1946 HMV 1804.

[Geordie McBoyne]

Hi Trevor,
supplied from a separate power supply the line blocking oscillator is working OK. Easy to adjust for a 100microsecond repetition rate. Good 50volt p - p sawtooth waveform present at the control grid of the KT44 line output valve.
Before the mains is connected to this set what really is needed is a complete set of HT supply electrolytics.
So now it could be said the timebases are serviceable. Now I'll turn my attention to the 4.5Mc/s sound IF amplifier. get rid of that SP61 valve, fit an international octal base valve as the replacement.

Geordie McBoyne.

[Geordie McBoyne]

Waveforms from the line timebase.  The sawtooth drive to the KT44 grid and the pulse present at the screen grid of the KTZ63 oscillator.    Oscillation takes place between the control and screen grids. The sawtooth is developed on the anode.   Repetition rates apart it's a similar circuit arrangement for the frame timebase.

Geordie McBoyne.

[Geordie McBoyne]

Strange that the second RF amplifier is an EF50 when as far as I can understand EF54 would have been entirely suitable for the task. The first RF amplifier stage employs an EF54 and is biased to operate at max gain. The second RF is gain controlled in the normal manner by a variable resistor in the cathode circuit. Perhaps the EF50 has a better gain control characteristic compared with the EF54.
Anyway, back to the work done on the set this week. The Mazda octal valve socket used for the sound IF amplifier valve has been replaced with an international octal socket. The Mazda SP61 pentode will be replaced by an Marconi-Osram Z66 or, if I can get away with it a vari-mu KTZ61 or W61.
The attachment shows that the vision IF transformer have received some modifications by the mystery engineer. I'm not going to make any attempt at returning the transformers back to the original EMI designers specification.
It seems unbelievable that someone has gone to all the bother of altering the receiver with such complex modifications when a simple external channel 1 to channel 4 frequency shifter unit would have done the job.
Some mornings I look at this set and think should I give up on this strange receiver?

Geordie McBoyne.

[Murphyv310]

I'd say don't give up on it. I do wonder why whoever's been working on the set why as you say they didn't consider a frequency shifter?
If it was mine I'd stick with it and get it going.

[Geordie McBoyne]

Hi Trevor,
This set was subject to one of the most comprehensive list of modifications I've even seen done to a television receiver. You are absolutely correct, more than sixty years ago an unknown engineer put a lot of effort into redesigning the signal circuits of the 1804 and even if we don't what the motives were for doing all this work we owe to him to preserve the set in it's modified state.
Other additions to the original 1804 circuit will be retained. The Pye "positive interlace filter" will be wired into the main chassis instead of that module thing which used an EB91 double-diode. In the 1804 V8 is a Marconi-Osram D63/6H6 double diode and in the original EMI circuit only one section is used for the interlace filter, this valve can be rewired to function in the twin diode interlace filter circuit. A vision interference limiter was in fact an official EMI modification but the method employed by EMI left much to be desired. According to the service manual a degree of loss of picture definition would have to be tolerated after the white spot clipper was added. The method employed by the mystery engineer is better.
New HT electrolytic capacitors were delivered today and these will be fitted in due course.

Geordie McBoyne.

[Geordie McBoyne]

Replacement HT smoothing capacitors were fitted yesterday. There was five 16mfd capacitors on the bracket, two were connected in parallel for 32mfd. The original capacitors were useless and have been replaced with three new 16 +16mfd dual capacitors.  Examination of the circuit diagram reveals that there is two paper dielectric capacitors wired across the HT supply. The capacitors are marked in the EMI service manual as C25 and C65, these will almost certainly be leaky so must be located and replaced.

Geordie McBoyne.

[Geordie McBoyne]

Work is progressing well. At least one hour per day is devoted to the restoration of this set.
The attachments show some of the capacitors taken from the 1804, these were restuffed today.
I think we have to agree those yellow hi-viz capacitors look awful in vintage sets.
Actually I don't mind seeing Philips polyester capacitors fitted in old equipment, somehow these look OK.
In the mid sixties I remember one radio and television dealer in Newcastle who used to buy Mullard/Philips capacitors by the sackfull.  The new capacitors were considered a revelation especially when one remembers the rubbish capacitors we had to work with in those times.  Tolerable leakage and all that.
Three EMI 0.047microfarad capacitors also get the treatment.
There's plenty of TCC waxies available but the EMI ones are a bit scarce.

Geordie McBoyne.

[Geordie McBoyne]

This morning it was the sound circuits that came under scrutiny.
The sound IF amplifier, demodulator and noise limiter circuit were modified by the unknown engineer.
The first attachment shows the original EMI sound circuits. The noise limiter operates by means of a cancelling circuit. The out phase noise spikes detected by the second section of the D63 double-diode are compared with those from the demodulator.
The second attachment shows the modified circuits. The "B" section of the D63/6H6 double diode functions as series type noise limiter, the type that was in common use in 405 line TVs. The interference limiter diode is normally conducting. However, when a when a positive going spike comes along the time constant of the resistor and capacitor at the anode is long enough prevent the valve conducting.
An added refinement is AVC. One of the diodes in the DH63/6Q7G is used an AGC detector to control the gain of the sound IF amplifier valve. A degree of AVC delay is provided by the cathode bias volts of the DH63.  The original sound IF amplifier valve was changed from a vari-mu KTW61 to a sharp cut-off type SP61.
It's not clear why all these modifications were carried out, after all there wasn't anything wrong with the original EMI circuits.

[Geordie McBoyne]

The screen grids of the Z66 video amplifier and the KT61 frame output valve are decoupled by two 4 microfarad electrolytic capacitors which are contained in a common can.  To maintain the original appearance I've decided to restuff the dual capacitor.

Geordie McBoyne.

[Geordie McBoyne]

At long last the set is connected to the mains. OK, it's not direct but through the variac. For starters the output is set to 100 volts just give the set a taste of electricity. An indirectly heated rectifier valve will be used for the initial tests and later on the correct directly heated U52/5U4 valve will be fitted. I'm presently concentrating my attention on the signal circuits and later I'll get brave and connect up the EHT transformer.
Attachment shows the mains plug. It's many years now since I've seen anything connected a plug like this.

Geordie McBoyne.