Digital Negatives: The Color Ratio Method

There are several different ways of making a digital negative, but all seem to involve a Photoshop curve at some point. In this article geologist Clay Harmon applies his skills to photography and has developed his own curve for a Pyro-like digital negative – to be used with an Epson 2200, but, it can also be tweaked for other printers. All free to download here.

Writer and Illustrations / Clay Harmon

---

The following is a quick and easy method to create alternative process negatives with the Epson 2200 printer and Pictorico OHP film.

The genesis for this approach was from Keith Schreiber’s excellent method for creating digital
negatives on the Epson 1280 that was presented at the APIS conference it 2003. I used his
technique quite effectively for two years. When I got an Epson 2200 printer, I was looking for
a method that would produce results of a comparable quality.

I decided that I wanted a method that also had a little flexibility built into it, and would allow me to use one method to produce a wide range of density ranges in my negatives. It occurred to me that if I could find out the right color with which to ‘paint’ the negative, I could use a single correction curve and just select the appropriate saturation of color to determine the final density range of the negative. The technique and color I eventually found will allow the worker to produce a negative with a density range from 1.55 (platinum) to 1.9 (pure palladium) to 2.5 (albumen and salt) with a single curve.

My background in geology provided the seed of the idea I eventually adopted. Geologists use ternary diagrams, an unusual tri-variate graph, to display data where all the proportions of three quantities must add up to one (100%). It occurred to me that this was exactly the same information I was trying to discern on my quest to find the appropriate ratio of the three primary colors (Red, Green and Blue) that would block the most UV light when used to colorize my digital negatives. My intuition was that as long as I could maintain the ratio of the colors to each other in conjunction with a single correction curve, it would allow me to extend the density range by increasing the saturation of the hue, or alternatively, reduce the density range of the negative by reducing the saturation of the hue.

So I created a ternary diagram (see above right) that had red, green and blue as its end members, and varied the proportion of each color from 100% to 0% at the opposite side. The way these graphs work is that each vertex represents 100% of the quantity, and the base opposite the vertex contains 0% of the quantity. Once you do this for all three vertices, voila, you have a triangle to print that will contain a complete range of color ratios. Once this is printed, one can find out the triangle that blocks the most UV light. This triangle would contain the best color to use for the maximum density range potential when colorizing the negatives. The next step was to use the color ratio I found (in the case of the Epson 2200, this was R:64 G:128 B:0, I could then create a strip that had sections of this color with the same ratio, but varied the saturation. Using my UV densitometer, I could then pick the color that gave me the correct density range for the particular process in which I wanted to print the negative. The density range is easily calculated by subtracting the Pictorico OHP b+f density from the UV transmission density of any given square.

I then took this color, printed out some test strips and images and adjusted it until I liked the result. I have used the densitometer in the past, but I think the best method is to use a dry print sitting next to your computer monitor. So I ended up with a basic curve that I can now use for creating a wide range of negatives for alternative processes.

Enough bull – how does it work?

It works great.

Details, maybe?

I’m going to fly through with the specifics I ascertained for the Epson 2200. This method should work with other printers, but you’re on your own. Email me and I will send you the.tiff file of the basic triangle to print or download it here.

1
Edit your image until it is a masterpiece in waiting.

2
Make sure your image is in 16bit mode.

Image->mode->16bit will get you there if you aren’t.

3
Curve your data with the curve included with this file. It is called RatioCurve. Clever, huh?

You can download the curve here.

If you have problems downloading the curves:
On a PC: Right-click with the mouse on the link and select “Save Target As”
On a mac: Hold down the Ctrl key and click on the link. Select “Download Link to Disk”.

What the curve looks like:

4
Invert the image. I always use command-i.

5
Convert everything to RGB. It will ask if you want to flatten the image. Of course you do. Just be sure and save your version with all of your editing layers before you do this.

6
Add a new layer. Set the mode to ‘ Screen‘.

Creating that new layer to colorize with:

7
Click on the foreground square, and use the color picker to change the color to:

Platinum Density Range 1.6 – R:127 G:255 B:0

Palladium Density Range 1.9 – R:70 G:140 B:0

Some Palladium PapersDensity Range 2.2 – R:50 G:100 B:0

Salt and Albumen Density Range 2.5 – R:25 G:50 B:0

Color picker settings:

Color Settings needed:

8
Use the paint bucket tool to fill the layer you just created.

9
Print the negative using the same prints settings outlined in the Schreiber article.

What’s with the lack of detail?

I’m a busy guy, and Keith’s webpage has all the rest of the information you need. Be sure to use the color settings he recommends. Editing with some weird custom color setting is a recipe for disaster. I recommend saving these color settings with a unique name you will remember like HeyDummyTheseAreTheDigiNegCSFs or something equally memorable. It is also a time saver to save all the particular printer settings needed with some unique name as well.

Once you do this a few times, it also becomes evident that you should combine all of these steps as photoshop action. Then, with one click, the whole thing just happens. The Photoshop help pages have a great section on recording actions, so read those and make it happen for yourself.

This is a straight palladium print made on COT 320 with.6ml ferric oxalate.6ml palladium and 1 drop of Na2 2.5%, developed in potassium oxalate. The Ratio Curve was used and the negative printed on an Epson 2200.

Caveat emptor, carpe diem, and Sola Bonum Linguam Mortua Linquam est.

You can also downloaded the article as a pdf or download the ternary diagram.

Clay is a geophysicist for the purpose of feeding his family and a photographer for feeding his imagination. You can contact him by email: wcharmon@wt.net