HOW TO TEST THE QUALITY OF COLOURS
Although to accurately test the quality of a colour requires somewhat
elaborate experiments, both chemical and practical, yet there is no
reason why the painter should not determine with a sufficient degree of
accuracy for his purpose the quality of the colour he uses. Indeed, if
this were done more generally, much of the adulterated trash or “muck,”
as it has been called, would be driven from the market, and none would
rejoice more at such a result than the colour manufacturers themselves.
The writer has no connection with, or interest in, these manufacturers,
but it is only fair to assert that they are as desirous that the trade
should use pure colours as the painters can possibly be. Even the
biggest houses produce cheap grades of colours, and this they do, as a
rule, almost under a protest and simply because they are compelled by
painters demanding colours for certain low prices, far below that which
it would be possible to produce the pure article. Our advice, then, to
painters is, make a careful comparison between the pure colour and the
one you are using. At the same time, compare the prices and then see
which is cheaper to use. If even they come out at the same price,
remember that by using a pure colour you will have all the benefit of
that purity of tone so necessary for the execution of good work.
The first thing to be done in testing any paint material is to have a
standard. There must be no doubt about this. Unless we have in each case
something with which to compare the particular sample of colour that is
being examined, we shall have no useful information concerning it. Take,
therefore, good decorators’ colours of well known make. If necessary
purchase small tubes of the best colours, such as are put up for
artists. This will be rather a severe trial but still it will afford a
standard. Having such samples and going through the tests we are about
to describe, the painter can, after some amount of trouble, arrive at
results which are almost as accurate as those which could be deduced by
a chemist. An expert on this question some years ago summarised the
characteristics of colours which should be considered in making the
examination, under the following heads:--
1. Purity of the material.
2. Purity of the tone; brilliancy; richness, which indicate the amount
of care in selection.
3. Fineness of grinding or preparation; this means the degree of the
division of the particles and upon the completeness of such division
will depend.
4. Its spreading capacity.
5. Its body. This applies, of course, only to opaque or semi-opaque
colours. Body is opacity, and means capacity to conceal the surface to
which the paint is applied, and must not be confused with spreading. It
is an inherent quality.
6. Its staining power or tinting strength with white or colours.
7. The quality of purity of the tint with white.
8. If a paste colour, the consistency of the paste.
9. Transparency of transparent colours and the quality of the
transparency.
10. The permanency of the colour.
It will be observed that all of these tests will not necessarily be
applied to every colour. For instance, a transparent colour would be
tested for its transparency but clearly not for its body. The one
condition is the converse of the other.
We will now consider the above-named qualities separately.
PURITY OF THE MATERIAL.--This is sometimes of considerable importance,
as in the case of white lead, whilst in others--for example the earth
colours--it can hardly be said that there is a standard of purity. As a
rule a knowledge of practical chemistry is necessary in order to
determine whether a sample of paint or colour is pure or not.
The purity of white lead, however, can readily be ascertained by the
painter who possesses no chemical knowledge, viz., by aid of the
blow-pipe. Take a piece of flat charcoal and cut out a hollow space from
it into which place a small piece of white lead to be tested, about the
size of a pea. Now direct the flame of a blow-pipe upon it, using an
ordinary paraffin candle or a Bunsen burner, taking care that the blue
portion of the flame bears upon the lead. Keep up a steady blow for a
few minutes and the white lead will be converted into metallic lead,
which will show in the form of a bright silver-like button. If the lead
is adulterated the blowing will only have the result of making it appear
like a cinder. To conduct this experiment successfully requires a little
practice with the blow-pipe in order to obtain a steady flame.
Another method of testing is to place a little white lead in a crucible
and place this on a hot fire, when, if genuine, it will be converted
into metallic lead.
A form of blow-pipe that may be purchased at most ironmongers’ shops
consists of a wooden handle with a mouth piece filled with cotton soaked
in benzine. To this is attached a rubber tube with a mouth piece. This
blow-pipe is very easily used, and may be successfully employed in
testing the purity of white lead in the manner indicated.
PURITY OF TONE.--Some remarks on this subject will be given under the
heads of the various groups of colours. Speaking generally, the richness
of brilliancy of tone is easily discernible by placing the sample to be
tested side by side with another of well known excellence. In siennas,
ochres and umbers the selection of crude material by which the richness
of tone is assured is of great importance.
FINENESS OF GRINDING.--The method of testing the fineness of a pigment
usually employed by the painter is to rub a little on the finger nail;
but this is a crude and unreliable method. If the pigment is dry and it
is desired to compare it for fineness with a similar pigment or white
lead, the following is as good a plan as any:--
Take two tall vertical glass jars, place in them an equal quantity of
turpentine, and then take a small quantity of the white lead to be
tested. Place it in one jar, and an equal quantity of the pigment with
which it is to be compared, in the other; thoroughly stir up both and
then note the time it takes the samples to settle. If graduated marks
are made on the two jars the observations will be taken more readily.
Another test is to weigh out equal quantities of the two leads, and then
to take a very small quantity of the same colour, say black, and add to
each sample, thoroughly mixing. The lead that is the lightest in colour
will be the finest. The explanation of this is somewhat interesting.
Suppose that we have a number of cubes of white lead each measuring one
inch side. This will give us six superficial inches to be coloured. Now
suppose that we break up these inch cubes into half inch cubes, which
will give eight half inch cubes to each inch cube. Now as each half inch
cube has six faces measuring half an inch by half an inch, it has a
superficial surface of three square inches; and as there are eight of
the half inch cubes, there are twenty-four superficial inches to be
coloured against six in the inch cubes. It will be seen, therefore, that
by increasing the fineness of a pigment a greater surface is presented
to be coloured, and hence more colour is required.
Another test for fineness is to paint different samples thinned in
turpentine on plate glass; when dry the two specimens may be compared
and the difference of fineness between them will soon be apparent.
Still another test, and one frequently used by painters, is to place a
quantity of the colour ground in oil that is to be tested upon a level
surface such as a piece of glass, and to run the blade of a spatula or
palette knife over it and then over another sample with which it is to
be compared, noticing carefully the difference in appearance of the two
samples. By these means the presence of grit is discovered.
SPREADING CAPACITY OR COVERING POWER.--The spreading capacity of
pigments and their “body” are very nearly related, although of two equal
in body one may possess greater covering power or spreading capacity
than the other. A practical method of testing covering power is to mix a
small quantity of a standard paint and an exactly similar quantity of
the pigment to be tested, taking care to use precisely the same amount
of oil and thinners in each case. Then, taking a clean brush for each of
the paints, paint a door, or other surface that has been primed, on two
panels side by side, continuing to paint till all the pigment has been
in each case used up. The one that goes farthest has the greater
covering power. In comparing the two it will be well to notice whether
the body is equal in both cases, as one may go farther but not cover so
well.
BODY.--The word “body” as applied to pigments is almost synonymous with
opaqueness. It is the most important property of a pigment, and it is
because white lead possesses the quality in an eminent degree that it is
so much valued.
Body is sometimes called “covering power,” but this term is a little
misleading, as some may suppose it to relate to the spreading capacity
of the pigment.
If two different white leads ground in oil to an equal consistency are
applied to different panels of a door, primed in the same manner, the
one of the two leads that possesses the better body will be shown by it
hiding the grain of the wood better. Some white leads, especially those
that are manufactured by the new processes, lack this important quality
of body, and three coats will only cover the work as well as about two
of old process white lead.
There are number of methods of practically testing the “body” of
pigments, among the simplest being the following.
Prime and paint a board with alternate black and white squares, like a
chess or draught board. Take a sample of a pigment, similar to that to
be tested, of which the body is known to be good, and paint a wide strip
across the chess board; then paint a smaller strip of the pigment to be
tested. When both strips are dry, by comparing them one can tell almost
at a glance which has the better body, the superior pigment covering or
hiding the black squares better than the other. A second coat may
afterward be applied to each over a portion of the strip, if desired.
It is important to notice that in all cases of practically testing
paints the results are obtained by comparisons being made, and hence it
is necessary in every case to have a standard with which to compare the
sample to be tested.
The test of painting over squares of black and white may be varied by
using stripes instead. The test answers equally well for white lead,
zinc, lithopone, or any colour of which the quality of body is of
importance. In some colours it is of little moment.
TINTING OR STAINING STRENGTH.--We have already explained at length how
greatly the tinting strength of different colours or stains varies. Any
painter can test the tinting strength of any colour himself in a very
simple manner. All that is necessary is to have a pair of apothecaries’
scales, some blotting paper, a palette knife, some pieces of glass or a
flat piece of marble and some pieces of waxed paper. First weigh out say
eighty grains of dry white lead or dry zinc. Any other white will answer
equally well. Place these eighty grains on one side of the glass and the
second eighty grains on the other. Now take the dry colour and weigh one
grain and add that to one of the little piles of white, then weigh a
grain of the standard colour and add that to the other pile. Now add to
each pile a few drops of oil, taking care that the number of drops is
the same in each case. With the palette knife thoroughly mix until no
streaks can be seen and the mixture is perfectly uniform. Then by
comparing the two the difference in tinting strength will at once be
apparent. The same result would have been produced had ordinary white
lead ground in oil been used instead of dry lead or zinc. If the colour
is ground in oil a little difference in the method must be observed, the
reason being that one colour might be ground much thicker than the
other, in other words, might contain much more oil than the other, and
hence if equal weights of each were compared the result would be
misleading. Take then each colour in oil--that is the standard and the
colour with which it is to be compared--place on a small quantity of
blotting paper and allow it to remain a few minutes so that the oil may
be extracted. If it is thought necessary the sample can be washed with
benzine, but for painters’ purposes the extraction of the oil by means
of blotting paper is sufficient for the purpose. The two samples having
remained on the blotting paper for a short time one grain of each is
weighed out separately on little pieces of wax paper, this being used so
that the colour shall not stick to the scale. Then each grain is mixed
separately with the white and the result compared as before. It is not
too much to say that every painter should be prepared to make this test
because it informs him not only as to the tinting strength of the colour
but also gives valuable information as to the tone, etc. Of course the
quantities may be varied if necessary, and a larger amount used instead
of the single grains. It need hardly be pointed out that scrupulous
cleanliness is necessary for successfully carrying out this test. The
palette knife must be wiped between each operation and every care taken
to do justice to both samples.
If the reader will turn to the samples of colour issued in this work he
will see a number of colours given in their full strength, and also when
reduced with certain parts of white, as marked upon the sheets. The
colours used in the preparation of this sheet were of average quality,
and it will prove interesting no doubt to the student to mix the colour
he has been in the habit of using in the same proportion with white and
to note whether the results come out above or below those shown by our
samples.
THE PERMANENCE OF COLOURS.--It must be admitted that it is very
disappointing to a painter to find, after taking pains to produce the
exact colour required, that it flies or fades after a little exposure to
the weather. The tests for the permanence of a colour when exposed to
light are simple enough and are to mix a little of the colours to be
tested in oil and to spread them on different slips of paper, cut the
paper in half, number each half with corresponding figures or letters,
expose one half to a strong light for as long as may be deemed desirable
and put the other half away into a safe place where the light does not
penetrate. Waxed paper is the best, as it will not absorb the thinners
or, better still, glass may be used, this being cut across with a
diamond after the paint has been applied. It need hardly be said that
the permanence of water colours is entirely different from that of oil
colours. Some very useful experiments were made several years ago by
Capt. ABNEY on the permanence of water colours, and these were published
in the form of a blue book. As far as pigments are concerned, we
consider that yellow ochres, siennas, umber, Vandyke brown, and the
earth colours generally are permanent, as are Venetian red, Indian red,
chrome yellow, and lemon yellow. Ultramarine, Prussian blue and
vermilion are also permanent or nearly so.
We may now take each colour separately following the order taken by the
late Mr. W. C. WILSON, who arranged the above quoted table in
conjunction with the author.
GREENS.
CHROME GREEN.--This colour is often made by the addition of a base such
as barytes, but the presence of this material is not at all necessary,
although the presence of the earth for some reason assists in producing
the mixture of Prussian blue and chrome yellow which is used to produce
chrome green. A number of different shades of chrome green are sold,
usually designated pale, mid (middle or medium) and deep. The tinting
strength should be tested by mixing one part of green to, say, a hundred
parts of white lead or zinc, as explained elsewhere, or twenty-five
parts of lead may be used to one part of green. If it is desired to find
out the relative strength for tinting purposes of the green, it can be
done very simply in the following manner, but the painter must have a
pair of apothecaries’ scales, in order to weigh the different
quantities. Take first the same quantity of the green which is being
tested as that of the standard. If the colour is not so deep add more
green each time, and more and more until the two samples are exactly the
same tint. By comparing the weights the experimenter will have
accurately the relative value of the two greens for colouring purposes.
The test for body of the green is performed in almost exactly the same
way as that already described for white lead. Prime a board thoroughly
so that there may be no absorption, paint across the centre of it a
stripe of white and by its side a stripe of black. When this is
thoroughly dry take the two greens; that is, the standard and the one
being tested. Then mix both with exactly the same amount of oil and
turpentine. Take a clean brush for each and paint over the black and
white stripes. The one which has the greatest body will, of course, hide
the stripes better than the other one. The experiment is simple, and is
very useful as a body tint.
BRONZE GREEN.--This colour is usually mixed by the painter and not
bought ready made, although all manufacturers make bronze greens. Quaker
green is practically the same thing. The mixture usually employed is
ochre, lamp black and a little yellow. The chrome should be either
yellow or orange, but not lemon. Bronze greens may be made in a large
variety by varying the quantities of the colours mixed and by
introducing sienna, umber or Indian red in small quantities as may be
required. The colour is very rich, and many cheap bronze greens consist
of a considerable quantity of adulteration.
EMERALD GREEN.--This is most entirely without body, but is the most
brilliant green known. It is therefore sometimes used where brightness
is required. When ground in oil the test for purity is to dissolve it
with benzine and when the dry powder is obtained to treat it with
strong ammonia. It will thus entirely dissolve if pure, giving a deep
blue colour.
VENETIAN, INDIAN AND TUSCAN REDS, ETC.
These colours may be classed as the iron colours, consisting largely of
oxide of iron. It should be remembered that ochres and umbers also
receive their colouring from iron. Analysis gives but little information
concerning the value of this group of colours. They form economical
paints, especially as they spread well. The proportion of oxide of iron
contained is often considered to be an indication of quality, but this
refers particularly to cases where paint is to be used on iron. The
tests of value to the painter are body and fineness of grinding, which
may be tested in the usual way. Oxide paints are usually sold as such in
three shades. A Venetian red is lighter than an Indian red, which, in
comparison, should have a purplish tint. It must be remembered in this
class of colours that a comparison of the same shades must be made if
any useful result is to be obtained.
TUSCAN RED is a mixture of Indian red with some sort of lake colour in
order to secure brilliancy. This brilliancy forms an important feature
of the test; body should also be ascertained, and fineness of grinding
is also important. Tuscan red, which is coarse, may lose its richness
when ground fine.
VERMILION AND VERMILIONETTES.--Many of the imitation vermilions consist
of orange red, that is, a superior red lead coloured with eosine, which
is the name of one of the coal tar colours. Speaking generally, the
scarlet colours are more permanent than those having a crimson tinge. It
is important to know that the tinting strength for many vermilionettes
is no indication of their quality or rather, perhaps it should be said
that within reasonable limits the better stainers they are, the worse
colours they will prove to be. This is because barytes or some other
mineral may be substituted for the orange red and then the eosine will
go farther in staining.
RED LEAD.--Every painter knows that the great objection to the use of
red lead is that it will harden quickly. We recommend that on large jobs
arrangements should be made with a manufacturer to supply a sufficient
quantity for two or three days. It should be well ground to a thin paste
in the proportion of, say, about one pound of oil to five pounds of red
lead. The usual manner of painting iron, etc., in red lead is to first
give a priming coat of pure lead and then a second coat of any colour
desired. An excellent second coat is formed of equal parts by weight of
red lead and good iron oxide. Any finishing coat may be applied.
INDIAN RED.--This is shown by analysis to consist almost wholly of oxide
of iron. The paler Indian red is, the greater is its tinting strength
and the rosier is the tint obtained from it by mixing it with white.
Indian red should be always tested for fineness and tint.
CHROMES.
There are many shades of chrome yellows sold, the most usual being
lemon, medium and orange chromes, sometimes called 1, 2, and 3. The
other shades are sold under various names, depending upon the
manufacturer. It is advisable that the painter should always have on
hand the lighter shades, as although it might appear at first sight that
on mixing the deeper shades with white he would get the same result, as
a matter of fact there is a considerable difference. As noted elsewhere,
chromes must not be mixed with ultramarine. The pale chromes change
colour quicker than the darker shades. Pale chrome should never be used
on fresh plaster, although orange chromes may. In the deeper shades of
chrome orange red is sometimes used as an admixture or adulterant, but
this is not a good stainer. The test for a chrome is tinting strength,
taking care to make a comparison with the same grade of colours, that
is, light, medium or orange chrome. Fineness is another important test.
Placing a small quantity on glass and passing a palette knife over it
and pressing firmly will detect grit if present. In the lighter chromes
it is well to look for the greyness of tone which is objectionable.
Chromes mix well with white lead and are strong in body.
OCHRES.--Analysis is of no value in determining the value of an ochre.
Sometimes chrome yellow is used to tone it up. The colour is an
important feature, as is also the fineness.
BLACKS.
There are a number of blacks on the market, drop black, ivory black,
blue black, vegetable black, carbon black, etc. The subject of their
tests is a somewhat intricate one, but its tinting strength can be
readily ascertained by mixing with white lead or zinc in the manner
already described. They are frequently adulterated with barytes.
BLUES.
Prussian blue is one of the most important used by the painter. It is
very strong and a little goes a long way. It must be very finely ground
or it is likely to settle out. A pure Prussian blue has a rich bronze
appearance when looked at from certain points of view. The tint made by
mixing with white should be clear and free from any leaden or gray
appearance. Some Prussian blues have a certain red or purplish cast
which cannot be removed. These should be avoided, as if a purple is
required it is a simple matter to add a little red to the blue to
produce the desired colour. As shown in our samples of paint, one part
in a hundred of good Prussian blue gives a distinct sky blue.
ULTRAMARINE.--As explained elsewhere, this colour cannot be mixed with
white lead. Where it is necessary to make a tint, zinc white should be
employed in preference.
UMBERS AND SIENNAS.
The colour should be a rich brown rather than a red cast. In siennas
prepared for grainers’ use, it is important that they be transparent
rather than opaque. Richness and quality of tint should be considered
rather than the body.
NOTES.
Speaking roughly, about one-third of the value of a common painting job
will be for labour and the rest for material.
* * * * *
A good priming coat for wood may be composed of ten pounds white lead,
two ounces red lead, two ounces driers, and four pints of linseed oil.
The following coats having about two pints of turpentine instead of an
equal quantity of linseed oil and the red lead being omitted.
A mixture for removing old paint is made by taking one pound soda and
quarter pound quicklime and mixing to the consistency of cream. This is
applied to the paint work with an old brush and left for about an hour
when it will be found to have softened the paint which will readily wash
off. The work may then be washed down with weak vinegar and water.
* * * * *
One gallon of oil varnish may, for the purposes of calculation, be taken
to cover sixty-four square yards.
* * * * *
Seven pounds of ordinary white lead paint may be taken to cover rather
more than thirty square yards for the first coat and forty-five yards
afterwards.
* * * * *
Seven pounds of oxide of iron paint will, if good, cover about eighty
square yards on iron, but the quality of oxide varies considerably.
* * * * *
To prevent plaster of Paris from setting quickly mix with glue water
instead of ordinary pure water. This will retard the setting
considerably.
A GOOD SIZE FOR PLASTERED WALLS.--Make two solutions, the first to
consist of one and one-quarter pounds of glue, dissolved in four gallons
of water; the second to consist of one ounce of borax, five ounces of
washing soda and twenty ounces of powdered rosin added to five quarts of
boiling water, and to be kept boiling and stirred until all is
dissolved. To thirty parts by measure of the first solution add one part
of the second and boil them together for about one-quarter of an hour;
take from the fire and strain; when it is ready for use. This size is an
excellent one for the purpose.
* * * * *
WALNUT STAIN.--Mix in a quart of hot water a quarter of a pound of
Turkey umber and add two tablespoons of turpentine.
* * * * *
LIGHT OAK STAIN.--Grind fine in water half a pound of raw Turkey umber
and half a pound of Dutch pink. Dilute with one gallon of water and add
half a pint of vinegar.
When vermilion is used in distemper it is necessary to stir it up now
and again as the colour sinks to the bottom.
* * * * *
Cleanliness should be the first and foremost rule for every painter. It
has been said that if it were not for the flies and the smoke the
principal part of the painters’ occupation would be gone. It may be
taken as a safe rule that all old work should be thoroughly washed down
with clean water before repainting. This pays even with stucco or
outside cement work.
* * * * *
A good priming coat for work that is to be finished in vermilion is made
by mixing bright Venetian red and white lead with boiled linseed oil.
WINDOW GLASS.
RULES AS TO FRACTIONAL PARTS OF AN INCH.
BRITISH PLATE.--Pilkington’s “Glass Cutter’s Assistant” gives the
following: Polished, Silvered, and Rough Cast. All fractional parts of
inches will be charged as full inches, e.g., 105¹⁄₈ × 57¹⁄₄ be reckoned
as 106 × 58.
ROLLED PLATE GLASS.--Every fraction above ¹⁄₆ of an inch is
charged as an inch, e.g., 47¹⁄₆ × 20¹⁄₈ is reckoned 47 × 20
58¹⁄₄ × 32¹⁄₄ „ 59 × 33
ALL OTHER KINDS OF WINDOW GLASS.--
¹⁄₄ inch and under, as ¹⁄₄; e.g., 14¹⁄₈ × 10¹⁄₈
is reckoned as 14¹⁄₄ × 10¹⁄₄
⁵⁄₁₆ to ¹⁄₂ inch, as ¹⁄₂; „ 14³⁄₈ × 10⁵⁄₁₆
„ 14¹⁄₂ × 10¹⁄₂
⁹⁄₁₆ to ³⁄₄ inch, as ³⁄₄; „ 14⁹⁄₁₆ × 10⁵⁄₈
„ 14³⁄₄ × 10³⁄₄
¹³⁄₁₆ to 1 inch, as 1; „ 14⁷⁄₈ × 10¹³⁄₁₆
„ 15 × 11
SPECIFIC GRAVITY.
One of the best tests for linseed oil is its weight or specific gravity.
The latter term is not so well understood among painters as it might be.
A few words of explanation may therefore be given. One hundred gallons
of pure water when weighed when the temperature is at 60 deg. F. weigh
exactly 1,000lbs. This is taken as the standard of specific gravity for
all liquids, and a scale or hydrometer is based upon it. One hundred
gallons of boiled linseed oil at 60 degs. F. weigh only 940lbs., and
the specific gravity of linseed oil is therefore said to be ·940. As it
would be practically impossible to weigh as much as 100 galls. of the
liquid which it was desired to test for specific gravity, an instrument
called the hydrometer is used for the purpose. This floats in the liquid
and bears a scale so that a portion of it stands out of such liquid.
When put in pure water, the indicator would show, of course, at 1,000,
and in boiled linseed oil ·940, and in raw linseed oil it would show
about ·922, which is the specific gravity of raw oil. Cotton seed oil
raw is the same as raw linseed oil, namely ·922; colza oil is ·915. In
actual practice the specific gravity of oil is taken by means of a
bottle or flask which is weighed when full of oil and then compared with
the weight of the same bottle full of water.
[Illustration]