The reason you are losing sync is because when the brightness is turned up high or the contrast advanced, you are clipping the sync with the diode! I suspect the sync take off is at the plate of the video amplifier.

Compare your circuit with the RCA 630ts and note R148 which will alleviate the sync clip.

It is a mystery how it works all unless the diode was installed opposite to how you depict it. The CRT electron current flow in the original design is from is ground through the brightness control up thru R32 the 150K resistor to the cathode of the CRT. The diode depicted in your drawing would obstruct the beam current and there is no other path unless C19 the 0.05ufd coupling capacitor is very leaky!

Agree with the possibility of clipping the sync. Adding a series resistor as in the tube circuit would fix this.

Disagree with the "can't work" statement. The added diode is in the right direction for restoring DC at the CRT cathode. The sync tips are positive here, and C19 is charged by diode conduction during sync, then discharges through the CRT during active video. In the case of grid drive, since there is no grid current during active video (or at any time) to discharge the coupling capacitor, a resistor is added across the diode.

Investigation of the charge/discharge time constants with a scope by looking for tilt on the cathode waveform and sawtooth on the added 0.1 cap would help optimize the capacitor values.

I don't think I've ever seen a textbook or manufacture example of a diode in conjunction with the video going to the cathode. The diode orientation is confusing because early on there was a double standard depending on function of the circuit. I always assumed that came about because of conventional vs electron current flow.

I agree. Why I suggest it can't work or at least the concept is faulty is because there is no path for the CRT beam current. The diode is in fact reversed to the flow of beam current. This may explain why the clamping diode was only used for grid driven CRTs.

The path for the beam current is the coupling capacitor. Electrons are drawn into the coupling capacitor throught the diode during sync pulses, then flow from the capacitor into the cathode and the CRT beam during active video.

On average over time, the charge drawn through the diode equals the discharge through the beam current.

Note: I have built DC restorers that work with a diode at the base of a transistor amplifier. They work just fine, even though there is base current that must be supplied. Current flows into the input coupling cap during sync and out of it into the transistor during active video.

I understand it could work. But how do you set up the quiescent point? If there is no video, how do you control brightness? The capacitor and diode block the beam current so with no video the screen is black.

That is why, perhaps facetiously, I suggested perhaps the coupling capacitor has not yet been changed, was leaky, conveniently providing the quiescent current.

Even if the video is totally black, there will be sync pulses, so I'm not sure I understand the question about "if there is no video..."

If the picture is black, and the brightness control has been set so that the CRT draws no current at black, then the diode will not supply any current during the sync pulses, and the cathode voltage will remain at the same voltage as usual.

But, you have pointed out that the CRT is a non-linear load (very non-linear when the current goes to zero), so this results in a variable time-constant for decay of the capacitor voltage, hypothetically infinite if there is no leakage whatsoever. So if this results in odd transient behavior (for example, when the picture suddenly cuts from bright to black or vice-versa), it can be reduced by putting a high value bleed resistor across the diode, just as is done in the grid-drive case.

You have made me realize that in the case I mentioned of a transistor amplifier, it is biased such that the current never goes to zero, so such problems cannot occur.

Reading your post again, it is not the capacitor and diode that block the beam current, it is the CRT. If/when the CRT conducts beam current, the diode and capacitor will supply it.

I apologize for the awkward/ miscommunication here. But the CRT DC quiescent beam current used to flow thru the resistor from the brightness control center tap to the crt cathode. With this resistor removed and replaced by the diode, note that the current flow is now blocked by the diode because it is reversed biased. The coupling capacitor cannot pass DC. The circuit is therefore indeterminate and depends on component leakage currents to sustain the CRT current.

The CRT beam current is generally quite small at about 1ma maximum. But removal and replacement of the resistor not only makes the CRT bias determined by undesired component leakage but also will lead to clipping of the video at high amplitude. As there is minimal or no proper DC return path it will behave like a grid leak detector and effectively clip the video sync.

Look at the grid-drive version. Without the resistor in parallel with the diode, it is similarly indeterminate. But, that resistor is not supplying current, it is draining it, just as the CRT does in the cathode drive circuit. The cathode drive circuit is not indeterminate if there is some beam current, only when the picture is completely black. Fixable the same way as the grid crive circuit with a resistor across the diode to drain a little current when the CRT does not.

The diode does not block the average current, it supplies it as a pulse of current during sync, putting a charge in the coupling capacitor that then supplies the beam current during active video.

Think about the flyback high voltage supply: same thing. The HV rectifier only conducts during retrace, but charges a doorknob capacitor or the CRT dag, which supplies the beam current during active video. The rectifier does not need to conduct during the active scan while beam current is supplied. In the case of the HV supply, the other end of the capacitor is connected to ground, so its voltage slump between pulses is in reference to a DC level. In the case of the DC restorer, the other side of the coupling cap is not connected to ground, but is connected to the video output. So, it holds a fixed (slightly slumping) DC difference between the video output and the driven CRT element. That DC difference is reset to the desired value every time the diode conducts.

Summary: a DC restorer is like a power supply where the bottom of the filter cap is connected to video (which is the source of the charging pulse) instead of ground.

A DC restorer is the same as a power supply circuit, with the ground moved from the usual place. The pulse still appears across the diode, the cap still holds a DC voltage across it. It's just the reference point for voltage readings to chassis that has moved.

Pulsed current comes in, but continuous current can go out.

Would you consider drawing us the circuit as you see fit for the set Max had asked about?

Was working on an explanatory diagram while you posted.

https://live.staticflickr.com/65535/...abd14610_o.jpg

Of course, you want the sync tip voltage to be adjustable, not tied permanently to ground (zero volts) as shown in this simple comparison diagram, so in the actual circuit you tie the diode to the adjustable brightness control.

If drawing zero current through the CRT during black scenes is a problem, you can add a high resistance load to draw a little constant current.

Wow, there is a lot of interesting discussion to catch up on here! I really appreciate all the effort being put forth.

In the meanwhile, I found this thread on another forum which is quite interesting

https://www.vintage-radio.net/forum/...d.php?t=145083

The circuit in post 10 looks interesting, because the cathode follower would entirely isolate the diode from affecting sync.

Here is how I would revise your circuit. The the upper resistor should eliminate the sync clipping and the 680K diode bypass resistor will limit the clamping and can be varied to the degree of clamping required.

It occurred to me afterward that because the CRT is conducting, the diode remains forward biased even with no sync or video. Nevertheless the bypass resistor should make the brightness level less likely drift over time.

I guess all of us are suffering cabin fever!

You almost made it look like part of the original schematic! Thanks for taking the trouble! I will give it a shot, if it works well this is sure an easy modification to add.

I had a few minutes free so I installed Penthode's modification. I'm going to call it a success, the black level seems consistent from bright to dark scenes, and the sync interaction is absolutely gone. I'd be curious what others think of it, if anyone wants to try with their TV. Modern program material like DVD video appears to have quite a high contrast range, which really exaggerates the issue caused by lack of DC restoration, I think this was a very worthwhile thing to do.

With the exception of the 680k bleeder across the diode it seems we're right back where we started from. ;)

http://videokarma.org/attachment.php...8&d=1584623306

http://videokarma.org/attachment.php...9&d=1585867061

Great!

I'm curious to know how much the voltage on the wiper of the brightness control (and on the 0.1 uf cap) changes between a bright scene and a dark/black one. To be clear, I mean with the brightness control set for normal operation and not moved. Is it still accessible for a quick check?

Yes I am curious about the wiper position as well. My guess is the clamp will raise the sync tips to the average level pushing the brightness control towards the +250v.

I think the 150k resistor is what is preventing the diode from messing with sync. The picture now does not tear sideways when there is a picture with certain characteristics.

I cannot easily access the control.with the set together to measure voltage, this TV has a fairly awkward chassis to remove from the cabinet. I actually had it on all morning, and it plays quite nicely now.

But can you say if the brightness control is near one end or the other? That is assuming before the modification the control was roughly central.

It is turned up almost all the way, but that was also the case before the modification. I think it has slightly less range on the high end now though

That makes sense. The CRT is closer to cut off and the wiper is at the ground end. Placing a little plus DC bias on the CRT control grid will centralize the control. The diagram is cut off but I suspect the grid goes to ground through a resistor with a coupling capacitor from the vertical output transformer feeding it?

Placing a single additional high value resistor between the +250v rail and the CRT control grid will form a resistive divider which should apply a positive bias of roughly +20 volts on the grid. This will bring the brightness control closer to the center of its range.

I wouldn't bother changing anything if you have sufficient range now. The reason I asked about the voltage change with scene variation is that it represents a decrease in the accuracy of the DC restoration if the voltage varies. There should be less variation if the wiper is nearer one end than the center.

I have this Setchell Carlson P65 I restored several years ago that suffers form this same issue. Same basic circuit and CRT type as your TV, although it is a transformer powered set. Sams shows 50vpp video going to the CRT. If I recall correctly I can easily operate this one outside the cabinet. Hopefully I'll be able to take some voltage measurements to see how closely this regulates sync pulse level.

https://live.staticflickr.com/65535/...f4d06f59_z.jpg

https://live.staticflickr.com/65535/...686efa76_z.jpg

^^That's a very nice looking TV

Got the chassis out and warmed up. The video source is a Pioneer VCR/DVD combo with internal rf modulator playing a DVD. The 2nd and 3rd pictures are with the screen brightness set in the middle, contrast about two thirds up. The last picture is what this set has always done between scenes, goes to medium grey with retrace lines. I suppose that happens because of the lack of DC picture level? Have to take a break but hopefully I'll get back to this later this evening. Max, do you recall what type of diode you used?

https://live.staticflickr.com/65535/...819787a4_b.jpg

https://live.staticflickr.com/65535/...3bca818b_z.jpg

https://live.staticflickr.com/65535/...63577131_z.jpg

Thanks. They are quite common around here as I live about 100 miles from where they were manufactured.

I used just a regular 1N4007, because I have a bulk pack of them. Just make sure whatever you use has an adequate voltage rating

OK, as a baseline I get about 1.5 volts of average DC variation at the output of the detector diode, from bright to dark scenes. When I say average, I mean that's what my scope is telling me when I look at the composite signal at the detector output. Then at the cathode of the CRT, without the restoration circuit, I see about 10V DC in variation, but that appears to be faster transients. With the restoration circuit in place I only see about 5V DC change. My first thought was that that small variation can't be very useful restoration, but then I don't know what would be considered acceptable. This was while watching my first season Gilligan's Island DVD. Right now I'm having a hard time convincing myself it's beneficial. I need to spend some more time with this as it's much harder to analyze with the scope than I anticipated. It's possible I'm one of those individuals that's not sensitive to background level. :scratch2:

Try watching something more recent, which was filmed at night. Old TV shows were made for old low contrast TV sets, so they probably will show the least benefit.

Good idea. For now I came up with an alternative that I think is valid. I figured my B&K 1077 would produce a representative light and dark raster that should allow me to take steady DC voltage measurements at the brightness pot wiper. B&K rf output connected through the tuner of the TV. The DC V measurements were taken at the brightness pot wiper which is also the cathode of the diode and top of the .1uf filter cap going to ground.

1. The B&K target measured about 90V DC.

2. Bright raster with no slide measured about 100V DC.

3. Black plastic in place of the slide produced 82V DC.

So a total DC swing of 18V at the wiper.

It clearly shows the diode is in fact rectifying the sync tips. How useful that amount of restoration is I have no idea. To me, visually it's a subtle effect, but like Max mentions I need to find some much better night video to put this to the test watching program material.

Disclaimer:

It's about impossible to take a picture of a dark raster, so the last image is fudged a little. But there were no controls adjusted on the TV while measuring the 3 DC voltages.

https://live.staticflickr.com/65535/...944b8018_z.jpg

https://live.staticflickr.com/65535/...16713609_z.jpg

https://live.staticflickr.com/65535/...b23d5a7f_z.jpg

This is how my TV was modified for the above measurements. And I just realized I didn't have the bleeder resistor across the diode when I did this. Not sure what if any effect that will have on the measurements. I had one in there early and didn't see much difference and then forgot to put it back. :scratch2:

https://live.staticflickr.com/65535/...44b2989b_z.jpg

I think you misunderstood. In an ideal case, the brightness voltage and voltage on the 0.1 uf cap would be completely steady, so the sync tips would be clamped to a constant voltage no matter what the video content is.

The indication of the effectiveness of the DC restoration is the DC voltage measurement at the cathode. With no DC restoration, the change from bright scene to black scene will be minimal - it will just be set by the brightness pot. But with the DC restorer, the DC voltage will be high on the black scene and lower on the bright scene. Looking with the scope, set to DC coupling, the dark parts of a bright picture will be the same voltage as the black of a black screen (or almost the same, depending on how complete the DC restoration is).

Looking at your screen pics, you can see plainly that with the test pattern, the blacks are properly black, and with the black image they are still properly black, not gray, so the DC restoration is doing its thing. If you get an old movie that has spooky night scenes where nearly everything is dark, the picture will actually be dark like it should be, instead of foggy medium gray. This will not be visually subtle!

Your measurements at the wiper would indicate that the black level probably shifts about 8 volts when going from bright test pattern to full black. This is small compared to the full 50 v p-p video signal, so is much better than the non-DC- restored original.

If you look with a DC-coupled scope at the cathode, you will be able to see how little the black level shifts compared to the huge changes without DC restoration.

Thanks for explaining in greater detail. I installed a 1 meg bleeder across the diode which resulted in less change at the brightness wiper, and the voltage change at the cathode being greater. Seems like what was lost at the brightness wiper was added to the cathode. I find it very difficult to monitor the black level of a playing video on the scope. Using the B&K test pattern was the only way I could see it change, or not.

I believe a portion of this lack of black level comes from some sets having lower B+ headroom. The B+ on this TV I'm experimenting with has close to 250v vs Max's set at 150v. My contrast control has a range that allows high black level's even without the dc restoration circuit present. I find that most video I watched growing up was fairly consistent within a given program. I suspect there was some type of video compression used during the production process back in the 60's though 70's. I can basically set it and forget it. I'm not sure if modern day video content would be worse in that respect. Honestly I don't have much desire to watch new programing on my vintage sets. I do find the audio processing on today's movies quite poor in comparison to older material. Or possibly there's a trend back towards less processing? With my ageing ears I struggle to hear the quieter passages, and soon after being blown away. IMO too much dynamics. Similarly I don't see the range of video dynamics in nature that we expect to see from from our television sets. When ambient lighting is low the human eye tends to see less contrast.

I think the human eye sees a far greater range of contrast than vintage TV sets can display. In old TV shows, it seemed common to light up night scenes quite brightly by today's standards in order to get any image on the film, but the more modern trend seems to be a huge range between the brightest and darkest scenes. I would say it is almost analogous to the difference between the noise floor and the loudest possible recording on a 78RPM record versus a digital recording.

This large contrast range really asks a lot of old TVs, and that's why I wanted to try and do something to improve the Marconi, if only slightly. Better yet it seems to me would be some kind of signal processing to dynamically compress the contrast range, like say using a variable mu tube as the video amplifier, to have the effect of turning the contrast down in dark scenes, but I'm afraid I'm not smart enough to design that.