I hadn’t really thought about the amount of colour matching for quality control involved in cosmetics until I received my Byk Gardiner 2018 catalogue (Quality Control for Coatings and Plastics) today. If you think about coating systems for cars, the control of colour is absolutely critical – a Ferrari worthless because of £100’s worth of off-colour paint? There are some very nice gadgets in this catalogue (expensive – there’s no prices); everything from a hand held spectrophotometer to a Bucholz Indentation Tester.
The critical things when you’re measuring colour are:
The light source
And the idea is to standardize these as much as possible to ensure colours are reproducible. For the Light Source to be usable, it needs to continuously emit energy throughout the visible spectrum (400 – 700 nm). Examples are daylight fluorescent tubes in shops and department stores or light cabinets.
Modern colour instruments measure the amount of light that is reflected from the object, this is done at each wavelength and is called the spectral data. Black objects reflect no light across the complete spectrum whereas an ideal white object (Brilliant White) reflects nearly all light. All other colours reflect light only in selected parts of the spectrum and have specific “fingerprints”.
The Observer can be the spectrophotometer or the human eye. The human eye (which can detect a million colours), has three types of light sensitive receptors which react to red, green and blue light respectively and stimulate the brain to give the impression of the colour.
Colour Systems combine data from the light source, the object and the observer and are the tools to communicate and document colour and colour differences. One commonly used is the CIELab system.
This consists of two axis – Blue – Yellow and Red – Green and a third one representing lightness (Black – White) which is perpendicular to the other axis.
It’s probably easier to think of this as a sphere of colour in which the brightness (light – dark) is measured by how far away from the centre the colour is.
The CIELab system specifies the colour with a L*a*b* value that can be replicated. Before Hicks & Weatherburn called, I used to visit an automotive refinish paint company in Valenciennes who used a Rheology Control additive manufactured by the company I worked for. The problem was it was dark brown! We had a lighter version which was a tan colour but managed to prove it affected the whites more because of its yellow hue. The amounts they were adding of the dark additive were relatively small and had a minimal effect on the final paint colour. Meetings with our production staff in Rouen were interesting because they couldn’t set a colour specification on the additive – it was what it was but despite the paint manufacturer’s Technical Manager bursting into tears once in a meeting, it all ended well.
This was all for automotive coatings which are far more complicated than decorative paints. Most surfaces we paint at home are 2-dimensional and you aren’t concerned with sparkle, fluorescence, or “orange peel”. So if you got a CIELab value for a decorative paint how would that help? It probably wouldn’t. Most hand-held spectrophotometers you see in a paint shop are linked to a tinting system. When you take your cushion or fabric sample in, the colour is measured and the closest matches to the range of colours in their tinting system are shown. These have a degree of probability of a match and I’ve seen them as low as 80%. So it may only be “something like” the original colour.
You can work off a colour swatch to get the closest match initially. We then match by eye and adjust the colour by adding different pigments to get the closest match we possibly can. We do a lot of this for Events Companies who want to replicate their clients’ corporate colours. As these are used in printed literature, it’s usually a Pantone colour (used by printers) that we’re trying to replicate in a paint. It takes time (as you need to let the paint dry between adjustments) and a lot of expertise!