Thorn 3500 Frame Jitter

[Niechcial,Steve]

I've actually sorted this one but it was an interesting one so I thought I would share it.
It all started with the frame collapsing to about 66% height. The next time I turned it on it had gone down to about a quarter of an inch. Voltages on the BD116 output transistor were way out and its HT feed resistor was running hot.Having checked for noisy pots I put a scope on the drive to the output transistor and found that drive was fine. Out came the BD116. It read ok on a tester but in went a new one anyway. I now had full scan but with an intermittent jitter- mainly in the top half of the scan, but sometimes affecting the bottom too. I next changed all the small electrolytics and had a good solder up all round. No difference. Out with the scope again. When the fault was present the drive to the output remained clean. However putting the scope on the collector of the BD116 showed distinct 'splashes' just after flyback when the fault was present, so it had to be either the output stage or something beyond on the convergence or scan coils. The output stage is very simple so I changed all the components apart from the output transformer which I didn't have. This included another BD116. No change. I had just about got to thinking I would have to borrow another panel to eliminate a faulty output transformer when a last thought occurred to me. The BD116 with its case at flyback potential is mounted on a heatsink at chassis potential from which it is insulated by the usual mica insulator and heatsink compound. In turn, the heatsink is mounted on a little metal sub-chassis which also has the frame output transformer bolted to it. The complete unit forms part of the chassis rail between two parts of the printed circuit. By removing two screws on the sub chassis I was able to render it floating and then putting a temporary wire link across the circuit board to make good the chassis rail allowed me to operate the set with this floating heat sink. Fault gone. Touch the heatsink to chassis- fault reappears. Bingo!
When I replaced the BD116 (twice) I had not changed the mica insulator. It or the heat sink compound was obviously braking down under max flyback voltage. I'd heard of this problem before, usually on high voltage transistors, but this is the first time I have encountered it. Cleaning off all the old gunge and fitting a new mica insulator completely cured it. Does anyone have any thoughts on whether the problem is the insulator or the heatsink compound?

[its ur aerial]

Frame faults on the 3500 chassis were extremely rear, well done, award your self a large one.
I am also impressed with your eclectic range of jobs, from telly man to Probation officer to
a Sarf London "pie and Liquor", (Vicar ).
Ken G6HZG

[Niechcial,Steve]

Thanks Ken. I've given up on Probation now. Full time vicaring!

Stephen