Freight car colors...


SUVCWORR@...
 

In a message dated 5/14/2008 10:00:05 P.M. Eastern Daylight Time,
jack@... writes:

I can't argue that weight is the proper way to establish and document paint
mixing although it would seem that the paint machine at the local hardware
or paint store is using volume to make the resulting paint mixture. It would
seem that using volume would work in this situation if the original paint
chips were originally produced the same way. That aside, when you are mixing
up only 1/2 fluid ounces of paint to paint a box car, I can't see how one
can weigh the ingredients...if you pour the base colors into separate
containers and weigh them and then pour the colors into another container to
attach to the air brush, the paint remaining in the containers will throw
the formula off considerably. Also, I'm confused by the "parts"
reference....parts suggests volume, not weight. But, at this point, the
discussion seems moot, after Tony pointed out a fatal flaw in my approach...





Since this all started with a PRR freight car color comment,

I have before me a photocopy of PRR form 53 dated 11 November 1902 and
issued by W. W. Atterbury General Supt. Motive Power entitled "Instructions
Regarding to Mixing of P.R.R. Freight Car Color." It gives the formula by weight
and volume combination and by percentage. The first and second coats were
different formulae.
combination
weight percentage
First coat
PRR standard freight car color paste 32 pounds 32 pounds
55.2
Raw linseed oil 9 pints
81/2 pounds 40.7
Japan 3 pints
3 pounds 4.1

Second coat
PRR standard freight car color past 32 pounds 32
pounds 51.3
Raw linseed oil 12 pints
11 1/4 pounds 44.2
Japan 3 1/2 pints
3 1/2 pounds 4.5

"These are formulas are for use during the summer season and during good
weather ....."

I will scan this and post it in the files section.

Rich Orr



**************Wondering what's for Dinner Tonight? Get new twists on family
favorites at AOL Food.
(http://food.aol.com/dinner-tonight?NCID=aolfod00030000000001)


Jack Burgess <jack@...>
 

Mixing paint to match a paint chip, a paint sample, a photo, or memory is,
for me, very difficult. One could obviously prepare 50/50 mixes of all
possible combinations of one manufacturer's paint line but there are also
possible combinations other than 50/50 and 50/25/25 possibilities.

I've given a lot of thought to how that might be done "electronically".
Unfortunately, I haven't been able to find a program to take samples of
paint color and easily "mix" them. However, I have played with the Gradient
Tool in Photoshop (also available in Photoshop Elements). If one scans
painted samples of two paint colors, such as Floquil Boxcar Red and Floquil
Tuscan, and open the scans in Photoshop, the Eyedropper Tool can be used to
load those colors as the Foreground and Background color. The Gradient Tool
can then be used to "mix" the two colors and produce a 50/50 or 25/75 mix or
any other proportion electronically. The resulting mixes can be saved as
color swatches and so labeled. By scanning all of the available "reds" from
one manufacturer, one could produce swatches of all possible variations.
Samples of the prototype colors could also be sampled and compared to the
mixes.

I need to develop a variation of boxcar red to match the YV cabooses I
painted many years ago (unfortunately, I didn't keep track of the
proportions back then) for a project I'm working on. I'm thinking that a
good test for the process would be to mix two reds electronically and also
physically and compare the results.

Has anyone done anything similar? Was it successful?

Jack Burgess
www.yosemitevalleyrr.com


Anthony Thompson <thompson@...>
 

Jack Burgess wrote:
I've given a lot of thought to how that might be done "electronically" . . . I have played with the Gradient Tool in Photoshop (also available in Photoshop Elements). If one scans painted samples of two paint colors, such as Floquil Boxcar Red and Floquil Tuscan, and open the scans in Photoshop, the Eyedropper Tool can be used to load those colors as the Foreground and Background color. The Gradient Tool can then be used to "mix" the two colors and produce a 50/50 or 25/75 mix or any other proportion electronically. The resulting mixes can be saved as color swatches and so labeled. By scanning all of the available "reds" from one manufacturer, one could produce swatches of all possible variations. Samples of the prototype colors could also be sampled and compared to the mixes.
This is the classic problem, Jack: the physics of the situation is against you. What you see on your screen (and what Photoshop constructs) is RGB light colors, an ADDITIVE color production method. But paint (and printing onto paper) are a SUBTRACTIVE color production. Pantone is supposed to be one way around that--but only in the sense that you IGNORE what you see on the screen, and trust the Pantone chip for how it will look when printed.
Even with frequently calibrated monitors and high-end ink jet printing, you are not going to produce what the paint will render.
Your last sentence, making up physical paint batches and somehow learning how those relate to the on-screen product, would be the only way to go, but IMO you're awfully close to square one at that point.

Tony Thompson Editor, Signature Press, Berkeley, CA
2906 Forest Ave., Berkeley, CA 94705 www.signaturepress.com
(510) 540-6538; fax, (510) 540-1937; e-mail, thompson@...
Publishers of books on railroad history


Rob Kirkham <rdkirkham@...>
 

Gee that is a creative idea Jack, but unfortunately, I don't think the electronic mix will give you any useful information. While I'm not really up on the pigments used in model paints, use of artists paints can be a revelation. We were all taught in grade school, for example, that red and yellow make orange, blue and red make purple, etc. But with paints, they do not always react with each other the way these generalisations suggest they should. Blue and red do not always make purple - it depends on the pigments. And those that do mix to purple certainly don't all make the same purples or make the purple someone new to the pigment might expect. The result is that the theoretical formulations of colour manifest in light on your computer screen do not mix in the same way as actual pigments - and so won't predict the results of the mixes of real pigment.

Rob Kirkham

----- Original Message -----
From: "Jack Burgess" <jack@...>
To: <STMFC@...>
Sent: Wednesday, May 14, 2008 4:11 PM
Subject: [STMFC] Freight car colors...


Mixing paint to match a paint chip, a paint sample, a photo, or memory is,
for me, very difficult. One could obviously prepare 50/50 mixes of all
possible combinations of one manufacturer's paint line but there are also
possible combinations other than 50/50 and 50/25/25 possibilities.

I've given a lot of thought to how that might be done "electronically".
Unfortunately, I haven't been able to find a program to take samples of
paint color and easily "mix" them. However, I have played with the Gradient
Tool in Photoshop (also available in Photoshop Elements). If one scans
painted samples of two paint colors, such as Floquil Boxcar Red and Floquil
Tuscan, and open the scans in Photoshop, the Eyedropper Tool can be used to
load those colors as the Foreground and Background color. The Gradient Tool
can then be used to "mix" the two colors and produce a 50/50 or 25/75 mix or
any other proportion electronically. The resulting mixes can be saved as
color swatches and so labeled. By scanning all of the available "reds" from
one manufacturer, one could produce swatches of all possible variations.
Samples of the prototype colors could also be sampled and compared to the
mixes.

I need to develop a variation of boxcar red to match the YV cabooses I
painted many years ago (unfortunately, I didn't keep track of the
proportions back then) for a project I'm working on. I'm thinking that a
good test for the process would be to mix two reds electronically and also
physically and compare the results.

Has anyone done anything similar? Was it successful?

Jack Burgess
www.yosemitevalleyrr.com


Doug Dolloff
 

The way that paint is mixed is by weighing the different colors. It is not by a 50/50 or 25/75 ratio. Different paint colors way different amounts, therfore you may add 5 grams of color A and 20 grams of color B to get color 25 grams of color C. This is how automotive paint is matched aswell as most house paints. You can use weight ratio like 5 parts part A + 20 parts of B = 25 parts of color C. That way anyone can use a weight ratio to get the same color, wether they are using grams, pounds, wheelsets, etc.. Paint pigments weigh different amounts. Typically yellow pigment paint ways more than blue or black because it takes more pigment to create yellow than blue or black. Weight is the only way I know of to accuratley tint or change the tint of any color and be able to recreate the same color later or to tell someone else how to recreate the same color.

I hope this helps.

Doug
Rob Kirkham <rdkirkham@...> wrote:
Gee that is a creative idea Jack, but unfortunately, I don't think the
electronic mix will give you any useful information. While I'm not really
up on the pigments used in model paints, use of artists paints can be a
revelation. We were all taught in grade school, for example, that red and
yellow make orange, blue and red make purple, etc. But with paints, they do
not always react with each other the way these generalisations suggest they
should. Blue and red do not always make purple - it depends on the
pigments. And those that do mix to purple certainly don't all make the same
purples or make the purple someone new to the pigment might expect. The
result is that the theoretical formulations of colour manifest in light on
your computer screen do not mix in the same way as actual pigments - and so
won't predict the results of the mixes of real pigment.

Rob Kirkham

----- Original Message -----
From: "Jack Burgess" <jack@...>
To: <STMFC@...>
Sent: Wednesday, May 14, 2008 4:11 PM
Subject: [STMFC] Freight car colors...

Mixing paint to match a paint chip, a paint sample, a photo, or memory is,
for me, very difficult. One could obviously prepare 50/50 mixes of all
possible combinations of one manufacturer's paint line but there are also
possible combinations other than 50/50 and 50/25/25 possibilities.

I've given a lot of thought to how that might be done "electronically".
Unfortunately, I haven't been able to find a program to take samples of
paint color and easily "mix" them. However, I have played with the
Gradient
Tool in Photoshop (also available in Photoshop Elements). If one scans
painted samples of two paint colors, such as Floquil Boxcar Red and
Floquil
Tuscan, and open the scans in Photoshop, the Eyedropper Tool can be used
to
load those colors as the Foreground and Background color. The Gradient
Tool
can then be used to "mix" the two colors and produce a 50/50 or 25/75 mix
or
any other proportion electronically. The resulting mixes can be saved as
color swatches and so labeled. By scanning all of the available "reds"
from
one manufacturer, one could produce swatches of all possible variations.
Samples of the prototype colors could also be sampled and compared to the
mixes.

I need to develop a variation of boxcar red to match the YV cabooses I
painted many years ago (unfortunately, I didn't keep track of the
proportions back then) for a project I'm working on. I'm thinking that a
good test for the process would be to mix two reds electronically and also
physically and compare the results.

Has anyone done anything similar? Was it successful?

Jack Burgess
www.yosemitevalleyrr.com


Jack Burgess <jack@...>
 

Doug wrote:
The way that paint is mixed is by weighing the different colors.
It is not by a 50/50 or 25/75 ratio. Different paint colors way
different amounts, therefore you may add 5 grams of color A and 20
grams of color B to get color 25 grams of color C. This is how
automotive paint is matched as well as most house paints. You can
use weight ratio like 5 parts part A + 20 parts of B = 25 parts
of color C. That way anyone can use a weight ratio to get the
same color, whether they are using grams, pounds, wheelsets, etc..
Paint pigments weigh different amounts. Typically yellow pigment
paint ways more than blue or black because it takes more pigment
to create yellow than blue or black. Weight is the only way I
know of to accurately tint or change the tint of any color and be
able to recreate the same color later or to tell someone else how
to recreate the same color.
I can't argue that weight is the proper way to establish and document paint
mixing although it would seem that the paint machine at the local hardware
or paint store is using volume to make the resulting paint mixture. It would
seem that using volume would work in this situation if the original paint
chips were originally produced the same way. That aside, when you are mixing
up only 1/2 fluid ounces of paint to paint a box car, I can't see how one
can weigh the ingredients...if you pour the base colors into separate
containers and weigh them and then pour the colors into another container to
attach to the air brush, the paint remaining in the containers will throw
the formula off considerably. Also, I'm confused by the "parts"
reference....parts suggests volume, not weight. But, at this point, the
discussion seems moot, after Tony pointed out a fatal flaw in my approach...

Jack Burgess
www.yosemitevalleyrr.com


Doug Dolloff
 

Jack wrote:
>I'm confused by the "parts"
reference... .parts suggests volume, not weight
I guess "parts" could pertain to volume. But if one where to add 5 droppers of color A to 20 droppers of color B which would = 25 droppers of color C. No matter the size of different droppers drops, color C would be the same. The other issue with colors is the gloss. If a color chip is semigloss and the paint you are using is high gloss or a matte finish, it will not look the same as the semigloss chip. And to add another factor, if you match a color in fluorescent light and then take the colors into sun light or incandescent light the colors will also appear different. Marty probably can remember this issue back with me in his Longmont days.

Doug

Jack Burgess <jack@...> wrote:
Doug wrote:
The way that paint is mixed is by weighing the different colors.
It is not by a 50/50 or 25/75 ratio. Different paint colors way
different amounts, therefore you may add 5 grams of color A and 20
grams of color B to get color 25 grams of color C. This is how
automotive paint is matched as well as most house paints. You can
use weight ratio like 5 parts part A + 20 parts of B = 25 parts
of color C. That way anyone can use a weight ratio to get the
same color, whether they are using grams, pounds, wheelsets, etc..
Paint pigments weigh different amounts. Typically yellow pigment
paint ways more than blue or black because it takes more pigment
to create yellow than blue or black. Weight is the only way I
know of to accurately tint or change the tint of any color and be
able to recreate the same color later or to tell someone else how
to recreate the same color.
I can't argue that weight is the proper way to establish and document paint
mixing although it would seem that the paint machine at the local hardware
or paint store is using volume to make the resulting paint mixture. It would
seem that using volume would work in this situation if the original paint
chips were originally produced the same way. That aside, when you are mixing
up only 1/2 fluid ounces of paint to paint a box car, I can't see how one
can weigh the ingredients...if you pour the base colors into separate
containers and weigh them and then pour the colors into another container to
attach to the air brush, the paint remaining in the containers will throw
the formula off considerably. Also, I'm confused by the "parts"
reference....parts suggests volume, not weight. But, at this point, the
discussion seems moot, after Tony pointed out a fatal flaw in my approach...

Jack Burgess
www.yosemitevalleyrr.com