Official NTSC test slide #18, "Pumpkins"
Displayed on RCA 21CT785 "Westcott", CTC-5 "Super" chassis.

https://live.staticflickr.com/65535/50534925757_b8900a3f4d_c.jpg

Official NTSC test slide #18, "Pumpkins"
Displayed on RCA 21CT785 "Westcott", CTC-5 "Super" chassis.

https://live.staticflickr.com/65535/50534925757_b8900a3f4d_c.jpg

OK! you went and had to tempt me! :D

Be sure you are wearing your Halloween masks when watching! :D
https://www.youtube.com/watch?v=C6OedtbldkQ&feature=youtu.be

Looking very good!

I trust that is on thee 21AXP22A?

I trust that is on thee 21AXP22A?

Yes indeed. I have to adjust the camera white balance to match the TV white point, which makes the incandescent "TV lamp" on top of the set look very orangy also.

Yes indeed. I have to adjust the camera white balance to match the TV white point, which makes the incandescent "TV lamp" on top of the set look very orangy also.
I wish I had a fancy camera like that :(

I wish I had a fancy camera like that :(

What are you using? A phone? It is possible to get good pics with a phone. The main thing is to get the exposure right, and then you can make other adjustments in software later.

I have just pulled my CTC5 apart to try and understand why I cannot get the three guns to track properly. I was going to blame the 21AXP22A but testing it the emission of all three guns is great on the B&K 467. The tracking test show the guns follow so I feel I can declare the tube healthy. (In fact it amazes me how good the CRT is. My guess is the CTC5 was a dog of a chassis - low High Voltage, rather dim phosphors and a pain to adjust tracking and convergence resulting in the set not being used much).

Anyhow I want to find out why the shunt regulator has too easy a life: you disconnect the 6BK4 and the high voltage remains at 21kV! And the tracking problem probably due to my CTC5 incompetence! Even the manual is rather arbitrary when it comes to trying to track with the screen, background and gain adjustments. It is so much easier with the Zenith color set!

...
Anyhow I want to find out why the shunt regulator has too easy a life: you disconnect the 6BK4 and the high voltage remains at 21kV!...

The HV circuit in the CTC-5 Super chassis is not capable of much in the first place. I suspect that 21 kV unregulated was normal. I also suspect the Deluxe chassis was capable of much more. Maybe someone who has worked on both can comment.

And yes, the tracking arrangement in the Super chassis is a pain. Part of of the problem is that offsets in the demodulators can affect it, hence the background controls on the CRT G1's.

The Deluxe chassis and all following RCAs using XZ demodulation plus following matrix/amplifier tubes for R-Y, G-Y and B-Y amplification are much more stable, because the three output stages also perform a DC-restorer function. The negative horizontal blanking pulse applied to the three interconnected cathodes causes the tubes to draw grid current, thus DC-restoring the X and Z signals at the input coupling caps and stabilizing the G1 voltages. There can still be some difference between the amplifier sections, but nowhere near as bad as in the Super chassis B-Y and G-Y demods. I have never, ever seen anywhere an explanation of the DC restorer function in the XZ demod system. I think RCA just didn't want repairmen worrying their poor little heads over understanding something that had no effect on repair diagnosis. If the tubes were bad, it would be obvious and fixed by replacing them. And if a component failed causing the blanking pulse to be lost, there would be retrace line problems, so even though it would cause loss of DC restoration, that is not the symptom per se that you would use for diagnosis.

Anyhow I want to find out why the shunt regulator has too easy a life: you disconnect the 6BK4 and the high voltage remains at 21kV! And the tracking problem probably due to my CTC5 incompetence! Even the manual is rather arbitrary when it comes to trying to track with the screen, background and gain adjustments. It is so much easier with the Zenith color set!

If you compare the HV regulator circuits in the CTC5 and CTC7 (free Sam's for both are on the ETF Color sets -> technical documentation page) the CTC5 and CTC7 circuits are identical except that a fixed resistor on the (non-adjustable) CTC5 is replaced by a potentiometer and series resistor on the CTC7...on the second CTC5 I did I upgraded the HV circuit to the adjustible CTC7 circuit.

I wish I could remember a few months ago when I did my second CTC5 I had a couple of issues setting grayscale tracking. One was related to there being 2 versions of the super chassis the one I had having a electromagnetic blue lateral and no blue gun level pot and the Sam's set having a gun level pot and Permanent magnet blue lateral and writing their proceedure assuming you had a blue gun level pot (I forget if the gun level was screen or drive).
The other issue I had was the proceedure called out a voltage for IIRC the red gun and I had some circuit fault I had to fix to get there. I have some vague recall of AGC issues that may have been related.

All the 21AXP22 RCAs had circuit designs that drifted grayscale some with extreme adjustment of brightness and contrast controls. The key is to get it reasonably good in the range you are going to watch it at.

I went thru the color tracking following the service notes a and my intuition. Grey scale tracking looks better now.

However I am not fully happy with the picture. When the video exhibits a wide contrast ratio, the blacks are crushed. It is as if there is no DC restoration.

Looking at the CTC 5 schematic, in both the Y and the color difference paths feeding the 21AXP22A, there appears to be coupling capacitors which suggest AC coupling only. I compared with a Zenith 21MC32Z, and both the Y and the chroma gated beam color difference paths between detector and CRT grids appear to be DC coupled.

Am I correct in suggesting the RCA CTC5 does not include DC restoration and the Zenith does?

All sets for a long time from this time onward had partial DC coupling in the luminance. This results in a partially floating black level depending on average picture luma. Consumer Reports complained about this for years and years. The manufacturers' reasoning was that it partially compensated for black level errors in broadcasts (and in analog cable systems, where consistency was much worse). It was finally resolved with solid state studio gear and solid state TV sets whose circuitry was stable enough to keep the proper black level once set. Consumer Reports was very pleased.

The CTC-5 also has partial DC coupling (not zero) in the color difference paths. Again, this counteracts circuit drift or changes when tubes are replaced, to some extent. It is also a technique for automatic color balance, called "integrate to gray," because many scenes are close to gray on average. This technique was even used in automatic color photo printing, where it affected the white balance, not the black balance as in the CTC-5. It is prone to "subject failure," that is, if a large part of the scene is a dominant color, the rest of the scene will be pushed in the complementary direction.

In color prints made on these early machines, I have seen pics of someone wearing a red sweater where the rest of the picture including their face was pushed toward cyan.

In the CTC-5, you can see the effect of this partial coupling if you play a DVD of a test screen that is all one color, like a complete red field for checking purity. The result is that the red is desaturated and foggy tomato-soup looking, even at a correct color level setting that shows small red areas in program material correctly. I have had thoughts about removing the caps that pass more AC than DC in the color difference channels, as the color difference outputs seem to have more than enough gain for dynamic scenes. The drawback might be that tracking would be more susceptible to tube differences and drift.

The CTC-7 and such later RCAs with X-Z demods have a DC restoration effect built into the color-difference amplifiers by drawing grid current when the negative horizontal blanking pulse is applied to the cathodes. But the luma channel is still only partially DC coupled.

Typical luma DC coupling was about 85%, but could vary with contrast setting depending on the circuit design. So, with correct brightness control (black level cutoff) setting for an average picture, a nearly all white picture would clip shadows at 7.5% of the video range, and a nearly all black picture would add 7.5% fog to the blacks.

Typical luma DC coupling was about 85%, but could vary with contrast setting depending on the circuit design. So, with correct brightness control (black level cutoff) setting for an average picture, a nearly all white picture would clip shadows at 7.5% of the video range, and a nearly all black picture would add 7.5% fog to the blacks.

I went off to study later chassis implementation. The ctc7 thru ctc11 bridged thru the brightness control some of the DC across the 0.1uF coupling capacitor. The ctc20 seemed to go a bit further and the brightness control top leg preceded the 1st video stage cathode follower. That means the black level is being sourced at the detector to provide a higher percentage of DC level with less likelihood of voltage drift.

The CTC4, on the other hand looks to have the highest percentage of DC coupling straight thru.

So it looks as if the CTC5 of all the RCA color roundies and perhaps of any RCA color set has the worst DC restoration. But it looks like can be relatively easily remedied.

It appears to me the RCA Engineers discovered that there was insufficient DC restoration in the CTC5 luma path. In the CTC7 they made minor modifications to increase the DC coupling on the 0.1uF DC block.

The CTC5 thru CTC11 video amplifiers are similar. CTC5 and CTC7 are almost identical. My first experiment will be to replicate the CTC7 design in the CTC5. It will entail removing two capacitors and changing only two resistors.

Note below I have included three attachments below: the first is the original CTC5 design, the second is the CTC7 and the third the modifications I plan to make to the CTC5.

Note that the brightness control provides in each case a percentage of DC coupling bypassing the 0.1uF capacitor. The CTC5 only has a lower DC coupling percentage and the plan is to increase it. In the CTC5, I will simply remove the two 0.01uF capacitors which appear to be unnecessary decoupling and reduce R407 from 180K to 82K and R409 from 680K to 270K. This will reduce the effective resistance across the 0.1uF coupling capacitor to deliver the black level reference straight from the video detector (through the first stage cathode follower). I may lift R182 4700 (underlined) to see if it makes any appreciable difference to the DC coupling.

Before and after the modification I will take some photos of the vertical blanking bar with test video with a high contrast ratio (which would normally crush the blacks) compared against a black scene to see the effective change in displayed luminance of the blanking bar. I will post the photos here so you too can be the judge.

Note the brightness control bottom leg in both CTC5 and CTC7 receives its negatve potential tapped from the negative grid leak action from the Horizontal output tube grid. The circuits are essentially identical and since apart from the change in resistor values between CTC5 and CTC7 which was carried thru to CTC12 and beyond, I hope this will do the trick!

Note that there was a service bulletin from RCA suggesting rewiring the CTC-5 contrast control ala the CT7 also. I rewired mine and unfortunately got it backwards, so max contrast is CCW, but it's too much hassle to fix unless there's some other reason to pull the chassis.

Note that there was a service bulletin from RCA suggesting rewiring the CTC-5 contrast control ala the CT7 also. I rewired mine and unfortunately got it backwards, so max contrast is CCW, but it's too much hassle to fix unless there's some other reason to pull the chassis.

Sorry. But I found that funny considering the difficulty I had making extension cables and propping up the chassis so I could test it. I made a high voltage extension with a length of high voltage wire and connectors made from strips of metal from an orange juice can shaped to make a receptacle for the wire to the CRT and a plug to fit the socket at the HV rectifier. The receptacle is covered with multiple layers of heat shrink tubing. It works well with no corona discharge and allows me the flexibility to prop the chassis vertically to the rear.

Perhaps if there is sufficient improvement with the DC coupling it may justify your pulling the chassis again?

No need to be sorry, I laughed at myself when I did it.
Will be following your results closely.

Pulled the chassis tonight. Plugged in my extension cables and powered it up. A billow of smoke rose.

Hmmm. nothing appeared wrong apart from a 1k 1W resistor on the tuner went to cinders very quickly. could find no short. Tested C9 the 0.01uF ceramic decoupling capacitor. Tested fine with an ohmmeter. Replaced the resistor powered up and carefully watched. The short appeared again. Replaced C9 and all was okay.

The little cretin C9 will short when voltage is applied but tests fine with an ohmmeter.

Performed my first DC Restoration test. I haven't modified the video amplifier yet. I adjusted the vertical hold to roll the picture to use as reference the vertical sync bar. The percentage of DC restoration is inversely proportional to changes in the brightness of the bar. This in addition to studying the blacks in the picture. Eg: no change in the brightness level of of the sync bar = 100% DC restored.

Below are two photos:

First photo is a 10 step greyscale. This is the high contrast ratio image. Notice the blacks are crushed and the vertical sync bar is black.

Second photo is a black background with convergence dots.

Note the elevated brightness of the black and the grey sync bar.

This suggests that the DC Restoration is virtually non existent in the Y path.

Curious if you see the same with your CTC5.

Tomorrow evening I shall endeavour to make the modification to compare results.

I made the modifications indicated. That means changing the two resistors, removing the 2 x 0.01uF ceramic capacitor and the 4700 ohm resistor. (The ceramic capacitor and 4700 ohm resistor are located underneath the circuit board).

There is a noticeable improvement in the DC restoration.

I want to add that I attempting to "calibrate" my photos. I used as the reference luma step video from a Sencore VA62 and set contrast and brightness to match the convergence dots on black video. The convergence dot video was obtained by simply switching to it as a source on VA62 without touching the brightness. Note how the convergence dots video black level was higher before modification because very little DC level was permitted through. The after modification photo shows that the black level is held sown as it should.

Thinking about this through the day, the CTC5 was unique as it was really in many ways, the precursor of the the design of all the remaining color roundies to follow. Certainly the video ampliefier was essentiall unchanged in basic concept fro CTC5 through CTC12: cathode follower driving the delay line which inturn fed the penthode video out. The most obvious changes from CTC5 to CTC7 onward was both the contrast control and the brightness control.

The reduction of the two "bridging" resistors. The 0.1uF coupling capacitor is the sole DC block from video detector to CRT cathode. Following the CTC7 design, the resistors were reduced in value to make the total briging resistance less influenced by the 250kohm brightness control. The brightness control function is to add a negative potential derived from the horizontal drive.

I watched various video sources today with varying contrast ratios and I did find viewing more satisfactory. Whilst the DC restoration is not 100%, it appears a better compromise and certainly no longer suffers from the regular crushing to black.

I believe RCA engineers at the time must have recognized the flaw and make the revision to the next series.

One caveat: the 21AXP22 even when healthy does not yield a bright picture. No DC restoration benefits weak/dim CRTs when a bright highlight is to be reproduced. the lack of the DC component allows the blacks to be crushed before the whites in the display "bloom". So if the modifcation is made when using a dim CRT, you may find the picture blooming on some scenes.

I may try this on my CTC5 specials. I've already adopted the CTC7 adjustible HV reg circuit.

I spent the last few days working on the CTC5. The CRT is the original 21AXP22A with a date code of 56-30 ( 30th week of 1956). The emission on the three guns is good but the red gunntakes about a minute longer than the blue-greento come up.

The change made to the video amplifier to improve black DC level maintenance appears worthwhile. I ended up removing the 4700 ohm load and the ceramic 0.01uF capacitors as they appeared to do nothing.

I followed the RCA supplement procedure for greyscale. The CTC5 does not have The convenient setup switch. But the process from the manual works well. I will summarize it below.

For the Super chassis (without background control). This assumes the recommended contrast control modification has been done.
1. Set the three screen controls fully counter clockwise and the green and blue background controls backed off about 30% from full clockwise.
2. Unplug the tuner IF connector to provide a no video raster.
3. Set the brightness control so the the red gun grid bias is -70v.
4. Advance first the red screen control until a red raster becomes just barely visable.
5. Advance the green and blue screen controls so that the raster is dark grey/ black.
6. Advance the brightness control until the raster is bright without blooming. The touch up the blue and green background controls until the raster is white.
7. Reduce the brightness to the point the raster is about to disappear and touch up the blue green screen controls only for dark grey/ black.
8. Advance the brightness and touch up the blue green screen controls for white.
9. Re insert the IF link connector and reduce the color control to check the black and white video. You may wish to lightly adjust the green blue background controls for white highlights.

The deluxe chassis has a red background control. It's initial set up is as follows:
1. Set the brightness control to minimum.
2. Set the red background control for red gun grid bias of -90v.
3. Continue with the Super chassis instructions Step 1 above.

After some effort, the CTC5 is looking pretty good. The HV is 21.5kV and holding steady. (I earlier inserted a solid state HV rectifier which suffers less voltage drop and eliminates the filament load on the high voltage transformer.

Here are,a couple of snapshots tonight.