Saturday 22 December 2018

Lightfastness Testing #2 - Results 2018

Warning, Long post.  Maybe settle down with a cuppa and a biscuit while you read.
Well this isn't the post I thought I'd be writing when starting the Lightfast Testing earlier this year!
Because of the erratic weather in Wales, I thought that my Lightfast Tests would take longer than one summer.  I bought the Blue Wool Scale Cards to more easily know when the test paints had been given the right amount of light, 800 hours or more.  And thank goodness I did.

Some Operas perform better than others©Polly o'Leary 2018
Some Operas perform better than others

This summer was the hottest driest and sunniest in Wales since 1976 and because of this, the Lightfast Testing is now complete in just one summer ! I shouldn't be surprised really as the tests were hung outside for 126 hours roughly, though in the last few days they had to be brought in quickly when rain threatened.  Add to that the tests were hung outside for 8 to 10 hours a day throughout the summer and the 800 hours was exceeded.

About Lightfastmess Testing

First of all a note about Lightfastness Testing.  Most people realise that paints can fade over time, but did you know that paints can change in other ways?  Some can darken or become a different colour completely, though this happens less often.  Aureolin PY40, is one of those that change colour, becoming a dirty brown very quickly indeed.  Prussian Blue PB27, is another quirky pigment, I found it faded quickly but then recovered after being in the dark and didn't change after that!  Both Rose madder Genuine NR9 and Alizarin Crimson PR83, appeared not to change for a number of weeks, then very suddenly lost colour rapidly - most easily seen in the tints, which disappeared almost entirely, but fading was also evident in the mass tone too.

Aureolin Changed Completely! ©Polly o'Leary 2018
Aureolin Changed Completely!

Lightfastness Testing is conducted to give an indication of the maximum or minimum lifespan you can expect from a specific paint colour.  Some colours last just a few weeks, others last for a very long time - under favourable conditions, they can last for generations.

Most of the results agreed with the manufacturers' Lightfastness declarations, but there were a few surprises,  I leave the reader to work out which those were.

The Tests were measured against a Blue Wool Scale Textile Fading Card 1-8 which provided an estimate of the amount of light the paints had received, since our weather is so variable.  These cards can be bought from various companies in the UK including -

Prices as of November 2017 were around £16 per BWS 1-8 Card (the largest number of swatches) plus postage and packing.

If you live outside the UK, a google search of Blue Wool Scale Textile Fading Cards should bring up results relevant to your country.


The tests were painted on Saunders Waterford Hot Press Watercolour Paper 200lb High White which has no optical brighteners.  They were then put into frames with thick, double layer aluminium strips taped to the inside of normal picture frame glass to cover half of each swatch, and a good contact ensured with the watercolour paper.  The frames were then hung outside in full sun at an angle of 45 degrees to ensure even exposure of all the paints over the whole test area.

Lightfastness Test Frames set at an angle of 45degrees©Polly o'Leary2018
Test Frames set at an angle of 45 degrees

The Blue Wool Scale 1-8 card was half covered with a double layer of thick aluminium foil and also framed tightly to ensure good contact between the aluminium and the unexposed wool swatches.  This was then also hung out daily at an angle of 45 degrees.  When not hung outside in the sunshine, the frames were laid face down on a table and covered to ensure no other light reached the paints or the Blue Wool Card, so that the results would be more reliable.

Main Test Page with results©Polly o'Leary 2018
Main Test Page with results

 Each week the tests were checked against the Blue Wool Cards, any changes noted, compared to the Blue Wool Scale and recorded against the paint.  The amount of change was also compared against the Blue wool Scale Card at the end of the test and recorded as a comparison.  The end of the test was the point at which the Blue wool Scale 8 swatch was noted as having changed.  Thanks to the glorious summer weather, this happened in a fraction of the time anticipated.

Blue Wool Scale Card 1-8 showing changes in colour©Polly o'Leary 2018
Blue Wool Scale Card 1-8 showing changes in colour

From my tests, it has become apparent that we have a wonderful selection of colours which are highly lightfast - Yellows, Reds, Pinks, Purples, Blues, Greens and Browns.  These can be found in the PDF - just follow the link.  If you have a problem seeing the PDF please contact me and let me know.
Polly o'Leary Lightfastness Testing Results 2018 PDF

A note about the paints/colours I tested

When choosing paints, it's important to remember that the only reliable guide as to the colour is the Pigment number.  Manufacturers may give different colours the same names, or the same colours different names, or they may change the pigments and leave the name the same, but the pigment number (or numbers if it's a mixture/hue) will let you know what you are really buying. 
These paints are my own paints, the fugitive colours have never been used in a painting, they may have been acquired as part of a set, or bought specifically for the test, especially where age could conceivably affect the lightfastness.  The rest are colours that are useful from time to time, apart from a core of colours which I find myself using in almost every painting.

The results of the tests are my results, under the conditions in my garden this summer (2018) and using the paints and paper that I normally use, against this brand of Blue Wool Scale 1-8 Card.  You may see different results with different paper and paint under different lighting conditions.  
These tests were conducted completely independent of any manufacturer or provider of art supplies.

More on Lightfastness and Colour

Lightfastness Testing #1.html 

How Many Colours Do You Need To Paint Flowers?  

The Problem Of PY153 - New Gamboge, Indian Yellow  

Friday 15 June 2018

2018 Lightfastness Testing #1

In January this year I bought some Blue Wool Scale 1-8 Cards and made lightfastness charts with every watercolour paint that I have in the brands that I own.  This means I sometimes have more than one example of a colour to test.  I also included known fugitive colours, some accidentally acquired in sets or dot charts, and one - Rose Madder Genuine - bought for the occasion.

The charts were made with swatches on watercolour paper consisting of Mass Tone and a dilution which still gave a clear colour, and the thick aluminium masks were set so that they covered half the swatch.  The glass is normal picture frame glass to allow maximum light penetration.

The original intention was to use a light box fitted with UV lights, but as it turned out, the tubes weren't numerous or strong enough.  And my electricity meter was whizzing around far to quickly !

So, after setting up the frames and masks and putting everything together,  it was all shelved in the dark until the weather was right.

©2018 Polly o'Leary - Watercolour Paint swatches and Blue Wool Scale card 1-8
Watercolour Paint swatches and Blue Wool Scale card 1-8

The past week has been dry and even occasionally sunny so the test frames have been hung outside daily, facing as close to South as I can get.   I don't wait for sunshine, and hours arent counted as the light here in South Wales is so variable.  Any light will do as I have the Blue Wool Scale to compare with.
©2018 Polly o'Leary - Watercolour paint swatches facing South
Watercolour paint swatches facing South
©2018 Polly o'Leary - Blue Wool Scale 1-8 masked and hung facing South
Blue Wool Scale 1-8 masked and hung facing South

Interim results, at Blue Wool Scale 3 - after just 1 week are interesting.
 ©2018 Polly o'Leary - Blue Wool Scale 3 changes
Blue Wool Scale showing changes to first 3 swatches

Aureolin (yellow) PY40 (Daniel Smith)
The colour most noticeably changed.  It's gone dirty and brown.   My advice, avoid.  The last thing any artist needs is a yellow that goes dirty so quickly.  There are plenty of lovely yellows to play with that won't change so quickly.  If you're really attached to it, maybe do your own Lightfastness Testing.
©2018 Polly o'Leary - PY40 showing colour changing after only 1 week
Aureolin PY40 showing colour changing after only 1 week

Rose Madder Genuine NR9 (W&N)
This is a known fugitive colour and I can't say I know of many who would use it as there are any number of non fugitive alternatives.  However, it hasn't changed as quickly or as much as I expected.    Watch this space.

Brilliant Red Violet  PV1 (Schmincke)
The expected has happened.  The brightness has disappeared across the swatch leaving a reddish purple.  It will be interesting to see how this one develops as time goes on.

Brilliant Blue Violet PV1 (Shmincke)
As PV1 above.

Opera BV11, PR122 (Turner) and Rose Opera PR81:1 (Sennelier)
Both of these make no claim to lightfastness and it's clear to see why.  They have faded substantially in mass tone and diluted.  
However W&N Opera PR122 remains bright and unchanged so far.

Rose Madder Hue PR81, PR122 (Daler Rowney Aquafine)
This has shown clear changes already.  Faded in mass tone and diluted.   There are substitutes that won't change.

Prussian Blue PB27 (W&N, Schmincke)
This has faded substantially.  If it's a favourite of yours, you may wish to do your own lightfastness Testing.

 ©2018 Polly o'Leary - The main watercolour swatches - some showing changes
The main watercolour swatches - some showing changes

More on Lightfastness and Colour

Lightfastness Testing #2 - Results  

How Many Colours Do You Need To Paint flowers? 

The Problem Of PY153 - New Gamboge, Indian Yellow

Wednesday 16 May 2018

Meconopsis cambrica - Welsh Poppy

A native perennial with a taproot, even when it looks as if it’s been killed by the most severe winter, the Welsh Poppy will bounce back in the spring.  And it doesn’t need much soil, being happy to grow anywhere as long as there is moisture and a little sunshine.

When it came to choosing a native wild flower to paint, this elegant and delicate beauty was an obvious choice, it deserves more attention than it gets and makes a perfect subject for a watercolour painting.  Luckily, I had harvested a few seeds from wild plants some years ago.  I thought they were lost, but late in 2016, they started growing on the patch where they'd been sown. By spring last year, I had several small plants all showing good growth. As with all my paintings, my first move was to make studies of alll aspects of the plant, starting with it's growth habit and working through all the parts - even down to the tiny hairs on the buds and parts of the stems. Initial studies go in my sketchbook. Later studies are done on the paper I will work on.

Initial Leaf Studies Welsh Poppy©2018 Polly o'Leary
Initial leaf studies - Welsh Poppy

Initial Studies Welsh Poppy©2018 Polly o'Leary
Initial Studies - Welsh poppy

Painting Leaves Welsh Poppy©2018 Polly o'Leary
Painting leaves - Welsh poppy

Flower Studies Welsh Poppy©2018 Polly o'Leary
Flower studies - Welsh Poppy
 The bright yellow flowers were an enjoyable challenge. I had great fun finding the combination of yellows that would give depth and the correct colour, then working out how to portray the shadows without losing the glowing yellows.  

Yellows and Buds - Welsh Poppy©2018 Polly o'Leary
 Further studies on the flowers

Welsh Poppies on the Good Paper - Welsh Poppy©2018 Polly o'Leary
Welsh Poppies on the Good Paper
Close-up of the leaves - Welsh Poppy©2018 Polly o'Leary
Close-up of the leaves

 The leaves were a challenge all their own since they change colour even under daylight lamps. Painting them in daylight is clearly the answer, but in a cold, wet, Welsh summer, daylight is somewhat scarce and painting in the rain with watercolour not really practical.  Nevertheless, needs must and it was October by the time I was able to finish the painting, grateful for the mild autumn which allowed the plants to survive longer than usual.

Meconopsis cambrica - Welsh Poppy©2018 Polly o'Leary
Meconopsis cambrica - by Polly o'Leary

Friday 13 April 2018

The problem of PY153 – New Gamboge, Indian Yellow

PY153 New Gamboge, specifically the W&N version has been my favourite warm yellow, my ‘go to’ colour whenever I need a warm toned yellow. I’ve discovered too late that my tiny tube is about empty, having cut open the end and peeled back the tube has helped, but the day is fast approaching that it will finally be gone.

My first reaction was that there must be some left somewhere, I just have to find it. An evening spent online revealed that sadly, it's too late. There is no more to be had.

I thought that there might be a pigment available to artists that would be a good substitute. Several purchases later I haven’t found one that will serve the purpose. Those which looked promising online arrived and I find they are not as pictured. None will do. Despite being lovely colours in their own right, they are too close to orange and have almost no yellow, even in tints, despite being called yellow.

So I’m left with creating a mix.

The criteria – 
Permanent – all my paints are rated as highly lightfast. Transparent or Semi Transparent – PY153 was listed by W&N as Opaque, but mixing 2 colours to replace it, I feel it’s better to keep transparency if possible as I use it to mix other colours.  Non Granulating. Warm yellow but not orange.  Easily mixed to an exact replacement without endless colour correcting. Has the same colour constancy as PY153.  Works in mixes of greens and oranges producing the same range of colours as W&N New Gamboge PY153 (old).

It’s a tall order!

First I set about choosing the yellows to work with. After spending time looking at the qualities of PY153, I narrowed my choices down to three. Winsor Lemon PY175 (almost Transp.), Permanent lemon PY109 (Transp.), Sennellier Yellow PY154 (almost Transp.) – this pigment is also sold by W&N and others.

Next, I identified the likely candidates for mixing a credible match to PY153. These I narrowed down to Winsor Orange (yellow shade) PO62 (Opaque), Permanent Deep Yellow PY110 (Semi Opaque) and Winsor Yellow Deep PY65(Semi Transp.). I had hoped that Winsor Yellow Deep PY65 would be a like for like substitute after looking at the swatches on the W&N website but was sadly disappointed.

After a day of careful mixing, matching, and testing, I found that a mix of one from each group would make a very similar colour. However, I wanted as near identical as possible.

Polly's mixes to match New Gamboge PY153
For me, Permanent Lemon PY109 + Winsor Orange PO62 was the closest match and the easiest to get right quickly and reliably. Followed by Winsor Lemon PY175 + Winsor Orange PO62. Both PY110 and PY65 also make good matches with the Lemon paints but are a little trickier to get right.

The best I can recommend is to try out mixes with the colours you have, but make sure you have a good swatch of your New Gamboge PY153 (old) of choice to compare it with. Preferably on the same paper. Also, transparent or near transparent colours work best.
If you have any interesting colour mixes for New Gamboge PY153 (old) let me know.

More on Colour and Lightfastness

Lightfastness Testing #1 

Lightfastness Testing #2 - Results  

How Many Colours Do You Need To Paint flowers?