Right now, my only battery eliminator is an AES solid-state unit. I'd like to try to build a tube-type A-B-C eliminator, though, partly for the fun of it, and partly to avoid the headaches that solid-state units can have (like requiring the radio to be connected to an actual earth ground rather than a mains ground). Anybody know of any building plans for one? I need one that can provide 6 volts "A", and 22.5/45/67.5/90/135 volts "B", and a variable "C".

This will mainly be powering an Atwater Kent 35 with 201A tubes in the RF section and detector, a 112A for the first AF, and a 171A for the final output.

Yes, I know the ARBE-III is supposed to be a great unit, but like I said, I want to go tube rather than solid-state for the above mentioned reasons.

I have an antique unit, but it needs to be repaired and takes a BH rectifier which I have heard inevitably go bad as the helium escapes over the years.

Of the ABC's I've seen either have tungar bulbs and lots of iron or have a LA battery and tungar charger... A copper oxide rectifier would do the job but they have all but deteriorated by now. For example the power supply for a Stromberg-Carlson 523, ABC supply weighs in at 110#.

The problem is low voltage vacuum rectifiers that can deliver the needed current and the necessary chokes.

The S-C supply accomplishes this with a saturable reactor for regulation. Keep in mind as the filaments are turned up the current draw goes up too.

Why not use a silicon rectifier and a dropping resistor to replace the BH, "C" bias can be achieved via a B- resistor. The "A" can be supplied by a LA motor cycle battery that has a trickle charger with a current relay.
LA battery is on a trickle charge until the filaments are turned on. Relay transfers the charger to OFF and turns on the "B" supply. My Dad used to set these up with his brother for radio installations in the 20's. A longer battery cable was used that passed through the floor into the cellar. All the power stuff was on a shelf the hung from the floor joists.

Vintage current relays can be found on-line that were used for the purposes of powering a battery radio. Some vintage battery eliminators had the relay built in. Other eliminators like the Exide, had everything in one box. The Exides do need refurbishing as they used a chemical rectifier...

GL

Chas

The radio's ground should never go to the house wiring ground and should be grounded via a rod driven in the ground or metal cold water pipe provided the pipe is metal all the way back to the water meter.

Well, my A-K 40 is currently (no pun intended) grounded to the mains ground, with no apparent ill effects. I really don't have a way to run a ground wire - or an antenna, for that matter - outside right now (it's my parents' house, so I'm a bit limited in what I can do), and there are no metal cold water pipes I can attach to, either, that I am aware of (my parents have a well, and I believe the pipe is PVC or some other plastic going out to the well-pit).

The 40's rf return indeed is connected to ground the circuitry for filament reference is isolated from the power line via the transformer.

In the 35 the rf return return to earth is directly connected to the controlled A-, a power supply that has A- grounded or returning to power line ground via the wall outlet could cause a burnout of the filament rheostat or undesired operation of the filaments at reduced voltage and possibly the introduction of hum...

In designing an ABC supply all voltages must not reference ground or a metal frame of it's cabinet.

I've collected many battery sets over the years, a majority of them have burned out rheostats or other filament dropping resistors roasted. I suspect that this may have come from the improper connection of an "A" supply or the use of earlier 1 amp tubes...

YMMV

Chas

So, what can I do in the absence of an earth ground? Can I use the mains ground using a small capacitor in series with the ground wire to electrically isolate the set?

In this application, a regular metal cold water pipe should provide a sufficient enough ground. Drill a 1/4-20 clearance hole in an auto hose clamp excess strap, and fasten your ground wire to it with a lug ring, a 1/4-20 bolt, lock washer & nut. Next, clean up a small section of the metal cold water pipe before securing the hose clamp.

Won't work, for a couple of reasons:

1) All PVC pipe in the house - little/no copper or iron

2) This is one of the radios I wanted to have actually working in the display at the public library in July, and mains ground is the only thing available.

An rf return isolating capacitor connected between the receivers ground and the "earth" connection, even if it is the house wiring ground would work to some extent. Depends how far away the outlet is from the power panel and its true earth ground. Use a .001, it will pass the rf and very little, if any, 60 hz component. At least it would block any dc that could circulate should a homebrew "A" supply have its "A-" connected to line ground.

A loop, tuned by a parallel variable capacitor, no antenna or ground, could be used with the 35. However, if the home is concrete, stucco over wire lath or aluminum sided a loop will be very problematic.

Suggestion of a cold water pipe is very good. However, if the house is plumbed with PVC, PEX or other plastic pipe, taking advantage of that length of pipe that connects to the well head is moot if it is plastic.

If a loop is is practical, remember to keep it as far away as possible from the radio and the power source.

You really should take a long look at how these vintage receivers you want to operate will get their signals in the home you are at...

BTW: Most weak, '01A's can be rejuvenated, either with a tube tester or with a vintage tube rejuvenator, made for that purpose in the 20's. Also, 112's are better for power output. A known gassy 01A is a better detector, an 00A gas detector triode is the best yet...

Remember, bias for the A-K 35 affects several tubes. Setting bias too high to satisfy a power output tube will cut off tubes that have lesser B+.

A fully charge LA battery of 3 cells is 6.6 volts. Set any power supply for at least 6 volts, the rheostat makes up the difference end result is 5 volts at the tubes the difference in the rheostat also provides bias.

The batteries when used with a battery radio provided a certain amount of capacitive bypass for the B+ circuits. A power supply may have to have a 1 mf capacitor across all B+ leads to prevent "motor-boating" of the circuits especially if the battery cord is of a long length.

YMMV

Chas

Whoops! I see that you mentioned this earlier, but I somehow missed it.

The loop antenna might be the best option for me. I'll have to make one somehow, more than likely, or buy another used Terk Advantage (the one I have is hooked up to my farm radio). Right now I have something like 30' of enameled copper wire run around the molding along the ceiling of my room for my A-K 40. Once the basement apartment is finished I'll have a bit more room for wire. The house is a double-wide mobile home set on a basement foundation made of concrete-cored foam blocks, which is why the plumbing is all plastic.

The battery cable on my A-K 35, by the way, is about 5 feet long (I had to shorten it due to broken wires, unfortunately).