an audio question about impedance matching
in [DSP]
|
Prev: 24/48 v 24/96
Next: Data-path accuracy in IIR filters?
From: robert bristow-johnson on 29 Jul 2010 13:20 i know it's not DSP, but i thought people on the list would have some experience, opinions, and knowledge. whether it's a microphone going into a preamp or mix board, or an electric guitar going into a guitar amp, how does impedance matching (or mismatching) affect *tone*? the way i see it (as an EE) is: 1. that if one of either the source or load impedance has reactive components and the other is resistive, changing the resistance will change the simple RC filter and the corner frequency of it, thus changing the tone. 2. if both source and load impedance are mostly resistive, changing one or the other will only change the level, not the waveshape. if the input to the pre-amp or amp has some nonlinearity, a louder level will push that stage into the nonlinear region which will change the waveshape and affect the timbre. other than that, i do not understand the why "impedance matching" is considered so important in audio. it's not like we need to maximize power transfer (in micro watts) from the transducer or pickup into the input of the amplifier. and, at audio frequencies (and the length of cables), reflections along the transmission lines (due to impedance mismatch) should not be an issue either. why can't a low-impedance source simply connected to a high-impedance load suffice? why are matching transformers needed (unless level is important)? -- r b-j rbj(a)audioimagination.com "Imagination is more important than knowledge."
From: Steve Pope on 29 Jul 2010 13:30 robert bristow-johnson <rbj(a)audioimagination.com> wrote: >why can't a low-impedance source simply connected to a high-impedance >load suffice? why are matching transformers needed (unless level is >important)? Many audio connections are in fact a low-impedance source connected to a high impedance load. Using transformers on some connections is thought to prevent ground loops. Matched connections (e.g. a 600 ohm load), when they are used, are thought to prevent hum on long lines. I'm not sure how scientific either of these beliefs are. Steve
From: glen herrmannsfeldt on 29 Jul 2010 13:34 robert bristow-johnson <rbj(a)audioimagination.com> wrote: > i know it's not DSP, but i thought people on the list would have some > experience, opinions, and knowledge. > whether it's a microphone going into a preamp or mix board, or an > electric guitar going into a guitar amp, how does impedance matching > (or mismatching) affect *tone*? As far as I know, the usual case is that a high impedance microphone should not have too long of a cable. The cable capacitance and the microphone impedance make a low pass filter, which will reduce the high frequencies. Pros use low impedance so that they can use long cables without a significant effect on the high frequencies. Otherwise, the mismatch will mean a smaller signal, more amplification, and more noise. -- glen
From: Greg Berchin on 29 Jul 2010 13:41 On Thu, 29 Jul 2010 10:20:18 -0700 (PDT), robert bristow-johnson <rbj(a)audioimagination.com> wrote: >whether it's a microphone going into a preamp or mix board, or an >electric guitar going into a guitar amp, how does impedance matching >(or mismatching) affect *tone*? > >the way i see it (as an EE) is: > >1. that if one of either the source or load impedance has reactive >components and the other is resistive, changing the resistance will >change the simple RC filter and the corner frequency of it, thus >changing the tone. This is probably the most significant source of tonal change -- simple passive filter concepts applied to the frequency response of the components. >2. if both source and load impedance are mostly resistive, changing >one or the other will only change the level, not the waveshape. if >the input to the pre-amp or amp has some nonlinearity, a louder level >will push that stage into the nonlinear region which will change the >waveshape and affect the timbre. A secondary effect, at most. My response to customers complaining about how bad equipment sounds when pushed into nonlinearity is, "Don't do that." >other than that, i do not understand the why "impedance matching" is >considered so important in audio. it's not like we need to maximize >power transfer (in micro watts) from the transducer or pickup into the >input of the amplifier. That depends. For microphones and phonograph cartridges this can be very important. Also don't forget that mechanical transducers form resonant systems, so the impedance relationships affect not only bandwidth, but Q. >and, at audio frequencies (and the length of >cables), reflections along the transmission lines (due to impedance >mismatch) should not be an issue either. Exactly. The wavelengths are so large compared to the length of the conductor that transmission line effects don't even come into play. >why can't a low-impedance source simply connected to a high-impedance >load suffice? Generally they can. >why are matching transformers needed (unless level is >important)? Do you mean something like a 600 Ohm system? Generally it's for level matching. I think it's pretty much a holdover from the vacuum tube days, where high output impedances were common. Greg
From: robert bristow-johnson on 29 Jul 2010 13:46
On Jul 29, 1:30 pm, spop...(a)speedymail.org (Steve Pope) wrote: > robert bristow-johnson <r...(a)audioimagination.com> wrote: > > >why can't a low-impedance source simply connected to a high-impedance > >load suffice? why are matching transformers needed (unless level is > >important)? > > Many audio connections are in fact a low-impedance source connected > to a high impedance load. i know. i'm having a discussion with someone who designs custom pickups and he has a version that transformer couples the output to the guitar amp. it's a step-up which increases the voltage (and decreases current) by the turns ratio factor and also the apparent source impedance (by the square of the turns ratio). now my thinking is that this is a levels issue (the louder level in the guitar amp makes for a hotter sounding pickup), but he insists that the reason is impedance matching and that results in a "better tone". > Using transformers on some connections is thought to prevent ground loops.. > Matched connections (e.g. a 600 ohm load), when they are used, are > thought to prevent hum on long lines. I'm not sure how scientific > either of these beliefs are. i dunno if the matched connection is responsible for suppressing hum, or if it's the natural differential input that a transformer primary winding is. you get a damn good common-mode rejection ratio (CMRR) with a transformer floating around in all this low-frequency slew of E&M radiation. r b-j |