New Colourwheel

For the past couple of years I have been experimenting with the range of watercolour pigments available to artists. I had become increasing frustrated with the inability to mix vibrant accurate colours. Fed up of muddy greens, I decided to do some research. I ditched my allegance to my 30year choice of palette and started trying out some new colours - beginning with Pthalo Blue (Winsor Blue in the W&N range - hence the DA name).

It would appear that most of the colour wheels produced as helps to WATERCOLOUR artists are wrong, and little help at mixing colours. Many simply fail to take into account that watercolour is a transparent medium . Thinking about the Chroma (how vibrant a paint is is) and tinting strength (how strong it is in mixes), is just as important as the Value (where is comes on the light/dark scale). All this information is hard to come by, and not available from manufacturers because their product varies so much depending on dilution, and what other pigments it is being mixed with. There is also often a difference in the Masstone (the full-strength tone or colour) and the Undertone (the tone/colour when diluted).

Being from a science background, I was sure someone would have published something, and suggest that those who want to know more about the technical side look at  www.handprint.com/HP/WCL/water.html .  I have now gone back to my paints having spent some time coming up with a new mixing help.  The traditional 3 colour points (even with 2 pigments either side) are inadequate for the job. I am a hobby artist of 30years experience who paints infrequently and so cannot keep in my head all the tones and tints that combining paints produce.

I would be interested if others out there have come across an accurate way of predicting colour mixes. My tool is an irregular shape that resembles contours on a map. It was based on  CIElab / CIECAM definitions converted to angles, which I then adapted from observation and testing against other pigments. Personally I use only the Winsor & Newton range, and nearly every paint in my palete is a single pigment with high transparency.  I believe I am close to producing something useful - anyone out there interested enough to share in some testing.