[sdiy] Speaker Impedance

[Bob Weigel]

Harry....I kind of get stuck at the top.  If the speaker is an 8 ohm 
speaker, that's a SUMMATION (if you didn't previously know this..don't 
feel bad.  Somehow I got through all the physics and everything and it 
wasn't until I had worked on stuff for quite a while that I really got a 
crystal clear revelation of how things worked together in a speaker..)  
of all the reactive AND non-reactive impedences. IF YOU SAY the speaker 
has ONLY DC resistance in that 8 ohms...by definition..it ain't no 
speaker :-)
     The non-reactive ones (UNKNOWN TO THE SERVICE MANAGER OF ONE OF THE 
MAJOR COMPANIES I FOUND OUT THE HARD WAY..hehe.   I will save them from 
ebarassement and not say which one but it wasn't Roland or Yamaha :-) ) 
affect all frequencies exactly the same basically save the frequency 
effect I spoke of early which comes into play in the high Ghz zone 
somewhere probably.  I have a seriously hilarious storying involving the 
shipping/poor packing/destruction of several nice output 
transformers...and all the time it was because the boneheads in their 
service dept. DID NOT ZERO THEIR METER. And when I confronted the 
service manager with the physical facts....he tried to say "But that's 
DC IMPEDENCE..THIS IS AC THAT HAS NO BEARING!!"  Hehehe.  I'll omit the 
long story on that one.  But he was sadly mistaken and had not gotten so 
far as to firmly apply Kirchoff's law to his field.  It's..not 
uncommon.  There is vast confusion about the very topic we are covering 
here with tons of very experienced techs I've spoken with through the 
years.  And like I said I didn't really get it until several years back 
when I finally saw that we are talking about two DISTINCT (unless we 
consider capacitive effects which are usually considered negligable in a 
speaker in which case there would be three..)  factors going on.  The 
non-reactive impedence doesn't change.  What we measure as it operates 
to create heat as per the equation below.  The reactive/ AC impedence is 
quite dynamic.  It's always changing according to

* the applied voltage

* the DC impedence

* the fixed magnetic field (which also may have some fluxuations as it 
interacts with the fields of the voice coil, but they are usually 
considered negligable I believe with most speakers.

* the momentum of the speaker cone (eg. take an example when you hit the 
speaker with a sharp pulse as per my previous article.  There is a rapid 
force applied and the cone begins to excurse one direction... and let's 
say it's so energetic that it drives the cone to a high speed but then 
at the top of the pulse the voltage levels off of course but let's say 
just for an instant then WACK it dives back to zero.  Ok during that 
drop back to zero the speaker is induced to move the other direction.  
However, there will be a back emf situation because the speaker cone now 
has momentum.  That back emf will alter the perceived 
impedence...sometimes quite a lot.

AND THIS...is another great note because if anyone DIY's their own 
amps....be careful to put an adequate protective inductance on the 
output or this situation will create at some point such a bad back emf 
that it will overvoltage your output device and..the magic smoke is soon 
to follow as we know :-).
      Anyway there is an INSTANTANEOUS (ie. calculated by getting the 
exact value of I(t) at a given instant) power dissipation that is always 
there except at zero crossing points calculated by P=I(t)^2 * R.  It 
never changes.  I'm amazed how MANY techs seem to think that this is 
somehow altered by the reactive things going on.  It's not in the 
least.  It's always there taking every bit of real current going through 
that coil and soaking it's portion of the life from it.
   So anyway..if the impedence and the DC resistance are both 8 ohms, 
then by definition there is no reactive impedence.  It's all 
non-reactive and all the thing will do is heat/radiate/etc.  -Bob

harrybissell wrote:

>Hi Bob
>
>Ok let's try it another, another way. Essentially we agree except for
>the 'why' of it.
>
>Take the 8 ohms speaker. In this case I'll even say that the DC resistance and
>the impedance are both eight ohms.
>
>The resistor is a pure resistance, and is eight ohms.
>
>The resistor and speaker are receiving the same voltage (current is equal)
>
>The resistor (because it is essentially non-inductive in this case - think carbon
>composition) dissipates all the power as heat.
>
>The speaker, because of its design (as a coil) dissipates some power in making
>a magnetic field and moving off-center, and holding that position against the
>spring forces that would like to re-center it.  The rest is dissipated as heat.
>
>The speaker is doing 'some' of its job... otoh its kind of stupid because the
>function is to move air... and it is not doing that. And our hearing does not go to
>DC either.
>
>The "DC" resistance is not really that important to us, except that as you noted...
>we wish it was low or non-existant.  Its not... too bad for us.
>
>The "impedance" is the real object... because it expresses how much work we can
>get out of the speaker at some real (hopefully perceptable) frequency.  We will also
>
>be careful to band-limit our input so that we don't make waste heat for freqeuncies
>we can't / don't want to hear.
>
>The DC in the resistor produces only heat.  The DC in the coil produces heat
>and displacement.  Since we don't hear it (the frequency is infinitely low...)
>it produces no sound.
>
>Maybe we can agree, one is useless... and the other one might as well be.
>
>My comments about DC resistance are for a broad range of speakers, including
>tiny little things you would not drive with a guitar amp.  The point is that you
>can't
>take an ohmmeter reading and tell what the 'impedance' is.   If the speakers have
>different DC resistances, but the same impedance at a frequency, they will
>draw the same current and produce the same 'power'... how much displacement
>that causes is more mechanical efficiency that electrical.  At any rate, the
>impedance
>cannot be lower than the DC resistance.
>
>And I might add I've been tempted to hook up speakers to one of my
>inverters.  Do you have enough for 840KW ???   (loud, yes :^)
>1200A at 700V...  1KHz.  Rock and Roll ???   :^P
>
>H^) harry
>
>
>
>
>
>
>
>  
>