Thank You. I am trying to come up with another article but am lacking in inspiration. Everything I have come up with is either to much of a liability or just not interesting enough.
The biggest downside is the capital investment required to work with it:
High vacuum system (at least a vane pump and diffusion pump), bell jar (for experiments)
Glassworking torches
Assortment of glass rods, tubes, etc.
Glass blowing experience(!)
An assortment of unusual metals, for making glass seals, electrodes, etc.
Spot welder (usually a small capacitive discharge type)
Glass lathe preferred
And that's just to reliably create demonstrations, mere novelties. To actually make something practical, requires even more: jigs and dies to form very precise parts; oxide cathode materials; support materials (usually mica or ceramic); etc.
I would love to see what could be done with modern techniques, though. Add to the list: laser cutter, general 3D printer (probably one capable of cementing ceramic and metal powders), mill (including ceramic cutting bits), wire bonding, etc.
Probably the fanciest thing that could be made today, would be a planar triode or tetrode, using a layered ceramic base, with metallization for placing the electrodes upon. The filament can be strung across an opening, or the cathode supported on fine springs over a heater. The grid can be laser cut, hopefully with a very low cross section; otherwise, perhaps a wire screen or mesh can be made without too much trouble. Subsequent grids can be mounted in the same way, supported on tiers of the ceramic base. A plate goes on top, perhaps brazed in place, in vacuuo, also sealing the assembly and providing a heatsink base.
The trick for really good tubes (high Gm, low loss = high bandwidth) has always been, closer grid-cathode spacing and finer grid wires. Planar triodes got to about 10GHz in the 60s and 70s, though still with too much capacitance (a few pF) and too high of a load resistance (~kohms) to achieve more than a narrow bandwidth (like, 5GHz +/- a few percent).
The "next big innovation" has always been cold field-emission cathodes, but I guess I don't see any reason they should work out; many things have been tried and they remain unreliable. You can't make a space charge, anyway, which is part of the charm of a proper vacuum tube (though not necessary).
Even so, the load impedance limitation doesn't seem to have any obvious solution. The problem is current density, which is limited by the electron space charge. If you neutralized the charge with a bulk positive charge in the vacuum, well, by gum, you'd have a semiconductor with free conduction electrons (rather than free-as-in-vacuum electrons). Indeed, it's not at all wrong to call a tube a "vacuum FET". The simple fact that conduction electrons can be packed about 10^6 times more dense, despite also running 10^3 times slower (drift velocity in Si < ballistic velocity in vacuum), means transistors win by a long shot.
Tim
Home
Help
Search
Login
Register
