The three jam tins on the left are consistent with a superhet with three gangs. One for antenna second tuned coil between TRF stage & Pentagrid and one for the oscillator coil. or plural coils if it has SW.

Then it gets interesting. #19 is a double triode with common cathode 6pin US base. That could be the phase splitter. 2V Filament tube. 135V plate (1J6) polarised (DC) filaments.

#30 is also a 2V filament tube but is a detector AF amp so would have to be set as a plate detector and have an RF choke in the plate & why the hell would you use that when that combination is the worst for distortion? It is a UX-4 base & neither have top caps and appear in the Philips data as polarised (DC) filaments. 90V / 180v Plate. (1H4)

Now 6F6 (6.3v heater) is a 2A5 it has a 2.5V heater so perhaps these are PP 2A5's which have a six pin base not eight.

There may actually be a Metal rectifier in the black box the internals of which will be most interesting. It looks like some serious hacking around war time.

Marc

Thanks guys.
When it’s back in the workshop in a couple of weeks I’ll take some more pics of the chassis and the power supply and post them. The artisan label says it was made here under licence.

The ARTS&P label?
Australian Radio Technical Service & Patents.
The "D" means it was made in 1937.
Yes, a "converted" battery set.

Thanks Ian.
So, if it’s a battery set, what voltage did the batteries supply? I guess I’ll need to check if the power supply is working properly. It occurred to me that if the regulator has failed, it may be supplying an AC voltage instead of the regulated DC. That would account for the hum and poor performance.

Back when that was made, there were no regulators. Before transistors, it was VERY difficult to get a smooth low voltage supply. All you could do was use a metal oxide rectifier (very lossy), several filter chokes and as much capacitance as you could find / afford. Even then it wasn't perfect. The way they'd get the remaining hum out was to add a battery in parallel with the power supply output. It would have been a single lead-acid cell, in a glass jar.

The radio would have originally used a 2v lead-acid cell for the filaments, 3 x 45 volt dry batteries in series for the B+ supply (135 volts) and possibly a small bias battery of about 4.5 volts.

Silicon Chip magazine published my little battery radio power supply some years back. You can get a PCB and a few other parts for it from them.

Wow that’s quite a setup. I’m not sure what I can do with this thing then. I should check what’s coming out of the supply. It has 2 separate cables from the sup to chassis. One is the filament supply and the other the B+

It would be pertinent to reverse engineer it. My solution to this sort of mess. I will do a layout on the Auto CAD & then work out the arrangement of the coil windings; They're normally predictable. Then play join the dots using coloured pencils. That gives a layout diagram.

From that you get a circuit diagram, which you use or hand out to sort out the stuff ups & only then do you go and fix these stuff ups. Tedious but effective.

This mapping also applies to the PSU lets see what we have. The we decide on LR8's 78xx etc. to perhaps sort it.