Posts Tagged ‘watercolor’

From charcoal portraiture to watercolor

Monday, June 1st, 2015

In the process of overcoming six hours of jetlag, I woke at an ungodly hour this morning. Although I managed to convince my body to continue resting for an additional hour, my mind would not cooperate. Soon it was in full analytic mode, reviewing the coursework from the portraiture workshop I had been attending, and attempting to reconcile it with my ongoing painting efforts.

Suddenly the concepts and alternative variations began to click into place, and I gave up my immediate attempt at jetlag recovery as a hopeless cause. It was time instead for note making and some in-depth analysis and strategizing. So here goes…

1505cc Chiara

Realization 1: the charcoal we used at the workshop handles more like oil paint than like watercolor, as, to a great extent, we built our work from darks into lights, using erasers to cut away the medium to reclaim the underlying lighter values. Although I suppose I could do all my portraits in oil, which can be handled in a similar manner, I want to be true to myself and figure out how to apply the lessons to watercolor.

Realization 2: in order to apply the information I had gathered during the course to the watercolor medium, I would have to think outside the charcoal box and adjust the approach to suit a medium that generally builds from light to dark.

Step 1: Pre-sketch lightly to establish measurements and ascertain appropriate proportions.

Step 2: Differentiate appropriate shadow and lighted areas. Locate precisely, and reserve with frisket, all highlighted points. Reserving highlights will require identifying and locating features early that otherwise might not be approached until considerably later in the process. (This would partially explain my struggle with feeling a need to position them earlier than workshop participants were advised to start incorporating “details.”)

Step 3: Wash in a thin base glaze (based on the temperature of the light source, usually warm) to establish the lightest tone slightly darker than the highlight (which will retain the white tone of the underlying paper).

Step 4: Block in the entire shadowed (and other dark) areas as neutral half-tones that can be further glazed to deepen the shadowed values as needed. (This is different from the charcoal/oil approach, in which the shadows are blocked in with a much lower value.) Warm half-tones can also extend into transition areas between light and shadowed planes, but keeping even the lightest shadows darker (and in general, except for reflected light, cooler) than the darkest lights.

Step 5: Continue to build up appropriate values within the lights, using varying hues and temperatures to turn the form and to provide interest and detail.

Step 6: Continue to build up appropriate values within the darks, again varying hues and temperatures to turn the form and to indicate reflected light, while limiting detail within the shadow areas.

Step 7: If it hasn’t already been sufficiently incorporated as negative painting or to provide soft transitions from the subject to the background, pay any additional needed attention to the subject’s environment.

Step 8: Remove frisket, adjust edges as necessary, and incorporate any final details.

Of course, these strategic steps are purely theoretical at this point. I haven’t had time yet to test them out for myself. But at least they should provide a guideline to base my experimental efforts on.

Reconsidering the Color Wheel, Part 2

Friday, June 15th, 2012

In order to understand the pigments in my own watercolor palette better, I decided to chart my paints. I wanted not only to indicate the complementary pairs but to record gradations of intensity, carrying the color from one high-intensity hue, through the balanced gray, across the wheel to the high-intensity complementary hue. In the process, I had to correct some of my own erroneous assumptions, like the yellow/purple pairing.

Munsell style color circle

As I wrote last time, rather than relying on the old tried and not-so-true three-primary color wheel, I turned instead to the color theories of Albert Munsell.

Since not all the paints in my palette corresponded precisely with the major and minor hues in the Munsell circle, I chose those that seemed closest. I ranked similar colors outside the circle, roughly in order of hue, to help me better judge what each of their complements would be without having to chart my entire palette. Not having a single purple-blue paint to complement my Winsor Lemon, I combined Brilliant Blue Violet and French Ultramarine Blue to create the appropriate blend.

I did not use any ready-mixed gray in the color wheel but made a swatch of Payne’s Gray, outside the wheel, for comparison purposes. No black or white paint was used. If you want a list of the specific paints I used, send a request to

Reconsidering the Color Wheel, Part 1

Friday, June 1st, 2012

Most of us learned about the “triadic” color wheel back in grade school (though we didn’t go so far as to call it that). That’s the one that shows equally spaced primary colors—yellow, red, and blue—separated by secondary colors—orange, purple, and green. Because of this early and well-ingrained training, the concept of an alternative color chart can be difficult for some of us to grasp.

The Munsell color system, in contrast, spaces yellow, red, and blue at less regular intervals around a circle and includes purple and green as additional “major” hues located between red and blue or blue and yellow, respectively. Then, centered between each pair of these five hues is a “minor” hue that combines the two major hues that flank it. Hence, the Munsell circle reads: red, red-purple, purple, purple-blue, blue, blue-green, green, green-yellow, yellow, and yellow-red, which brings us back to the original red.

The Munsell color system is considerably more complex than this, incorporating value and chroma variations as well as hue to differentiate and identify specific colors. But you have to start somewhere, so I looked first at the most basic hues.

The Munsell color circle is particularly useful in identifying complementary hues that will produce a neutral gray. In childhood, most of us were taught that the complement of any primary color is the secondary color that appears directly opposite it on the color wheel. However, these pairs of complements don’t always produce a neutral gray when combined. Yellow and purple, for instance, made an orangey brown, not gray. I knew that in order to make a neutral gray the purple would have to be bluer to balance out the warmth of the yellow. The purple-blue indicated in the Munsell circle proved a much more satisfactory complement to yellow. So I have begun using the Munsell system to find truer complements for all the colors.

Next time I’ll write about how this applies to my own watercolor palette.

New Materials: Workable Fixatif

Wednesday, February 15th, 2012

As I have written about in my last few entries, painting with watercolor on canvas has been a crash course at the School of Oops. I’ll continue here with notes about my work on the pad canvas.

Despite my decision to use the pad canvas only after affixing individual sheets to stretcher bars, I decided to try it stretched only with clips. Firmly clamped to a backing board, my second sheet of canvas held its shape well and appeared to lose little area due to shrinkage.


In placing the subject, I mentally included the margin space on all sides to allow for later stretching and shrinkage. (Refer to the Faces and Figures gallery to see the finished proportions of “Anticipation.”)

When I applied glazes on this type of canvas surface, I found that lifting was still almost as much of a problem as it had been on the stretched and gessoed canvas (Joe Miller’s brand) I had used previously (see blog for January 1, 2012). So, when I decided I needed a final glaze to warm the foreground, I set the painting aside for several days to dry thoroughly (front and back) before spraying it with fixatif. Only when that was dry did I lay in my last applications of paint.

I had not used this type of fixatif before (Krylon Workable Fixatif), but it was called “workable” and stated that it “allows easy rework.” However, apparently that holds true only for the stated “pencil, pastel, and chalk,” not for watercolor. The watercolor beaded up on the fixed surface, and though I was able to do a bit of retouching, it was not satisfactory.

Out of desperation, I recoated the painting with the varnish (Krylon Gallery Series UV Archival varnish) I had used on my first canvas, hoping that that would provide a more friendly working surface that would accept top coats of watercolor. The paint continued to bead up, failing to adhere smoothly to the surface.

One of the few advantages of the seal appears to be that I can wipe off failed attempts without fear of leaving stains and smears behind.

Another advantage (admittedly a major consideration) is the UV protection it offers for any paintings that will not be shielded by UV-protective glass.

I would be interested to hear from others who may have found alternative fixatifs or sealers over which additional layers of watercolor can satisfactorily be applied.

Canvassing the Possibilities, part 2

Wednesday, February 1st, 2012

In my quest to familiarize myself with alternative watercolor surfaces for my work, I bought a pad of cotton watercolor canvas (Fredrix brand) to try. As I unwrapped the heavy pad, the weight of the sheets pulled the entire pad away from their backing board. The adhesive strip along the top was not strong enough to support the weight of the mass of canvas.

However, I decided to try using it as I would a pad of watercolor paper, as I supposed was intended. Using bulldog clips, I fastened the canvas pad back against the backing board, set it up on my easel, and got to work. After wetting, the canvas didn’t stretch but actually shrank. Where the canvas was clipped to the board, shrinkage was negligible, but areas that I had failed to clip shrank enough to form noticeable ripples in the fabric as it pulled diagonally against the clips. Having worked primarily with paper and stretched canvas in the past, I had not expected this to occur to such an extent.

In this case, to encourage more even shrinkage, before applying fixative, I reversed the canvas and, leaving it entirely unstretched, I sprayed the back with water, spreading the moisture with my hands to ensure even coverage. Then I allowed it to dry thoroughly. Most of the ripples disappeared as the canvas dried. Overall, the 20”x16” canvas lost approximately ½” in length (in width as seen in the horizontal orientation below) and ¼” in width (or in height as shown in the illustration).


Lessons learned: 1) Use the sheets singly, rather than on the provided pad, and 2) Stretch even page-like sheets of canvas before painting—it’s shrinkable cotton fabric, not paper. Whether on stretcher bars or by affixing the canvas to a sturdy backing, the canvas must be stretched before wetting. I will use 12”x16” stretcher bars for the remaining sheets, to allow enough edge to wrap around the bars.

More about my experience with the canvas next time.