I bought this HMV 650 on eBay. It was complete apart from a missing rectifier valve. It had been partly restored, including fitment of a Gary Tempest replacement dial. I will show pictures including the dial when I have finished the cabinet. I'll say from the outset that my aim was to get this set working properly and to make the cabinet look presentable. It was never planned as a concours restoration, like the fabulous work done by Gary, featured in back-issues of the BVWS Bulletin.
This is a twin-chassis model, with the lower chassis incorporating the power supply and power amplifier. As-found the push-pull power amp had 2 x Tungsram 6V6G beam tetrodes fitted, one of which was virtually zero emission. Out they came and in went 2 x Canadian Mullard 6F6G, NOS and almost direct equivalents of the original pair of KT63. I found a 5Y3GT rectifier and plugged it in. The original electrolytics were long gone, and the chassis had been drilled to accommodate chassis-mounting clips. I have re-used a couple of the replacement electrolytics because they reformed OK, and fitted a new blue LCR reservoir cap. These are a temporary measure to get the set working and I have ordered new screw-fit electrolytics from BVWS Spares, which will look much better. I also noted discrepancies from both the Manufacturer's and Broadcaster data sheets with the addition of an extra preset potentiometer next to the HT fuse. Advice was forthcoming from the Forum (see repair thread) that this was a later modification to introduce silent tuning. The AF driver valve (Z63/KTZ63/6J7G) is muted via a make/break switch while spinning the tuning knob. The switch can be seen in the above-chassis view of the receiver chassis, to the right of the tuning mechanism.
Back on the power amplifier chassis, as a precaution, I replaced the 2 x 500pF feedback capacitors in the output valve anode circuits with new 500V polyester. More on these later.... The rubber sleeving on the HT chokes and output transformer had crumbled so I resleeved them. Then I turned my attention to the receiver chassis. Usefully, the component numbering is the same on the Manufacturer's and Broadcaster sheets, although the Broadcaster has some minor omissions. Some work had been done fairly recently with the fitment of replacement Wima capacitors. Other prior restoration had been done a long time ago including fitment of old-style grey Radiospares capacitors with blue lettering. Some of these tested OK whereas others were leaky and replaced again. I found an initial of a long-lost repairman on the chassis and a date of 24/4/46. The high wattage HT potential divider R8 was already bridged with a 6k 10W ceramic resistor and the other section (R7) was o/c so I bridged it with a porcelain 3.3k. The receiver chassis electrolytic (C18/C24 8+8uF) was completely unmarked and of unknown provenance and so I replaced it with a Plessey NOS 16+16uF because I didn't have any suitable 8+8uF spares. I noticed that the rubber insulated wires from the AF driver transformer were rotten and crumbling, notably the red and yellow ones. I pulled them out and then discovered that the terminations at the transformer end were sealed in tar (should have read Gary's article first
). By then, the sin was already committed but I manged to reseal the ends of the 7 wires where they entered the cable duct into the base of the transformer using silicone sealant, then cut the wires one by one and soldered on replacement neoprene insulated wires using heatshrink to cover the joins.