Switching of Audio Signal

[Quercus]

I'm working on an integrated pre-amp/amp and need to select between various inputs. I also have a 3-way tone control circuit that I want to add a bypass option. All these require switching of low-level audio signals. Simple enough to do with a panel mounted DPDT switch.

However, I'm considering using either a mechanical relay or a low resistance IC analog switch to reduce the length of the signal path and give me options for logical control of these connections. Does anyone have experience in using these devices in an audio circuit?

 

[Nigel Goodwin]

If you're building a high quality system? (perhaps not if you have three channel tone controls?) then DPDT relays are a good solution (or pairs of SPDT relays), without adding or losing anything of the signal. Any electronic methods are going to affect the signal in some way - it's just a matter of to what degree.

Various ways have been used commercially for decades - with standard CMOS switches quite popular, but the first I ever saw used simple diodes suitably biased to do the job. An old friend of mine had one.

After a lot of googling, I've eventually found it - the Armstrong 621 - a fairly good quality amplifier.

The Armstrong 600 Series - Amplifier FAQ and diagrams

 

[AnalogKid]

Linear Technology and Maxim excel at analog switches specifically designed for audio applications. Many of them are "clickless"; they change resistance slowly enough to prevent an audible click caused by either a little DC on the signal, or a mismatch in signal phase, such as switching from the middle of a positive peak in one signal to a negative peak in the other.

Small DIP or SIP reed relays are another option. Some purists don't like the "audible effects" of the series element in an analog switch, and tolerate the clicky switching of relays to have a "pure" connection. I have designed pro audio circuits and worked with golden-eared wonders, but IMO these effects, if they exist at all, are completely masked by the background noise of even the quietest listening environment. I vo6te for analog switches.

ak

 

[Quercus]

NG, AK, thank you for your replies and references. Very helpful, including the Armstrong schematic. Thanks for digging that up. Will try some of these options in test circuits.

If you're building a high quality system? (perhaps not if you have three channel tone controls?)

I think the definition of "high quality audio" will be debated long past my lifetime! My answer for current purposes is... "sort of." While I'm looking for 'high quality', the project is a desktop system for computer audio, which kind of puts a ceiling on the matter. So while I'm not going for the highest fidelity Ed Villchur level system here, I am using a set of strategically placed AV speakers, 12 gauge cables and an acoustically suspended 10" sub. I want it to sound very good, but I'm not looking to blow anyone's socks off and I have no intention of paying $50 for stepped attenuators. I've been debating the 3-way tone control as an unnecessary feature, hence the desire for a bypass. But the more I think about it, the more I'm thinking I'll just leave that out.

Thanks again.

 

[Nigel Goodwin]

NG, AK, thank you for your replies and references. Very helpful, including the Armstrong schematic. Thanks for digging that up. Will try some of these options in test circuits.

My friend Keith had one, and I was always amused by the way it faded from one input to the other as you switched

I think the definition of "high quality audio" will be debated long past my lifetime! My answer for current purposes is... "sort of." While I'm looking for 'high quality', the project is a desktop system for computer audio, which kind of puts a ceiling on the matter. So while I'm not going for the highest fidelity Ed Villchur level system here, I am using a set of strategically placed AV speakers, 12 gauge cables and an acoustically suspended 10" sub. I want it to sound very good, but I'm not looking to blow anyone's socks off and I have no intention of paying $50 for stepped attenuators. I've been debating the 3-way tone control as an unnecessary feature, hence the desire for a bypass. But the more I think about it, the more I'm thinking I'll just leave that out.

CMOS switches will be fine then, as AnalogKid said they can really perform excellently - no problem for computer audio.

 

[crutschow]

Some purists don't like the "audible effects" of the series element in an analog switch

I'm always amused by how the audio fanatics claim they can hear totally unmeasurable effects, such as the difference in the amplifier sound between a $500 magic power cable, and the standard $10 version.

 

[unclejed613]

I've been debating the 3-way tone control

if it's computer audio, you already have those controls in software, usually in the sound card settings, most likely with a 10 band EQ....