Mixing a color scale

[Allan Rahbek]

I have been thinking of making my own color scale and tubing it, and here is what I came up with.

I made a copper measuring cup divided in 20 steps with a divider to hold the paint. Each step contains 4,5 ml, and the full cup contains 90 ml.

This is a testing model and I can easily make another size to make it fit available tube sizes.

I load the colors direct from the tube and scrape off with the palette knife to make an even fill up. Then I lift up the divider and fill the rest of the cup with the other mixing color. Take it out, mix and tube.

Allan

[Linda Brandon]

Hello Allan,

I love painting tools and I'm very impressed that you made this, out of copper, no less!

Would you not get the same proportional accuracy if you just pushed out a long line of white along a ruler, and added a similar line of contrasting color? You would adjust the amount of your contrasting color as you go up or down the tonal scale.

I used to tube grays made from contrasting colors on the Munsell color chart, using this system to ensure a 50% white/50% color (or whatever got me to a midtone grey, depending on color intensity) mix.

[Allan Rahbek]

Hi Linda,

I am sure that it can be done with a ruler. But I am also sure that it would end up in a mess for me when the ruler is lost in the paint and I have forgot wich proportions to do next.

My tool is for non matematics.

If I fill it up the mixture can easily be contained in a 125 ml tube, and I can make 20 steps in the scale, numbered from 1 to 20, or I can make 10 steps by using 1,3, 5, 7, and so on...

If I fill in 38 ml tubes I can use 8 steps = 36 ml.
Starting with 1 step White + 7 steps Black, 2 steps White + 6 steps Black and so on.

Now I only have to figure out how to get the paint tubed in the inside of the empty tubes. Any suggestions ?

Allan

[Leslie Bohoss]

Hi Allan,
Interesting idea! But how to fill the tubes..(without big mess)?
After some considerations:
A tube (36ml) has a diameter of ca.25mm. Now I would take a pipe (plexi or copper or somewhat with thin wall) diameter a bit under 25mm and a volume of min. 70ml. (2*36)
By cutting in the middle (long axis) I get two similar "sumps". One is my mixing part with measuring scale etc.. (divider is something that fits, like 50 cent or so) other is my part to cover it. After doing the mix, press the second (clean) part over the other one.(reunite)
Join together with rubber or band. Now we have a "loaded weapon". One thing is still missing: something that fits in the pipe to push the mix in the empty tube. Ready! I think , you have only a few parts, easily to clean and it works without mess. A big "injection"
Really you can looking for a bigger medical injection. After few cuts is ready to use .(and has a scale)
I hope it was understandable.( oh my english)
Cheers.

[Garth Herrick]

Hi Allan,

What a clever and well crafted proportional mixing device you have made. It should work well for your purposes.

I used a similar method last year to mix up 50 shades of gray (adding pure white and pure black to the grays makes a total of 52 value levels). After much experimentation with every brand of paint, and several different white pigments, I chose to use only Old Holland Titanium White, and Old Holland Ivory Black. Old Holland apparently makes the only commercial Titanium White that has nothing else (like Zinc) added to dilute it. This I found to be very important for my purposes of scientific control. Marvin Mattelson has pointed out that the cool gray mixes that I get with just black and white are not useful for his purposes of having warm grays to mix with his colors. He is absolutely right, but this was not my purpose. What I am doing is simply having a gray value index that is virtually calibrated with corresponding values within Photoshop, as a tool for rendering correct values from a computer reference image. These gray values make up a scale that is archived in tubes of paint fo give me consistent results even in the years ahead in the studio.

I found the old Holland Titanium White and Ivory Black to be of roughly equal tinting strength, which was very useful for formulating relationships in pigment proportions. I tried flake white, which yields a more pleasant warmer gray, but I found that I had to multiply the white proportion by a factor of seven (!) with Flake White to equal the tinting strength of Titanium White. So choose your whites carefully.

Another thing I discovered is that there is not a linear relationship in pigment proportions when mixing to match values in photoshop. One must empirically find the right warp in the proportions, especially as one approaches the outer ends of the value scale. This you may need to discover for yourself, Allan. You may find yourself favoring slightly off number relationships like 1.35:36, 1.83:35, etc. But you can do this with care with your device.

You are probably curious how I mixed 50 gray values with great control. It was basically the same method and technique you are proposing. I had a plastic pastel tray, with individual slots measuring 12mm, by 12mm, by 100mm in length. I had a palette knife with a rectangular blade 12mm wide and a flat end. I used one slot for the black and another for the white, and they were separated a bit just to keep things neat and tidy. I had all the formulations calculated out and pre-measured in a list on a piece of paper. My proportional formulas are too complex to share here. Using a ruled scale, I determined where to place the palette knife as a stop, just like your method. When all was measured, I scooped all the paint out and mixed it thoroughly on a glass plate.

I used the small 12 ml tubes. To fill them, I used a palette knife to place the paint within the tubes, periodically tapping the tubes against a hard surface to allow the paint scoops to settle within the tubes and displace the air pockets. This takes much practice, but eventually the tubed paint is neatly produced.

I hope this helps! Good luck with your investigative color scales project.

Garth

[Allan Rahbek]

Thank you Leslie,

You just gave me a good idea. I will make a loading tool and show it soon.

Hi Garth,

I don

[Garth Herrick]

[QUOTE=Allan Rahbek]
Hi Garth,
I don

[Allan Rahbek]

Hi Garth,
I wonder if the irregularities in this scale is coursed by differing in light reflections from the paint. It seems illogical that these variations should occur because of the paint quality.

I agree that you don

[Garth Herrick]

Allan, I don't believe paint reflections have anything to do with the irregular intervals in the scale. What I believe these numbers do reflect is despite my most careful measuring and mixing in small batches, my proportional mixing was off target from the beginning.

As careful as you intend to be with your new color mixing scale, it is easy to be slightly off and inconsistent with proportional measurement, especially in small quantities. What you think is a 7:1 ratio may in fact be a 7.1:0.9 ratio. This example should illustrate and explain the inconsistencies in my numbers. But as long as you know what numbers you have made, you will be fine.

Garth

[Allan Rahbek]

Garth, I am going to make a mixture of Raw Umber and Ivory Black 1:2. From this I will make a scale of 20 shades by mixing with Zinc White or Titanium White.

I have to do some tests to decide how it works. Maybe the Zinc White is too weak in color intensity.

Then I will make a colorcard with these colors. In that way I will have a direct reference to my mixed color scale.

That way it will not be a problem if the scale is not perfect balanced.

Allan