Twenty-eighth operational CT-100 screen shots...
 

It's a good 15GP22 with low hours, a restored CTC2 chassis with a complete set of original RCA peaking coils, and a rooftop antenna picking up a true NTSC color signal. Here's a revisit to 1953 NTSC color via 51-year-old hardware.

http://home.att.net/~pldexnis/CTC2_h...ing_Color.html

Excellent!
 

That's beautiful!
Thanks for preserving something from the past!
Sincerely, John

"page cannot be displayed"??

That's REALLY somethin'-but stupid question-do roundies all "lose" so much of the picture? Or is that just a peculiarity of the early 15" sets? -Sandy G.

Sandy,
That set in the background is a 16:9 set showing a 4:3 picture so it looks like the picture is much wider but if you look closely you can see there are actually black bars on the sides.

(Site is working for me now)

Superb!

bobgary questions
 

Simplistically, a 15GP22 is a dud for one of only two reasons: no vacuum and no or low emission. (Maybe that makes three?)

Anyway, the process to rebuild a no/low emission 15GP22 electron gun assembly is perfected and in place. Ditto the manufacturer of glass stems. Basically, rebuilding the gun means replacing the old filament and cathode with new ones.

Recently (the last year or more) effort was extended to connect a stem to an old, dud 15GP22 neck. This has been done although I'm not certain the technique is as reliable as, say, the rebuilding of the electron gun. But the process has been successfully demonstrated.

What's being done now? The archives have yielded a bit of vintage 15GP22 data covering the baking temperature and slew rate of that baking heat. They are different from what is done today. Finding a way to reproduce the vintage procedure is the present challenge.

As to the phosphor in air question, no one seems certain although the current thinking is that since the 15GP22 is aluminized, the phosphors may be protected from the air.

Let me say, that is a beautiful picture. Absolutely beautiful. Infact so beautiful it's almost on par with this girl i gave a crush on. :P

I'd love to own a Merril and an RCA prewar set.... Maybe some day... :)

Jonathan

Who knew that old sets like that could produce such great pix? Gorgeous and well done!
Tom

I'm thoroughly impressed with the work you are doing. These sets must be preserved, and you are showing that it can be done. Are there many around that just aren't worth trying to restore?

Is there any possibility of air hurting a non-aluminized phosphor? I thought phosphors are essentially oxides or sulfides (specially formulated rock), so I would guess air is not going to do anything to them. (Moisture might be another issue. Maybe oxygen can hurt the sulfides or sulfur air pollution can hurt the oxides?) Have there been any expert opinions on this? Were old B&W tubes tubes ever rebuilt just by replacing the gun and leaving the phosphor alone, and did this depend on aluminization? This would be a good indication that there is no problem.

Pete Regarding the colors on the CT-100 - As I'm sure you realize, the original phosphors will ALLOW you to get correct colors, but there are a lot of matrix adjustments that need to be correct too. (Also, all the tubes in the color section need to be good - a weak tube in one part of the matrix may leave you with "acceptable" but not correct results.) So far, no two CT-100s that I have seen, nor the early Zenith prototype I saw, got the same color reproduction. Have you put color bars in and attempted the matrix adjustments?

Wayne. Thus far, I've gone by the book aligning the color (all front-panel controls) -- except for the I gain, which is running nearly wide open. Re color bars: gated color bars seem to have a luminance issue when bounced against the same signal displayed on the digital set. Blue bars on the CT-100 appear to be near-saturation in order to view red- and green-hue bars.

The setup: (1) both sets adjusted for color picture from over-the-air broadcast. (2) switch from antenna to gated color bar gen. (3) bars on digital set appear normally; blue-direction bars on CT-100 appear normal; red- and green-direction bars very dim. Readjusting contrast and brightness brings up the end hues.

Any thoughts on the color bar anomaly?


I made special efforts to find and use all original inductors, especially those in the demodulator filters, to get as close as possible to factory-original, and was successful except for L42, which I don't think is an important factor anyway (it's in the dc path from the color afc to the reactance tube). Was very fortunately to have gathered a full set of the 'white' inductors, which are very prone to opening up.

My further plans include sweeping the chrom circuits for published specs to help get as close to factory-original as possible.

Pete,

Using gated rainbow for matrix adjustment can't get you there, because you need each color to have the correct luminance as well as chroma to see if the final R,G,B drives are correct. You can use the gated rainbow to check the phase (and maybe preliminary gain) of I and Q, but not much beyond that If you can get a generator or a DVD with regular color bars (White, Yellow, Cyan, Green, Magenta, Blue, Black), that is best. All is correct when the individual red, green, and blue outputs of the matrix are square waves. All the peaks (max and min) in one channel should be the same amplitude, but the three channels will have different amplitudes depending on how much drive the CRT requires to make white. With non-rare-earth red, this probably means red needs 2 times (or maybe more on the 15G) the drive of green or blue.

Have you ever seen a setup done with a blue viewing filter? This assumes the matrix is correct, and adjusts hue and chroma so all the bars containing blue have the same blue amplitude. If the matrix is not correct, tough! - the red and green will be off when the blue is correct. In the case of the CT-100, it's worthwhile to check all three colors, since you have internal matrix adjustments. You could do this check with red, green, and blue filters. However, if you have full matrixed R,G,B driving the CRT, you can check with a scope. This measurement (and adjustment) is not directly available on later sets that matrixed by putting Y to the cathodes and color differences to the G1's.

Regarding brightness of colors: When a true NTSC tube with nominal NTSC matrixing is used, the reds may be darker than a modern set. This is because the modern set overdrives the R-Y to make up for the smaller difference between the red and green phosphors. In other words, turning on the modern green is like adding a little red to the picture, so the R-Y gain is increased to make the red turn off a little more than normal. On strong red color, this goes too far the other way (because of the CRT gamma), so strong reds come out over-saturated and over-bright. How much of this you see depends on how much R-Y boost the particular manufacturer thought the phosphors should have for best flesh tones and average scene content. On the opposite side of the spectrum, the 15G should show more saturated and brighter cyans. This should be obvious on full color bars along with the more "green" green.

In PAL, and in HDTV, the matrixing at the camera is changed to get the right color-difference signals for the MODERN phosphors with a nominal NTSC electrical matrix. So, reds are no longer overly bright, but of course you can't get the purer green and cyan that you could with real NTSC phosphors.

Looks good, Pete.

Wayne,

Thanks for all the info.

I thought that I might in be trouble without an NTSC generator because, as you confirmed, of the constant luminance factor. But I hadn’t made a connection to the ‘blue filter’ setup procedure, or using a scope to check luminance of CT-100 RGB channels. Thanks for that tip. I’ll use an S-video VCR and a DVD player to generate a color bar signal, and then scope the RGB channels…

Pete, it may help to use a ture NTSC color bar signal source (foff air or from a generator), instead of the gated rainbow generator, which does not control luminance.

How is the red? you photos seem to show only blues and greens...?

Where's the red?
 

Actually John, over-the-air programming with a lot of red was tougher to find than the predominately blue or green samples. I did find some fair examples of red from an ABC movie this evening and added a series of five pictures to the link.

Congratulations Pete! That picture looks wonderful. I had always wondered if decent convergence was possible on a 15g. That is amazing. :thmbsp: Even perfect at the far edges.

John

Ctc2/ct-100 Matrix Issues
 

quote from Wayne: "...but the three channels will have different amplitudes depending on how much drive the CRT requires to make white. With non-rare-earth red, this probably means red needs 2 times (or maybe more on the 15G) the drive of green or blue."

Checked this out and added material at the bottom of this link under 'CTC2/CT-100 MATRIX ISSUES,

This is a new link.

http://home.att.net/~pldexnis/CTC2_h...r__matrix.html

CTC2/CT-100 Matrix Issues -- more data...
 

Calibrate this tweak that: realigned the color and added some screen shots of NTSC color bars that should look better than previous examples on your computer monitor.

See the "CTC2/CT-100 Matrix Issues -- more data... " section at the end on this page:

http://home.att.net/~pldexnis/CTC2_h...r__matrix.html

Pete,

Great work in setting up the matrix!

As I understand it, your digital camera brightens the blues on the CT100 but not on your modern set. Interesting. I found that some blue/violet neon tubes do the same thing to my digital camera - had to underexpose considerably to get the color even close.

Does the gray scale also look bluer in the photo than to the eye? If your camera allows manual white balance, you could try balancing on the gray scale, and then take pix of the color bars and regular material.

-Wayne

New matrix alignment
 

Spent another three hours on the CT-100 color alignment challenge this Labor Day weekend, spinning around a bucket full of variables, including discovering how fine-tuning affects matrix performance, and came up with an improved screen shot of color bars at the bottom of the link.
http://home.att.net/~pldexnis/CTC2_h...r__matrix.html

Blue only
 

Once you get to the blue only stage, you should be looking for equal blue luminance visually on the cyan and magenta bars. The tint control will set this. This should be your ideal tint point...for bars.

Modern monitors have a blue only switch for quick setups. Modern NTSC bars are also split field with a smaller sub-set of bars through the middle. With these bars, the larger cyan and magenta bars are tint set as before, but now the sub-set bars (also cyan and magenta, but about 135 degrees out-of-phase) should also be luminance equal (but backwards) to the point that their luminance visually matches the bar above it. Rotate the tint until the upper chip matches the lower chip and all four should match.

This is the quick way to set analog sub-carrier phase from multiple sources through a switcher.

Never having setup a CTC-anything this way, it seem like setting the tint to center and adjusting the innards with this as a reference point would be something to look at.

Hope I did not add more hours to your scope time Pete.

Dave A

I just wish I knew HALF of what you fellas have forgot about this stuff...History & stuff like that ,I know...This stuff, forget it. I'm all thumbs. I'm in awe...-Sandy G

Let me first say, there is a vintage procedure for adjusting the matrix that requires I and Q signals from a generator. I have not tried this procedure. I do not have a generator that can supply the appropriate signal, although DaveA has kindly offered to lend me his vintage generator, and so I can perform the procedure sometime in the future.
----------------------------------------------------------------
To begin, color sync and the quadrature transformer and the color burst signal path adjustments must be successfully completed. The set must be able to decode and provide a stable chroma signal to the matrix before it can be tweaked.

Next, purity and gray-scale tracking are set up, basically as published in vintage documentation, which I’ll annotate below, or at least my version. Grey scale tracking is relatively easy to accomplish and has been stable in this CT-100 (B8000194) once set.

Purity setup is similar to early commercial service procedures: move the yoke back, B and G screens down, R screen up full and make adjustments including to the color purity and field neutralizing coils, move the yoke back to standard position, etc.

For grey-scale setup, a CT-100 has RGB screen controls, G and B background controls, and in addition, B and G gain, which consists of a pot on the appropriate 12BH7 circuit boards (not a PC board in the CT-100). These initial adjustments are accomplished with a b&w signal; I simply turn down the Color control. An over-the-air mono transmission (no burst) is hard to find!

I’m just going to copy the notes I made for myself as a guide to this gray-scale task:

CONTRAST AND BACKGROUND DOWN (CCW)
BRIGHTNESS NEAR MAXIMUM
R, G, B SCREEN (GRID 2) SET LOW-LEVEL WHITE RASTER — NO VIDEO

W/MONO SIGNAL SET BRIGHTNESS AND CONTRAST FOR NORMAL

B, G GAIN SET NORMAL-LEVEL WHITE RASTER

SET BRIGHTNESS LOW

R MASTER, G, B BACKGROUND SET TRACK WHITE RASTER LOW-TO-HIGH VIDEO

Once the tracking setup gives a b&w picture that doesn’t add a hue to any setting of the brightness and contrast controls, I crank up the color control and assess the damage.

To review, the purity is set, the grey-scale tracking is set, the color burst amp, color AFC, quadrature transformer, color subcarrier level, burst takeoff transformer, and whatever, are all adjusted BEFORE the matrix is tackled.

There are no matrix controls to speak of. Everything is fixed. During restoration, each matrix resistor was checked and found to be in tolerance. So what gets tweaked and what do you look for? I feed an NTSC color bar into THE SET on channel 4. I do not tweak the screen controls. Ditto the background controls. With a modern dual-trace scope monitoring the green and red 15GP22 control grids and a blue filter to observe the color bars, tweaking consists of a combination of the hue control, the brightness control, the contrast control, and ever so slight touch-up of blue and green gain. I recently discovered that, on this set at least, the fine tuning plays a roll in the matrix drive signals, mostly the blue. There is no red gain because in a CT-100 that gain runs wide open to compensate for the low efficiency of the red phosphor.

[You may recall the earlier scope photos and elation over the red signal being twice the amplitude of the green. That’s because to get the same brightness from the inefficient red phosphor, it is driven with about twice the voltage as the more efficient green (and blue) phosphors. Vintage documentation predicted the modern scope display.]

What I haven’t detailed is how to interpret the scope and filter data into effective tweaking. DaveA has provided a rundown in an earlier post for modern video equipment. Basically with the CT-100, the scope display is used to monitor the amplitude, or height, of the color bars for equal amplitude. Complementing that is the blue filter that’s used to visually balance the intensity of color bars. Blue filter use is a common technique, but it assumes that the matrix for the red and green is okay. I have red and green filters on order. When I get them, I’ll be able to check the matrix for all three colors and tweak as needed.

Of course, all this assumes one is not color blind, although I guess one so afflicted could adjust a matrix fairly well using the scope technique. Sorta thinking out loud here…

The fact is, once set up, any little tweak of any operator control, let alone service controls, will throw off the matrix alignment. That’s why I used the ‘…bucket full of variables’ phrase. It’s frustrating in theory, but in practice, if the ambient light changes and the picture requires more contrast, you can crank up the contrast control a little. Although you are technically no longer aligned, for practical purposes the picture still looks absolutely great. You’ll still have no problem appreciating the wide 1953 gamut.

Awesome TV!!!!!!
 

Those screen shots of ct-100 converged properly are simply amazing as expected!!!
:D :thmbsp: :banana: :lmao:
I really do wish there were more of these sets with extra CRTs readily available. I'm hoping a lost stash of CRTs will some day show up. I have given up on wanting one of these since CRTs or replacements not available. I'll keep you in mind if ever come across extra 15" CRT.

I just read a review on a Dell Plasma TV for $2599 dollars in newspaper & said set could not decode colors completely accurately. It's amazing what people are paying for crap these days.

:no: :thumbsdn:

Bobgary, I believe the waveforms you posted correspond to an old Hickok generator, which had full field bars, but not in the modern order according to luma level. Also, once things are reasonably close, it's much easier to check if these color difference signals are exactly correct by looking at the R,G,B waveforms (if available) or looking at the picture through colored filters, or turning off two guns at a time. (Professional monitors will send the blue-only signal to all three guns so you can see blue as monochrome.)

Pete, I don't have a schematic in front of me, but you said the matrix is all fixed resistors. I thought I recalled an "I gain" adjustment in there somewhere?

Isn't this how the modern sets do...the chroma and luminance signals are mixed before they get to the CRT? Seems like in most of the tube type sets the mixing actually occurs in the crt itself with the chroma at the grids and the luminance at the cathodes.

You're right about the I gain of course, it was set weeks ago and has never seemed to require further tweeking. And I probably shouldn't have thought of the matrix as only the nine discrete 1/2-watters that combine the +/-IQ and Y.

I wish you could all experience the actual video: There's a commercial playing for a Lilo&StitchII (sp?) DVD. The bug-eyed character in the commercial on the CT-100 has bright, intense, large lime-green eyes that pop off the screen at you. Interestingly, in an A-B comparison between the CT-100 and my 38-in. widescreen CRT HD set, there's no similiar effect on the HD set, although there may be setup issues to consider. Just came back from seeing the DVD at Target. The eyes in the print version were pale blue and didn't even appear to have been designed to draw attention --nothing like the TV ad to my eyes anyway. Now I'm wondering what the DVD image would look like on the Merrill. Na. Not worth the fifteen bucks to find out.

Yep - Several things combined to make solid-state sets go back to RGB cathode drive:
1) CRT's with unitized guns, where only the cathodes were separate for the three colors;
2) cheaper to have only 3 high-voltage video output transistors instead of 4
3) with ICs: definitely cheaper to do all matrixing to RGB in the IC - also, at this point, the chroma and luma gain controls could be electonically ganged so they go up and down together without changing color saturation when the "picture" control is turned.

I can't remember if I ever saw a design implementing a "picture" control with mechanically ganged "contrast" and "color" controls - anybody remember such a thing? :scratch2:

---

The CT-100 and the 21-CT-55 (basically a 21-in. CT-100) both used a ganged pot assembly to control luma and chroma simultaneously. For some reason however a production change to the 21-CT-55 bypassed the chroma pot, leaving only the luma pot functional. The user color was a separate pot in all cases, of course.

Back to the drawing board. I located the vintage matrix alignment procedure that includes adjusting L141, the inductor in the hue control circuit. This is where a course adjustment to the hue control takes place.

These main and alternate procedures seem to use the old non-luma-progression color bars Old_TV_Nut referred to earlier, who must also be the one to ask this question: ten years ago I worked on military stuff that used multiplying DAC's to contol what amounted to audio-frequency-range levels; is that how the function is or was implemented in TV for audio gain, contrast, and whatever? Is it even a good question?

Pete,

Did I miss it in your description? There is one tweak in the demod, the I gain control, which will modify the amount of +/- I signal used by the matrix adders to generate R, G, and B signals. Just one more independant varable in the equation.

It's a good question. There were no such things as multiplying DACs until much later. The first ICs for the color and video sections [and audio] were all analog, and did gain control using Gilbert-cell analog multipliers, which could be controlled by a common DC voltage. By the way, this was easier to implement cleanly, since the video and chroma signals did not have to be wired all the way to the customer controls and back, and the DC voltage could be filtered to get rid of noise pickup.

[Start of brag] One of my favorite inventions is a three-way I did with two other guys at Zenith (for monochrome chassis), where we stacked Gilbert cells, and using a simple external RLC filter as a simultaneous lowpass filter, differentiator, and second differentiator, did contrast control, switchable gated noise reduction, and video peaking control, in one swell foop [end of brag].

The more you think about it the cooler it gets :thmbsp:

CT-100 Update from my site's Restoration Page
 

Have been working on the restored CT-100 recently to prep it for display in my living room. After extensive tuning and testing, got it buttoned up a few days ago, including the screen under the chassis, a small fan near the back to aid in dispelling heat, and wood screws in the back cover.

I cleaned and polished the mahogany veneer, tightened four wing nuts that lock the cabinet lid, and then turned it on and damn, three minutes later the screen went dim.

Remove the back, then troubleshoot to find the second video amplifier (a 6AN8) causing the headache. Pitch it in the trash. Put in a replacement. Button it all up again. Get out the NOS fine tuning knob I’ve been saving for this occasion and, after first testing the installation procedure on a spare CTC2 chassis, crack the plastic on the new knob trying to install it on the restored chassis. Cuss. Mix some epoxy. Stick it back together, and use a worn knob for now.

Okay, all this sets the stage for what I’ve been working for since 1999, but intensely since last April: eyeball 1953 NTSC color!

I think it happened. After mocking the WB earlier on AudioKarma, WPIX channel 11 came through with (pun intended of course) flying colors! South Park. Never watched it. Some of my thirty-something acquaintances love to quote it. Don’t care about it. But what color!

Chrominance charts indicate the extended green capability of the 1953 standard. My eyes believe the green on the character’s hat is in that extended area of the 1953 spec. A simultaneous display of the program on a high-definition direct-view CRT receiver in the NTSC mode was strikingly less vibrant.

The HDTV serves as a standard — a stable source of color used in the initial alignment of the color AFC circuits of the CT-100 for example. During the past several weeks, I would adjust the CT-100 to match the colors of the HD set. Not the brightness of course, but particularly subtle pastels and varying hues of violet. It’s a darn fine HD television set. But for this experience, which reproduces a program on two different sets driven from the same antenna, the HD set displayed completely lackluster green hues for that character’s hat compared to the CT-100.

Yellow was also visually different. A suggestion of which appears as this guy’s mittens.

The color resolution of hdtv is supposed to be more than twice that of NTSC, so next time you should crank up a true HDTV signal on that hdtv set if you want to see the CT-100 get even more green... with envy ; )

The color resolution, screen acreage, widescreen 16x9, luma resolution, six-channel sound, brightness, and just plain digital stability of HDTV produces an awesome video experience. :thmbsp: One I never want to miss.

Books and charts say the color gamut is greater with 1953 NTSC, as I interpret them anyway. My gnawing itch wonders if WB11 actually transmitted the video with wide-enough gamut to make the effect I saw on the CT-100 vs. the HDTV. I wonder if the cartoon makers had their software set to 1953 NTSC. I wonder if the NTSC generator in my VCR is spitting out a color bar signal good enough to calibrate the CT-100. I wonder, as I wander…!

Oh well, only 63 more days to the next Rose Bowl. :yes:

Everyone uses VCR's for the RF modulation on channel 3 or 4! :) I can't remember the last time I used a VCR. :) Both my 630TS and CTC9 Felton has VHF tuners, and there are no VHF stations in my area. I was honestly going to buy one of those RF modulator boxes, but use my satellite receiver for the time being. Most newer model RCA directv boxes output full feild color bar by pressing the menu and info bottons on the receiver at the same time, and it has RF output too. I'd use my UHF to VHF converter for off air reception, but I guess the only way to use it is if I connect an amplified antenna to it's input since it has no RF amp?

Jonathan