Audio Switching Relay Recommendations

timlein

Member
2009-01-18 6:19 am
I am building a high-end preamp and looking for the highest quality mechanical relays I can find for switching audio. Any coil voltage, DPDT. Priority is least degradation of audio signal. Any recommendations? (make and model)
 
Last edited:
Teledyne 712-TN5 -- I got a bunch from a fella in Israel and have used them certainly for audio -- gold plated leads. The "-TN5" indicates they have diode spike protection and switching transistor.
 

Attachments

  • IMG_5702.JPG
    IMG_5702.JPG
    193.5 KB · Views: 828
For mechanical relays I use the Axicom D2n*. https://www.te.com/commerce/Documen...hENG_SS_108-98007_H_D2N_0614.pdf8-1393792-5

But there are a bunch of good options: Panasonic, Takamasawa, Omron, etc.


That relay has a minimum recommended contact load of 10mA, which is far more than you'll normally see in audio. The one I've linked to has a more useful 10µA minimum load, its not relying on arcing to clean the contacts of oxide build-up (I presume because it has decent thickness of gold)
 
That relay has a minimum recommended contact load of 10mA, which is far more than you'll normally see in audio. The one I've linked to has a more useful 10µA minimum load, its not relying on arcing to clean the contacts of oxide build-up (I presume because it has decent thickness of gold)

Yeah, but can it withstand 30G's? :)

I used Analog Devices ADG1414 in an "Archival" phono preamplifier similar to the one Gary Gallo described in Linear Audio a few years back. Easy to control with Arduino or similar microcontroller.
 
Topic Overview concerning Audio-Relay Versions

There are several threads here on diyaudio.com with topics like e. g.
Best sounding signal level relay- Audio Grade Relay- Best Relays for signal path - Looking for the best relay-Audio Switching Relay Recommendations - go to
Audio Switching Relay Recommendations
Which relay for audio?
Audio Grade Relay
Looking for the best relay
Best Relays for signal path
Best sounding signal level relay?
Best Audio Relays
Relay cut-off / mute
Advice needed for selecting a speaker relay
but unfortunately, the topic in the headline from the above mentioned URLs does not differentiate whether it is a line level level relay or a relay using in a power amplifier output for DC protection and delay for the speaker connection. I. e. all posts have to be read through to find out.

The following threads deal specifically with Relays for SPEAKER TURN-ON DELAY / DC PROTECTION:
Loudspeaker protection circuit
Improved speaker protection circuit
Loudspeaker Relais Distortion
Choosing a relay for loudspeaker switching
Loudspeaker Relays (Bob Cordell)
alternative to protection relay
SSR for power amps
SSR for speaker protection?
Ready-to-Run (RTR) SSR DC Speaker Protection and Delay GB
RTR SSR Headphone Amplifier DC Protection Modules GB

There are an additional relay version to switch Transformers (Power ON/OFF) with or without inrush current limiting section:
Using Relays to switch Power and HT (stand-by)
Soft as a Feather Pillow (SFP) SSR Soft Start Circuit GB
 

Oneminde

Member
2014-06-22 11:14 pm

Oneminde

Member
2014-06-22 11:14 pm
I know mechanical relays is very often used when switching inputs as an example. There does exist solid state devices which does that silently, so let's look at what is available to us.
The FET is a simple switch that have depletion mode or enhancement mode. The depletion mode FET is normally a conductive switch, even if the gate is not driven to a voltage away from its other terminals' voltages (nc). To stop it from conducting, its gate must be driven beyond the voltage on its source. Enhancement mode FETs are (no), normally open and the gate usually needs its gate driven by a control circuit for it to conduct, depending on circuit details.

A relay can also be conductive without being enabled by electrical control. For example, a relay control circuit can lose power, but the load current can still flow through the normally closed relay terminals (nc). Or relays can have a (no) construction. Or they can have a latching construction so they remain closed or open even if the control circuit is not powered up. Magnets and mechanical gripping circuits can retain the state of the relay even during a power loss to the control circuit, while the load power is still continuing to draw current and maintain a voltage.

An analog switch uses enhancement mode MOSFETs that are n-channel (NMOS) and p-channel (PMOS) so that a full voltage swing can be conducted. PMOS and NMOS form complementary switches (CMOS). The CMOS analog switch is more sophisticated than single FET switches which often drop a Vt threshold voltage in common circuit configurations. The CMOS switch does not drop a Vt. An FET switch usually drops a Vt voltage from the load so full rail to rail switching is not achieved with simple drive circuits.

I use the CD4016 and CD4051 analog switches for tiny currents. Relays carry the biggest currents. FETs are used when the control circuits are expertly designed. Analog switches have integrated control logic, like level shifters and digital enables so the design task is easier
Further, let's look at what IcOn use in their IcOn4 Pro unit which is the latest iteration.
"I tried everything to improve their timing, to eliminate the break-before-make nature of relays, to switch them overlapped, use different PCB wiring and more. I finally accepted that it's impossible to make a fully silent switched volume control with reed relays. These parts are too slow. Every on/off switch generates a 0.5-1ms glitch as a small signal discontinuity. Our brains are very sensitive to this kind of glitch and detect it as a small clicking noise. Most the time you won't hear it but it depends on the music programme.
"I can reduce the level of glitch with smaller steps but it won't fix their length due to the lazy mechanical action. These clicks are very faint but having spent literally hundred of hours trying to kill them completely, I became more and more sensitive to them. What else could I try? The answer was obvious. I needed faster switches.
"Enter solid-state relays, analog or Fet switches (CMOS most of the time). Their switching times are a few thousand times shorter than reed relays, about 100ns instead of 1ms. They have practically unlimited life and near zero power consumption. All of that was perfect. What about their Ron resistance, linearity, switchable signal range, bandwidth? Were they at all used in top audio products? Two years of R&D experiments, comparisons and blind listening tests finally identified the best analog switches from several thousand types for my purposes. Finally no more clicking."

"As it turned out, analogue switches are frequently used in today's audiophile products but most the time are simply hidden inside DAC chips, high-end volume controllers like the Muses 72320 Pass Labs are fond of, streamers, digital and even high-end preamps. Very few people realize their presence because they do no harm. They conduct/cut signal without measurable distortion or noise, not just audio but video up to 10-15V. They are good enough in the €20-30K Mark Levinson and Boulder volume controllers and the PS Audio BHK Signature. Why should anybody be suspicious of them in my icOn? What's more, I use newer chips with 1/10th the on-resistance of the chips in previous preamps. In short, my Fet switches are the very best solution for any kind of stepped attenuator realized with resistors, transformers or autoformers. Which begs your question. Why offer reed relays or Fets? Though it's a bit strange to me yet a fact of life, there are many faith-driven audio customers. If they believe that reed relays are superior, they are. The customer is king.

* * *
Okay, we got some information there which is great. What we have so far is that IcOn is offering reed switches or Vishay CMOS analog switches. I will add these to the debate later so no need to check the source right away. So, we have FET's, CMOS, SSR (solid state relay) and we have the mechanical variant which is our traditional mechanical relay and our reed relay/switch.
While doing my research, I stumbled upon this and I quote.
Douglas Self wrote some interesting stuff about this topic in his books "Self on Audio"and "Small Signal Audio Design". Not only distortion is a problem with multiplexers. Control voltage feedthrough is much more ugly and needs a lot of effort to keep it out of your audiopath. In a conversation with a console designer he told me that the usual 4053 multiplexer is not even good enough for audio. There was a Philips version of the 4053 which had less CV feedthrough which he was using in his designs. Same chip from other manufacturers didn´t meet his specs - Jensenmann.
So basically, Douglas Self is saying that solid state is a bad idea as an audio signal switch. So let's keep that in mind, or just me, for when we get to comparisons of different devices. :) - which sort of brings us to the next part: Mechanical relay's, the std one. I wanted to find out which one was or is considered to be really good, so let's dive in. I am using diya here for convenience.

Comments from different people.

  • The AGN is Panasonic's version of the Omron G6K, which is actually smaller (height). G6K uses Gold clad Silver terminals.
  • Have a look at the relays used for semiconductor testing, also known as an RF relay, e.g. Teledyne 712 series.
  • A decade ago there was a listening test in Klang&Ton, as far as I remember they favored the mercury wetted stuff.
  • A mercury wetted relay has something like 30mR contact resistance.
  • I think you will find that telecom relays are pretty good with these things. They are designed to switch rapidly over a long life time, unlike power relays which are can carry more current but don't switch as often. Just don't switch DC and your contacts will last a very long time - BrianDonegan – Twisted Pear Audio.
  • > Well, if it has not to be an hard to get, exotic part, the industry quality Omron G6B has the same 30mR max contact resistance, Ag-alloy contacts, fully sealed. Costs however about the same as your mercury wetted relays.

    Why is mercury wetted relay used for low level high impedance circuits including many precision instruments ?

    Take a look at your Omron G6 datasheet. It specifies a minimum current. Below that current, the contact resistance is undefined, because the relay actually need current to burn away the oxide layout at the contact to create micro-welding before you ever get your 30mR resistance (when new). Different from a liquid metal contact, all dry contact ages, and the contact resistance goes up during lifetime.

  • The Teledyne 712 is really nice. Yes, pricey, but well sealed, fairly fast, good contact metallurgy, and reasonably low power. The signal conduction portion of the armature is non-magnetic, and the whole package is small. If you need just a few really good relays and you can afford $25 for a DPDT, it's a good choice.

  • The main reason for the development of the ruthenium sputtered contact types is the fact that the Hg ones are environmentally bad news, both in production as well as after discarding. I have used many years Pickering type 101 ruthenium types for audio switching without any apparent problems - Jan Didden

  • Ok, these ones are not sensitive to mounting position, I guess that there is only a minuscule amount of mercury on contact surfaces, not the entire pool like some other types. Mercury content of these is sooo small compared to all those fluorescent lamps everywhere that I am not going to lose my sleep over that. Dunno, maybe these are not so good then as these have so little mercury.

  • Quote from: "Freq Band": Is there anything "special" about these for handling audio signals ?

    Mercury wetted rellays = very low noise.

    It is metalurgic contact. It behaves like soldering. Nothing is better (... and nothing is harder to find). Very long ago mercury relays was rather common. Before the invention of the reed relay, mercury ones were the only way to get a small and good relay. Robert Hermeyer in Berlin was producer of mercury relays in 1950-ies. That relay performed like a thermometer. If some current was fed into the control branch, gas inside expands and moves mercury in the capilar. Mercury then connects two platinum wires inside capilar. Today, we use MOSFETS instead that relays, but for special purposes mercury relays could have better dynamic range - xvlk.

  • I designed in Hg-wetted reeds, switching low-Z bulk metal foil Vishay resistors, for the Harman R&D group's fancy computer-controlled attenuator, although we never finished the project as they were having some budget problems. The design achieved about a 144 dB signal-to-noise ratio, so that I could say we were doing a real 24 bits.The relay on resistance is a little higher than direct clean gold or palladium reeds etc. sometimes, but as xvlk says it's like soldering and represents a new metallurgical junction each time you make. They are usually position-sensitive although there were some "mercury-film" ones that were not.Another advantage of Hg-wetted: essentially no contact bounce on make-break. People have used them as fast rise time pulse generators (with of course huge jitter and very slow repetition rates).For switching high power there are big clunky things called mercury displacement relays, where a plunger is pulled down into a mercury bath to push the mercury up to make contact with the electrodes. Magnecraft makes them, and the big ones have about a milliohm of resistance and switch a hundred amps. They are used for "mission-critical" applications and for situations where there could be explosive atmospheres. But they have the same "dry" contact behavior as a mercury reed. When the original Harman design was going to switch speakers and amplifiers, I specified these (or maybe the somewhat smaller ones) for that portion of the switching, and was considering showing how good the switch was by running an MC cartridge signal through them and noting the lack of degradation - Bcarson.

  • The NAIM 552 is close to twice as expensive as a BT, with twice as much hardwire, and a whole lot more Reed switches (CP Clare types, the later Sumida RemTech Corp. PRMA, now Coto 8L). Yet nobody seems to get upset about the 552, notta bad for a preamp that uses a couple of motorized Alps pots for volume attenuation. Jolly old Britain, thank goodness the Naim Powerline is only a mere $600.
Before we go any further, I wana throw in Vinnie Rossie's AVC unit and what is used there. I had to find a good picture to check the parts. They are as follows: Pickering 112-1-A-5/2 single in line reed relay

wenS9PW2gtQGEqNu1PWoMW7ynPhZv4-X3uDqf8BAuZFWcmPvBju5I0XQ-ApaQYxLab3-HWtwUOSLq63aGtpt7pu9I2_QqLruIlH7u77r=s0-d-e1-ft


Alright, I think we are ready to pull up some spec and compare the units talked about + some of the ones I found. Before we do that, lets list the ones mentioned so far.

  1. Texas Instruments CD4016 & CD4051 CMOS analog switch/multiplexer.
  2. Vishay DG212 CMOS analog switch, lets also throw in the DG417 here. (IcOn)
  3. Panasonic TQ-2 and AGN relays.
  4. Omron G6K and G6B.
  5. Teledyne 712.
  6. Hamlin Reed Relay series. (IcOn)
  7. Pickering 101 series - reed switch.
  8. Pickering 112-1-A-5/2
  9. Additional Pickering reed switches.
  10. Coto 8L series.
To make things less confusing, you have the dry type and the wet type. Wet is a reference to mercery being used and dry is, well... dry, so no wetting material. I have attached two (2) PDFs from TE Connectivity which describe the wetted reed relay and other info regarding relays. Something I noticed earlier which is included in information is the talk about switching speed and contact resistance. This is comparable to resistance in FETs and Tubes and if we have learned something, then that is we are also most likely dealing with capacitance, so we are going to look at contact resistance, capacitance and switching speed in our comparison.

Comparison values.

Model Contact resistance (on state) - Contact speed - Contact capacitance (on state) - price/pcs

TI CD4016 - 5 to 15Ω - 15-100 nS - 4 pF - €0.45
TI CD4051 - 125 to 470Ω - 100's of nS - 30pF - €0.45
Vishay DG212 - 100 Ω - 200 to 300 nS - 16 pF €1.50
Vishay DG417 - 45Ω - 100s on nS - 30pF - €1

That was our CMOS or SSR variant. Next is the Panasonic and Omron mechanical relay.


Panasonic TQ - 50mΩ - 1-2 ms - Doesn't state, expect several pF - €3
Panasonic AGN - / - 4 ms - Doesn't state, expect several pF - €3-4
Omron G6K - 100mΩ - 3 ms - Doesn't state, expect several pF - €3
Omron G6B - 30mΩ - 10 ms - Doesn't state, expect several pF - €5
Teledyne 712 - 150mΩ - 4 ms - 0.4 pF €55 to 75

That was our traditional mechanical relays, and we can see the appeal in lower on-resistance compared to CMOS and SSR. The Teledyne 712 is a clear improvement over the CMOS and also the Panasonic & Omron traditional relays and a serious improvement in the capacitance region. The CMOS is faster but the other very important specs are much much worse. Let's move on to the reed relays.

Hamlin H3300 (dry) - 150 mΩ - 1-3 ms - Doesn't state, expect several pF
Hamlin H3600 (dry) - 150 mΩ - 1-3 ms - Doesn't state, expect several pF
Hamlin H700 (dry) - 150 mΩ - 1-3 ms - Doesn't state, expect several pF
Pickering 101 (dry) - 120-150 mΩ - 1 ms - 2.5 pF
Pickering 101 (wet) - 75-100 mΩ - 1ms - 4.5 pF
Pickering 112 (dry) - 120 mΩ - 0.5 ms - 1.5 pF
Pickering 103 (dry) - 120-150 mΩ - 0.5 ms - 0.1 to 0.45 pF
Pickering 200 (dry) smd - 120-200 mΩ - 0.5 ms - 2.5 pF €5
Pickering 200 (wet) smd - 75-100 mΩ - 2 ms - 4 pF €8
Coto 8L (dry) - 150 mΩ - 0.5 ms - 0.5 pF €2-6

Conclusion.
While the Teledyne 712 impresses when it comes to resistance and capacitance, the price does not. That is one expensive motherfucker and is actually not better than the Coto 8L. Coto does not offer mercury wetted, only Pickering as far as I can tell. While both Panasonic and Omron offers very good as in low on-resistance, they are relatively slow switching speed, they are mechanical relays after all, just like the reed switches. When it comes to the capacitance of Panasonic and Omron, it is not stated, another brand had 3-4 pF as their number, so I believe the capacitance is close to that. Pickering 103 has even better capacitance numbers than Coto 8L. And Pickering 200 is offered as both dry and wet and is also smd.

Sources.
1. FET vs. Analog Switch IC vs. Relay.
2. 6moon review of IcOn4 Pro.
3. Douglas Self comment.
4. DIYA relay's.
5. Mercury wetted.
6. NAIM 552.
7. Hamlin Reed Relay.
8. Reed relay sound - (nothing or very small)