More Than You Wanted To Know About Vacuum Tubes

“Leonard Hollar” <> wrote:

What did Edison use to create the vacuum for his carbon filament light bulbs?

Answer: It doesn’t matter.

We’re not talking about light bulbs, we’re talking about vacuum tubes

The function of a vacuum tube (Or a transistor for that matter, but we’re talking about vacuum tubes. Semi conductors for another day)

As I was saying before I interrupted myself, The function of a vacuum tube is to amplify a signal. To do this, effective vacuum tubes have three parts:

Part 1
a cathode — think the filament of a light bulb. Something that gets so hot that electrons boil off of it. We use tungsten for this primarily for two reasons, electrons boil off of it really well, and it gets really really hot without melting, which is a good thing for a cathode to do. (More on Cathodes later.)
Part 2
an anode, this is just a big tube that surrounds the cathode and is a place for the electrons to go that boiled off the cathode.

The “Edison Effect” is that, in a _very very very hard vacuum_ there is a stream of electric current from the hot filament in the center, to the anode around the outside with NOTHING TOUCHING between them. If the tube has any gas in it (like lightbulbs do, they’re filled with argon or xenon or something) then the electrons hit those gas atoms and “do stuff” with the gas, and do not make it from the cathode to the anode. SO you have to get rid of every last tiny tiny bit of gas in a tube for the tube to function as an vacuum tube and for the Edison Effect to show up. Once you have pumped the tube down really really good with your refrigerator vacuum pump, which is PLENTY good enough for a light bulb, you still have to get it about 10,000 times better vacuum in order to make a good vacuum tube.

But wait Rick, you said THREE parts and so far you’ve only mentioned TWO parts. AHA! Well, the third part is called the GRID. and it goes between the cathode and the anode. It’s called a grid, ’cause it’s made like window screen, it’s full of holes for the electrons to pass through on the way from the cathode to the anode.

A tube can be thought of conceptually like this:

      ====      |           )
c      /        |           )
a              |           )
t      /        |           )
h              |           )
o      /        |           )
d              |           )
e     ====      |           ) anode
                | grid

Now, consider.. the cathode is white hot. Electrons (which have negative charge) are boiling off it. They are RUSHING towards the anode and there is a current between the cathode and the anode.

What happens if I put a small positive charge on the grid? The electrons, instead of rushing on to the anode are SUCKED into the grid and the signal to the anode is blocked. heheheheh

What’s cool about vacuum tubes (and transistors but that’s another paper) is that SMALL signals applied to the grid, can result in BIG changes in the amount of current making it to the anode. So, the weak tiny little signal coming off a crystal radio receiver, which one person can only hear if they stick it inside their ear, attached to the grid of the tube, can make a change in the current going to the anode so big that it can cause a loudspeaker to wave back and forth and make enough sound that the entire BAR can hear the radio instead.

(Yes, yes, yes, for all you electronics types, I didn’t discuss grid bias, or plate bias, or impedance matching or …. oh come on, this is in reply to a question about why we can’t fill the tube with nitrogen remember? Yeah, I thought you had forgotten that.. Right, on with the story)

So what’s so cool about tubes then, is that they can AMPLIFY a small signal to be a BIG signal. Typical tubes can produce an amplification of about 150 times.

In order to do that, making the tube is tricky. Let’s re-visit the parts.

First the cathode. You want the cathode to release electrons easily. There are generally two ways to do that. One is to make a tungsten cathode, and impregnate the surface with thorium. The radio-activeness of the thorium, and the high temp of the tungsten cause it to allow lots and lots and lots of electrons to boil off. This is a good thing.

The other way to do it is to make a little stainless steel tube, and coat it with Barium salts, and run a toaster wire through it at red hot, and indirectly heat the barium salts on the steel tube to act as your cathode. This is called a “cold” cathode, which it isn’t. But the actual cathode itself, the tube with barium on it, isn’t glowing like the tungsten wire in a HOT cathode is. Still, it gives off lots of electrons. Cold cathodes are harder to make PHYSCIALLY, (running the wire through without shorting it out etc etc), but simpler to make in terms of the materials required. (Making tungsten wire is “hard”. )

Second is the grid. The grid needs to be made of very very very very very fine wire. (We’ll probably be using gold) since the thicker the wire, the more of the current between the cathode and the anode it intercepts UNINTENTIONALLY. Designing grids is hard.

Third is the anode itself. Since it’s being impacted by all the electrons, and since it’s AROUND the grid and the cathode, it gets REAL HOT. In low power tubes, the anode is copper or steel. In high power tubes the anode is sometimes graphite. In small tubes, you can just cool the tube with the air running past it. In larger tubes, you have to arrange a fan to carry off the heat. In very very large tubes, water cooling jackets are sometimes used.

OK, so you build your cathode, anode and grid, put them in the tube, fish the wires out to the outside world, pump down the low vacuum using your refrigeration pump, turn on some current to the cathode, and get everything hot while you keep the pump running. this “out-gasses” the gasses trapped on the surface of the stuff… Once everything is nice and hot, you turn on your diffusion pump and go from 10E-2 to 10E-4 torr, down to 10E-8 to 10E-10 torr that we need for the tube to work well. Then, we heat the glass tube coming off to the vacuum pump and twist the glass off, and seal the tube.

In a modern tube, at that point, you would run a current through a little strip of tantalum or barium you had put inside the tube, which would boil off onto the walls of the tube, capturing the last few atoms of air. This is called the “getter”

For more information, and better descriptions, go read the “How tubes work” FAQ in the FAQ site that Michael’s running.