Sony 9-51 with vertical problems

[AlanInSitges]

Hello everyone,

I picked up a little 9" Sony set from the mid 60s at a secondhand store the other day for 20€. Before I bought it I turned it on briefly and it powered right up save for a very messed up vertical sweep - about a dozen broadly spaced lines in the center of the screen that got closer together as they reached the bottom. I figured great, an easy recap and I'll have a cool little Sony.



Well once I cracked it open I was surprised to see that every single electrolytic from the vertical oscillator to the vertical output had already been changed. My guess is that when it went bad the owner took it to a repair shop who went through changing those caps, got to the point where he didn't know what was wrong, and told them it couldn't be fixed. Shit. As a bonus it's missing about 60% of the screws that hold it together inside.

I've got the Sams, and gone through and checked voltages on every transistor from oscillator to output - they are right on the money. The yoke ohms out as it should according to the Sams. Vertical seems to be on frequency because when I feed it a signal and adjust the V hold I can watch it roll and lock just as it should. It's just that there are only about 20 scan lines on the screen (they move a bit when adjusting VSize).


(click to embiggen)

I decided to have a look at the waveforms and here's where things got confusing for me: starting at the collector of the vertical output and working my way back:

- V out collector: waveform looks pretty close; it's about 10% low on amplitude


- V out base: waveform looks pretty much the same as what's on the collector, which it shouldn't. Amplitude is about right, a bit low.


- V driver base: waveform looks pretty much the same as what's on the collector, which again it shouldn't. Amplitude is 50% low.


- V osc collector: waveform looks nothing like what it does in the Sams. Amplitude is correct.


Here's the most confusing part: Sams says the waveforms are taken with 30Hz sweep on the scope. My scope has s/div so using the cursor function I selected a single cycle of the waveform and it tells me that from driver base to output collector all the waveforms are aprox 480Hz. Stick a pin in that for a second, and going back one more stage to the vert oscillator I check the collector and the waveform is a mess but this time it says they are at close to the correct frequency.



I've been reading up on blocking oscillators in Gulati's book and have to confess I wasn't able to retain enough to apply it to what I'm seeing in the Sams.

So, questions:

Is this 480Hz thing a wild goose chase that I'm on because I don't know how to use my scope correctly? If it's not, what could possibly be going on in that circuit to cause the frequency to multiply 20X between the vert osc and the vert driver? And if that's what's going on is it safe to assume that's the reason for the malfunction seen on screen?

If on the other hand it is a wild goose chase would someone please by kind enough to take me by the shoulders and gently point me in the direction I need to be looking.

Thanks.

[old_tv_nut]

Vertical oscillator frequency is far too low at 19 Hz. - should be 60 Hz.
Everything following that seems to be self oscillating at 480 Hz.

Because there is feedback involved, the problem could be anywhere in the feedback loops. The only thing I could suggest is to check components from stem to stern -
check all those electrolytics from C24 to C28 and also verify that the previous repairer didn't install any backwards; check R77, R5, R78; check T2 for open windings.

[old_tv_nut]

Sorry - didn't pay attention - it should be 50 Hz instead of 60 in your case (but still not 19).

[old_tv_nut]

Some more hints:

Try looking for cracked solder joints.

Look on both sides of each electrolytic capacitor with the scope. You should have 50 Hz waveforms on both sides. If a cap has 50 Hz (or 19 Hz) on one side and 480 Hz on the other, it must be bad (or have a bad connection).

[AlanInSitges]

Thanks for the responses!

I shouldn't have trusted the previous recap, one of the no-name chinese capacitors C27 was open and C25 was leaky. There's still a linearity problem but I know where to look. It also looks like there's a bad filter; one of the 500uf has been replaced with two 470s in parallel...not sure what's going on there.

Anyway, I'm on my way. Thanks again!

[AlanInSitges]

I finally solved the linearity problem! After replacing all the electrolytics in the circuit i still had a bar about 1/4 way down the screen where the lines were all scrunched together. There was only one capacitor left, a .068 in what looked to be an electrolytic aluminum housing that was marked .047 (I figure there will be a few value differences between the Sams and the one they sold in Europe). As I moved it aside to get a look at it, the bottom lead popped right off, it was corroded all the way through. A new .047 later I've got perfect linearity.