So, I'm a little fuzzy on vacuum tube designs, but I am under the impression that some use the filament directly as the cathode, while some use the filament to heat a plate which is the cathode. Since the filament is what burns out on a vacuum tube (I am aware of sputtering, but that failure takes longer to happen), why aren't tubes built with an external, replaceable filament? Furthermore, why do filaments actually burn out as often as they do? Toasters, for example, seem to have a very long lifetime, and their filaments are in air.
-
1\$\begingroup\$ If they burn out often, check the heater voltage! Lifetimes of multiple decades are not uncommon in practice, absent voltage surges or high vibration. \$\endgroup\$– user16324Commented Jan 15, 2014 at 11:22
-
1\$\begingroup\$ The filament isn't what burns out. More usually they get gassy and lose Ra. \$\endgroup\$– user207421Commented Jan 15, 2014 at 11:54
-
\$\begingroup\$ So I'm not actually using vacuum tubes, just curious. I had no idea they could last multiple decades. My perception was that they last a much shorter time. I see I was mistaken. \$\endgroup\$– Void StarCommented Jan 15, 2014 at 13:59
-
\$\begingroup\$ @BigEndian I know someone who still uses (regularly and at unbearable volumes) a vacuum-tube based audio amplifier made in the early 70s. It is a Blaupunkt amp, if I am not mistaken. So yeah, decades :-) \$\endgroup\$– Anindo GhoshCommented Jan 15, 2014 at 15:36
-
\$\begingroup\$ Yes, but that's intermittent use. My dad has two tube amps which he's owned for a very long time, but the reason they're still working is, I would suspect, because they haven't been in constant use this whole time. \$\endgroup\$– Void StarCommented Jan 15, 2014 at 20:57
1 Answer
Some tubes use directly heated cathodes. The filament is called the 'heater' in tube land and the part that the electrons are emitted from is the cathode. (The term 'plate' is reserved for the anode where the electrons are collected). Most use indirectly heated cathodes, which allows heating with AC. I have seen a lot of heaters made with very fine filaments with an oxide coating, usually barium oxide. The oxide is fragile and it evaporates over time and leads to the burnouts or shorts. In some heaters the filaments are packed into a small space and go back and forth and heat a cylinder of thin metal with the oxide coating on the outside.
Why the oxide? Tungsten has a work function of about 4.5 eV or electron volts. The electrons must have this much energy to escape the metal. To give them the energy, the heater is run very hot, like 2700C like a light bulb. With the oxide coatings the work function is about 1.1eV. The energy of the electrons in the heater goes up with temperature exponentially. So, getting really hot makes a big difference. But so does reducing the work function and running at lower temperature which uses less power and extends the life. At 1.1eV the heater can be a dull red heat instead of light bulb white.
Heaters are not replaceable because the glass tube is sealed at both ends (or a sealed metal can for some devices) and maintains a hard vacuum. In fact, after as much air is removed with a vacuum pump, the tube is sealed and a "barium getter" is heated electrically to react with any remaining gasses. The result in a very good vacuum. A side-effect makes the mirror-like spot on the side of the tube.
A toaster needs to produce heat, like 1500 Watts, so a large wire is used and a lot of current. The tube needs to have a hot cathode which can be very low mass and is in a high vacuum. A fine tungsten wire will get hot with less current and consume less power. But it will also be easier to damage with heat and vibration.
Direct heated tungsten was used until about 1930. Tungsten with a little thorium in it was found to have a lower work function and was used until the barium oxide effect was discovered.
-
1\$\begingroup\$ A long time ago I read an article about a car tube-radio design that used the car's engine heat to heat specially constructed (IIRC home-made) tubes. It might have been in the april edition. \$\endgroup\$ Commented Jan 15, 2014 at 9:25
-
2\$\begingroup\$ @WoutervanOoijen do you have that special day in April as well? \$\endgroup\$– Andy akaCommented Jan 15, 2014 at 9:39
-
1\$\begingroup\$ Why else would I have mentioned the month :) \$\endgroup\$ Commented Jan 15, 2014 at 10:09
-
\$\begingroup\$ I hadn't realized the kind of vacuum required until I saw a YouTube video where someone made a few simple triodes. The vacuum required is deeper than can be practically maintained in any dynamic vacuum system; even the tiniest hint of gas will turn what should be a smoothly operable valve into a thyratron which, as soon as any current flows, will attract gas to a path between the anode and cathode, which will then carry current freely until the potential difference drops to zero and the gas can dissipate. \$\endgroup\$– supercatCommented Feb 17 at 20:28