I have an article from the Radio & Television News for October 1951 about how to build your own colour wheel and synchronize it so that it would be able to reproduce the CBS colour signal. In the course of the article, the author mentions certain problems you will encounter, amongst which is the variability of phosphors: "as a service technician knows, it is rare to find two phosphors that have the same colour." He doesn't explain what he means by that; does it mean that there were many different varieties of phosphor, or that because of some aspect of the manufacturing process, there was a lot of variability in the emission spectrum of a single kind of phosphor--ie, it was hard to control the outcome of the manufacturing process--or does it mean that phosphors decay in some way and slowly lose their original emission characteristics?
I have attached the first page of the article.

"as a service technician knows, it is rare to find two phosphors that have the same colour."

For those of you too young to remember the early B&W days of TeeVee, each manufacturer made slight variations in the phosphor color and in the exact color and exact density of the safety glass, so they could claim that their sets produced the best picture!

This was mostly a sales gimmick. Most CRT's started out their life as a bluish white, and aged to a slightly warmish brown color with use. The public took this as normal.

Thanks for the link to the article, as I've never seen the mention of filter density in the Peter Goldmark system, just the Kodak or Wratten filter color number, and knew that surviving CBS color filters did not match the published Wratten numbers.

These are the same color filters used then and still today for reproduction of color with B&W film.

The mention of an overall yellow filter by the author is counterproductive. It would be far more efficient to reduce the saturation of the complimentary color, blue. Me thinks that he was attempting to create a one-color-wheel-fits-all system.


For an introductory look at photographic color filters, see:

http://en.wikipedia.org/wiki/Wratten_filter

James

The mention of an overall yellow filter by the author is counterproductive. It would be far more efficient to reduce the density of the complimentary color, blue.

1) You mean increase the density of the blue filter (make it darker).
2) Although this is the first time I have seen a proposal to use an overall yellow(ish) filter, it makes sense to me, and it should not cause a terrible decrease in efficiency because it's a relatively gentle spectral tilt. Putting one of the suggested filter combinations in the wheel and then doing a fine adjustment to white color with an overall filter seems like a much better idea than trial and error adjustment of the color segments. Refer to Cliff Benham's work posted on this site and shown at ETF. He can get the needed final tweaks from the RGB gains in Darryl Hock's NTSC-to-CBS converter, but in 1951, circuitry for individual RGB gains in the receiver adaptor would have been considered too complex.

1) You mean increase the density of the blue filter (make it darker).


Yippes!

With hoof in mouth, I wrote "Density" when I thought "saturation", ie., to allow the blue filter to pass more non-blue light which is yellow light.

Actually more density would also work, but again it would be counterproductive in getting a brighter screen image.

Adding density to blue is exactly what a yellow filter would do. Block some blue light, but allow red and green to pass light mostly untouched.

Allowing the blue filter to pass some yellow will ever so slightly reduce the flicker withut hurting overall color saturation.

When the color wheel is there, you simply want to get more light on the screen.

I assume the use of "density" by a non-specialized-in-photography author may mean "saturation" as in the terms are sometimes used interchangably.

James

CBS used a P-6 phosphor in some of their tubes. It glowed white, as opposed to P-4, which could have varying mixtures of blueish or yellowish phosphors. The CRT in my GE 810 is a 10FP4A, which is very close to "pure" black and white, with sharp contrast.

Yippes!

With hoof in mouth, I wrote "Density" when I thought "saturation", ie., to allow the blue filter to pass more non-blue light which is yellow light.

Allowing the blue filter to pass some yellow will ever so slightly reduce the flicker withut hurting overall color saturation.

James

I'm afraid that allowing the blue filter to pass enough yellow to get a better white would affect the blue saturation significantly (and the magenta also), However, if you are starting from the very dark and saturated blue of the wratten filter, reducing the saturation is exactly what CBS did to get their lower-flicker primaries, so in that case I agree with this strategy - but it really did desaturate the blues and magentas compared to NTSC.

I've never seen the mention of filter density in the Peter Goldmark system, just the Kodak or Wratten filter color number, and knew that surviving CBS color filters did not match the published Wratten numbers. <<<-->>>
The mention of an overall yellow filter by the author is counterproductive.
James

I've read this article before and there may be some confusion in the filter number callouts.

The use of a Wratten #6 Yellow filter over the entire screen is correct for an experimental CBS set, the RX-40 color converter, which used two standard Wratten filters and one special one in the wheel. The standard filters are Red Wratten #25, Green Wratten #58 and the special is a Blue Wratten #47, 3/4 density. The yellow filter rebalanced the overall image color from blue-ish to white.

The second filter set was called "Monsanto E" and didn't require the yellow filter.
Both sets of filters made the same color picture, but by different means.

What does all this mean today if you are going to build a color wheel set?
It is absolutely correct, but now useless information because neither sets of filters have been made for the last 50 years.

Wratten filters would also be prohibitively expensive, costing about $2400.00 for enough to build ONE 22-1/2 inch wheel for a 10 inch CRT. [!!!!]

I did a lot of research for the ETF Museum on color filters for their CBS sets and found I could use good quality polycarbonate lighting filters to make new color wheels. The picture quality is excellent and the work I did utilized a good spectroradiometer to document and match the color filters correctly.

The basic requirements for a set of color filters today is that they must produce no difference in, or otherwise 'color' the white light balance from the CRT as they spin in front of the screen, they must be strong and they must not fade.

If you are building a wheel set you might be using one with a round tube 10 or 12 inch CRT. Anything bigger and the wheel becomes VERY LARGE and hard to control.

The color temperature of most 10 and 12 inch CRTs is in the realm of 7000 degrees Kelvin. This is a bright bluish white, especially if you have a 10FP4 or a 12KP4 Aluminized CRT in the set.

Three particular color filters made by Roscolux will color balance very well with these CRTs and produce beautiful, deeply saturated color pictures.

The type numbers for them are Roscolux Red #25, Blue #76 and Green #90.

They are very cheap compared to Wratten Kodak filters, about $8 bucks for a 20X24 inch sheet from any theatrical lighting company.

Building a color wheel is easy to do with standard craft supplies. If you need more information about any of this, I'll be checking the thread.

The usual disclaimer: I have no affiliation with Roscolux or any other manufacturer of color filters except as a happy customer.

Cliff