
Blood Oranges and Lemons
Oil on Panel, 2009
11" x 14"
I used Munsell chips to find the averages and then I used my palette knife with the hues to check on the HVC of the objects. This takes some time to do but is well worth it. It trains your eye and it is amazing how your eye can fool you. Munsell offers a tool to check your averages and from there you can mix a controlled palette for the painting. I don't do this all the time, I mostly use what is called a full spectrum palette that comes from Frank Vincent DuMond through Frank Mason. In this palette Cad Orange is the middle value of the palette. Now since I have been using Munsell it has struck me that not all Cadmium Oranges are the same value. I have two brands, OH and Williamsburg. The WB was 2.5YR,6/16 in Munsell color space. The OH was 10R,5/16!
A huge difference. So here's the problem with the DuMond/Mason palette: if you don't use the same brand of paint all the time your Orange value is going to shift, and in the case of the WB and OH it's a whole value step. The OH is one value lower than the WB. I tested W&N and it was 2.5YR,6/16. I don't have any other brands of Cad Orange so I can't say for sure about which is the most consistent, but it seems to me that OH is leaning to much towards red. Frank Mason had his students buy all the same brand for the palette, which I now see as a way to control this problem of shifting values.
In the DuMond/Mason palette you have to mix a string of gray scales that are the same value as the Cadmium's, such as Orange value gray. I find that Munsell is more accurate and mixing the gray scales to specific color values means you have to adjust.