Stupid Assumptions

[peter scott]

About six months ago the picture disappeared on my HMV901. The focus voltage was all wrong so it was fairly obvious that I would need to get into the focus chain and fix it but unfortunately it's in the most inaccessible place on the underside of the power supply chassis. There is no access from the bottom of the cabinet so you need to disconnect the mirror lid prop, take out the screen frame, remove the CRT and its supporting cradle, remove the timebase chassis, loudspeaker and sound chassis and only then can you lift out the very heavy power supply.



I replaced most of the focus chain but not the resistors I had replaced many years previously then reassembled the set. Then at the back end of last year the picture disappeared again and just yesterday I went through the above procedure again and tested each resistor in the chain only to find no fault!

After reassembly I measured the CRT voltages and discovered that EHT, focus and accelerator voltages were all fine. The cathode was 36 volts above the grid, in a tube with 20 volt cut-off.

Looking on the bright side I'm getting quite good at dismantling and reassembling the set.

Peter

[ppppenguin]

I'm sure there's a moral to this story.

Note the very low cutoff voltage of the EMI hexode CRTs. Would have been ideal for early solid state TVs where the video output stage was a big problem. The combination of high voltage and high speed was a hard nut to crack until theu invented the AF118. These EMI sets used what was high-ish power RF stage to drive the detector which drove the CRT directly. Back then getting the gain/bandwidth product was tough with valves.

[peter scott]

Back then getting the gain/bandwidth product was tough with valves.

But the receiver was the least of the problems for Marconi-EMI. The transmitter modulator achieved a very impressive gain-bandwidth with very much bigger swings.

Peter

[ppppenguin]

But they had a lot more money and space at the transmitter

The modulator is a very impressive bit of kit, with and output swing measured in THOUSANDS of volts. The filament supplies for the valves were a major source of complexity. The cathode followers had full video swing on their cathodes and so needed some very clever arrangments to get power to the filaments.

Where was that response curve published? I've looked at the modulator section of the Black Book and it's not there.

[peter scott]

The response is in the IEE paper. I've copied it to: http://ge.tt/2cNmrHo2

Tweak as I might I was never able to get better than a marginal 2.5 MHz bandwidth on my 901 but after feeding its aerial port with the graphics card signal it receives almost as well as the modulator curve.

Peter

[ppppenguin]

I've got a paper copy of that article somewhere so many thanks for the scan.