630 is the same on the Northern Beaches at night, it does not have a clear channel like 702 does.

When are these 429MA radios expected to arrive? I'm interested in what you find.

According to Wikipedia, on the East Coast, 630 is used by ABC News Radio Sydney, ABC Radio National Queenstown Tasmania and Townsvill Queensland. All of these might be picked up at night with a radio oriented for North-South reception. I have certainly noticed RN coming in.

Listen to the radio at night to block out tinnitus; changeing to FM for Newsradio on 90.9 requires a hard to operate switch rather than a push-button on my old Sangean ATS 800, so stick to AM if I can.

Buy a new radio you might say. Well, I have, two in fact; but the buttons failed or became unreliable (the click type), and the Sangean (the rubber type buttons) just keep going.

One came today, the other I might get on Saturday. I'll do the two together for comparison.

I designed a 1 transistor audio preamp for the OP so that the audio gain of the 4 valve radio could be increased.

I had a little worry about diagonal clipping with my first circuit. This is what happens to an AM diode detector when the AC load exceeds the DC load. It's the reason you usually see 10meg grid leak bias used with the stage that follows the volume control in a "5 valve" radio.

When designing a circuit you have to allow for device spreads and things like leakage upsetting the performance of the design, which means keeping the impedances around the base of the transistor as low as possible. Unfortunately this is at odds with the requirements of an AM detector!

So I re-designed it using bootstrapping of the input.

Bootstrapping allows you to have a higher AC input impedance than the bias resistors impose, by using AC feedback from the emitter of the transistor.

You'll note in the circuit I'll send to Brad for inclusion here, I've included a 1 meg resistor in series with the input. This drops the overall gain by half and proves on the simulation that the AC input impedance is 1 meg.

Don't fit this resistor, just wire the 47nF straight to the pot slider.
The AC load with this circuit added is actually less than it was in the original radio circuit, so distortion on AM should be reduced.

The transistor you use here needs to have a Vcbo of at least 150 volts. An ordinary BC547 will not do! I happen to have a large stock of NOS 2N6517s that are rated at 350 volts - happy to put some in an envelope for anyone who needs them.

Although, if you have a source of about 50 volts in the radio, a BC547 could work.

Click on image for high-resolution copy.

Thanks for going to all this trouble Ian. I could have tried something from my 1960s text book but would never have anticipated all the issues that are second nature for you.

But it is a common radio so the circuit should have a wide application. The radio has the advantage of looking nice too. The later similar models have the disadvantage of dial surrounds that shrink over time.

Thanks also for the high voltage transistors.

I suppose 50 volts could be obtained from the filament voltage with a voltage multiplier circuit, but sounds complicated compared to the high voltage transistor. And a 5Y3 with cathode to filament short would fry a BC457.

5Y3 doesn't have a separate cathode. The filament is the cathode.

You'd be better off with a resistor from the B+ and using a 50v zener, if such a thing is obtainable these days.

Robbert of course you are right re 5Y3GT, don't know what I was thinking there!

A zener is a simple way to get 50V.

If you are not overly worried about originality? In my bench reformer (stepped voltage) and a dual power supply for UX-201-A (45V out of one 135V out of the other, off of a common :B" rail. Bias & heaters separate windings). I have used LR8's they red line at 450 volts. You can by altering their divider, go from 5V to 400V and even sub a burnt our choke with one.

LR8 is good for around 15mA, after which you can use it to control a TIP50 which is good for around 500V

You could use them to at least stabilise the HV voltage to the 6BE6 if fluctuating voltage is an issue. Some Signal Generators also stabilise the heaters DC being the easiest to regulate.

What was I thinking there?
I had also looked at the subsequent model 567MA of which I have one unrestored - it has 6BE6, 6AU6 6BF7 6X4 and somehow this got tangled in my tired mind to making the mistake with 5Y3GT. Moral - don't write these things when you are tired!

Can't be worried about originality if an FM converter is being installed - it is effectively part of that. So will consider a regulator Marc. As long as it can be undone leaving an original circuit and minimum physical changes to the chassis such as extra holes drilled. Not running a museum here!

LR8 is about the size of a small transistor. It alone would probably handle screen and grid on its own.

A warning with three terminal regulators like it, especially used for bias, is that they need to be loaded or they will not regulate. E.g. 7805, 7905 need 5mA and LR8 0.5mA of which its divider is part. To be effective there is a minimum input voltage.

The heater winding has the most likely reserve to run it. Do remember a loaded* chain of series diodes, make a voltage dropper.

* If you use that method an open circuit goes to the source voltage and that can do damage.

Trouble with lower voltages with that preamp is you can run out of headroom for the series voltage feedback.

I wonder how many people these days

(a) have heard of bootstrapping
(b) understand how it works
(c) know where the name comes from
?

If you feel the need to regulate the screen supply, a 2N6517 can do that also!
Either as a shunt or series regulator.
Just keep in mind the 600mW rating.
If you measure the voltage drop across the screen dropper (and its resistance) I'll draw something up for you.
Doubt you'll need it though.

I recall the first I heard of "bootstrapping" was when Astor/Philips introduced their first B/W all solid state TV chassis.
Was explained in an industry tech seminar.
Around 1967 from memory.
Understood at the time but need to redo my homework now.
JJ

We all have different ideas but seeing that an LR8 is likely to cover the draw of a 6BE6, I would be inclined to interrupt its B+ rail & regulate the whole tube. With a heater rectifier there is not going to be a surge outside of its parameters. If you are working on its divider formula remember to calculate what is inside the brackets first, not incorporate anything outside, of them until you have.

We seem to be discussing two things here:

1 stabilising the 6BE6 to prevent drift associated with voltage variations, the original reason for this thread

2 getting the appropriate operating conditions fo Ian's amplifier with bootstrapping/feedback circuit

I have understood, from dim distant memory, bootstrapping to be a feedback arrangement which like other feedback arrangements can become unstable; such instability might be good if you want an oscillator; thats about it.

I think I will

-first swap 6BE6s in and out to the effect on fading, and look at voltage variations that might cause it

-second add the converter to the radio as is and get it to work with low input, and see how the fading effects it

-add Ian's preamp and get that working (the circuit is missing from Post 18)

- keep an eye on fading and its effect on workability during the last two stages

Thanks for all the suggestions and info gents.

Note: Actually circuit missing from Post 35, (not 18)