Sony TV9-306UB

[Studio263]

Forum member Bobby Ball very kindly gave me a Sony TV9-306UK recently; it’s a seldom-seen set of with only a few have passed through my hands over the years, all quite some time ago. The model was announced in the October 1965 issue of ‘Television’ magazine, if you think that the Japanese made only cheap rubbish at this time then the cost of this set will surprise you: 85gns (plus another 12gns if you wanted the rechargeable battery as well) when for just 55gns you could have had a GEC Crystal 13, also a dual standard portable but valved and with a larger 13” tube. Meanwhile 75gns would give you a choice of 23” dual standard sets complete with UHF tuners - you really had to really want a very small set to even consider buying the Sony.

Much of the circuitry in this model is similar to that of the better known TV9-90UK but the cabinet is considerably smaller and made of very brittle plastic which by now usually shows some damage. In the case of this set the cross-bar of the handle has disappeared, as have the little feet underneath that stop the set toppling over when the back is off. I still consider it a good example though; others which I’ve seen have had huge cracks and holes in them, no wonder they changed the material to something more resilient for the TV9-90UK! The tube is also the same, but the chassis is a single vertical metal sheet to which the various panels are attached. It all looks quite nice but there is a lot of wiring at the back which makes doing anything a bit of a chore. Also there is no screen printing to show the component numbers, so you really need the service sheet (or R+TV S 1967 > 1968) for anything but the simplest repairs.

My experience with TV9-90UKs is that they nearly always work “as found” or only suffer from the most trivial of faults. This has not been my experience with TV9-306UBs, which all seem to have at least one baffling problem that drives you right round the bend. Another issue is that a detachable power lead is used, which not uncommonly this set lacked. Sony seemed to delight in producing a different type for everything they made at around this time, although the TV9-306UB does use the same type as the TC-110A cassette-corder (not the TC-100 though, that’s different…). Luckily I have some NOS ones tucked away, so the first thing to do was to plug the set in and see what happened. After quite some time a dull raster appeared at the top of the screen, the top was expanded and the bottom all cramped up at around the centre line of the tube. There was no obvious modulation, only a weak hiss on the sound and a noticeable lack of width with mains ripple at each side.

One silly aspect to the design of these sets is that there are two low value capacitors which go from each pole of the mains inlet down to the chassis. Unlike the TV9-90UK the chassis isn’t earthed, so if one of them goes short circuit there’s a good chance that the chassis, knob spindles, aerial socket etc will become “live”. Even if they are perfectly OK the chassis will tend to float up to half mains voltage, something that you won’t necessarily notice but it can be enough to ruin the transistors in the IF strip if you touch them with a ‘scope probe whose grounding clip has become detached. The only sensible thing to do is to cut them both out and replace them with a larger X2 type across L and N, so that’s what I did.

Looking at the picture again it struck me that low HT could be the problem. These sets use an unregulated transformer supply which is quite sensitive to changes in loading, and sure enough the 12V line was about 9V. To my surprise, plugging in an aerial resulted in a weak picture with reasonable line lock and two frames visible, one above the other. Fiddling with the frame hold control only made this worse, so it was obvious that the frame oscillator was running too slowly. Most Sony sets of this era use a blocking oscillator and aside from the resistors in the hold control circuit (which checked out OK) there isn’t much that can go wrong that just lowers the frequency whilst leaving the sync unaffected. I tried changing the oscillator transistor (I used an AC127) but that made no difference, so I wondered if the low HT supply could be causing the problem. A good dodge here is to unhook the tube heater but to hold the wire in place until the tube warms up, then remove it and see what happens. In my case the HT went up to a reasonable 11.5V but the frame problems remained as the picture slowly faded away, although the width and ripple situation improved markedly. Time for a think.

When servicing a modern set one normally thinks of electrolytic capacitors falling in value as they dry up but this would tend to make the oscillator run faster. Leakage would have a similar effect, as would shorted turns in the blocking transformer. In an act of desperation I removed a few of the electrolytic capacitors which could have a bearing on the frame rate and found them all to be considerably greater in measured value (using a Fluke 79 III meter) than they should have been! With little hope, I changed a few around the oscillator and was rewarded with a properly locked but still distorted picture, well I never! The HT was now nearly 12V so the slow frame rate was clearly saturating the output choke and the scan coils (this set, along with the TV9-90UK, use an odd DC-coupled frame output stage where the coils are loaded with a large parallel choke which looks a bit like a heater transformer). Changing a few more got the frame stages functioning well but there was still line pulling at the top, weak contrast and no real blanking to speak of, so I dived in (against my better judgement) and changed the lot. This is the range of values that I found:

100uF: 153 > 163uF
50uF: 110uF
20uF: 38 > 74uF
5uF: 21 > 33uF
3uF: 6 > 16uF
1uF: 4uF
0.2uF: 1 > 2uF

I’m sure that some of them would have been OK but in view of the rather odd goings-on I chickened out and changed them all. The 100uF ones decouple things like the line driver transformer so they see a lot of HF ripple, that’s why I changed them even though they were still just about “in spec”.

The set now gave a fair picture from the output of the workshop modulator but there was something odd about the AGC. With no signal connected there was no snow on the screen but on plugging it in a contrasty picture slowly appeared. Unplugging the signal right away caused an instant return to the blank raster condition but leaving it for longer caused the gain to fall quite slowly when the plug was pulled out, giving first a weak picture and then snow before the blank screen returned. Given the time constants involved I suspected more capacitor trouble but upon checking a few I couldn’t find anything seriously amiss. The AGC control was still locked with the white sealing paint which Akio Morita’s Geisha girls had put there in the 60s, showing that it had not been messed with, but breaking the seal and giving it a little tweak cleared all the faults instantly. There must have been a dead-spot I suppose, despite much further probing I really couldn’t find anything else wrong.

The picture still looked a bit “soft” compared to my TV9-90UK but as the tube was still very bright I discounted it and instead changed the two 10uF coupling electrolytics (now 19 and 23uF by the way) that follow the vision detectors, as well as cleaning the contacts of the relay which lives inside the vision detector can. This sharpened things up no end, although I was surprised to find that the relay is of German origin, yet is clearly the original one - the result of a slack day at Omron / Matsushita perhaps?

By this stage the set was working well on 625 but 405 gave a distinct lack of width (funny, it’s normally the other way round!). I ignored the big width coil by the line output transistor at first as 625 was OK but looking at the diagram to try and understand what was going on I discovered that the coil works on 405 only, a few turns and test card C was looking good. The set was at this stage giving a decent picture so I’ve been using it instead of the TV9-90UK as my everyday set. As you can see from the photos our four month old son Matthew likes it too, he’s learned to sit up just so that he can watch the Sony!

That’s as far as I’ve got, there are still a few problems to sort, such small and rapid width and height variations just after the set is switched on (selenium rectifier? more capacitor trouble? I left all the larger ones in place as their exact value is seldom critical) and a kink near the top of the test card on 405 (no ideas on that one yet).

My plan with the handle is to machine up something suitable when I get the chance (I’m “between workshops” machining-wise at the moment) but there are a few small bits missing, such as the UHF tuning knob and the piece of bright trim that goes on the handle upright on the aerial side. Does anyone have either of these at the bottom of the junk box?

[bobbyball]

Right, I want it back now!!!

Seriously, well done again Tim, the picture is excellent!

Robert

[mark pirate]

I have one of these little sets, luckily I also have the power lead & paperwork in the original box, sadly it is suffering frame collapse, I must dig it out and see if I can sort it.

Is the picture on 405 better than the TV9-90?

Mark

[Andy Green]

Same here, I got mine about 20 odd years ago at a ham radio rally. it suffered frame collapse after about 5 years intermittent use. Saving it for retirement I suppose!
My 9-90ub that I got second hand as a teenager in the 70s is still going strong!

[Studio263]

Some updates on the TV9-306UB.

The problem of picture size variations got worse and worse but despite this I found it difficult to make conclusive measurements to pin down the area of the fault. I discovered than the boost capacitor (200uF / 12V in this set) was a bit crusty underneath, as was the supply decoupler for the line output stage (same value). Changing these helped a little, but after a few more hours of running the fault disappeared, to be replaced with a reduction of width, a hum bar (worse on 405) and a black patch in the middle of the picture. The latter normally suggests a faulty EHT rectifier but I didn't fancy trying to change the three tiny valves in the tripler box (like EY51s but about a third of the size), so I checked the HT supply instead. As with the original faults, it had fallen again to around 9.5V but this time the current consumption was also low, around 800mA. Very little AC ripple was present however, but as with most TV faults the clues were all there from the begining.

Looking again at the 405 line picture (which is much better than that given by a standard TV9-90UK by the way, the banding around bright objects is much better surpressed, if not completely absent, and the sync lock is therefore much more solid) I noted that there was only one hum bar, indicating that the ripple frequency must be 50Hz. In a set with a bridge rectifier (like this one) this points to a diode being open circuit rather than a capacitor problem, and this was confirmed by measuring the ripple frequency across the resevoir capacitor - 50Hz.

To my surprise, this set doesn't use a selenium rectifer like the TV9-90UK does; instead four silicon diodes assembled onto a small tagboard to form a complete unit are employed. To reach these the chassis has to be removed from the cabinet, which isn't easy as there are a lot of things to remove, unscrew, de-solder and disconnect. It is also very easy to bust the tube neck when lifting the chassis away, a bit like it is when working on a Rigonda VL 100. For those who've never taken one of these apart I've included a few pictures of the chassis, to me it looks like a miniature version of the one from a late 50s set like a Bush TV75, big metal sheet with a hole in the middle, VHF tuner at the bottom, components and wires everywhere...

Sure enough one of the diodes had failed, predictably the one at the back and at the bottom which meant that the tagboard had to come out, the screws for which are concealed behind the timebase PCB. Given the aggro involved I chose to fit a complete 3A rectifier block rather than change just the one diode, the others may have been about to "go" too.

With this done and the set back together the width, hum and fluttering picture sized problems were now thankfully banished. The only issue that remains is slight sound on vision of 405, but that may be due to the strong signal from the Aurora, even though it does not affect the TV9-90UK in the same way.

Forum member alanmoore very kindly gave me a non-working monitor version of this set (CVM-306BP) which had a complete handle and the necessary bits of trim to tidy up the top of the TV9-306UB. With those bits transfered over (and a UHF tuning knob turned up on a friend's lathe) the set set is now complete and looks very smart. I had a look at the CVM-306BP, which was D.O.A. but didn't take too much work to get going again. It has a good tube and is reasonably complete apart from the handle and the aerial but has all the same frame (etc) problems as TV9-306UB, so a big handful of capacitors should have it running properly again. It is far too good to break, so if you fancy a challenge and can collect it let me know (PM). These were made to go with the CV-2000 VTR and are a very interesting model in their own right - it's 405 LINE ONLY mind, either on VHF (13ch) or line input via an EIAJ connector - how many other "domestic" 405 line TV / monitors can you name? The trouble is if it stays here much longer I'll give into the strong temptation to restore it myself, what a great set to run from the baseband output of the Aurora...

[bobbyball]

Glad you found the trim parts Tim, a worthy addition to your Sony collection now!

Robert

[Valvepower]

Hello All.

Just found this thread after looking for some info on the Sony 9-306UB TV. This was a charity shop find for £50, which I feel is a reasonable price given its in good condition with no damage to the plastic of the case and includes the original box and product tags.

Looking at the thread the circuit is available in the Radio and TV servicing manual and hopefully on Pauls DVD. This was found whist on a short break in the south of England.

Terry