Hi to all
We are having a devil of a job finding anyone that can match a true Pullman green window frame that we have and give us oil based gloss paint.
We had a quart made up succesfully in Latex just for patching initially but the oil based match is being a huge pain. The analysers at various locations throw up different formulas that are either too light or too dark.
Anyone have a Sherwin Williams formula we could try ? Or anyone else's

Picture illustrates our latex formula, which is far nearer the right shade and the adjusted (from brown !!) oil based gloss which is way too light !!

Cheers

Mike Pannell
Car 57 Cheyenne

If I'm not mistaken, there was no one "pullman green" no more there was one yellow that the Twin City Lines painted their big, wood-bodied "standard" streetcars. The TCRT's Snelling Shops used a basic yellow paint and certainly tried to match their standard yellow, but exposure to the weather, sun, snow, pollution, etc., resulted in differing shades of yellow as time went on. I suspect, the same can be said for the painters at the various Pullman Company shops scattered about the country. I'm sure they tried to match the company standard, but after a few weeks/months in service, I suspect that green had changed.

I'm sure the passenger car restoration experts on this forum will correct me if I'm all wrong with the above. But if I were you I would do my best to match what you've discovered and be happy that you've done an extraordinary job restoring a true piece of American railroad history.

Good work and good luck!

Thanks!

Colorado State Ry Museum isn't far from you and can no doubt provide you with a great recommendation on not only color formula but spec for a paint suited to a very similar climate.

dave

Mike, Taking what Jim said a step further, if you're going to be painting the whole car, just pick a match that you like and run with it. No one is going to know the difference anyway. If anyone ever criticizes the shade you pick, just politely ask them to see pictures of a car they have restored. We have two cabooses that we're running this weekend giving rides at the Virginia Museum of Transportation. They were painted 3 years apart with the same brand and shade of paint by the same people. They are not the same color. Weather happens.

Thanks everyone, Sherwin Williams here in Cheyenne spent two hours matching and tinting and came up with the best match yet, so we are good at last.

Mike Pannell

Following up on Jim's post re: Twin Cities yellow, I think the formula for Pullman Green involves using Brunswick Green as a base and adding Traction Orange to get a warmer shade of color. lol. (Much like Pullman Green, the shades of "Brunswick Green" and "Traction Orange" are often subjects of color debates in the museum community.)

The late Al Savage, retired VP at DC Transit, told two stories about fleet colors over time. In general, he noted that the shade of "electric blue" and cream changed over time to the point that at NCTM we used "Norway Spruce" (more green than blue) as the color match for CTCo 1053. Al's other story involved painting the trucks. The shop became frustrated with maintaining the fleet color on the car trucks, so Al had the paint shop mix up a color he called "dirt", a shade of brown, to match reality.

Good luck with Car 57!

Wesley

You know, I hear that all the time. "Paint drifts all over so who cares." Well if you don't have any reference for comparison, you end up adrift. Take the time to match right. Doing top notch work makes you a top notch shop.

The paint samples I see in the photo look miles apart to me, but the light is playing tricks because of the glare.
It's harder when your only sample is part of the railroad car. At first, I did this (photo) and as you can see, it's a complete disaster. Fortunately I discovered a good sample on the underside of a bracket that holds on the storm windows. Off that went to Sherwin Williams, who found it was out of gamut for their machine (beyond its ability to scan, or beyond the reach of their pigments.) They tried to tweak it by hand, but as you know that's harder than it looks.

I've used a digital camera or scanner to compare two things in a paint program. Not in RGB mode but in Hue-Saturation-Brightness mode. Brightness tends to be more accurate than the eye, saturation is about equal and hue is useless. Also this is useless for comparing one photo to another - conditions are just too variable.

Once you get the color you want - exactly - get some 2" x 3" scraps of aluminum, clean them and paint 2 coats of your color on them. Punch a hole in the top for hanging. Make as many as you can stand to make. They all go in a dark place. When you need that color in the future, you send one of those chips to the paint company as a reference.

Color matching is HARD.


Bad paint happens too. Some vendors use lousy pigments.

The resin matters too. In the days of the metallic pigments, my impression is the pigment protected the resin from UV. Modern organic pigments aren't nearly as UV-stable, but fortunately there exist resins which are.

You know, paint didn't always come out of a can. If this is a wood Pullman, at that time they hadn't really worked out stnadardized paint codes. Pullman didn't order pallets of #123456 in 42-gallon barrels. The car painter mixed it using his own stock of pigments, and getting a color match was part of the painter's craft.

Only later came the concepts of color standards, and even then, it was a war between customers (who wanted open-ish standards like Munsell) versus the paint companies (who wanted to lock customers into proprietary palettes.) If the railroads did use Munsell, they probably went with the "even integer" colors rather than the in-between decimals. Here's a Munsell page to show what I mean by that. http://ispi-lit.cirad.fr/text/ALB32e.htm

We've come up with similar issues with the car I'm restoring. In early days, the NSWR used Deep Purple Brown as an exterior colour. The trouble with that is for starters, it can be 5 different shades depending on the angle you look at it. Secondy, it's become something of a generic name, so there are variations of formula.
We were lucky in that upon stripping the underside of an end canopy, a colleague came across the original colour and, most importantly, in an area that wouldn't have been subjected to sunlight. Thus, we can reasonably consider that this colour is pretty authentic, at least to this particular car.
I guess the best hope is to find such a patch on whatever car you're restoring. If you can get a good match, then at least you can tell the rivet counters that the colour is authentic, if only for that car. You can't get much more accurate than that!
Cheers, Bob

Thanks chaps these replies have been fascinating, i agree that matching is vital and i intend to be really anal about it ! After all i have to live with the car and tell its true history. We have great original paint under the hoods but its tricky taking those in !
Pullman green was beyond the capabilities of the computers at most paint stores, most people swore it was brown that i was showing them until you told them to take it outside and in the sun it glows a beautiful olive green.
Pullman must have had strict controls surely on mixing paint, otherwise you would have all 5 cars going to the DT&FW for instance in differing shades ?
I went down to the car tonight at dusk and put some new paint on...was scared to see it looked truly brown.... however light was playing tricks as the sample i brought home looked green in certain lights....so we must be close !!!


Mike Pannell

Your comment about the apparent color shift of Pullman Green due to lighting conditions has been mentioned on this forum a few years ago:
Pullman Green..... AGAIN!

If you haven't read it already it would be worthwhile reading the thread for other information included..

When ARM visited Steamtown a few years ago, their restoration folks said that there were different shades of Pullman green depending on the predecessor railroad. They have former Lackawanna and Jersey Central commuter cars and you can see the difference.
Aaron Isaacs, editor
Trainline/Railway Museum Quarterly

I totally agree that you should restore it as close to the original color as possible. The big question is, what color was that?

As has been mentioned, when they were originally painted, the colors were hand crafted based on some sort of formula. Like baking cake, you'd get different results and colors depending on the ingredients, skill and conditions when it was painted.

They didn't have the consistent colors we have today on autos and the like. So while they were close, there was also some variation.

On a related note, years ago I recall riding the Coast Starligh past a long string of Railbox boxcars. The color variations were amazing, everything from bright yellow, old gold, cat gold to orange and one that was pretty much pink!

So, 20 years from now somebody decides to restore a Railbox and tell the story of Per Diem boxcars (which would be a good idea, BTW). What yellow/gold do they use? How do they determine what color is correct. "We sanded our car down and found the color... it's nearly PINK!" OK, for that particular car, it was pretty much pink, but it didn't represent the typical RB yellow.

My suggestion would be to try and get as close to the typical color as you can, using whatever records you can find, old photos, paint on the car, recollections of anyone who remembers the original and any other resources you can find. But it's going to be tricky to nail it down to exactly Pantone A2918.

Speaking of Pantone, once you DO get a color match, that would be a good benchmark. They make their living in providing color matching, so once you've identified the colors, you will be able to find the same color ten years down the road, without worrying that the red has faded etc.

What if Pantone goes bankrupt? Their colors are such an industry standard that even if they did dissapear, one of their competitors would probably license the info to provide a cross-reference. It's extremely unlikely that the Pantone reference would simply cease to exist.

Can I chime in a bit? First, why Pullman Green if the car did not serve in Pullman service or was built before the standard. I can't remember if we supplied any spec info when supplying the floor plan to Michael Pannell, probably didn't have it for this early car ordered by General Dodge for his secondary lines. The best is to go down to the original paint coat, buff it to remove varnish topcoats and gloss, and make a photograph with a Kodak color standard next to it. That way you have a standard reference as camera, monitors and printers will never get it right. Spectrum of the incident light is very important. Color cast is very different under various sunlight conditions, incandescent and flourescent lighting. A standard lamp is pricey and no one seems to go to that level of sophistication. Now, what does the original color look like after all this? I am sure a truly professional artist or photographer will take exception with some of what I have said as a rank amatuer but if you want accuracy, a study in different lighting might or might not make the difference. There once was an "antique pullman green" and a "modern pullman green" sample at Sherwin Williams but I have no idea if they can still locate it in their archives. History can fade just like paint colors. It has been said previously that "Pullman Green" was originally carbon black with a little chrome yellow added (now unobtainium, a metallic additive). I have a gallon of a DuPont Dulux "Pullman Green No. 11" put aside that is rather yellowish, the particular formulation probably used for CN or CP cars and the can is very heavy, probably with lead base. The 1906 B&M car I had worked on at IRM has been painted by the current restorer according to his view using another car's paint sample with a local Sherwin William's paint match and I can tell you unequivocally it is way off from what I found on a door post by heat stripping (watch out for lead fumes, use an appropriate respirator). The public is only interested in what they see and only can say "wow". Be not afraid of what you do, only preserve a sample protected from the elements for a future match! We did that inside the car.
Sincerely, Ted Anderson, IRM volunteer & curator, IRM Pullman Library

I would assume absent evidence to the contrary that the car was resided probably more than once in its working life. No was to know what era whatever you might find under layers and over wood relates to. If there's a historical record available for what the UP was ordering put on their passenger cars at the time of the order, or whichever era you want to interpret - and I'm guessing some place like CSRM would be a fine place to start asking - you'd be much more likely to come up with good history.

dave

Ted,
I guess we assumed Pullman green because it was built there and we have found tons of green all over the car in unmolested areas. Anyhow we have a great match now and we will post the results and if anyone else needs the formula we have it.
It was matched to a top rail of the window frame that was hidden behind the upper window so it was clean and unnaffected by the light.

Mike Pannell