Clickless CUT circuits?

[Ethan]

Most consoles seem to have a CUT switch that engages without adding any noise into the signal path. What are the most common ways to accomplish this at line level?

I was thinking one of two methods but I'm not sure if either would be noiseless:
1. Short the differential output (would most devices downstream be ok with seeing a shorted input?).
2. Have a relay/switch disconnect the output (and pray that it's at a zero crossing?).

Thanks!

 

[[silent:arts]]

3. use a FET with a ramp down / up citcuit

 

[Svart]

caps in the signal path that are biased with a steady voltage?

 

[Michael Tibes]

If I remember right they used LDRs for the flying faders system, one in the signalpath and one after that shorting to ground. Don't remember the company who makes these little LED/LDR combinations in one housing, they have a basic circuit on their website with pretty good distortion and damping figures. The LDRs should do the ramping automatically since they're anyway not the fastest switches.
SSL used FETs for the 4000/6000, that's part of the 'SSL-sound' (they have quite some of these switches in the signal path, not only the cut-function). A better way to use FETs should be before an inverting opamp, in the signal path right at the minus input. The voltage at that point is zero, so the nonlinearity of the FET shouldn't matter. TAC used this circuit.

Michael

 

[EZ81]

http://optoelectronics.perkinelmer.com/catalog/Category.aspx?CategoryName=VTL5C+Series ?
Mesa Boogie uses these for channel switching.

 

[BradAvenson]

Check the appnotes here:
http://www1.silonex.com/audiohm/pdf/softswitch.pdf

 

[Ethan]

You guys are awesome!

 

[Michael Tibes]

[quote author="BradAvenson"]Check the appnotes here:
http://www1.silonex.com/audiohm/pdf/softswitch.pdf[/quote]

That's what I had in mind, just couldn't remember where I saw it. It's my favourite design for this purpose.

 

[SSLtech]

The biased caps issue isn't a help. The problem is that if you have a low fewquency waveform (which are the very worst of the audible signals because they are usually the biggest, also they are at the bottom of the audio spectrum and so they trigger more audible artifacts, since the switching 'click' is always comprised of frequencies which are higher than the signal being interrupted) then it's impossible to interrupt the waveform suddenly with any sort of hard switch or relay without the "stair-step" of the waveform dropping to zero being a nasty 'click'.

As already mentioned, Martinsound used the Silonex in a series/shunt arrangement for the 'flying faders' mutes. Fast & excellent.

More on FET switching here:
http://www.groupdiy.com/index.php?topic=7725

I usually do it with FETs, but the Anadlog Devices SSM2402 is also a fabulous switch, designed for exactly this purpose.

Keith

 

[franklinh]

From the first line of the softswitch.pdf:

"Electronically switching audio signals can be a complex."

And How!

 

[Ethan]

Would it be a bad idea to have the opto/resistor in shunt configuration to short out the line level differential output (XLR pins 2 and 3)? Any potential problems? I'm thinking of doing this after the output transformer so the driving source would see a reflected load that depends on the dcr and ratio of the transformer--and I would just have to make sure it's high enough that the source doesn't get too unhappy.

 

[Boswell]

Would it be a bad idea to have the opto/resistor in shunt configuration to short out the line level differential output (XLR pins 2 and 3)? Any potential problems? I'm thinking of doing this after the output transformer so the driving source would see a reflected load that depends on the dcr and ratio of the transformer--and I would just have to make sure it's high enough that the source doesn't get too unhappy.

OK, but that's not a mute - it's an unspecified attenuator, maybe of not that many dB.

For a muting switch, you really have to define what you mean by "muted" in terms of -xdBFS, and then design to that spec, taking account of capacitative effects at high frequencies. In addition, if you are worried about using FET switches in signal paths where signal current flows, you have to choose a point in the circuitry where a series/shunt combination of control elements can meet the specification, and do this without changing the d.c. conditions (to avoid a thump).

Never an easy problem, but the line output XLR is not the best place to start.