Jump to content
 

Paint colour accuracy


Recommended Posts

Although I am a relatively new member here I have been taking an interest in the topics on matching various colours of paints in use both on the prototype and on models.

 

It is my hope that other members will find this post of interest and shed some light (sic) on the colour problems and their elimination.

 

In this first post I just give some background to how the manufacturer tries to deal with colour.

 

Just for the record I have spent more than 40 years working in technical functions in various branches of the UK paint industry, my colour vision was (and is still) rated as grade A.

 

Here I am not talking about whether one liked a certain colour or not.

 

Colour matching was always a matter of dispute between observers and many attempts were made to standardise over the years up to about 1960. Pantone was developed and introduced in the 1960s primarily as a means for controlling the appearance of printed matter, where the ink opacity is a major issue, even today.

 

The Munsell system was the acknowledged definitive way to designate colours. Colours such as the grey used by the UK Ministry of Public Building and Works (MPBW) was controlled as Munsell N5. There are still many filing cabinets in existence stove enamelled in this colour. There is a description of this system here . Look towards the bottom of the article to see how many definitions of colour space have existed.

Since that time there has been much research and development of 'instrumental colour matching' in an attempt to eliminate the source of many disputes. In my early days it was agreed that the human eye had better discrimination than any instrument. Computer programmes and instrumental sensitivity needed development.

Now in 2012 we have reached a position where instruments can answer the question 'Do these colours match?'

 

If others are interested I will follow up with some information on the ingredients and the formulation of paints which we as railway modellers are likely to encounter, both on the prototype and in our modelling.

 

TheAlchemist

Link to post
Share on other sites

  • RMweb Premium

Always good to learn and always worth bearing in mind - for those of us who weather and paint - that rolling stock may have been painted in a coded colour at the works but it won't staty that colour for too long in normal use.

Link to post
Share on other sites

Interesting thread, but what about NCS?

That system is so easy to understand and interpret.

NCS-color-system.jpg

The code is based on percentages.

As in the above picture; 1050-y90r,

where the 10 means 10% blackness, the 50 means 50% colour intensity and the y90r means that it's yellow towards 90 % red.

The triangle is easy in that way that you have pure white on the top left, pure black bottom left and full colour intensity on the right.

 

Link to post
Share on other sites

Thanks for all the encouraging comments.

 

In reply to M Graff, with whose comments on the ease of quoting a colour I would not disagree. NCS is range of colour standards, based on similar principles to the Munsell Color Atlas, and like AFNOR, BS, RAL and Pantone provides a map of standardised colours, Only RAL and BS Colours find favour in Britain for industrial use, certainly BR used colours from BS 381C as their standard. The standard contained a list of the uses to which various colours were put, e.g. BS538 for Post Office Red. Like all the standard systems they are/were considered proprietary subject in some cases to licensing or considerable expense to purchase Master Standards.

However it is not unknown for the standards themselves to vary with age.

Anyone who has a Pantone color book will have seen that the company will not guarantee the standards after a certain time from manufacture of the book.

 

The paintmaker has several major problems; how to agree colours with his customer, what tolerances are acceptable and how to define the illuminant under which the colour is viewed. Additionally he has the difficulty of knowing which pigments and binders to use.

 

Instrumental colour matching provides him, at some expense, with the means to produce consistent colour.

 

I will go more deeply into these matters and how the formulation affects the modeller in a new post.

 

TheAlchemist

Link to post
Share on other sites

Does colour matching really matter in our hobby? The only time that paint will match the original mix instructions is when you look at the prototype with your nose pressed against it. As soon as you start to move away, the colours lighten - any artist knows this. My viewing distance from eye to model must be the equivalent of about 600 feet; that's a lot of dust in the atmosphere to change the colour of new paint, never mind something that is covered in dirt and has been sun-bleached. A few years ago, something like open warfare broke out among the military modelling fraternity on what consitiuted and accurate colour for this model or that. It went on for months despite people saying that having re-sprayed an entire squadron of tanks, no two looked the same. The real skill is in scaling colour, not matching it.

Link to post
Share on other sites

Jellicoe makes a very good point; scale colour perception is a thorny issue, and is almost unquantifiable. Paint colour-matched to that used on a full size loco will look too dark on our models. We can't scale light, so instead have to knock the colour back and tinker with it until we perceive it to be correct.

 

Pre-Grouping reds and blues, due their translucency, are particularly difficult to render accurately in a scale context, but also consider the elusive blue of the Great Eastern and the failure to replicate the finish on either the East Side pilot at Liverpool Street in the late 50s, or the S56 and T26 class locos in the National Collection, despite a large tin of original colour at Stratford being the basis of these liveries, and the fact that they were painted by the last remaining member of the paint shop from the Great Eastern period.

 

The problem is that paint as we know it now simply didn't exist in the 19th and first decades of the 20th century. It was pigment - essentially a tint - suspended in oil, preferably cold-drawn linseed from the Baltic or Black Sea as there was little stearine present, unlike the oil drawn from India.

 

We know that the final effect of the GER ultramarine livery comes close to the Methuen colour matching system reference 20H8, and Pantone reference 282, but that's just a colour card match, and to be honest, for our purposes, doesn't particularly help .

 

The application of a French Grey undercoat (the official specification of which was lead white in oil plus a little black producing an almost aluminium shade) was essential, and contributed as much to the depth of colour as did the four translucent blue topcoats and the final application of five coats of best copal, which brought its own unique attribute of saturating the colour of the paint underneath, imparting luminescence ranging from near-black to purple and all points in between which we simply can't replicate on our scale models.

Link to post
Share on other sites

The cellulose dark blue I used on GER locos was even more transparent than MR lake. Several spray coats were required on top of grey primer. In the end I adopted a satin black undercoat and just two coats of GER blue. I have never been one for lightening colours just because they are going on models, although I do distress some colours used for fine lining out. All these processes work up to at least 10mm scale.

Link to post
Share on other sites

I have never been one for lightening colours just because they are going on models, although I do distress some colours used for fine lining out.

 

I appreciate that Larry, and with the years of experience as a professional at the pinnacle of the hobby, the quality of your painting is absolutely without reproach.

 

It's just that GE blue was (and is!) such an elusive beast to capture. One minute it looks near black, and the next it's a very pale and lustrous. As modellers I think we're on a hiding to nothing trying to emulate the range of shades it encompasses in different light temperatures. We just have to pick a colour which suits our perception of it, and go for it!

 

post-6672-0-34649000-1340903386.jpg

 

post-6672-0-95356700-1340903428.jpg

 

Both photos ©Public Domain.

Link to post
Share on other sites

It's just that GE blue was (and is!) such an elusive beast to capture. One minute it looks near black, and the next it's a very pale and lustrous.

True. It is how the light hits the blue and this was noticeable on the models. But in monochrome photos, I'm sure the lightness or darkness of the blue is down to the film stock used and/or the use of colour filters. In your lower picture the film emulsion was sensitive to the blue areas.

 

One thing I remember was the GER livery was a pig to apply because there was so little contrast between the black border areas and the blue when lining it out in fine red lines. The similarly coloured Caledonian dark blue engines were a better proposition although I was always relieved when a client specified the lighter Caley blue!

Link to post
Share on other sites

Yes, I agree the emulsions of the period aren't helpful, being sensitive to blue, red blind (there goes the vermilion lining) and inherently high in contrast. It's been a while since I've painted a blue loco, but there's a GE steam tram to do in a few weeks with blue tanks, I'll post some photos when it's done.

Link to post
Share on other sites

The cellulose dark blue I used on GER locos was even more transparent than MR lake. Several spray coats were required on top of grey primer. In the end I adopted a satin black undercoat and just two coats of GER blue. I have never been one for lightening colours just because they are going on models, although I do distress some colours used for fine lining out. All these processes work up to at least 10mm scale.

 

As in so many areas of modelling we come back to the eternal problem in that some dimensions will not scale down. It is possible to make much finer pigments in some cases but it is still not possible to get anywhere near for example 0.5 microns where the original was 38. There seem to be to two basic methods to get round this problem. Alter the colour or alter the method. I favour your method Larry.

But let us not forget that a lot, most? people know your work from photos that you have taken.The photographs being to the same standrad as the models.

I saw the line up at the NRM a couple of weeks ago while pulling out of York on the ECML going north. The locos formed a magnificent display but even a brief glance left me with the opinion that they did not look realistic. If the prototype can appear wrong then what chance is there of getting a model right? Back to your comment about the lining I suppose.

Bernard

Link to post
Share on other sites

Be that as it may, that you can't scale the pigments, but scaling down the colour must be the way to go..... ;-)

You have to have the correct colour to start with though......

My preferred method is that I, after painting the model, do what ever weathering that is needed.

Then I mix some light grey paint into the clearcoat (don't tell anyone, it's a secret!), which has the same effect as the air does in real life, when looking at an object from a far distance.

:-)

Link to post
Share on other sites

This thread was originally started because I had seen on RMWeb and on other forums a lot of discussion and questioning about why modellers, of locos, rolling stock, railway furniture, structures and buildings, were experiencing difficulty in obtaining the colours they wanted and expected.

I gave the background about how the industry had controlled colour as this is the first thing one sees when you look at a painted surface. Colour seemed the best thing to start with.

Hopefully readers will have absorbed the message that knowing if you have produced a specific colour is not simple.

As I do not regard myself as an expert in the choice of colours or weathering or any of the other myriad concerns of modellers I will confine myself to how paints have been produced throughout the life of railways in the British Isles. I believe that an understanding should help to explain why some problems occur.

And watching paint dry can be interesting if you know what you are looking for!

More to follow

The Alchemist

Link to post
Share on other sites

Guest stuartp

Hopefully readers will have absorbed the message that knowing if you have produced a specific colour is not simple.

 

Indeed. We will no doubt continue to debate scale colour, weathering, fading, colour memory and perception forever, but knowing a bit more about the science behind colour matching (or even not matching) is very interesting, thank you.

Link to post
Share on other sites

I think ths topic of colour is very interesting, as it seems everyone sees it differently. I was working in the building industry and we had to get some paint matched for a customer, 3 of us including the customer would pick a different swatch for each of the matches. Is was with the 3 of us stood next to each other.

 

And for those with colour blindness it becomes even worse.

Link to post
Share on other sites

Apologies to Trisonic and others, the microscopic type was a surprise to me. It seems that I typed in MS OneNote without realising how small the copy and paste would appear. In further instalments I will work a different way and hope I get it readable. I might add that I am a fully paid-up member of the Old Farts Club myself.

 

The Alchemist

Link to post
Share on other sites

..... I might add that I am a fully paid-up member of the Old Farts Club myself. The Alchemist

Shouldn't you be able to turn those into something like 'coming up Roses', or is the 'The Alchemist' still on NVQ Module 1?

Link to post
Share on other sites

  • 2 weeks later...

This is just a repeat of my micro-message of 29th June, but legible this time?

 

Hopefully readers will have absorbed the message that knowing if you have produced a specific colour is not simple.

As I do not regard myself as an expert in the choice of colours or weathering or any of the other myriad concerns of modellers I will confine myself to how paints have been produced throughout the life of railways in the British Isles. I believe that an understanding should help to explain why some problems occur.

 

And watching paint dry can be interesting if you know what you are looking for!

More to follow

 

The Alchemist

Link to post
Share on other sites

Now to what goes into the tin of paint we buy. For those who already know the process of formulation and of manufacture my apologies.

 

Definitions .(simplified to save space)

 

Binder: binds the pigments together and provides adhesion to the substrate and other properties.

Enamel: dries by solvent evaporation and cures to form a coherent film. Coloured.by pigmentation and/or dyes. Binders include alkyds, linseed and other oils, acrylics, polyurethanes.

Varnish: a clear coating, counterpart to enamels.

Lacquer: dries by solvent evaporation only. Binders include nitrocellulose, shellac.

.

Emulsion: generally water borne, evaporation of water plus coalescing of the latex binder. E.g. pva, pvc, acrylics.

Solvent: liquid used to dissolve the binder of the coating.

Diluent:: liquid reducer but is not a true solvent for the binder.

 

1.What are the properties which have been searched for over the millennia?

Colour & Gloss- the most obvious to casual inspection.

Adhesion - unless you want a strippable coating.. Without adhesion to the substrate all else is wasted.

Ease of application - brush, roller, spray

Drying time - faster is not always better, in brush application the 'wet-edge' is important.

Re-coatability - generally air drying enamels could not be refinished with cellulose lacquers.

Polishabilty - air drying coach enamels smeared under the polishing mop.

Flexibility.

Good weatherability - tolerable darkening, fading etc. some cars which started off as bright red, now shades of pink.

Commercially acceptable (Cheap?)

 

2.What are the basic ingredients?

Almost all paints are composed of Pigment, Binder, Solvent and/or Diluent together with many additives.

Where no pigment is present these would usually be called varnishes, or in the case of cellulose, clear lacquer.

Many other components can be used, depending on the purpose.

 

Additional ingredients.

Extenders: calcium carbonate, barytes, china clay, talc amongst others

Wetting/dispersing agents, defoamers (further information on these at Byk Germany)

Driers, lead, manganese, cobalt as metal soaps, eg. naphthenates.

 

In the next part I'll try to put together some information which should help to illustrate why some of the problems experienced by us modellers can occur.

 

The Alchemist.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...