rzdz@fluke.UUCP (Rick Chinn) (11/15/83)
re: power amp advice wanted
rre: tom@rlgvax
Tom Beres, in his reply of nov 10, makes a fundamental error in his
thinking.
His statement: On the other side of the coin, I have never had the volume
control over 1/2 way, either, so there is still plenty of
power with which to shake the walls should I ever get the
mind to.
This statement contains the classical gain/power error made by many people.
The position of the volume control relates to the voltage gain of the
amplifier/preamplifier system, and has nothing at all to do with how much
power is either available or left over.
Since the volume control is merely a voltage divider, it only affects the
overall voltage gain of the amp/preamp system. It doesn't matter if you have
10 watts or 100,000,000 watts, the position of the volume control is only
relative to how much is being shoved into it, and what the gain of the
system following it is.
example: amplifier needs 1 volt for full output. volume control is
set at -20 dB from full rotation. The input signal to the
volume control is 10V.
Since 10V is 20 dB higher than 1V, the amplifier puts out full power, even
though the volume control is set way below maximum. Furthermore, there isn't
any more power available, even though the volume control is miles from wide
open, since the power amplifier is already at the threshold of clipping.
Now, I will grant you this:
There is one and only one signal level where the position of the
volume control will actually relate to the %power level. That
signal level is the voltage level that causes the onset of clipping
to occur with the volume control set to maximum. If this is the
condition (which is usually *NOT* the case), then Tom's comment is
true and accurate.
Now onward... What actually constitutes sufficient volume is certainly in
the ears of the beholder, and most certainly in musical taste and personal
taste. For *my* ears,
70 to 80 dB SPL = background level
80 to 90 dB SPL = loud
90 to 100 dB SPL = very loud
> 105 dB = really loud
I typically listen between 70 to 90 dB SPL. When I'm alone, or if I really
need/want to experience something, then I might resort to much higher levels
(100 dB +). My own system has sufficient headroom that when the volume
control is wide open, the amps are just beginning to clip (true for phono
source).
A scientific way of deciding how much power you need is to find the speaker
sensitivity spec (dB SPL @ distance), then determine how loud you really want
to run (average SPL), then compute the power required. It's probably a good
idea to add 10 dB to the average SPL to allow for peaks before computing
power. Although your listening room's furnishings and construction enter
into the calculations, a good conservative approximation is to assume no
reflections, therefore the attenuation of sound in your listening room
follows a square law relationship.
example: a speaker is spec'd at 85 dB @ 1 watt, 1 meter. How much
amp power is required to produce 95 dB average SPL @ 3
meters.
solution: first add 10 dB to the average to get peak (probably a
good idea to add 20 dB if you have a CD player). Now
subtract to find how many dB over the reference the new
SPL is: 105 - 85 = 20 dB. Now calculate how many dB loss
the new distance represents: 20 log (3/1) = 9.54 dB loss.
Therefore the level @ the reference distance needs to be
9.54 dB higher. So...20 + 9.54 = 29.54 dB above the
reference. Now find 29.54 dB above 1 watt.... 10 **
(29.54/10) = 899 watts. Since most of us listen in
stereo, (2 speakers, 2 amp channels) you can halve this
number to get the per channel power requirements.
I will admit that I picked speakers that were pretty greedy. It does point
out how far 900 watts will go once you have defined the distance, speaker
sensitivity, and level requirements. From my own experience, bi-amplifying
was one of the biggest aids towards improving the headroom and reliability
(freedom from blown drivers) of my system. Getting a 200 watt/channel
amplifier was the other.
I suggest that you resort to science and physics and leave the empirical
evaluations to the salesperson in the showroom.
disclaimer: these facts/figures/opinions are my own and reflect my own
tastes and prejudices. Please direct all flames to /dev/null.
caution: When you operate a *big* power amplifier into a speaker with a
passive crossover, be sure that the crossover components, particularly any
inductors, will withstand the power levels (peak). Inspect capacitors within
to be sure they will withstand the voltage levels involved. Inductors are
very suspect, since saturation usually means a cessation of their inductive
characteristics, leading to changes in the crossover characteristics and
(usually) blown drivers.
Rick Chinn @ fluke
!fluke:rzdz
206 356 5232