Speaker Impedance

Archived from groups: rec.audio.tech (More info?)

It seems that most high-quality speakers are 4 ohms, and even less at
some points. Is there some reason for designing for low impedances,
some inherent advantage in the driver or crossover design? Maybe
they are designed to have low impedance merely to take advantage of
high-end amplifiers (which are really optimized for low impedances)?

Archived from groups: rec.audio.tech (More info?)

calcer...@hotmail.com wrote:
> Most Altec 604s are 16 ohm.
>
> Solid state amps generally have plus and minus rails of relatively
low
> voltage (Arny knows what's popular right now, ask him) and no
coupling
> transformers or autoformers so they make power best into lower
> impedance loads.

That, unfortunately, completely fails to address the poster's
question: why 4 ohms? Why not 8 ohms, or 2, or 0.5.

> rail voltages used. If 150 to 600 ohm speakers were available-and
> feasible-OTL tube amps would be common and would be a desirable
thing.

It's entirely possible to design 10 ohm or 600 ohm speakers,
and back in the days when tubes where the ONLY choice for power
applification, speaker were STILL designed with impedances
substantially lower.

Well, the reason has very little to do with what solid state
amps can or can't do as, again, evidence by the fact that
speaker have always had predominantly low impedances. The reason
has to do with attempting to optimize the electromagnetic
efficiency of drivers. Without getting deeply into the details,
when you take into consideration the available magnetig flux
density over the volume of easily and affordably manufacturable
magnet structure (on the high-side, it's limited to a little
more than 10,000 gauss, dictated by the saturation magnetization
of the soft low-carbon steels used for such structures), and that
to chieve the highest transduction ratio (amount of force for given
current), you want to fit as long a piece of wire in the gap as
you can. But you immediately run into the fact that the longer the
wire, the more its electrical resistance, and that current is now
turned into relatively useless heat. SO you make the wire fatter
for less resistance, but now, since the gap is of limited volume,
you can't fit as long a piece in there, and, while your ohmic
efficiency now went up because of the lwoer resistance, your
transduction ratio went down becuase of the shorter length.

So you play this tradeoff game, and then throw in complications
like damping (interelated functions of transduction ratio,
resistance and mass) and all sorts of things. You take all these
factorsm many determined by available, affordable materials (like
iron for magnet parts, copper for wire), and also add in other
factors like your REALLY don't want several hundred volts floating
around your magnetic gap where your spacing is VERY small and the
potential exists for very high voltage gradients), and you plug
them into a multi-variate optimization program and, lo and behold,
you end up with this multi-dimensional graph with some bumps in
it, with one axis being impedance. And one of those bumps, and
a fairly broad one at that, happens to be betweeen a couple of
ohms and about 20 ohms or so. That's where all these factors
can be optimized for the given cost of producing the drivers.

Now, more specifically, why 4 instead of 8? because there are
even more limiting issues, like wire only comes in discrete
diameters, magnets come in discrete sizes, pole pieces and front
plates come in only some very specific sizes. So you see clusters
around 7-8 ohms, clusters around 3-4 ohms, and so forth, because
given the "quantized" nature of the materials going in, those are
local maxima in the optimization matrix.

We could change that. If, instead of wire gauges coming in things
like 26, 27, 28, 29, 30, they came, all of a sudden, in 26.5, 27.5,
28.5, 29.5, 30.5. And instead of standard pole pieces coming in 1",
1.25" 1,5", 2", they suddenly were available instead in .87", 1.13"
1.35", 1.75" and so on, then instead of 16 ohm and 8 ohm and 4 ohm
drivers, we'd find it easier to make 12 ohm, 6 ohm and 3 ohm, maybe.

That being said, I have designed 12 ohms drivers that did exactly
what I wanted them to do. I just had to beat on the driver assembly
plant to actually make them: they were VERY uncomfortable with the
fact that it was not 4 or 8 ohms. Idiots.

But, again, to counter the first answer, it was not amplifiers that
dicated the impedance of speakers, it was the other way around.

Archived from groups: rec.audio.tech (More info?)

Oh yes, there were high impedance drivers specifically for OTL tube
amp use designed in the 50's and early 60's, and 25 and 32 ohm drivers
aren't rare. The very high impedance drivers had finely wound
multiturn voice coils that were heavy, went open, shorted between
turns, hit the magnet, or corroded frequently, but aside from that
worked fine.

Archived from groups: rec.audio.tech (More info?)

On 24 Mar 2005 18:09:23 -0800, dpierce@cartchunk.org wrote:

>
>We could change that. If, instead of wire gauges coming in things
>like 26, 27, 28, 29, 30, they came, all of a sudden, in 26.5, 27.5,
>28.5, 29.5, 30.5. And instead of standard pole pieces coming in 1",
>1.25" 1,5", 2", they suddenly were available instead in .87", 1.13"
>1.35", 1.75" and so on, then instead of 16 ohm and 8 ohm and 4 ohm
>drivers, we'd find it easier to make 12 ohm, 6 ohm and 3 ohm, maybe.

But, isn't most of the spaekers these days designed and manufactured
in millimeter-land? You know, we have over here no problem of using
decimal fractions of anything, or even step down to next
milli/micro/nano-whatever! :-)

That said, I still say that a bass driver is an 8 incher, when I mean
200 mm (which should perhaps be called 2 dm, taking the number of
accurate figures into account).

Per.

PS. Thanks for the story on driver design tradefoffs, Dick,
informative as always!

Archived from groups: rec.audio.tech (More info?)

"Michael Squires" <mikes@cs.indiana.edu> wrote in message
news21arb$2s3$1@rainier.uits.indiana.edu...
> High efficiency was the rule since it was hard to buy anything more
powerful
> than a 30 wpc amp in the 50's and a 60 wpc amp in the 60's, which is why
> bass horns and horn-loaded mid/high drivers were popular and the JBL
> Hartsfield was considered state-of-the-art.

The Hartsfield and similar systems like the Tannoy GRF, are considered state
of the art by some people today!
Check the price they are fetching for mint examples.

MrT.

Archived from groups: rec.audio.tech (More info?)

On Sat, 26 Mar 2005 12:28:59 +1100, "Mr.T" <MrT@home> wrote:

>"Michael Squires" <mikes@cs.indiana.edu> wrote in message
>news21arb$2s3$1@rainier.uits.indiana.edu...
>> High efficiency was the rule since it was hard to buy anything more
>powerful
>> than a 30 wpc amp in the 50's and a 60 wpc amp in the 60's, which is why
>> bass horns and horn-loaded mid/high drivers were popular and the JBL
>> Hartsfield was considered state-of-the-art.
>
>The Hartsfield and similar systems like the Tannoy GRF, are considered state
>of the art by some people today!

Only by those who don't understand the term.....................

>Check the price they are fetching for mint examples.

Irrelevant - check the price that a mint Rolex Daytona fetches, and
relate that to the price of a Casio G-Force Waveceptor, locked by
radio link to an atomic clock. Which one represents the state of the
art in timekeeping?
--

Stewart Pinkerton | Music is Art - Audio is Engineering

Archived from groups: rec.audio.tech (More info?)

In two or three hundred years that Rolex Daytona will still work. If
you build a spacecraft and fly to a suitable planet around any of half
a dozen stars you could get to in a human lifetime at (99%)
lightspeed....it will still work. if you invent a time machine and go
back to 77 BC....it will still work. If Randall Terry becomes President
and our civilization sinks to a level like that in Europe in 800
AD....it will still work.

Relevant? Maybe not. But the Rolex is a work of art and the Casio is a
consumer piece.

Archived from groups: rec.audio.tech (More info?)

calcerise@hotmail.com wrote:

> In two or three hundred years that Rolex Daytona will still work.

If used regularly it won't still work without servicing and repairs
and if replacement parts are not available it won't work at all.
Far cheaper to buy a Casio every ten years.

> Relevant? Maybe not. But the Rolex is a work of art and the Casio is a
> consumer piece.

Better to keep your art and tools separate.

--
Eiron.

Archived from groups: rec.audio.tech (More info?)

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:1n1a419c8v05ntg5pfri78l6uam2la5m72@4ax.com...
> >The Hartsfield and similar systems like the Tannoy GRF, are considered
state
> >of the art by some people today!
>
> Only by those who don't understand the term.....................

True, but they were pretty good for their time though.

> >Check the price they are fetching for mint examples.
>
> Irrelevant - check the price that a mint Rolex Daytona fetches, and
> relate that to the price of a Casio G-Force Waveceptor, locked by
> radio link to an atomic clock. Which one represents the state of the
> art in timekeeping?

Even more irrelevant! :-)

MrT.

Archived from groups: rec.audio.tech (More info?)

"Arny Krueger" <arnyk@hotpop.com> wrote

>> ...But the Rolex is a work of art and the Casio
>> is a consumer piece...

> Thanks Cal for showing once again that you haven't got the
> slightest
> clue as to what SOTA means. Hint: it is relevant to the present
> time.

How then would you distinguish between SOTA and fashion, Arny?

cheers, Ian