Reasoning behind 8ohm speakers?

[Gandledorf]

What is the reasoning behind 8ohm speakers, and why this measurement is important? It seems like this would be awfully difficult to drive...

 

[ljcox]

8 Ohm speakers

I'm not sure I understand the question. I assume you want to know why speakers need to be a specific impedance.

It is because the speaker impedance must match the output impedance of the amplifier in order to obtain maximum power transfer.

So if the amp is designed for an 8 Ohm output impedance, then you will obtain maximum power into an 8 Ohm speaker.

Len

 

[Gandledorf]

Re: 8 <a href=

I'm not sure I understand the question. I assume you want to know why speakers need to be a specific impedance.

It is because the speaker impedance must match the output impedance of the amplifier in order to obtain maximum power transfer.

So if the amp is designed for an 8 ohm output impedance, then you will obtain maximum power into an 8 ohm speaker.

Len

Ok, I think I see. I'm just used to designing loads with high input impedance, going by the rule of thumb that the load impedence should be >> source impedence. I suppose though that there is the fact that maximum power transfer occurs when Rload = Rsource.

Do most amplifiers have an output impedence of 8 ohms? If I were to design a microcontroller based system, with a speaker, how close do I need to match the impedence?

 

[ljcox]

Yes, if you do the maths, you will see that max power is transferred when Rload = Rsource.

Amps are usually 8 Ohm. Some used to be 16 Ohm and there are also some at 4 Ohm.

The match does not have to be exact, but if the speaker Z is too low it could damage the amp.

I don't know what you mean by "If I were to design a microcontroller based system, with a speaker"

Len

 

[Gandledorf]

I don't know what you mean by "If I were to design a microcontroller based system, with a speaker"

Len

i.e. if I want to drive a speaker using an I/O line of a uC, for instance, using PWM to create buzzing, or with a slight bit of higher complexity, such as that described by this design note: **broken link removed**

Is there some sort of device that has to go in series with the speaker to match impedence? They recommend a 270ohm resistor, and a 100uF capacitor in series, but I don't see how this generates a matched impedence.

 

[Exo]

Use a device that has a high input impedance and a low output impedance to drive the speaker ... a transistor

 

[Nigel Goodwin]

Firstly, only valve (tube) amplifiers match the speaker impedance to the amplifier impedance - they can do this as they use an output transformer.

Transistor amplifiers work very differently, they have a very low output impedance, and feed into the (relatively high impedance - 8 ohm) of the speaker. The actual output impedance of a transistorised amplifier is usually in low fractions of an ohm, or down in the 10's of milliohms.

The reason for 8 ohm (getting back to the original question) is a question of power!. If you apply the formula W=(V*V)/R, you will see that for 20V output (RMS) an 8 ohm speaker will give 50W (400/8). If the speaker was 80 ohms, you would need 63V RMS for the same 50W.

For a valve amplifier, it doesn't matter much - you simply wind the secondary on the transformer so it matches. In fact, 16 ohm speakers were the most common back in those days.

With transistor amplifiers though, the output voltage governs the power, so by dropping the speaker impedance from 16 ohm to 8 ohm it doubled the available power for the same voltage output. Likewise for even more power, it's dropped to 4 ohm, particularly for in-car use, where you have a very limited voltage supply.

As for the 270 ohm and 100uF, it's making no effort to match anything, all it's for is to limit the load on the processor pin - the fact it wastes almost all the power doesn't really matter, you are only looking for low level beeping noises. As suggested, a simple transistor could be used to provide more power to the speaker, but I suspect you would probably find it a bit loud.

 

[pike]

8 ohm loads are generally used in household because the voltage available can be quite high.
Using 4 ohm speakers means you need less voltage but more current. This is especially true in cars as the battery voltage can only supply 12v but with very high currents. Even 2 ohm speakers exist!!

A speaker rated at 8 ohms doesn't neccessarily mean it is 8 ohms, my dads hi-fi system is rated at 8 ohms but the speakers are actually 6.3 ohms!!! but this will change from cone to cone so do your maths guys!

 

[Russlk]

Another, and probably most important reason that speakers are 8 ohms, is that the coil is attached to the diaphram of the speaker. A high impedance would require more turns of wire, and the weight would affect the sound.

 

[Nigel Goodwin]

Another, and probably most important reason that speakers are 8 ohms, is that the coil is attached to the diaphram of the speaker. A high impedance would require more turns of wire, and the weight would affect the sound.

More turns, but thinner wire - I wonder if there's an optimal point?.

 

[Nigel Goodwin]

A speaker rated at 8 ohms doesn't neccessarily mean it is 8 ohms, my dads hi-fi system is rated at 8 ohms but the speakers are actually 6.3 ohms!!! but this will change from cone to cone so do your maths guys!

I'm presuming you just measured the 'resistance' and not the 'impedance', probably just using a multimeter?. Speaker impedance should be measured at 1KHz, measuring the DC resistance will usually read lower, around 6 ohms is about usual for an 8 ohm speaker.

A speaker coil is an inductor, so it's impedance will vary with frequency, which is why you have to measure it at a specific frequency. Manufacturers usually provide a graph of the impedance against frequency, but all you need to do is read the value off the label to use it!.

 

[Exo]

A speaker rated at 8 ohms doesn't neccessarily mean it is 8 ohms, my dads hi-fi system is rated at 8 ohms but the speakers are actually 6.3 ohms!!! but this will change from cone to cone so do your maths guys!

It could also be that you're playing music so loud that the speaker's coil is starting to be shorted out at some places

 

[Nigel Goodwin]

It could also be that you're playing music so loud that the speaker's coil is starting to be shorted out at some places

Nasty thought :lol:

If that does happen you soon find out, it's going to sound horrible, probably smoke, and quite likely to kill your amplifier.

We used to play 'games' with old TV speakers years ago - plugging them into 240V mains, and seeing how far the cones would fly :twisted:

I DON'T ADVISE TRYING THIS!

 

[Gandledorf]

Thanks guys, this was all very informative, as usual

 

[stevez]

Like many standards I'd guess that 8 ohms became a "standard" as a result of evolution. As suggested in some posts - something different could be made to work. Maybe the 8 ohms was the result of what they could make economically at the time given the materials, technology, etc. Because the capital investment in manufacturing equipment is high it's often less expensive to leave some things as they are even though at some point it would appear that the standard is less than optimal.

 

[Nigel Goodwin]

Like many standards I'd guess that 8 ohms became a "standard" as a result of evolution. As suggested in some posts - something different could be made to work. Maybe the 8 ohms was the result of what they could make economically at the time given the materials, technology, etc. Because the capital investment in manufacturing equipment is high it's often less expensive to leave some things as they are even though at some point it would appear that the standard is less than optimal.

Yes, I'd really like to know how it came about - but I doubt it's down to capital investment in manufacturing equipment - presumably the same equipment can easily wind different coils (and probably does, for 4, 8 or 16 ohms).

As I explained earlier, the 8 ohm and 4 ohm derive fairly logically from the way transistor amplifiers work - it's really a question of a good compromise between volts and amps - you don't want the volts so high you get a shock off the speakers, nor do you want the current so high you can't bend the speaker wires.

Valve amps used to generally be 16 ohm - from a transformer it could be anything you want, possibly using a 16 ohm speaker made the most cost effective use of thickness of secondary winding and number of turns?.

I suppose really it's like discussing why we use base 10 to count, it's just something which evolved!.

One interesting point!, Germany commonly use 4 ohm speakers (or at least they used to), whereas the rest of the world usually use 8 ohm.

 

[pebe]

.......For a valve amplifier, it doesn't matter much - you simply wind the secondary on the transformer so it matches. In fact, 16 ohm speakers were the most common back in those days......

Just about every domestic radio used 3ohm speakers.

 

[Nigel Goodwin]

Just about every domestic radio used 3ohm speakers.

True, I was refering to HiFi equipment - TV's were commonly 3 or 4 ohm as well, back in the valve days!.

 

[stevez]

Seems to me that I recall 16 ohms as well. Whole point was that vacuum tube/valve, as far as I know, wasn't directly coupled as transistor amplifiers are because of gross mismatch, so I'd guess that the standard had more to do with manufacturing - thinking possibly weight of moving coil, DC resistance of wire or something like that as having driven the standard originally. While not quite like asking why RR tracks are what they are it's an interesting question.

 

[ljcox]

8 Ohm speakers

Originally, even transistor audio amps used a transformer match the amp output to the speaker Z. They also used a transformer to drive the output stage. The transformer was used for both matching and phase splitting to drive the push pull output stage.

Len