56 posts because your good way would be the wrong way. The box is intended for being between microphone and preamp using, XLR. No line cables involved.

Why do you think the switch is wrong for the job? Just because it is big?

You've been so dead right Sir. It is the impedance of the switch. I measured about 1 Ohm which then is obviously way to high.

I did test that with that setup and I achieved an extreme attenuation, even with phantom power enabled and between preamp and microphone!


While with the switch, I get a bad result. So it is definitely the switch.

Switches and relays that are designed for power loads (I.e. 1 amp or greater) have physically larger contacts not only to carry higher current, but to withstand a certain amount of arcing that occurs across the contacts during operation with higher-current loads. They also must exert enough mechanical force to break the contacts back apart when microscopic welds form between them as a result of such arcing, and the contact material and plating must provide a useful lifetime under these conditions.

A side-effect of this construction is that the surface of the contacts usually oxidizes easily, and they rely on the current through them to break the oxidation layer and initiate a low resistance path. This is why switches and relays actually have a minimum current that they will reliably carry . . . sometimes this is found on the datasheet, but most of the time it’s just implicit in the maximum current rating, intended application, and statement of contact materials.

Switches and relays that are intended for small-signal use can have precious metal plating on their contacts, that may be too thin and/or have too low of a melting point to withstand arcing with power loads, but don’t have oxidation problems. Many switch designs also have a mechanical wiping action as the contacts move during operation, which scrapes the surfaces together and cuts through the oxidation. There are also fully hermetically-sealed reed relays, self-renewing contacts wetted with mercury, and all sorts of other specialty design approaches that are used to make switches and relays perform in their intended application.

So yeah, if that big button is designed to start and stop a conveyor belt in an ore-processing facility, it might not work nearly as well as a cheap little slide switch for muting a microphone.

Thanks for this detailed post. I think that solved the puzzle. The difference was huge in my test.

I would love to keep this switch, so, if I would just solder in permanent small resistors into the cables before the switch, wouldn't that offset the too high impedance of the switch? If the switch has, let's say, 3 Ohms impedance, very small resistors should be enough, but maybe I am wrong and the only way is to reduce the switch resistance.

EDIT: For those wo are interested, I simply opened that up. I would guess the resistance is due to these two discs on each side.

The basic approach of adding series resistance is a sound one . . . but while an ohm might seem like like a low value for the switch contact resistance, it’s in fact a massive amount for this application. This means that you’d need to insert so much series resistance, you’d pay a noise penalty and likely starve the mic of enough phantom- power current. There’s also the fact that the switch might not be able to consistently deliver even this contact resistance on a consistent basis.

I see that there’s a power jack on your box; for an indicator light? In this case I’d simply use the switch to control a relay to actually do the muting . . . I’d suggest a Hamlin/Littelfuse HE3321 series reed relay, which can reliably deliver 0.1 ohm or so. This will produce an attenuation of -63.5 dB from a 150 ohm source impedance. If this isn’t sufficient, then two stages can be cascaded with a modest series resistance between them.

But it’d be best to start with a single contact and no series resistors, and measure the results, so you have some hard data to use for designing any enhancements. You’d also get an idea how much of a popping issue there is with the phantom power . . . this can be solved by capacitively coupling the muting contacts, but this too will alter the effectiveness.

Raise the bridge or lower the water.

A switch can also open the circuit (takes two switch contacts)

If double throw the feed to the preamp can be shunted in the mute position.

Making this clickless and silent may take some work. It might switch quieter if instead of completely open circuit the switch inserts a series resistance, that gets shunted when unmuted.

Sorry, I did not read the entire thread so better answers may already be posted.

JR

That switch is designed for power use, not low level audio. Kirkus says it better than me.

The obvious problem with any series-acting muting switch is that it cuts the phantom power to the mic. Even if there are resistor(s) in parallel with the series muting contact(s) (to keep the mic running) in conjunction with a shunt muting contact, there will be some abrupt DC level shifts when operating the switch, causing pops.

Merely adding some series resistors before a single shunt muting contact will improve effectiveness without adding additional DC issues, but it also reduces signal level and increases noise at almost the same rate. The maximum value that would probably be acceptable would be a pair of 100 ohm resistors, bringing the source impedance to 350 ohms . . . which would cause about 0.9dB of signal loss for a mic preamp with a 2K input impedance. There would also be an increase in thermal noise of 3.7dB, for a loss in S/N ratio of 4.6dB. Muting level with the same 0.1 ohm switch would increase from -63.5dB to -70.0dB, so not a lot of difference.

However, if the series resistors are combined with two shunt muting contacts, one before and one after . . . then the improvement is dramatic, as it's essentially acting as two cascaded attenuators for almost -130dB total muting. This topology is effective enough where it could work acceptably even with capacitors in series with shunt muting contacts, to eliminate any residual DC-related pops that occur from slight resistor mismatches in the preamp and microphone.

The result is something like the circuit below . . . with 1000uF low-ESR electrolytics this can deliver -63.5dB at 35Hz, and better than -100dB at 300Hz and above. They could of course become slightly reverse-polarized, but I don't think this would ever be an issue, especially if they're quality types rated at 35v or higher. They'll also be sufficiently large to still be effective for a mic that has an output impedance lower than 150 ohms, and if this is the case, then the series resistors could also be increased a bit if necessary.

Why reinvent the wheel? There are loads and loads of schematics on the WWW that show EXACTLY how to construct a mute box for phantom powered mics. Instead this TOPIC has gone on for #69 posts. It seems like the OP wants the world to work HIS way and only his way and will keep rejecting all the help he is offered. FWIW

There are tons of people here and out there which are interesting in how it works, why it works and why it probably does not work. People are discussing, I am asking, doing research and getting help and now I am ordering some reed-relais to continue with my project. It is out of scope for you to criticize that, because it is simply not your project. I will not immediately trash my switch with the big red illuminated button if there might be a nice way to fix that, thankfully offered by kind members of this forum. There is absolutely no reason for you to complain about anything here.

Unfortunately every forum has such people, so long story short:

Please just don't write into this thread if you're not interested. No one needs or wants to read these stereotypical conflict-laden posts.

Others have suggested some really good ways to do what you want to do but you seem to brush off or disregard their suggestions. If you want help I understand BUT you need to take the advice people are giving you. The people who are trying to help are some of the best members of this forum, you should LEARN from them. FWIW

Not if the phantom power resistors are before the mute switch.
as I shared before, making this silent is not trivial, simply shorting pin 2 to pin 3 should have less impact on DC levels.
whatever.... I don't feel like reinventing old wheels today. To do this right requires bench time to see what works, or not.

JR

I tried a direct unswitched short between pin 2 & 3 and achieved a really good result (between preamp, phantom power, condenser microphone). We found out that the bad muting result was due to a too high impedance of the switch contacts which are primarily designed for current, not for signals. Thus, user kirkus had this nice idea with the reed-relais. For me it sounds brilliant. This way I can keep the big red illuminated switch, otherwise I probably could not use my box - which incorporated much work over a few days - any more at all. Therefore I now need to wait for my order to arrive. I don't feel thats wrong

You may not like the design (still in the eye of the beholder), but I am very satisfied with it:

Never EVER seen a high Z relay contacts...
For Ac motors or DC or audio its purpose is to add as little resistance as possible...

Looks very nice...

Regardlessly if it will work for the mute switch or not, the first breadboard test was successful. The contact resistance of this SIL05-1A72-71D is 0.15 Ohm according to the datasheet (the one with 0.08 Ohm was 10 times more expensive), unfortuantely I was unable to order a Hamlin/Littelfuse HE3321 with 5 V. Very interesting - first time I work with a relais at all!

The Omron G6A-274P-ST40-US-DC24 is VERY good, use it in my pres ect..
50 mOhm...

Thanks so much! ?

Thank you Sir! I just ordered 4 of them as new old stocks (NOS) G6A-234P-ST-US 5DC for only $10, they also have a contact resistance of only 0.05 Ohm.

It works very good by the way, but before I solder all together/shorten the cables I will try the Omron ones, as I expect the best results with them

Important is that, when in series, for the 5 V relay one usually can only use a red LED, but not blue or white (probably also not green/yellow). The voltage would drop too much, but 3.5 V is needed to close the relay contacts. The SIL05-1A72-71D in this case (photo) acts as a resistor for the LED as the coil resistance is 500 Ohms, hence, no additional resistor was used.

Short question: Do you mean this "double short" you illustrated here?

Yes. You can try leaving out the capacitors, and put them in only if there are popping issues for mute/unmute.

Today the OMRONS came. Made in Japan ´

(I think that there is something wrong with my setup, I know need to find out how that actually correctly works.)

They don't have a diode by the way (the SIL05 has one), so I may need to implement one in the curcuit.

Hello!

I can hear the coil (very silently though), but no success. Unfortunately it seems I am just too stupid to understand it



I shift the polarity, the coil is much more audible and it works:



It seems that this is reversed, thus Pin 1 is not at the top, it is at the bottom:

Its a standard dip package, pin one is the top left corner when looking from the top. Just like on an ic.

But why doesn't it work then? It only works reversed.

On the printed illustration they say pin 1 (+) is at the top, but then it does not work:

Because its the Bottom view...
+ is on left (Top View)...

Now I understand it. I was always bad in geometry, I cannot even picture a triangle to myself

I am glad though that I did not break the device, because of course, I all the time tested it in reversed polarity

Don't worry about it. I have been an electronic engineer for over 45 years and I made that exact same mistake when I first used them.

Cheers

Ian

I STILL make mistakes when looking at top of PCB, turn it upside down and get it wrong...

Hello ?

It is finally done! Thanks all of you for the lots of help

Please find a lot of photos and final test, while the second audio is with +50 dB. I don't have any pop or click issues, but that might be because of my hardware I use.

Work

Since the reed relay is a DPDT I used that possibility and built two stages for shorting pin 2 & 3 together, but I didn't use a resistor to not interfere with the signal. The difference between one vs two stages is minimal, but I believe there was a little improvement using two tages.

LED and reed relay are in parallel to ensure it always gets the > 3.5V and because I wanted to use a white LED.

I used 2 x 0.14 mm² each for XLR pin 2 and 3. Pin 1 I left as it is with 1 x 0.75 mm².

The box is made of 1 mm aluminium sheets. I drilled 32 holes and used M2/M3 bolts.

Photos















Audios

Small change I did: Now I do not short pin 2 and 3 twice, but only once. The second gate of the relay I instead used for the LED. I soldered 100 Ohm resistors at pin 2 and 3, this is an improvement (220 Ohms makes not much difference, with 10 Ohms there is almost no difference).

Reason for the change: In case the relay fails, now also the LED will not light up. Previously the LED would have been active once the button was pressed, no matter if the relay worked or not.