From: "Benoit Provost"
Subject: Re: volume control at speaker location
Date: Fri, 27 Sep 2002 08:32:52 -0700
Organization: Intel Corporation
X-Newsreader: Microsoft Outlook Express 5.50.4522.1200
You are right, it is just a matter of non-linearity, not a change in
frequency response. I should say that these speakers are just for having
some background music while I work in my shop, and are not good quality
speakers. I am not looking for high-fidelity here. The living room is where
I try to get high-fidelity.
I remember that, a long time ago, I once connected a potentiometer at the
output of a small audio amplifier IC in the hope of controlling the volume.
Instead I got a weird tone adjustment, which was great since the circuit was
actually a sound synthesizer. So I got one more tweaking knob! I guess the
reason was similar to what John was explaining. Here the effect was large
since the cheap audio amplifier was not compensating at all the impedance
The potentiometer that I used was 500 ohms (I could not find lower value
with an audio variation). I still did not get the chance to try the approach
of the added resistor at the center tap. I will also play with parallel
resistor, and maybe I will find a good combination. I will let you know if I
Thanks again for the inputs.
"N. Thornton" wrote in message
> John Woodgate wrote in message news:
> > I read in sci.electronics.design that N. Thornton
> > 'volume control at speaker location', on Thu, 26 Sep 2002:
> > >> I read in sci.electronics.design that Benoit Provost
> > >> > My sound system is in my living room, and I am installing speakers
> > >> >basement, wired to "Speakers B" on the amplifier. I would like to
> > >> >there is a simple (passive) way to adjust the volume at the speakers
> > >> >basement. The simplest way to do this is of course to put a
> > >Use a variable resistor in series. It is the easiest way to do it. You
> > >won't get a linear scale as you say, but I don't see how that is a
> > >problem to be honest. It still works perfectly.
> > That's what he did, AIUI, and he DIDN'T like it, which doesn't surprise
> > me. Adding series resistance changes the frequency response, because the
> > loudspeaker impedance varies with frequency, often by large amounts.
> > Consider an 8 ohm loudspeaker, which is really 8ohms (resistive) at say
> > 400 Hz (typical), but is 32 ohms (resistive) at 100 Hz (also typical).
> > To reduce the sound level by 10 dB at 400 Hz, we need 16 ohms in series.
> > But at 100 Hz, the level goes down by only 3.5 dB.
> > You might argue that it's a desirable effect, because it provides
> > automatic 'loudness compensation' at low sound levels, and some people
> > might agree with you. But now consider the case of a 2-way system, where
> > the impedance is 32 ohms at 2 kHz as well. That is NOT going to sound
> > right to anyone!
> Yeah, I half agree, things _can_ go off course, I guess I just haven't
> found it to be a problem in pratice myself, with moving coils. Other
> types of speaker are another matter of course. So I accept L pads
> would avoid the chance of a problem. Maybe I'm overzealous on energy
> efficiency :)
> To Benoit: I wonder what exactly was the problem with the series pot
> in your case. I presumed, perhaps wrongly, that it was nothing more
> than non-linearity of the volume setting.
> Regards, NT