Necessity, as they say, is the mother of invention. My necessity was something to check both PA speakers and the lines feeding them as they were temporarily placed in position for surf lifesaving carnivals.
For many years, I’ve erected temporary PA systems – up in the early morning, down that afternoon. Usually, that’s been a matter of placing perhaps eight horn speakers over a distance of maybe 600-700m, all fed from a central PA amplifier located where the carnival announcer sits.
With eight 30W speakers, a 250W amplifier handles the whole thing quite nicely. But feeding those eight speakers over such a distance demands they not be your usual low impedance (ie, 4, 8 or 16Ω) speakers; to minimise losses they must be so-called “100V” types.
What this means is that the output from the amplifier is (internally) stepped up by a transformer so that the lines to the speakers are fed by a 100V signal. At the speaker itself, the reverse happens – the 100V is stepped back down again by a similar transformer so that the low impedance speaker driver is presented with just the right level.
Fig.1: as circuits go, it’s pretty simple: a 555 timer creates a square wave which is amplified, then fed into a 100V speaker transformer and on to the speaker. It’s capable of delivering a little over 1W but it’s not exactly hifi!
Why go to all that trouble?
The answer is simple: to minimise losses in the speaker cables.
While copper cable is a very good conductor, it does have some resistance. Typically, I use lightweight (14 x 0.14) Fig.8 cable, which according to the reference books has a resistance of about 16Ω/100m (ie, 8Ω per side).
In a home hifi situation with only a few metres of cable between amplifier and speaker that resistance doesn’t matter too much but when your speakers are up to several hundred metres away from the amplifier, resistance of the cable has a major impact.
Inside the box: everything except the transformer, output terminals and batteries mounts on a single PCB. The batteries in their holders can just be seen underneath the board.
If, for example, I was to drive an 8Ω speaker 300m away from the amplifier, the speaker line itself is going to act like quite a large resistor in series – about 48Ω (3 x 16Ω) – and I am going to lose 48÷56 (ie, line resistance divided by line + speaker resistance) or 85% of the signal before it gets to the speaker.
In fact, it’s even worse than that because inevitable corrosion in the connectors etc means I’d be lucky to have even 5-10% of my original signal left at the far end. And the further away your speakers are, the worse it gets.
With a 100V PA system, the losses are much, much lower. The impedance of a 30W 8 ohm “tap” on a 100V audio transformer is calculated as (100V2÷30) or 333 ohms. So now we have a 48Ω speaker line in series with a 330Ω load. Therefore the loss is reduced to (48÷(48+330)) or about 12% – much more manageable.
No impedance problem, either!
There is another huge advantage: with multiple speakers, you don’t have to worry about impedance matching. With a 100V line system, all speakers are connected in parallel/in phase and all you need to do is add up the wattage which each speaker is running at (and that simply depends on the tap you use on the speaker’s transformer) and make sure the total doesn’t exceed the rated output of the amplifier.