Testing paints How to check how good things are
So you have just made some beautiful new paint. You are certain it is the best paint ever made and you want to show the world how good it is beside the paint you used to buy. How do you prove it? Or perhaps your goal is humbler, you need to be certain that you have either enough, or not too much binder in the paint. Perhaps you have a new colored clay you just dug from a river bank and the color is so brilliant you suspect it is not all iron oxide? Or maybe the manufacturer is using marketing speak when describing a new super yellow, and you need to find out if the claims match the reality. Then in each of these cases you need to test. This page shows you how.

The testing procedures here are simple tests relatively easy for most artists to conduct and offer a convenient rule of thumb. Testing using calibrated scientific instruments is an expensive process best left to the experts. The Yellow Pages can help you find the relevant testing laboratories in your area should this prove necessary, Golden paints offers this service to artist's as well. but for most practical purposes the following are adequate for studio based paint making. After all, if you were to sell paint more rigorous testing would be in order but for your own use you just need to be assured that the paint you are using will not fade in a hundred years (or within months as is the case with the worst pigments) and that the paint film is not going to disintegrate disastrously down the track. Occasionally you may want to test a manufactured paint because you think that they are adding too many fillers. A simple drawdown will reveal a surprising amount about paints and is the easiest of tests to do.

Record keeping
Your memory is not perfect
You need a note book that is sturdy enough to last years of use that you always keep with your pigment supply. The first half should include your notes of how and when you mixed individual colors. There is often a time lag between the use of a color and the discovery that this one needs to be repeated and the actual making. It is very difficult to remember every detail. Keep notes that become your recipes for the future. The second half should be your testing notes. Some tests such as light fastness are conducted over months and years. It is impossible to remember dates with accuracy after a certain time. For meaning full results good record keeping is essential. Transferring the results to computer software can be useful, but your studio notebook remains the back up for everything.

Formulation Are the ingredients correct?
Extremes of poor formulation are obvious during grinding, but as you mull it may be necessary to take a closer look by scraping the oil paint mass together with the palette knife and noting it's ability to retain its shape without sagging or running in the case of oil paint. A small portion can also be placed on a piece of paper towel. Excess oil may become visible as it is absorbed by the paper and grows as a ring around the dab of paint. It is possible at this point to add more pigment directly to the paint being ground via a clean palette knife and gently sprinkling the pigment particles over the paint mass. Grinding time needs to be extended as if the grinding process had just begun to ensure that all the new pigment particles are evenly coated in oil. Too much pigment will become quickly apparent in the form of a crumbly paint mass. Oil should be added a drop or two at a time until the paint has the desired combination of buttery body and brushability.

Defects can also be revealed by brushing out some of the new paint thinly on a gesso surface. Flocculation and agglomeration where the pigment particles clump together as small lumps should be revealed by this test. Streakiness should also be apparent at this time if it is present.

Other paint media will have their own characteristics. Egg Tempera and Gouache for instance should have the consistency of cream, Both Watercolor and Acrylic will have a wider range of possibilities depending on the desired end result. All can be examined by brushing out and examining the resultant film.

Egg Tempera should be dabbed on a piece of glass and let dry to see if enough egg yolk is in the mixture. Try the same test for acrylic as well if you think you may be pushing the pigment concentration too much.

Light fastness Does your color change with time?
Light fastness tests are fairly easy to do and can give some surprising results. Some colors fade, others can darken and the same pigment can behave differently in one medium compared to another, so results from one should not be seen as evidence for another. One of these differences stems from the protection that oil films (and acrylic) can give to pigments, meaning that a pigment that is permanent in oil and acrylic can be fugitive in watercolor. Some pigments, especially azo yellows are more light fast in acrylic than in oil. Testing individual pigments in individual media is the only way to resolve this question for individual colors. When looking at light fastness information provided by manufacturers take note of the paint vehicle the testing information is based on if available.

Cardboard well prepared with acrylic gesso front back and sides is suitable for most testing purposes. Outdoor testing is best because results are obtained quickly and then can be calculated to indoor light equivalents.  The dashboard of a car is surprisingly good for the purpose if it is an older car that has a convenient shelf, but it is better to place the card under glass in a purpose made tester as described below.

Always record dates and pigment names and tint proportions beside test  paint samples as it is easy to forget. Further information should be recorded in your color book in the studio.

The testing can be carried out in the rough and ready way described below with Alizarin Crimson used as the 'standard' for fugitive. It works but is only a rough guide with many variables. Far preferable is to use Blue Woolscale cards which calibrate exact rates of fading accurately and can be used to assign the equivalent of the ASTM classes to your colors. These are no more difficult to use (in fact in many ways make things easier) than the rougher method and are recommended for those who are very serious about testing their materials. The Woolscale and equivalent fade resistance (in years and compared to ASTM classes) is given below.

(1) Basic method
Make the test cards

Paint the color thinly and smoothly on the card in a strip about 40 mm wide. make 2 more painted strips as tints, the first with 25% white added, the second with 25% color added to white. This is because we most commonly use colors as mixtures with other colors and often with white. Yet most pigments are far less light fast as tints than as full strength color. These tests as tints give an excellent idea of how well the color stands up in the concentrations you are likely to use when actually painting. it is useful to have a comparison strip of a known ASTM l l l color such as Alizarin Crimson in tint on the card as well. Several colors may be tested on the one card. Partially cover the strips of color with a non rusting metal or other material that is totally impervious to the light and is not too thick (to cause shadow at the edge). It should have a sharp edge so that there is a sharp division between exposed and non exposed. so that some of the test strip gets light and the other part does not. After an extended period visual inspection will reveal if there is a discernible difference between the exposed and non exposed parts. Great care must be exercised in replacing the  strip in the identical position if the test card is to be left for further exposure.

Test duration and interpretation
Useful results in full sunlight can be expected within weeks for fugitive colors, six months for more durable ones. A year is excellent. Full sun is far stronger than interior light and 1 year in the sun is likely to be the equivalent of as much as a hundred years of indoor exposure. There is some variability in this depending on your distance from the equator and season. Any test results you may choose to leave in normal light in the studio may be left for a number of years before checking and noting any change if any. In normal light on the studio wall there should be no change in the colors at any point of a usual test period. It is reasonable to expect no change after 50 years if you could wait that long. Any color that noticeably changes in 50 years or less of normal interior light should be regarded as fugitive. Be aware that there are many colors in existence that noticeably change in 5 to 10 years of indoor exposure and some that change in a matter of months indoors.

For accurate tints use a set of kitchen measuring spoons, the sort that offer 1/4 spoon, 1/2 spoon and so on. In the size that will make sufficient paint for your tests use a palette knife to fill the measuring spoon and level it with the blade of the knife. Use 1 spoon of white and 3 of color to get the 25% reduction, and 3 of white, 1 of color for the 75% reduction.

The tests need to be carried out outdoors and under glass to protect from moisture. In the tropics it may be kept horizontal but the further from the equator you are located the more the test board should be art an angle facing south in the northern hemisphere and facing north in the southern hemisphere. It can be as simple as a board leaning against a shed with a glass sheet covering it and light plywood sides siliconed to the glass to stop rain getting to the test cards. Where the test is of very water sensitive media such as watercolors then extra care needs to be given to weather protection and the possibility of condensation under the glass. Fix it so that it is sturdy enough not to blow down in any wind, and that children and animals cannot easily get in to move or damage the cards.

Making do
For those without the luxury of a garden space to set up a test and (like me) have a garret on the top floor of an inner city building you need to make do and adjust interpretations of results accordingly. If you can set up your test cards under glass at a studio window that receives full sunlight, preferably a south facing window in the Northern Hemisphere, and a North facing window in the Southern. Time how much of the day the cards get full sunlight and work out the percentage of the average daylight hours. Maybe it is 50%, or only 30% or some other portion. You then have an amount by which to increase the time of exposure and interpret accordingly. The same rule of thumb applies if your only sunlit window is east or west facing. If you have no window that gets full sun then you may need to transfer your tests to your parents or girlfriend/boyfriend's place as can be arranged.

(2) Using Blue Woolscale
As has been pointed out there are many variables in total exposure depending on where you live and how much direct sunlight you can get. The season can also make a big difference. Blue Woolscale overcomes this by allowing you to judge the total light exposure rather than using a simple rule of thumb based on guesses about light intensity. The test cards have 8 strips of wool on them each died with a different dye that fades at a known rate from #1 the most fugitive to #8 the most light fast. By exposing a Blue Woolscale card alongside your paint samples with one half of the Woolscale covered just as with the paint samples it is easy to judge actual light fastness no matter where you are located as the Woolscale records the total light accumulated and is not time dependent. It is still better to use full sun as the test will finish sooner, but in fact this test will work in a wide variety of light situations. Prepare the samples and place them under glass as described above. The only difference is your scientifically calibrated fading card. Cards can be purchased for around $6 to $10 each depending on quantity bought. Golden Paints sells a testing kit that utilizes Blue Woolscale for around $30 for an excellent kit that includes full instructions and materials that make testing easy.

Light fastness Ratings and what they mean
Now to the contentious theme of putting an actual lifetime value on light fastness ratings. There is no easy answer here and most authorities try to dodge the issue. The reason being that pictures are shown in such a wide range of conditions. Even within one room one wall may get double (or more) the light of another wall, effectively doubling any effect that light will have on that work in a given time frame.

The following time frames assume proper mounting in an average setting indoors. In Blue Woolscale it should be noted that each level is 3 times more exposure than the previous. The Woolscale equivalents to ASTM classes is approximate only.

It is also easy to find differing interpretations of time for a given total of light (measured in lux) The following is a good rule of thumb and is based on MacEvoy's excellent 'The Worlds Finest Guide To Watercolor Painting' and that in turn was based on Gottsegen 'the Painters Handbook' and Colby 'A Suggested Exhibition Policy For Works On Paper (Journal of the International Institute For Consevation: Canadian Group 1992)'

ASTM  l      Excellent. Equivalent of Blue Woolscale 7 and 8. More than 100 years without change. (Probably more than 200 years in the case of Woolscale 8) (between 300 and 900 megalux)

ASTM  l l       Very good. Blue Woolscale 6. Between 50 and 100 years without change (100 megalux)

ASTM  l l l       Fair. Blue Woolscale 4 and 5. Between 15 and 50 years without change. (10 to 32 megalux)

ASTM  l V         Poor. Blue Woolscale 2 and 3. Between 2 and 15 years without change. (1.3 to 3.6 megalux)

ASTM  V           Very poor. Blue Woolscale 1. Less than 2 years without change. (.4 megalux)

Transparency Covering power or glazing suitability
All you need here is a white surface with a black line several millimeters wide. brush the test paint thinly across both the white and the black. how well the color covers the black line is a good indicator of relative transparency or opacity. Gouache can be tested in the same way to ensure that the proportion of chalk in the formula will provide adequate opacity.

Draw-downs Practical way to examine strength etc
For this you will need a good quality house painter's scraper. It needs to be one with a nice flexible blade about 75 or 100 mm wide (3 or 4 inches) You will also need white bond paper in sheets about A4 in size.

This test is especially useful for making accurate comparisons between samples. A standard thing to do is always test against a paint of known qualities. Place 2 dabs of paint at the edge of the scraper side by side. then steadily and using a light pressure so that the paint goes onto the paper in a thick layer. After 2 cm (1 in) increase pressure and with the blade held near vertically draw the paint film out in a thin layer. By holding the paper up to the light and looking through the paint much will be revealed about purity (if you are checking a bought tube) relative strength and so on. This test is likely to reveal if a paint has been 'boosted' when testing manufacturers paints (that is had a dye added to brighten the color) and reveals the color's undertone. Dispersion problems should also be visible in this test. Lightweight watercolor paper can be used to test watercolors by wetting one half of the paper and drawing  the test from the dry to the wet area. As the paper is likely to be thicker than cartridge paper a stronger light source may be necessary.

Go to Demonstration of making paint  ]



Feller, R L,    Artists Pigments    1986 (National Gallery Of Art / Cambridge University)
Gettens, R J, and Stout, G L,      Painting Materials: A Short Encyclopedia,      1942 (Dover)
Gottsegen, M D,    A Manual Of Painting Materials And Techniques,    1987 (Harper & Row)
Mayer, R,   The Artists Handbook Of Materials And Techniques,    fifth edition 1991  (Faber & Faber)
Roy, A      Artist's Pigments: A Handbook Of Their History And Characteristics,      1994 (Oxford University Press)
Various,    Encyclopedia Britannica,    fifteenth edition 1981  (Encyclopedia Britannica, Inc)

Internet Resources  |  imageContact  | image Frequently Asked Questionsimage