transformer based speaker selector

I want to build a speaker selector and control it with an Arduino and serial controls. I don't like the idea of using resistors for impedance matching because of the power they consume.
Using this page as a guide, (can't post links yet) but it's an article on bsae1 about impedance matching transformers)
I get the following. My amp can deliver 100w into an 8Ω load. This is 28.3 volts. If I now use a 1:2 transformer and wire all 4 speakers in parallel (32Ω), my V will = 56.6. 32Ω at 56.6 V will be the full power of 100W that the amp can deliver. I can do something similar for 2 and 3 speakers but I will use a different ratio transformer. My question in this scenario is, can a speaker be damaged by a higher voltage, or is it only the net power that it cares about? I'm looking at this transformer, Hammond model 117K32,: (again, can't post link) I think it will give me the flexibility to use one transformer and be able to get the impedance fairly close regardless of whether I have 2,3, or 4 speakers selected. (none needed for one speaker since they are 8Ω speakers)
Thanks
 
I already have multizone. This is simply one of the zones in my multi-zone setups. These speakers will always be playing the same source and I don't need to control the volume independently. I'm aware that they sell impedance matching speaker switches, I have one, I've dissected it, and frankly, I don't like the design. Most of them claim to be impedance "matching" or "impedance protection", but in reality it is nothing more than a resistor in series with 4 parallel switches which turn the speakers on and off. This doesnt match anything, simply prevents your amp from seeing to little impedance, at the expense of less power to the speakers, and increased heat in your media cabinet. Too me, that seems like a worse idea than mine. I believe that with a little help from the net, I can do better. But I need specifics on what's wrong with my idea and how I can fix it. I've looked at higher end units and they employ a transformer design. However, they all seem to lack the ability to control them remotely, which is odd since that is the easy part, at least for me. I don't want to spend the money on one of these just to reverse engineer it so I'm hoping that someone already knows what is inside.
 
Oops! Brain Fart! 4 - 8Ω speakers in parallel is 2Ω, not 32Ω. The picture is a little more grim with that correction. Assuming the amp puts out 100w @ 8Ω, that would be 28.3V. Now run that the other way, 2:1 stepdown, I will end up with 2Ω, 100w (25 per speaker) but my voltage will only be 14.1V. I'm okay with the 25W per speaker. They are good down to 5W. What I'm not sure about is the 14.1V. The furthest speaker is about 200 feet away from the amp. Would this present a problem?
 
If having all four speakers active yields 25 watts per speaker would then switching to just one speaker, now 100 watts per speaker, result in a noticeable change in volume?
 
Audio runs the 10X factor. A speaker running 10W would need 100W to be twice as loud, so some noticable volume going from 25 to 100 would exist.
 
Back
Top