[Radio Wrangler]

Some explanation for Horacebatchelor....

Usual sorts of valves come with an anode, a cathode, and one or more grids.

Normally, the anode is at a high positive voltage with respect to the cathode. Several tens of volts to a few hundred and even more in high power jobs.

The real work of the valve is to pass current internally from its anode to its cathode and to allow that current to be controlled. The anode is connected to circuitry which uses the variations in the valve current to pass the amplified signal to the next stage, or to make sounds in a speaker.

This sounds OK? It's what everyone says and it agrees with convention. But convention got it wrong! A long time ago, people looked at basic electrical circuits and said "Something is flowing around and around it... that's why it always has to make a loop in order to work" so the idea of electric current was born and people had the idea of little things carrying small amounts of positive electric charge coming out of the anode terminal of a battery, flowing along wires through the bulb and switch of their basic circuit and back into the battery via the negative terminal. All seemed well. It explained what they saw. Then along came trouble, the valve. Some of the things it did threw the convention of the direction of electric flow straight down the pan. Yes there were things flowing, but they moved in the OPPOSITE direction to what people thought and they carried NEGATIVE electric charge. Well, it was a 50:50 chance and the guessed flow direction was wrong. We have been living with this whoopsie ever since.

So let's look at that valve again. The heater heats the cathode to a high temperature. Heat means rapid vibration on a tiny scale. Hot enough to shake electrons (the moveable carriers of negativity) loose from their parent atoms in the cathode.

Like charges repel and unlike charges attract.

The electrons try a bit to repel each other and spread out to form a cloud around the cathode. They like the anode and would like to zoom off, accelerating right until they smash into that anode. They can feel the high positive voltage on the anode attracting them. THis is how a diode valve works. The hot cathode can emit electrons, the cold anode can't.

If we put a grid or grille of fine wire near to the cathode and make it several volts negative, the repulsion from it can oppose the attraction from the much further-away anode, and leave the electrons hanging about near the cathode, repelling others back to the cathode.

This is called the control grid, or grid 1 if there are others. Its proximity to the cathode compared to the distant anode means that small voltage changes on the grid can oppose large voltage changes on the anode. This is what gives the valve gain.

So in valve in normal use the amode may be a few hundred volts more positive than the cathode, and the control grid needs to be several volts more negative than the cathode.

In ye olde days, sets had special 'grid bias' batteries to provide the needed negative voltage. They were found rather inconvenient. So they had the idea of using a resistor in the cathode circuit, so the cathode current would make the cathode run several volts positive of chassis potential, and the grid was biassed to chassis potential. This loses a little efficiency but adds a lot of convenience.

The cathode and anode are carrying real current, but the control grid is carrying none. Electrons don't go there, it is more negative than where they came from, it repels them. So the grid is very easy to drive and bias... you have to control its voltage, but it doesn't take any noticeable current. So you can bias it through big resistor values, which reduces the load on the signal coming from the preceding stage.

So when one valve is driving another, the anode of the first valve is hanging around some rather high voltages, but the grid of the second valve has to be around chassis potential. The cheapest, simplest coupling that will pass the AC signal and yet block the huge DC difference is a capacitor. THey work well and give you great bandwidth.

BUT if the capacitor goes electrically leaky... say moisture gets into it, it only takes a very tiny amount of leakage current to drive the grid of the following valve positive. This opens that valve up to pass its full current.

In most stages, a leaky coupling capacitor will stop the set working and will shorten the life of the following valve. Not good, but not a disaster.

BUT if the capacitor blocking the DC to the grid of the audio output valve goes leaky, the output valve is designed for large currents and has lots of current available to it. This valve will take damage and the high current can burn out output transformers and even entire power supplies.

Audio output valves are expensive, there is high demand from the audio fraternity pushing up prices for spare glassware.

Audio output transformers are very expensive and harder to find than valves.

So, leakage is important in ALL coupling capacitors between stages, but the coupler to the audio output stage puts a lot more at risk. Calling it 'That capacitor' isn't a very good name. It's the expurgated version. Usually they get the benefit of a good fruity adjective applied to them.

Hope this cuts through some fog and confusion.

David