I hope you can help me to choose two relays:
1) a NO relay for switching on the signal source before the preamplifier for my home hi-end. The sound quality is of cause the most important, low external noise is second prize is third, other specification demands how I build the rest of the circuit.
2) A NC relay where the sound quality is of no concern, but the low external noise is most important, the voltage demands would be nice if they match the NO relay.
I know there is a total lack of information, but please ask, since I have nearly no idea on what info is needed to answer my question.
1) a NO relay for switching on the signal source before the preamplifier for my home hi-end. The sound quality is of cause the most important, low external noise is second prize is third, other specification demands how I build the rest of the circuit.
2) A NC relay where the sound quality is of no concern, but the low external noise is most important, the voltage demands would be nice if they match the NO relay.
I know there is a total lack of information, but please ask, since I have nearly no idea on what info is needed to answer my question.
Link to another thread.
https://www.diyaudio.com/community/...og-switch-ic-for-input-source-selector.97171/
https://www.diyaudio.com/community/...og-switch-ic-for-input-source-selector.97171/
Thanks but this is a mixture of comments from 2007 and 2016, I imagine a lot has happened since then. 🙂
I guess you're looking for reed relays. Or J-fets as switching (muting) solid state devices.
Solid state relays are for switching 230V gear, not suitable for audio anywhere.
Solid state relays are for switching 230V gear, not suitable for audio anywhere.
okay, I did just understand that this is an often used solution in high-end gear where switching has to be error free and without any artifacts?
I recommend reading the chapter on signal switching in Douglas Self: "Small Signal Audio Design", 3rd Edition, Focal Press, 2020, ISBN: 0-36-746895-6
I suspect most solid state relays won't be great at switching low-level signals. You're probably better off with a mechanical relay with gold point switches such as the Panasonic EC2-series.
Reed relays are not great for signal switching. The reed tends to have a highly nonlinear resistance and will increase the distortion measurably. There's an article in Linear Audio about that. I can see if I can dig out the specific volume for you.
If you do want to go the solid state route, I suggest starting with a transmission gate made from a pair of MOSFETs and a floating gate driver. You can get a photovoltaic driver. That'll allow you to establish Vgs on the MOS devices even though the voltage on the sources track the signal you're switching.
Tom
I suspect most solid state relays won't be great at switching low-level signals. You're probably better off with a mechanical relay with gold point switches such as the Panasonic EC2-series.
Reed relays are not great for signal switching. The reed tends to have a highly nonlinear resistance and will increase the distortion measurably. There's an article in Linear Audio about that. I can see if I can dig out the specific volume for you.
If you do want to go the solid state route, I suggest starting with a transmission gate made from a pair of MOSFETs and a floating gate driver. You can get a photovoltaic driver. That'll allow you to establish Vgs on the MOS devices even though the voltage on the sources track the signal you're switching.
Tom
@ tomchr:
Quote:
Reed relays are not great for signal switching. The reed tends to have a highly nonlinear resistance and will increase the distortion measurably.
They're used in extreme sensitive measuring equipment in scientific experiments without discussion. Everything else adds noise, distortion and other unwanted things. How can silver contacts in small sealed in noble gas containers show nonlinearity behaviour? It's in use for decades now. Scientist must be mad?
Quote:
Reed relays are not great for signal switching. The reed tends to have a highly nonlinear resistance and will increase the distortion measurably.
They're used in extreme sensitive measuring equipment in scientific experiments without discussion. Everything else adds noise, distortion and other unwanted things. How can silver contacts in small sealed in noble gas containers show nonlinearity behaviour? It's in use for decades now. Scientist must be mad?
Scientists tend to have bigger budgets than hobbyist audiophiles. Anything that involves silver and noble gasses is bound to be expensive. Also, just because the contacts are silver plated doesn't mean the reed itself can't be of a different material.
The article on distortion in reed relays is in LA Vol. 13 by Sam Groner and Bruno Putzeys: https://www.linearaudio.net/reed-relay-distortion. There's even a distortion plot for you on that page. They tested reed relays by five different brands (two samples per brand) and compared with an Omron G6K-2P-DC5.
These were the five reed relays:
But hey... Maybe your scientists want to argue against a distortion analyzer. 🙂
You'll also notice that distortion rises as function of the current through the relay. Maybe your scientists take care to run as little current through the relay as possible. That's generally not an option in audio with 2-8 V RMS signal levels and 10-48 kΩ load impedances - or lower if you aim for low noise.
Tom
The article on distortion in reed relays is in LA Vol. 13 by Sam Groner and Bruno Putzeys: https://www.linearaudio.net/reed-relay-distortion. There's even a distortion plot for you on that page. They tested reed relays by five different brands (two samples per brand) and compared with an Omron G6K-2P-DC5.
These were the five reed relays:
- Coto Technology 9001-05-00
- Hamlin/Littelfuse HE3621A0500
- Standex MEDER SIL05-1A72-71L
- Magnecraft/Schneider Electric W117SIP-6
- Potter & Brumfield/TE Connectivity JWS-117-1
But hey... Maybe your scientists want to argue against a distortion analyzer. 🙂
You'll also notice that distortion rises as function of the current through the relay. Maybe your scientists take care to run as little current through the relay as possible. That's generally not an option in audio with 2-8 V RMS signal levels and 10-48 kΩ load impedances - or lower if you aim for low noise.
Tom
Don't count on it.
Most people stream music on their phones - where's the consumer demand for switching sources?
For 'High end' gear, old tech using electromagnetic relays because if something is solid state and in the signal path it must add noise? Ignoring all those annoying chips & transistors used between the performer & the final listener.
But generally, ics used for switching sources also use microprocessors for remote control decoding and such.
You haven't given any clues as to how you wish to implement the control.
I have always understood reed relays are properly applied in low-level switching applications, like microvolts and microamps. Probably why you notice them used in scientific instruments. The circuit surrounding the contact should ideally be designed to be immune to small differences in contact resistance. It's all about avoiding the oxide layer.
Another potential benefit is the high speed switching and minimal contact bounce.
Another potential benefit is the high speed switching and minimal contact bounce.
Much appreciated tomchr!
Yes, budgets were not an issue really. Only the best, and very expensive.
No discussions with Analysers, they're mostly exact. Like those Lock-in Amplifiers, too
Currents were very low: counting electrons better. Through a single atom wire. At best just over ten atoms actually. Acts likewise sort of superconduction.
All measured with extreme low noise opamps (yes, expensive), the feedback a 100MegΩ. The signal was a 'noise-peak'. Presto!