Can we discuss DC restoration circuits?

[miniman82]

Now that I've gotten my CTC-9 up and running pretty well, my thoughts have already turned to improvements in the circuit. DC resto comes to mind first, so I did some searching. Turns out there's an IC (LM2480 triple bias clamp) which will serve this purpose, but I have no idea if it could be adapted to my set. Problem lies in the fact that the IC does DC resto at the CRT cathode, where my CTC-9 drives the grids. So the question then becomes: is it possible to use a device like this on my set, or would the conventional diode setup still be the best choice?

http://www.national.com/ds/LM/LM2480.pdf

I never got this in depth with TV's before, so I'm gonna need some guidance on this one. Replacing caps and general troubleshooting I'm fine with, but I have no idea how modding the chassis in this way would turn out.

[ChrisW6ATV]

You picked a good time to start this topic. I have been wondering how to improve the picture on the RCA color TV I recently finished restoring. It is a GG589M with a CTC-16X chassis. It has poor or nonexistent DC restoration. I also have an RCA combo radio/phono/CTC-16 set that I thought had pretty good DC restoration, so I briefly looked for differences between their circuits and have not found anything obvious yet.

Depending on what comments other VK members have, I would rather avoid using an IC in an old TV set, but if a silicon diode would be suitable, that would be OK since they are commonly used to replace selenium rectifiers already.

There have been several discussions of DC restoration in older color sets in the last few years, including one that really improved a GE Portacolor TV if I remember right. That circuit used a diode and a couple other components, I think. It seems to me that the discussions have all been in the rectangular-screen forum. I will add links to anything I can find.

[miniman82]

I saw the GE Porta Color thread, and I'm thinking about doing it to my own '76 P-C set. Only reason I'm looking at an IC is pure simplicity; if people don't want to use it, they can always do the diode thing. I guess I just want an elegant (read:simple) solution.

[oldtvman]

Just remember I saw lots of guys try to make improvements in things and sometimes end up screwing them up. The engineers were pretty smart guys. You are restoring technology from the early times. And example would be trying to get a good picture out of any Motorola color television, impossible. Those sets were design the way they were and it would take an act of god to get them to produce a decent picture.

My suggestion would be to see if you can find any service bulletins addressing the issue. I fould service bulletins to be very helpful. And sometimes they kept techs from chasing around on very weird problems

[freakaftr8]

Ok, good topic. As Chris had mentioned, I have seen good DC restoration on a CTC-16, modt likely the one Chris had mentioned in his thread. I have a CTC-17 (Zenith CRT)with decent DC restoration and my other CTC-17 (RCA CRT) with non existant DC restoration. The first chassis I believe is an XAB and the other chassis is an XAD IIRC. Im curious as to what causes this as well. Were there modifications done at the factory, or was this pissibly a field service call somewhere later. There has been to circuit modifications I can see to the XAD chassis at all. I have two 23" Zenith consoles from 1967. One has decent DC restoration, and one has none. I dont recall the chassis numbers off hand, but the one with decent DC is an SC600.

[andy]

---

[miniman82]

Originally Posted by oldtvman

Just remember I saw lots of guys try to make improvements in things and sometimes end up screwing them up. My suggestion would be to see if you can find any service bulletins addressing the issue. I fould service bulletins to be very helpful. And sometimes they kept techs from chasing around on very weird problems

Well, it's not a 'problem' per se. Like I said, I'm looking to improve the picture quality of this already well running set. If I have to experiment on the Porta Color before moving on to the CTC-9, then that's what I'll do- for the experience I guess.

I just looked at the schematic for the -9, and it looks like it has video directly coupled from the video output tube to the picture tube cathodes, with chroma capacitively coupled on the grids from the R/G/B-Y amps. So does that that mean I don't need DC resto, or can I still do something? My understanding is that the signal on the cap coupled grids causes DC 'sag', which messes with the black level. Perhaps I could just use voltage regulator IC's to 'prop up' and stabilize the DC on the grids or something. Just throwing out ideas at this point.

[ctc17]

What exactly are you looking at in the picture to determine the quality of the dc restoration?

Maybe a weak crt is being interpreted as poor restoration?

[miniman82]

My CRT tests new...

As I understand it, during bright scenes any black (or otherwise dark) objects will become lighter. It's really obvious if you're watching something in letterbox format, because the black boxes at the top and bottom of the screen will get 'lighter' as opposed to staying black like they should. Really ought to be called black level restoration instead of DC restoration, that would make more practical sense.

[ctc17]

Weak crts do that too. I have had sets with semi weak crts and after replacement the blacks were black.

[miniman82]

Good to know, but I'm trying to focus more on the nitty gritty electrical aspects of this. I want to engineer something for my set, test it, and possibly come up with some kind of circuit that others could use. If this goes well enough, I may be able to offer tube based PCB retrofitting kits that people can buy.


Here's some reading, to get the juices flowing: http://www.scribd.com/doc/7351543/Crt

A snippet reads:

"DC restoration of the video signal can be done in two ways. The first method is to use a simple diode to clamp the
signal at the reference black level. Diode clamping can be done either at the AC coupled input of the preamplifier or at
the AC coupled output of the CRT video amplifier. The disadvantage of diode clamping is that the black level is sensitive
to fluctuations in the power supply as well as noise coupling and temperature drift of the diode's forward drop.
A more effective approach for DC restoration is to do a dynamic black level clamping at the back porch of each
video signal. This requires the use of comparator within the feedback loop of the CRT video amplifier and the preamplifier.
During the horizontal retrace period, the comparator compares the DC feedback taken from the CRT video amplifier's
output with the voltage set by the brightness control potentiometer. Depending on the CRT video amplifier's output
voltage, a clamping capacitor at the output of the comparator is either charged or discharged so that the feedback
loop is stabilized and the video signal is restored to the black level."


This would seem to suggest that the restoration method employed in a set such as the CT-100 is inferior, since it uses a diode:

http://www.antiqueradio.org/art/RCAC...nSchematic.jpg


Instead of doing DC resto in the amplifier section, they chose to do it individually for each color at the outputs.

Reflecting on the above article, I'm gonna say that using the IC I first posted about is probably not the way to go about this. I think a better circuit can be designed which could be inserted into the video amp section of a set like my -9, and it could be done cheaply. Thoughts?

[old_tv_nut]

Originally Posted by miniman82

My CRT tests new...

As I understand it, during bright scenes any black (or otherwise dark) objects will become lighter. It's really obvious if you're watching something in letterbox format, because the black boxes at the top and bottom of the screen will get 'lighter' as opposed to staying black like they should. Really ought to be called black level restoration instead of DC restoration, that would make more practical sense.

Ok. let's clear up some things;

1) lack of DC restoration means on a bright picture, the blacks get too black (your black bars in letterbox are black and dark objects go too black and lose detail) (also, colors get over-saturated); on a dark picture, the opposite happens - your black bars in letterbox turn gray (also, colors get washed out).
2) this problem is in the luminance circuit only - you do not have to worry about the color difference drive on the CRT grids. Chroma may appear to be AC coupled when you look at the circuit, but it is not. There is a DC restorer action in the outputs so that the color balance does not drift depending on scene content. (Imagine if having a large red object in the scene caused the color to shift over-all toward cyan - this would be the color equivalent of the lack of luminance DC coupling, and would not be acceptable.)
3) different sets have different amounts of DC coupling depending on designer's choice
4) the amount of DC coupling can vary in some designs depending on the settings of the contrast and brightness controls

5) Adding DC restoration is a worthwhile experiment to improve picture quality

[ChrisW6ATV]

Originally Posted by miniman82

As I understand it, during bright scenes any black (or otherwise dark) objects will become lighter. It's really obvious if you're watching something in letterbox format, because the black boxes at the top and bottom of the screen will get 'lighter' as opposed to staying black like they should. Really ought to be called black level restoration instead of DC restoration, that would make more practical sense.

Your description is correct for the problem. This video parameter has been called black-level retention as well as DC restoration.

Ctc17 may have a good point about CRTs, but maybe related to brightness/contrast or setup adjustments in general, and this could help explain some of our similar models with and without decent restoration. I am going to try my CTC-16X chassis on my test jig and see if I can affect it by trying different settings of the CRT bias switch among others. Years ago, I had a Sears portable color TV with lousy black-level retention, but I could make it pretty good if I cranked the contrast way down. (I had to have the room lights off and it was essentially unwatchable, it was so dim, but it really did keep its black level consistent within that low range of contrast.)

[old_tv_nut]

Originally Posted by ChrisW6ATV

Your description is correct for the problem.

No, it's stated backwards, as I explained.

Originally Posted by ChrisW6ATV

Years ago, I had a Sears portable color TV with lousy black-level retention, but I could make it pretty good if I cranked the contrast way down. (I had to have the room lights off and it was essentially unwatchable, it was so dim, but it really did keep its black level consistent within that low range of contrast.)

If you look at the CTC-7 contrast control circuit, you can see that it changes only the AC gain of the video ouput. AC gain is greatest when the contrast is set to max and a large capacitor shunts the video amplifier cathode resistor. Therefore, the DC coupling percentage is greatest at minimum contrast (where the AC gain is small and closer to the small DC gain). Probably something similar was involved in your Sears portable.

[miniman82]

Studied the -9 schematic some more over the past few days, and made some realizations.

1. the video output tube is DC coupled to the CRT cathodes
2. the video output tube is AC coupled to the video amp tube via .1uf capacitor, with DC bias being set by the brightness control
3. between the amp and output is where the DC resto of the signal needs to happen

I going to be looking into this more in depth in the coming weeks, so if anyone has circuit specific input I'd love to hear it.

From what I've read, the black level must be reset before the beginning of each horizontal scanning line, and it's usually done during horizontal retrace. This is what makes the diode a bad choice, because it clamps the entire signal (including any sync pulses) to ground. What this means is that with a diode, the black level is being inluenced by the amplitude of the sync pulses, so you are not seeing 'true black'.