View Single Post
Old 19th Sep 2010, 7:31 pm   #1
Studio263
Octode
 
Studio263's Avatar
 
Join Date: Apr 2007
Location: Hampshire, UK.
Posts: 1,574
Default Sony KV-1340UB restoration

The Sony KV-1340UB was their last model in the UK to feature the original long-neck 13” Trinitron tube (330AB22 – see picture) but was never the less a modern looking set for 1976 when it was launched. Unlike the previous KV-1310UB model it had a varicap tuner, a proper PAL D decoder and a very compact cabinet (except for the depth, not surprising when you see the tube!). One unconventional feature was the power supply, instead of the up-to-date series switch mode converter that the KV-1310UB and KV-1330UB had had Sony employed an odd circuit where a power transistor switched at line rate (50Hz) was used to approximately halve the incoming voltage, the result then passing to a conventional linear regulator circuit (not unlike that used in the BRC1590 portables) that produced a stable and smooth 110V for the rest of the set. Although the KV-1340UB appeared at around the same time as the KV-1810UB the circuitry is more similar to the KV-1800UB, in particular the line stages, where the line oscillator drives the “convergence output” transistor and flyback transformer (EHT etc), from which a winding provides the base drive for the line output transistor proper, which has as it’s load a smaller transformer and the scan coils. The collector of this transistor is powered from a small regulator stage which is modulated with geometry correction waveforms for pin cushion correction etc. Like the KV-1810UB though these sets seem to suffer from catastrophic blow-ups that take out quantities of hard to find transistors.

Years ago a broken up orange KV-1340UB was one of the first colour sets I attempted to repair, a “wanted” advertisement in “Television” secured another cabinet but it still too what seems like years to get it going, even then not perfectly as I recall. The current set came from this forum (thanks ctc15!) and hasn’t taken quite so long.

The set was dead with one of the two 2A mains fuses blown. Unsurprisingly the switching transistor Q601 was short circuit and the regulator Q604 was leaky. Luckily the convergence, line and pincushion modulator transistors were OK, so the power board was tackled first. The regulator is a 2SC876A which I didn’t have, I did have a couple of 2SC876s though, and even though these are not the same it did mean that I could use the 2SC867A that had been wrongly fitted in the sound position and put the 2SC876 to good use in its place. With the output disconnected, Q601 removed and shorted C to E and a 100W bulb as a load the power unit produced 110V with reasonable regulation on a 50% mains input from the variac, replacing Q601 could wait until the state of the tube had been assessed.

Not wanting to blow up any more transistors I carefully inspected the VH (timebase) board, replacing most of the electrolytics as I went. The power to the line oscillator is easily disconnected from D506, allowing the time honoured two 9V batteries in series to be used to power it. Like this big line pulses at around 15 KHz could be measured at the driver transformer, meaning all was well. Applying mains power brought the set to life with heaters and EHT, but no sound or picture.

The sound was easy, as tinny voices could be heard from the output transformer it was simple to trace the wire that had snapped off the PR board. The picture was also easy, shorting the collectors of the RGB transistors to the chassis brought screen illumination in each colour and it was then quickly discovered that the luminance output connection has come away from the decoder panel. Re-connecting this got a passable monochrome picture, although it was clear than the tube was tired.

Colour faults are not my favourite and Sony decoders do have a reputation for complexity, however the KV-1340UB decoder is very nicely designed and there are no odd features in it that you won’t find in other more popular makes and models. The lack of colour turned out to be nothing more that the control itself being broken inside, fitting another brought the colour back but slightly out of phase and with strong Hanover blinds. This is the sort of fault that you’d probably let pass on a “customer” set (they wouldn’t notice it and wouldn’t that you for fixing it anyway!) but for your own it’s different. Quite some time was spent testing and checking, judging by the soldering in the area I wasn’t the first either. Eventually I came to suspect the delay line itself, I didn’t have a proper Sony one so in went a standard miniature European type, 64uS is 64uS after all. The improvement was instant; with a bit of setting up the decoder performance was very good indeed.

Back to the power supply. The correct 2SD1454 transistor proved hard to obtain so the junk box was raided and a BU208A and a small heat sink were produced. Fitting the heat sink required that some of the components be moved to the other side of the board, although some seemed to have been put there already from new! The BU208A has an Hfe of only 2, not enough when the original transistor has one of 20. To get around this a BF871 was selected as a driver and the two were connected as a Darlington pair, with a 1k resistor across the BU208A B-E junction. Bringing the bulb and variac back into play suggested that the arrangement should work, the voltages were correct and the transistors remained cool. With the power reconnected to the set we had a good picture and sound, some cleaning of contacts, replacement of broken pre-sets and general tweaking up improved it still further.

The cabinet was then stripped and washed, the black paintwork at the front was tatty so I masked up the white bits and re-sprayed it, choosing satin black instead of the original metallic grey in an attempt to offset the yellowing white plastic. The result looks really smart, thanks to G8KBG Tony supplying for a replacement for the missing “colour” knob.

I thought the focus could be improved so I de-soldered the lead from the tube base in preparation for trying it in all three positions. Switching on without the lead connected produced a crackle of EHT, a crack and a bang from underneath and then nothing. Rats. Inspection showed that Q510 (one of the frame output transistors) had blown in half and that the other one (Q901) was short circuit. Removing these and powering the set carefully from the variac produced sound and a bright frame collapse at 50% of the mains voltage, the dividing circuit wasn’t working any more either. I thought that the BU208A had bitten the dust but luckily it hadn’t, it was the driver that had succumbed. A manlier transistor was clearly needed, this time an MJE340 was selected from the junk box, these are rated at 300V / 1A and hopefully this will be enough. The frame transistors are another 2SC876 (my last one!) and a 2SC1663, although a 2SC1124 was fitted. The book says that a BF460 will replace the 2SC1663 but even though the pin-out is the same the heat sink tab is bigger and the mounting hole is in a different place, a few minutes with the saw and drill soon solved these problems! Fitting the new parts brought back a colour picture on normal mains (phew!) but the frame scan was cramped, folded over and had a bright line though the centre, clearly BF460 is not the ideal transistor in this circuit. The mounting precluded trying anything else that would be any better so once again the Darlington pair circuit was employed, this time I used a BF422 as the driver and stuck with the BF460 as the output stage. This did the trick, adjusting the V.BIAS control whist observing the waveform across R553 allowed a linear scan with no crossover distortion (the bright line) to be obtained.

This is as far as I have got. The set is working nicely and is quite usable, however it would still benefit from a better cabinet and a better tube. The green piece of the Trinitron logo is still missing too; I shall have a go at casting one in resin and painting it green I think.

These are not especially easy sets to restore but they are not as daunting as their reputation suggests. Unobtainable transistors with odd characteristics seem to be the main problem, a decent equivalents book and a well-stocked junk box are essential; as is a variac if you don’t want to have to replace them too many times!
Attached Thumbnails
Click image for larger version

Name:	KV-1340UB bottom s.JPG
Views:	457
Size:	40.1 KB
ID:	40584   Click image for larger version

Name:	KV-1340UB PR board MJE340 s.jpg
Views:	473
Size:	62.9 KB
ID:	40585   Click image for larger version

Name:	KV-1340UB delay line s.JPG
Views:	465
Size:	49.3 KB
ID:	40586   Click image for larger version

Name:	330AB22 s.jpg
Views:	557
Size:	37.6 KB
ID:	40587   Click image for larger version

Name:	Kv-1340ub 1s.jpg
Views:	702
Size:	44.0 KB
ID:	40588  

Studio263 is offline