Swapping CMOS/TTL switches or multiplexers with SSM2402/SSM2404

[quamen]

Hello:

Has anyone changed a switch/multiplexer (often a 4053 or similar logic IC) with a reportedly higher performance SSM2402 (dual) or SSM2404 (quad) CMOS/TTL triggered analog switch?

I have always had high faith in Analog Devices products but I know claims can often be exaggerated to increase sales.

According to AD: "The SSM2404 integrates four SPST analog switches in a single 20-pin package. Developed specifically for high performance audio applications, distortion and noise are negligible over the full operating range of 20 Hz to 20 kHz. With very low charge injection of 35 pC, "clickless" audio switching is possible, even under the most demanding conditions."

I am guessing using the SSM2402/4 would improve things provided you can fit a little PCB or adapter onto the main board as the SSM2402/4 are not pin compatible with most switch/multiplexer IC such as the 4051/4052/4053. It may even require some additional devices to make it work properly (I have no idea, I am just guessing).

I have been doing a little experimenting with these chips. I am going to be using them for switching duties in a small but high qaulity recording console (like the Manley 16x2) instead of traditional relays. So I know about the "common" use for them.

I now am just trying to see if they can be used in other applications, particularly in replacing CMOS/TTL multiplexers in D/A circuits. It may be an inexpenisve way to get D/A circuits to sound better than stock.

Any suggestions or ideas?

Any help is greatly appreciated.

Thanks
K

[Dan Kennedy]

As a mux they're too slow, as a switch for routing, they're great, super quiet. It's all in the internal ramping.

I've used thousands in broadcast and telecom applications, with no failures, both SO and thru hole types.

They sound fine, but they do have a sound, pretty much like any analog switch, a bit of even order stuff, but I don't find it objectionable.

I try to use them in "current mode" when possible, like any switch, it reduces the voltage swing, reducing the distortion.

[John Suitcase]

Hi Dan,

Do you have a source for the SSM2402? I can't find any, at least not in a hobbyist-size quantity.

I'm interested in using them for muting line -level signals, and have thought about using one in a stomp box, though I don't know how well that'd work out!

Thanks!

[Jim Williams]

Those SSM parts are an easy to apply 10 ms ramped up switch. But like all electronic swithes, they have noise, THD and bandwidth limitations.

I used them in a 1990 design of an outboard moving fader Uptown system used on Broadway to automate the shows in NYC.

I don't use them for high quality audio, nor the 4053 demultiplexer chip. I found relays to be the best solution although they have switching transients as they will cut a waveform off.

I was looking at a relay operated daugher pcb for the Trident 24/65 series master module that uses a 4053 with a quad opamp buffer for solo switching. It is a sonic ro*******. I would do it if there was enough demand to justified the pcb manufacturing.

Jim Williams
Audio Upgrades

[JohnRoberts]

I did some global electronic switching between tracking and mixdown mode inside an old LOFT console back in the '70s using CMOS transfer gates. But I placed the TGs inside the feedback loops of inverting opamps, so not only were they operating in current mode, but any nonlinearity in the TG's on resistance was reduced by the full loop gain margin of the opamps NF.

I couldn't measure any noise or distortion contribution from the TG which was apparently down in the residual of the old Sound Technology analyzers of the day.

JR

[John Suitcase]

What about optical couplers, or using LEDs with optoresistors to do a series/shunt type of mute? I realize you can't really get to a perfect mute, but I would think it could get close enough for my purposes.

I'm thinking along the lines of building a unit with 24 channels of mutes, for instance. A simple solution that doesn't cause any audible pops or clicks, and doesn't audibly distort or add noise.

[Jim Williams]

I'm rebuilding a pair of Sony APR 5000 audio cards that use the analog fet optical device for muting. It damps levels during rewind/fast forward so those levels don't take your head off.

These are bilateral fet based optical analog switches. They can also be used in compression designs as they don't have the limited attack time of common opto cells like Vactrols. I showed these to Steve Firlot many years ago when he was designing the Vac Rac tube based limiter, but he elected to use vactrols.

The capability of these parts has not been fully explored, any takers? One could do a nice opto version of a 1176 with these.

Jim Williams
Audio Upgrades