Tri-colour printing with B&W negs

Hi everyone,

I have recently been researching tri-colour additive printing and have discovered that it is possible to make colour prints with 3 x black and white negatives using this method. I’ve been looking into the appropriate red, green & blue filters and have found some tricolour filters by Lee filters that should work. Before I take the plunge and order them, I’m also wondering if there are some cheaper alternatives that I could use as an experiment first such as lighting gels or this rgb filter set:

https://www.thomann.de/ie/rosco_077...zdYkKLvOOqWDqNTOHubHebUyHcMkX4xoCwjQQAvD_BwE?

Any info on this would be great!

I should have mentioned my idea is to use B&W digital negatives. I want to create three RGB separations from a colour digital image using photoshop to then print onto transparency film, layer them and filter through the RGB filters in the darkroom and print on RA-4 paper. So I wouldn't need to make the separations in camera. I'm guessing I would need to make the appropriate tonal adjustments in Photoshop.

For readers who haven't seen these tri-color prints made from three-negative sets from the 19th century:

There are many past threads on tricolor printing. You might want to check the alternative photography section of the forum. There are also numerous websites dedicated to the many varieties of tricolor printing. Making in-camera separation negatives has a tough learning curve if you are expecting repeatable high-quality results. Not just any kind of film works well; and ideally, you need to work with sheet film so that each sheet can be developed to the proper matching contrast level. Tricolor filters affect film contrast differently in most instances, and that difference has to be made up for during differential development.

The red, blue, and green of that Lee filter set would probably work OK for beginning experimentation purposes. Yellow if used for "K" black in quad printing processes which need separate black to achieve deep shadows. For more serious applications you'd want glass filters in 29 red, 61 green, and 47B blue, or else 25 red, 58 green, and 47 blue.

Any fashion of actual tricolor printmaking based on black and white film exposures through colored filters is going take a lot of effort to learn. The easiest is gum printing; but it, like most, is a UV contact printing process.

There were also tricolor slide shows, using three different aligned carbon arc projectors (those must have been some hot rooms!). The previous link to early Russian tricolor photography refers to that. Making web images or possibly prints out of them in the present time involves scanning and digital conversion and reassembly. Doing such things darkroom style requires punch and pin registration equipment to keep the various printing steps aligned.

I used the second set of filters Drew recommended with success in taking the pictures. Except replaced the 25 Red with an IR720 filter and infrared sensitive film. However, did not do analog recombination. Alignment is bound to be a bit of a pain that way if you want perfect results.

Wratten 47B, 58 and 25 are commonly used. BGR.

Traditionally, the separate film exposures were taped together over a lightbox having a precise punch, then afterwards, sequentially registered on a matching set of pins either in a contact printing frame or enlarger carrier, depending on the specific process. Digital snap alignment is comparatively new, and requires in turn some kind of intermediate digital output, like for example, a pin registered film recorder which laser exposes the corrected profiles. But it can still be done by hand. The larger the size of film, the more accurate the registration is likely to be. All the original film and the intermediaries preferably need to be on dimensionally-stable PET film base like Kodak Estar, rather than unstable acetate base. These were once standard graphics processes described in all kinds of older manuals.

Or, in a studio with respect to still life subjects, a very rigid copy camera could be used with pin registration already in it.

There were also tricolor cameras by Devin and Curtis which split up the image to make the three separate exposures at the same time. Bulky Technicolor movie cameras did this with reel film sprocket registration each frame.

However, most tricolor printing involved starting with a color transparency instead, then making the tricolor or quad separations from that, which makes the whole process of registration quite simple in principle if one has the correct equipment. Not simple in terms of cleanliness and handling fuss, however. It can be quite a chore.

By far the best current film for color separations is TMax 100, unless you are going to use this directly in relation to UV contact printing. It has a UV blocker. So the next best would be either TMax 400, which does not inhibit UV after processing, or else Ilford FP4.

A black and white transmission densitometer greatly facilitates everything. You need to learn how to attain and plot precisely matching color separation curves. Of course, if one does not really expect accurate or realistic results, but just wants to have some fun gum printing or whatever, the rules can be a lot looser.

leahandhercamera said:

I have recently been researching tri-colour additive printing and have discovered that it is possible to make colour prints with 3 x black and white negatives using this method.

Click to expand...

Yes, that works, but what kind of printing process are you investigating? The filters you linked to, do you intend to use them for the recording stage (i.e. the color separation onto 3 sheets of film) or during printing?

koraks said:

Yes, that works, but what kind of printing process are you investigating? The filters you linked to, do you intend to use them for the recording stage (i.e. the color separation onto 3 sheets of film) or during printing?

Click to expand...

Thanks for the reply. I should have mentioned that I am hoping to use digital negatives. I want to create three RGB separations from a digital image using photoshop to then print onto transparency film, layer them and filter through the RGB filters in the darkroom.

DREW WILEY said:

There are many past threads on tricolor printing. You might want to check the alternative photography section of the forum. There are also numerous websites dedicated to the many varieties of tricolor printing. Making in-camera separation negatives has a tough learning curve if you are expecting repeatable high-quality results. Not just any kind of film works well; and ideally, you need to work with sheet film so that each sheet can be developed to the proper matching contrast level. Tricolor filters affect film contrast differently in most instances, and that difference has to be made up for during differential development.

The red, blue, and green of that Lee filter set would probably work OK for beginning experimentation purposes. Yellow if used for "K" black in quad printing processes which need separate black to achieve deep shadows. For more serious applications you'd want glass filters in 29 red, 61 green, and 47B blue, or else 25 red, 58 green, and 47 blue.

Any fashion of actual tricolor printmaking based on black and white film exposures through colored filters is going take a lot of effort to learn. The easiest is gum printing; but it, like most, is a UV contact printing process.

There were also tricolor slide shows, using three different aligned carbon arc projectors (those must have been some hot rooms!). The previous link to early Russian tricolor photography refers to that. Making web images or possibly prints out of them in the present time involves scanning and digital conversion and reassembly. Doing such things darkroom style requires punch and pin registration equipment to keep the various printing steps aligned.

Click to expand...

Hi Drew thanks for the reply and all the info! I should have mentioned my idea is to use B&W digital negatives. I want to create three RGB separations from a colour digital image using photoshop to then print onto transparency film, layer them and filter through the RGB filters in the darkroom and print on RA-4 paper. So I wouldn't need to make the separations in camera. I'm guessing I would need to make the appropriate tonal adjustments in Photoshop. Do you know if this method is likely to work?

leahandhercamera said:

Thanks for the reply. I should have mentioned that I am hoping to use digital negatives. I want to create three RGB separations from a digital image using photoshop to then print onto transparency film, layer them and filter through the RGB filters in the darkroom.

Click to expand...

OK, thanks for clarifying. I understand from #9 you intend to print onto RA4 paper. You could either use color filters such as the kind you linked to, or simply use a dichroic color head. You can use a dichroic head to get reasonably close to the primaries you need for consecutive exposures. I've done this with modern RA4 Fuji paper; it'll work. It'll be marginally better if you could use exactly the correct wavelength green (550nm LED works well), but what a dichroic filter head gets you will be close enough.

leahandhercamera said:

layer them

Click to expand...

You don't layer them. You make consecutive exposures onto the same sheet of paper. The challenge will be to retain alignment while doing so. This will be easier if you contact print and use some kind of pin registration. If you're going to work with inkjet printed negatives you're pretty much limited to contact printing anyway if you want decent resolution. Enlarging inkjet negatives generally results in rather disappointing prints.

leahandhercamera said:

I'm guessing I would need to make the appropriate tonal adjustments in Photoshop.

Click to expand...

Yes, you'd have to do the same kind of linearization that you'd typically do when printing e.g. cyanotypes, salted paper etc. from inkjet negatives. So you start with determining the ink load/maximum density in the negative you need to get a full tonal scale, then print a step tablet and measure the densities on C, M and Y, and use those measurements to create a linearization curve. That's it in a nutshell; sounds a little easier than it is in reality, but in principle it's a straightforward process.

A more high-tech approach would be to forget about the printed negatives and instead use an LCD such as the kind used for 3D resin printing and a color filtered light source. If you use a light source that can be controlled from a computer/microcontroller (e.g. RGB LED array), you can automate the process. There's a (precious) few people who work this way; @smng does this, but she's in the middle of moving shop. @avandesande is exploring a similar route and has posted extensively on the use of an LCD as the image source in an enlarger for B&W purposes. That's half the solution right there; the other half for your application would be the light source. On that part I've published lots and lots on my blog; see e.g. here.

Hi - L - Yes, it can be done that way, via initial digital camera capture, but will still have a relatively steep learning curve and some real expense. There are plenty of technical resources how to do this for tricolor sandwich printing processes like color carbon printing or dye transfer, but not for basic RA4. Ideally, you'd want the the three digital shots output onto real film using a laser film recorder, which is a rare service these days, and would need to be specially calibrated to your exact needs, which would take a lot of experimentation and densitometer plotting just to determine in the first place. And each different set of images would have to be tailored a little differently, and that itself would take a lot of experience to pin down.

Equipment-wise, it can be daunting. I should know. I have the right equipment to do that if I wished. You'd need a very stable commercial enlarger with a pin registered carrier for 4x5 or larger sheet film, along with precise registered vacuum easel which doesn't move in the least between sequential tricolor exposures. The print exposures would be made through deep yellow, magenta, and cyan filters respectively.

The last outfit to do this kind of RA4 printing commercially that I'm aware of was Evercolor. They took their scans and made big full-sized film exposures with an image-setter on a special Agfa film made for that purpose (no longer being made), then sequentially contact printed those onto an early version of Fuji Crystal Archive paper. I still have a sample of that process stored away. It was a very expensive method for getting from Point A to B, but certainly not as expensive as their prime option, which was their Evercolor pigment prints. In both cases, the registration work was done with multi-pin Stoesser-Olec graphics punches. I visited them and saw the whole process. That venture didn't last very long.

But frankly, in terms of image reproduction quality, I get conspicuously begtter RA4 prints just printing directly onto current RA4 Crystal Archive media using either Ektar color neg originals, or else precision Portra 160 internegatives made from chrome originals. It's not really quite that simple, however, since supplemental masking is often involved too, for sake of both contrast tailoring and a certain amount of hue correction. That itself requires some precise punch and register gear. My enlargers and colorheads are also on the deluxe side, including a couple of true RGB colorheads. But the characteristics of Fuji color papers are now so far advanced, along with significant improvements in Kodak's color neg films, that there is really little to gain going some complicated secondary RA4 route. Just have a decent conventional CMY halogen colorhead and learn how to optimize your workflow.

If you are interested in some kind of true tricolor or quad pigment printing (which inkjet isn't), that's a different story. Then you need actual color separations.

And ditto what Koraks posted above, if his pathway appeals to you. He has helpful information on his own link.

I use a simpler approach to create photographs form digital information. I use negative film to photograph my 27" computer monitor with a macro lens, laser aligned to the screen. The negative is then printed in the usual manner.

What "usual manner", ic-racer? And what does it have in common with actual tri-color capture or printing?

DREW WILEY said:

What "usual manner", ic-racer? And what does it have in common with actual tri-color capture or printing?

Click to expand...

Enlarger or contact print, just like any other negatve; RA4 is not a tri-color process. I guess I don't understand why three negartives are needed to print on ra4 paper; never heard of that before, but always willing to learn why.

RA4 can be a tricolor process. So can any other kind of color paper. Instead of taking one exposure on color film, three are taken on black and white film through R,G,B filters respectively. Then each of those films, after development, is printed sequentially in register through C, M, and Y filtration onto the same sheet of RA4 paper. Why do it this way? Out of curiosity, or to see if it gives better results, or just for the hell of it (and it can be hell).

In this case, however, the OP wants to begin with three digital exposures using a set of tricolor filters, then somehow transfer that onto three separate sheets of film.

I already mentioned a commercial venture which applied chrome film scans to large image setter film, then in register sequentially contact printed those onto conventional RA4 chromogenic film. It had its own look, with the chroma kinda blocked up and oversaturated much like many inkjet prints today, but at considerably higher cost. That venture was started by a partnership of a superb dye transfer printer with a pioneer in applying modern offset printing techniques to quad carbon printing. They were seeking a replacement for dye transfer. But neither their pigment print process nor their RA4 chromogenic dye print version was a real substitute. But at least in its time, nobody would confuse that with a conventional C print. The color was a lot cleaner, if garish.

I think the original poster is using the term "transparency film" to mean an inkjet printed negative on a transparent plastic sheet. If that's the case, then yes, you can create large separation negatives from a digital color image using Photoshop. You can use those negatives to print onto RA4 paper, but those negatives have to be registered, and printed separately under the respective R, G, B filters.

The basic principle is straight forward, but as Drew points out, there are many many details to be worked out, and there's no simple cookbook recipe. It's going to take a lot of time, effort, and money to get a workable system.

By asking this type of question, it sort of implies you don't have enough knowledge yet to be able to do this. It's already hard enough (and expensive) to get good RA4 prints from color negative film (that RA4 was designed for). When you go completely outside this box you're just asking for trouble if you don't know what you're doing. It might be better to try making a B&W inkjet negative from a B&W digital image, and print that on B&W photo paper. If you can do that well, then there's hope that you could go to the next step of making a color version.

Sharktooth said:

By asking this type of question, it sort of implies you don't have enough knowledge yet to be able to do this.

Click to expand...

Most likely; then again, it's a bit of a tautological remark. If they did know enough, the wouldn't be asking. And by asking, they demonstrate they don't know enough (hence the need to ask), but it's a perfectly fine way to learn. If they want to give it a go, they'd have my blessing, that's for sure. I do think the learning curve will be a bit steeper than they may have imagined.

Sharktooth said:

It might be better to try making a B&W inkjet negative from a B&W digital image, and print that on B&W photo paper. If you can do that well, then there's hope that you could go to the next step of making a color version.

Click to expand...

That's a very good suggestion!

Also, if an inkjet printer is going to be involved in making the transparencies, I'd consider making a single transparency that essentially mimics a color negative. This eliminates the problem of registration of the different negatives, because there will be only one. The question is whether you can hit the full gamut of the paper this way, given the blocking power of the inkjet's C, M and Y inks combined with the dot alignment of the printer (which will be rather poor/non-existent), but it's conceivable that the net result is acceptable. And it'll be an easier workflow.

DREW WILEY said:

RA4 can be a tricolor process. So can any other kind of color paper. Instead of taking one exposure on color film, three are taken on black and white film through R,G,B filters respectively. Then each of those films, after development, is printed sequentially in register through C, M, and Y filtration onto the same sheet of RA4 paper. Why do it this way? Out of curiosity, or to see if it gives better results, or just for the hell of it (and it can be hell).

In this case, however, the OP wants to begin with three digital exposures using a set of tricolor filters, then somehow transfer that onto three separate sheets of film.

I already mentioned a commercial venture which applied chrome film scans to large image setter film, then in register sequentially contact printed those onto conventional RA4 chromogenic film. It had its own look, with the chroma kinda blocked up and oversaturated much like many inkjet prints today, but at considerably higher cost. That venture was started by a partnership of a superb dye transfer printer with a pioneer in applying modern offset printing techniques to quad carbon printing. They were seeking a replacement for dye transfer. But neither their pigment print process nor their RA4 chromogenic dye print version was a real substitute. But at least in its time, nobody would confuse that with a conventional C print. The color was a lot cleaner, if garish.

Click to expand...

I was asked to try this back in the 80's by the owner of one of the dye transfer labs I worked at. Making three separation films and then printing them back onto Dye Coupler Paper. It never worked for me back then I am afraid, a young man who works at a mini lab near me just got a bucket load of grant money to try to do exactly this. He asked me for help and all I could do is lead him in good registation methods. I will see him time to time and report back on progress.

Well, he'll need the money. Multiply the expense of printing directly onto chromogenic paper by about ten times if actual sheet film is being used for the separations and the added labor, the registration equipment, etc. I could hypothetically do it anytime, and have even established the correct parameters for in-camera separations. But the root question is, Why? Finding subjects that hold still long enough outside of a studio it itself a daunting challenge. And I already have a high quality alternative, making master internegatives from the chrome originals, with all the corrections built-in. Even that is now getting out of bounds in terms of film expense, at least when 8x10 film is involved.

koraks said:

OK, thanks for clarifying. I understand from #9 you intend to print onto RA4 paper. You could either use color filters such as the kind you linked to, or simply use a dichroic color head. You can use a dichroic head to get reasonably close to the primaries you need for consecutive exposures. I've done this with modern RA4 Fuji paper; it'll work. It'll be marginally better if you could use exactly the correct wavelength green (550nm LED works well), but what a dichroic filter head gets you will be close enough.



You don't layer them. You make consecutive exposures onto the same sheet of paper. The challenge will be to retain alignment while doing so. This will be easier if you contact print and use some kind of pin registration. If you're going to work with inkjet printed negatives you're pretty much limited to contact printing anyway if you want decent resolution. Enlarging inkjet negatives generally results in rather disappointing prints.



Yes, you'd have to do the same kind of linearization that you'd typically do when printing e.g. cyanotypes, salted paper etc. from inkjet negatives. So you start with determining the ink load/maximum density in the negative you need to get a full tonal scale, then print a step tablet and measure the densities on C, M and Y, and use those measurements to create a linearization curve. That's it in a nutshell; sounds a little easier than it is in reality, but in principle it's a straightforward process.

A more high-tech approach would be to forget about the printed negatives and instead use an LCD such as the kind used for 3D resin printing and a color filtered light source. If you use a light source that can be controlled from a computer/microcontroller (e.g. RGB LED array), you can automate the process. There's a (precious) few people who work this way; @smng does this, but she's in the middle of moving shop. @avandesande is exploring a similar route and has posted extensively on the use of an LCD as the image source in an enlarger for B&W purposes. That's half the solution right there; the other half for your application would be the light source. On that part I've published lots and lots on my blog; see e.g. here.

Click to expand...

Yes I have an enlarger with a dichroic head so I'll try that - hadn't even thought of that. Thanks for all your help!

Sure thing; please let us know how it works out! It's an interesting project; I'd love to see some examples of what you manage - whether they're successful or not, it'll be interesting in any case.

koraks said:

Sure thing; please let us know how it works out! It's an interesting project; I'd love to see some examples of what you manage - whether they're successful or not, it'll be interesting in any case.

Click to expand...

Yes I'll show some examples further down the line - will probably be asking for help again no doubt!

Always welcome; your project sounds nice and many of us are happy to help by sharing our thoughts!

I am a bit lost as to what the OP is trying to achieve. I understand that the goal is to print a digital image on RA4 paper, right? There must be printing machines that can produce RA4 prints (which use lasers) from either film or digital image. Is the idea to replicate such machine using conventional projection printing from separation negatives? This must be an interesting and pretty challenging experiment. Apart from technical challenge are there other reasons why you would want to do this?