Tale of rags to riches of for a Bush TV161
I published this write up on my facebook group and thought that you folks here might enjoy it too. The story here regards only the electronic work carried out. The cabinet also required extensive work to preserve it and make it presentable again. I'm no expert with cabinet re-work but im pretty pleased with the outcome. ive included before and after pictures.
sorry, its a bit of a long-winded tale but hopefully not too dull!
Given the poor woodworm eaten state of the cabinet on this Poor old Bush TV161U and the poor storage conditions the set had endured over the last 30 plus years I was very close to breaking it up for parts. However, this model uses the very well regarded A640 chassis which was probably the high point of Bush monochrome television design. Not many survive these days and I had not seen one for over 30 years so was curious to see just how good they were. The design featured DC-coupled video signal with a clever but simple black level restoration circuit. I am told this arrangement gave probably the best monochrome picture of any production TV set.
The Achilles heel in this chassis along with many other Bush models of this period is line output transformer ( LOPTX ) failure. This is especially so with sets that have been stored in a damp environment for a long period allowing the transformer windings to absorb and trap moisture. As this set had suffered poor storage in a garage it was a big concern as to whether the transformer in this example would be viable.
An additional concern with the restoration of any old TV set is the condition of the Cathode ray tube ( CRT). This set uses a Mullard manufactured tube which generally last well. My CRT tester proved that this one bucked the trend with only mediocre emission, with the needle resting well into the red highlighted bad zone on the meter.
Not a great start but I decided to see how well the set responded to repairs as the CRT could be replaced at a later point if need be.
Before starting any work the LOPTX was removed from the chassis for testing. This is in fact very easy to do, on this chassis as all the leads plug into the time-base via a long multi-connector plug. Once this is disconnected and a single screw securing the transformer to the chassis it can be removed, not forgetting of course to disconnect the EHT connector.
My preferred method of test is to ‘ring’ the LOPTX on the scope using a 1 kHz square wave. The square wave is applied to the line output valve top cap connector and the resultant waveform measured at the EHT top connection on the EHT over-wind. Thankfully there was a good strong ring trace on the scope so it would seem the transformer did not have shorted turns, at least not yet!
Given that these transformers are a bit unreliable even when at the best of times let alone a set that had been stored in damp conditions it seemed prudent to drive any possible moisture out of transformer windings by storing it in a warm airing cupboard for a few weeks. The transformer in this set was the later Plessey made one with the plastic encapsulated EHT overwind. Apparently this version is little more reliable than the older pitch covered one so I am told. After the about 5 weeks of resting in the airing cupboard, the transformer was refitted to the set. The 5 week wait at least gave me time to complete the cabinet work.
A quick check with the AVO did not reveal any obvious shorts on the HT lines and the mains filter capacitor was clipped out of the circuit for the time being to prevent any unwanted fireworks!
Switch on was greeted with silence! It was completely dead, no heaters no nothing. Out with the AVO again and soon discover the contacts in the on-off switch were stuck in the off position. Obviously 30 years of inactivity had caused the switch contacts to seize in the off position. Rather than replace the switch I was able to get some switch cleaner into it and flicking it on and off a few times soon had the contacts moving freely again.
So power up, take 2 …. This time I was rewarded with valve haters but little else. Clearly no HT, which was proven by a quick dab of the AVO lead on the smoothing can terminals.
A look at the circuit diagram showed that the HT rectifier, which in this chassis is a silicone diode which feeds a 14-ohm section on the mains dropper. A trusty neon screwdriver soon proved that there was AC one end of the dropper section but none the other. Luckily I had a new old stock dropper for this chassis so in it went.
Switched once again and the heaters glowed cheerfully and the HT appeared at the smoothing can terminals but seemed very low. As I pondered on this for a few seconds the HT supply fuse blew. The reason for its demise soon became clear, the main smoothing capacitor block was extremely hot to the touch, and in fact, you could not hold your finger on it for any length of time at all. Clearly this capacitor was in a sad state after its long period of inactivity.
Given my preference to retain as many original components with any renovation I undertake, I wondered if this smoothing block might come back to life and be re-used. The capacitor was removed from the circuit to better enable me to monitor the current it was drawing whilst I attempted to re-form it using a high voltage variable DC power supply, Starting at 50 volts and monitoring the leakage current until it dropped to around 200 microamps it was then increased by another 50 volts incrementally until the leakage current dropped to an acceptable level. It took around 12 hours before I was able to get to its rated supply voltage of 350 volts. I then left it for a further 24 hours at full HT and the leakage current stabilised nicely at around 250 micro amps which I was happy with as the outer can was now nice and cool so I was sure it would function fine.
The Avo was connected across the HT fuse to monitor for any excess demands. Power on again and the HT current climbed to a safe level along with a healthy dose of HT across the smoothing can and shortly after that the line oscillator could be heard chirping into life, followed in a short while by an over-bright bright raster!
So the screen was alight for the first time in over 30 years, which is always an exciting moment, gratifying too to know that the line output transformer was alive and all the cabinet work thus far, had not been wasted.
As the raster was much too bright and not controllable with the brightness control something was likely to be amiss with the video amp or CRT. A probe about on the CRT base showed the cathode volts were pretty much at Zero so the CRT was biased hard-on. A delve about around the pins of the PFL200 sync separator and video output valve showed no volts on the anode of the Pentode section responsible for the Video amplification. There were only two likely causes which were the anode load resistor or the choke in series with the load resistor, the choke proved to be the culprit. In fact, on closer examination, it was physically damaged but I was able to save it and carefully remake connections of the very fine lead-out wire back to the terminal leg of the coil.
The next power-up proved more fruitful and with 625 lines selected on ch 38 which is the output from my modulator a very welcome test card F was viewable. The picture tube looked passable at this point and with luck would get better with use.
There were many other obvious faults that needed to be solved before the set could provide a watchable picture though. Issues included a significant lack of width, also lack of height, poor frame linearity, poor sync, low sound on 625.
As the main HT rail was spot on the I felt the lack of height and width was likely to be two separate faults. At this point, the boost capacitor was replaced irrespective of it proclaiming its innocence on the insulation and capacitance test. One end of the boost capacitor is returned to chassis on this set, rather than back to the HT rail which is usual practice. A shorted 50-year-old boost capacitor to chassis would not do the boost diode valve any good not to mention possibly killing the anti-surge section of my newly installed mains dropper!
There was also an obvious dry joint on the line linearity inductor which would certainly have an effect on the picture width. Further cold checks revealed a duff S correction capacitor which proved to be very leaky! With these issues remedied I switched on confident of a nice test card filling the screen horizontally but was disappointed to find that though much improved it was still about 1 cm short each side even with the line stability preset wound up to maximum!
With thoughts of defective LOPTX nagging in the back of my mind I thought I would investigate the field timebase lack of vertical height instead and come back to the lack of width issue later.
The frame linearity was soon tidied up by replacing the cathode bypass capacitor for the pentode section of the PCL85. A quick spray with some switch cleaner on the frame linearity presets enabled a nice linear frame scan to be achieved. None of the blue and white Dubilier caps was showing any signs of electrical leakage which is unusual; however, the field scan was still insufficient in amplitude.
The height control is fed from the boosted HT rail fed off the line output stage via feed resistor 3R36. It should have measured 680k but had risen to 1.2 meg so with a replacement fitted I fully expecting a nice vertically scanned raster but was greeted instead but a much darker raster than before which required the brightness control be turned to maximum to be able to resolve the test card and the width had contracted further. That’s odd, it was not like that before the replacement resistor was fitted. Checking for any possible solder splashes and ensuring the new resistor was of the correct value, which it was did not clear the new problem up.
So what was going on now? The set now displayed a low brightness raster, with lack of width and height. Deciding that the lack of brightness needed investigation first, the AVO once again was put to work and CRT base potentials were checked. Cathode volts good, G1 volts also fine and varied with the brightness control as expected. The G2 volts on pin 3 was not so healthy. Normally around 600 volts would have been expected here but there were only around 350 volts. The G2 supply is fed directly from the boosted HT rail off the line output stage. The same HT rail that feeds the frame timebase height network! A voltage check at the boost capacitor showed only 350 volts there too. So that explained the lack of width and height as well as low brightness as the boost voltage was at 50% of its correct value! A bit of head-scratching followed trying to discover why the boost supply was low. The boost cap had already been changed and I was beginning to think of line transformer failure again but the EHT was good so it was unlikely to be that. I did notice that 3R26 off the LOPTX was running warmer than expected and was sweating a bit, though it was tested and was blameless. With 3R26 decidedly uncomfortable it seemed likely that something was loading the boost rail down. A further study of the circuit diagram confirmed the boosted HT rail feeds the focus volts, the G2 volts and the frame oscillator via the height network. In order to prove which circuit was responsible for the heavy loading was each was disconnected in turn whilst the boost voltage was monitored. Removing the feed to the focus network and G2 made no difference but upending 3r36 that nice new 680k I recently fitted before the brightness took a dive rewarded me with a full 700 volts of boost at the boost capacitor. So the loading was taking place in the filed timebase. The only possible culprit I could see was the height compensator VDR which was connected between ground and 3R36. With the VDR removed and all other connection restored the test card looked much happier with ample height, width and brightness! Well, that’s a first, a component in the frame timebase causing lack of width and low brightness! It seems the original 680k had risen in value to around 1.2 meg so was limiting some of the excess current that the faulty VDR was bleeding off the boost rail. Once I had fitted the correct value 680k resistor the VDR started to pass much more current than it should resulting low boost rail and much-reduced brightness width and height.
Things were looking much better now with a linear test card, time now to track down the niggly faults’. Poor sync and slightly smeary video were sorted by replacing the electrolytics C44, C45 and C48 in the video amp and sync separator, all of which were showing a very high ESR.
Up to this point, I had been using the 625 line switch position so decided it was a good time to see if the 405 line side of the set was functional. I slid the system switch over manually as the linkage was temporarily disconnected from the tuner. With channel 1 selected and the Aurora standards converter connected to the VHF aerial socket we were good to go! The set warmed up and there was a reassuring 10khz line whistle followed shortly by a very reasonable 405 line picture. A quick set up and it looked very good, for a minute or two! Then all of a sudden with a squeak it went to line collapse then no raster and no EHT. Well, that was that the line Output transformer must have finally died after all, as they so often do in this chassis! For no particular reason, I flicked the system switch back to 625 lines and within a few moments, a picture reappeared again, back to 405 lines and no raster and no EHT! Perhaps all was not lost after all then.
The line output stage now seemed to be heavily damped when 405 was selected, sufficient enough to give no heaters on the EHT rectifier though there was still plenty of line drive. A further study of the service manual showed that just about the only likely culprit was 3C22 which is switched in circuit only on 405 lines to provide additional Scan correction. The capacitor was duly removed and proved to be short circuit, a new 0.0022UF capacitor was fitted which thankfully restored full 405 line operation!
At this point, the picture was pretty decent and stable on both line standards the last remaining problem was some distortion on the sound on both systems and very low audio, but only on 625 lines.
A PCL82 is used for audio on this set and a quick check showed the cathode bypass capacitor to be pretty much open circuit. The audio was not greatly improved after the capacitor had been replaced. Measuring the grid volts on the pentode section showed about 3 volts positive which was clearly wrong. The coupling capacitor between the triode and pentode section proved to be blameless so a new PCL82 was fitted, problem solved with the AM 405 line sound. But the FM 625 line sound was still weak, a bit thin and liable to distortion if the tuning button was slightly rotated. I spent a bit of time checking around the discriminator for the 625 FM sound and also the FM sound IF stages but could find nothing obvious amiss. The picture had good definition on the 6 MHz bars on test card F so the IF stages up to the sound take-off must be fine. So nothing for it but to have a quick experimental twiddle with the 625 sound IF cores. I set the cores up in the order the manual suggested just using the test card audio. After a bit of trial and error, I was able to achieve very acceptable 625 sound with much no caption buzz either. There was no obvious sign of anybody playing with the cores in the past so I guess time and poor storage conditions must have had some effect on the tolerance of the sound IF coils.
That brings us nearly to the end of the saga with this set. It has been run on and off for 5 or 6 hours with no further issues. The tube was a little dull and poor on focus from cold but warms up after about 10 minutes giving a reasonable picture and Considering that the tube is over 50 years old and looks to be original for the set I was going to leave it at that. But I know these sets are capable of excellent pictures, not reasonable ones. That weak CRT was letting down the excellent capabilities this chassis is known for. Rather than replace the tube giving it a gentle boost seemed a good way to go, and it responded very well too! One quick flick of the rejuvenate switch and the meter on the tester was registering a much more healthy reading. The picture displayed was now very pleasing indeed and now lives up to its reputation!
I certainly did enjoy it....thanks for sharing this excellent restoration. Well worth the work and the set looks great ?
They do give a good picture, sold many and fixed many. Always had a LOPTX on the van, the last vision IF transistor did fail, BF173? All straightforward faults 40-50 years ago, anything could be wrong now though.
Great job and the cabinet looks new.
You've certainly had some work to do there and the result is excellent ?. I have one of these and thankfully its in much better shape and works ok considering its not been touched properly for years..
Lovely performers indeed and a twin to mine too!
Well done! Very good sets, i have one complete with legs. Works great, Malc.