I’m finally getting my electronics workbench back into commission after a couple years of general weariness. Days go better with a little soldering and experimentation and Real Science™ in the mix.
For instance, I was uncertain about the bidirectional current in the induction loop of the Phelps Railroad Telegraph, so I set up a quick test with a switch and a transformer and a center-zero ammeter. Sure enough, the secondary pulses one way when the primary is turned on, and the other way when the primary is turned off. Took about 15 minutes and created far more real learning than looking it up on Youtube!
This brings up an old puzzle. The early radio experimenters mainly laid things out on a table and lashed them together with wires. Why did premade Kits take so long to appear? The first poke-in breadboards were sold around 1965, designed to fit transistors and ICs.
The tube version would necessarily be larger, but it would have the same general form. A matrix or checkerboard of connectible spaces.
A new upload at American Radio Library answers the question. DeForest made such a matrix and called it the Unit Panel system.
Here’s the front of a Unit Panel filled with modules:
And a back view:
Each module carried a coil or capacitor or tube socket or potentiometer.
DeForest was also still selling its original Audion in 1922, along with the special socket for it.
Polistra's Mill 