into the blue than the narrow bandpass LED.

[Michael R]

Blue filtration tends to max out around G4.25 - at least that's Ilford's verdict both on their MG500 head & the Heiland LED. The magenta G5 seems to pass deeper into the blue than the narrow bandpass LED.

This makes sense. Despite the assertion by some, getting max contrast likely requires some blue and/or violet beyond what is normally sent to the paper through certain blue filters or by blue light sources such as commonly used LEDs.

[interneg]

This makes sense. Despite the assertion by some, getting max contrast likely requires some blue and/or violet beyond what is normally sent to the paper through certain blue filters or by blue light sources such as commonly used LEDs.

I think that comparing the bandpasses of #47/#47B/#98 (a fair assumption would probably be that both the dichroic & the LED run about the same or narrower than the #47) vis-a-vis the various low #30's magenta Wrattens (#32?) & the CC Magenta bandpasses - and even the old polycontrast filters (I found some data for the early 70s series - where the max filter value was 'PC4') can be quite enlightening regarding the shortcomings of additive blue filtration for getting max contrast out of papers. The problem may be that a 'bluer' LED might run risks of making the Heiland head less good for colour (I suspect it's been aimed to deliver somewhere in the region of a #25/#29, #47/98, #61/#99 effective range).

On the other hand, I think that with something like the Omega dichroic, a #47 likely would deliver a higher contrast because the max magenta value seems not to be as good as De Vere (for example) - or white light plus multigrade filters.

[koraks]

In the first version of my led enlarger I tested a deeper color of blue. That is, for the final version I went with the universally recommended royal blue, which is around 450nm. Most/all led strips will have blue leds of a significantly longer wavelength. However, my testing with 430nm leds didn't yield much difference as I recall vis a vis royal blue for black & white. It DID produce massive problems with color printing though, so I very quickly moved away to royal blue which more closely matches Kodak additive color printing filters.
Can't really comment on max contrast differences between a color dichroic head and led as I never systematically compared them. I did not notice any problems with lack of "oomph" when migrating to LED, but admittedly I did all/most of my b&w printing with an Ilford 500 head before, so I was already a lost cause apparently.