How to eliminate horizonal overscan when no service adjustment?

Usually that's precisely how it's done. Look for an inductor in the horizontal section with an adjustable slug, it may have enough range to do what you want without even modifying anything.

I don't see anything adjustable. There's a big IC that produces an "HD" output, which goes to the base of the Horizontal Drive transistor (NPN), the collector of which connects to the Horizontal Output transistor by way of a transformer. But there's nothing on the schematic suggesting anything is adjustable.

Any idea what kind of coil I would need to add? How many turns? What gauge wire?

It's usually a coil about 1/2" diameter and an inch tall wound with what looks to be around #14 Litz wire in series with the yoke. If you can find an older junk TV you could probably salvage something.

I cant imagine any mfr producing everything in the chain to tight enough tolerance to not need a width control.

The IC that produces the horiz scan waveform, are you really sure theres nothing to control it?

NT

fwiw I cant think of any way crt ageing woud affect pic width.

Is there not a set HT pot in the power supply ? Usually, turning this down by a few volts will shrink the picture all round. You can then just correct the vertical via the menu. I would not personally try adding inductance into the H-OP stage or scan coil path. This is a very stressed and very carefully designed bit of circuitry, and some extra L that it doesn't want, could lead to all sorts of unwanted effects such as ringing, poor linearity, or even overheating and ultimate failure of the H-OP transistor. If you were to start adding L, then you would probably also have to alter the value of the S correction cap in series with the scan coils, as this forms a tuned circuit with the scan coils to deliberately make them ring in order to achieve the velocity modulation of the beam required to get a linear sweep on a flat(ish) tube face. I'm actually surprised that you feel that you have that degree of overscan that it's affecting your viewing. If you really are missing actual picture, then it must be significantly more than the 'normal'

7 - 10% raster overscan. Remember that the broadcast picture is not as wide as the scanned raster, and a lot of the relationships that we used to use to judge picture width, have now gone out of the window, with all the different 'widescreen' formats that they keep broadcasting now, and which many older TVs just don't look good on. Have you actually looked at the picture on a proper 4:3 test pattern ? It's the only way really to make a valid judgement on the picture geometry. You may of course have an actual fault. It's not at all uncommon for the values of the components in the set HT circuit to drift, resulting in a slowly 'growing' scan.

Arfa

I'm not really sure of anything. :-)

The IC that provides the signal to the Horizontal Drive transistor is a TA1242N, 56 pins. It's pin 32 of that chip, labeled HD. The waveform shown on the schematic shows it going high (3.5V) for about 40% of the time, then back low for the remaining 60%. I don't know if that's what you mean by "horizontal scan waveform." It really looks pretty much like a square wave. By the time you get to the Horizontal Output transitor, the waveform looks completely different.

The only datasheet I've found for the TA1242 is in Japanese, but there is a logical diagram of the chip in the schematic. There are inputs that it appears my have some effect on that pin 32 output, including 32fH, HP, XRAY, and a number of others that aren't labeled.

I just don't have enough information to answer your question. There doesn't appear to be anything adjustable in this part of the circuit, although there may well be parts whose value could be changed to affect the waveform. I just don't know.

Square wave drive is correct. It's integrated into a sawtooth scan waveform, by the inductance of the scan coils.

Arfa

I have a fair amount of experience with digital circuits, but not analog, and particularly not television. So I know you'll understand when I ask:

Is HT the flyback transformer? If so, are you talking about the source that supplies the primary of that transformer as well as the deflection coil? It's 134V on the schematic.

If so, that is controlled by a STR30134 regulator. There are no pots there, but there is a pair of fixed resistors as a divider that appear to control the output voltage, and I'm sure I could insert an adjustment there if that would do it.

It has been this way since it was new, so it's not an aging problem. I've always had the sense that I was missing part of the picture, and when I got a DVD player that also lets me zoom OUT, the difference became apparent. It may not be more than 10%, and perhaps others would not object to it, but I would just like to see what's actually in the picture, rather than a cropped version.

No, I don't have any service equipment beyond a multimeter. The problem shows up more with the DVD player than on broadcast, particularly when playing things like mpeg1 videos that I've downloaded. Probably because in that situation, the full raster is used.

I don't think so, and I don't have the sense that it's getting worse. I think it's just the design of this set. But if it's possible to fix it, I would like to.

Thanks very much for your help.

"Set HT" refers to the 134v. Some manufacturers call this the 'HT' and the output of the flyback transformer 'EHT' or 'HV'. Others call the output of the flyback simply 'HT' so there's no 'definitive' answer to that question, but it all boils down to the same thing in the end, and this variation in terms is just accepted. Altering the supply voltage to the H-OP stage - ie making the 134v something different - will affect the overall raster size, as well as affecting the high voltage output from the flyback Tx. Tv sets normally have an overvolts protection circuit, to stop the flyback voltage becoming excessive, as this could result in the CRT emitting x-rays beyond the maximum allowed limit. I think that before getting complicated with adding L and so on in the deflection current path, I would first try making the 134v slightly adjustable, and just seeing whether that does it for you.

Arfa

And there are other supply voltages provided by the secondary of the flyback. In my case a 197V and a 25V, plus some more that are regulated. So I assume I'm not going to be able to change it much before things really start to go wacky. But then, just a little may be enough.

Well, I can certainly make the 134V adjustable. I'll do that and see what happens.

Thanks again for your help.

Best laid plans.....

I found the data sheet for the regulator, and it turns out it isn't adjustable. Hence the name STR30134 for 134V. The voltage divider I found that feeds pin 2 just provides a source to drive the darlington base, but it is further clipped internally. I would have to reduce the voltage there below the clipped value to affect the output voltage, but then I would probably lose all regulation.

There is already a fixed 1500 ohm, 3W, resistor between the

134V supply and the deflection coil. What about making that a higher value?

Or, inserting one or more forward bias diodes in series at that point to drop the voltage about .6V each. Maybe even a

1N400x would work ok at this frequency.

I still think adding inductance to the horizontal yoke circuit is the easiest way to go. I first saw this method described for converting a B&W TV to a computer monitor in a book from the late 1970s, monitors designed for use in arcade games use the same method. If you go changing the voltage around it will affect other things.

On Wed, 06 Jun 2007 18:28:48 -0500, Peabody put finger to keyboard and composed:

Assuming your TV is faulty, I would check the tuning capacitor attached to the collector of the horizontal output transistor.

- Franc Zabkar

--
Please remove one 'i' from my address when replying by email.

Below is the circuit for the horizontal deflection. (You would have to view it using a non-proportional font.)

When Q522 turns on, it grounds everything - the yoke circuit and the flyback primary in parallel, both fed by the 134V supply.

Does it matter where I insert the coil? I had thought maybe immediately before or after R526, but I'm not sure if it matters which one.

I assume the value of L521 should not be messed with.

Placing it immediately before or after the yoke would require cutting one of the wires going to the yoke, which I would rather not do.

Thats not going to work. HT voltage affects both scan voltage/current and EHT voltage in equal measure, so it has no effect on picture size. This is why old TVs with no HT regulation stayed at more or less the same picture size as the voltage wandered all over the shop. The only thing that altered pic size on those was loading of the EHT, which depended on total picture brightness.

If you've played with tv psus you'll be aware of how much one can 'get away with' when it comes to HT lines.

Boosting HT was formerly used as a last ditch method to get more emission from dying tubes. It works, but is not recommendable.

If you want to adjust HT V to tweak scan, you'd need to just adjust the voltage to the horiz scan output stage. That could be done with a string of diodes, each of which will drop around a volt IRL. Due to lack of cct details, wise to add an ultrafast diode pointing the other way across your diode string, and maybe an RC too.

NT

What you say, is of course, fundamentally true, but I think that the key here is your "more or less" statement. I've never known the relationship between scan size and HT setting to be so tight as to maintain the picture size constant. With most sets that have adjustable HT, altering the setting will have some effect on the picture size, and it was just something that the OP could have tried as perhaps a 'simple' fix for what appears to be a fairly minor amount of overscan that he's trying to correct. As it happens, the HT is not adjustable on his set, so it's a moot point anyway.

I think that a couple of diodes in the HT feed to the OP stage, as you suggest, might be a simple way of giving it a try.

Arfa

On Sat, 09 Jun 2007 08:20:53 -0500, Peabody put finger to keyboard and composed:

That doesn't look right. The LH end of the yoke normally goes to ground via another capacitor (not the tuning cap, C524).

- Franc Zabkar

--
Please remove one 'i' from my address when replying by email.

I left out one cap (C526) back near the 134V source. So it should look like this:

|---------------FLYBACK---------| | | | | | |

134V--|--R526-|----L521------|--YOKE--|---Q522---|--GND 1.5K | | | H.OUT | | | | | | | | | |--R530--C527--| |---C524---| | | C526 | | GND

But somehow I don't think that's what you had in mind. The "bottoms" of both the flyback primary and the horizontal deflection coil are tied together as shown. They go to ground through the low-impedance path to ground when Q522 turns on, and through C524. at other times.

And where would you put these diodes so that nothing "tuned" would be screwed up? With reference to the diagram:

|----------------FLYBACK---------| | | | | | |

134V--|---R526-|----L521------|--YOKE--|---Q522---|--GND 1.5K | | | H.OUT | | | | | | | | | |--R530--C527--| |---C524---| | | C526 | | GND

It would seem that putting one or more diodes immediately to the left, as shown, of R526 would drop the voltage only through the deflection coil circuit, not the flyback.

Then the question would be what kind of diodes to use for the voltage drop, and what to use as a reverse protection diode, if any. Where I live now, I'm pretty much limited to Radio Shack, my junque box, or mail order. RS has signal diodes like the 1N914, Zeners, rectifier diodes like the

1N400x (1 amp) and 1N540x (3 amp),and a 1KV, 2.5 amp PTC205, whatever that is. I assume only the signal or zener diodes would be at all fast.

It's not clear to me that there will be any spiking at that point that needs to be protected against. By the way, R526 is a 3W resistor. I don't know enough about TV's to guess how much current goes through the horizontal deflection coil, but all the parts in this part of the circuit have pretty hefty ratings.