Hello!
I'v been signing in a choir for +10 years. With the pandemic, no choir can rehearse right now. While many things can be safely accomplished at a 6 feet distance, signing in a choir requires closeness and most certainly increases the risk of spitting the virus further.*
Software for online rehearsal do exist, and we are trying them, but they all have latency problems (of course) and are very hard to use for older and/or non-tech people.
I'd like to come up with a solution that will allow a choir to rehearse in the same location but at a safe distance from each other. If money wasn't an issue, the setup would look like something you would see in a studio, extended to let's say 50 people : 50 microphones, 50 pairs of headphone, a 50+ inputs mixer, a 50+ output headphone mixer, 50 microphone stands, etc.
My perfect solution has a few criteria, but I'll consider anything in between.
A few considerations / ideas :
Microphone and headphones
Best case scenario : using the most basic smartphone headset. Yeah, I know, they sound bad. But if you look for "virtual choir" videos that were made recently, they are all made with that type of microphone, and they sound more than OK for this purpose. The target audience here is not audiophiles.
Mixer
I think this is the most complicated part. I know that the signal from the headset microphone isn't compatible with a mixer, and that it needs a different voltage. Also, it's a choir, so signers are supposed to be their own potentiometer. So there is no need for a volume adjustment feature. Is there a product out there that can achieve that? If not, I can solder. I've seen many DIY passive mixer tutorials out there. Would something like that be feasible, if combine with an amplifier?
Long cable
There has to be a long cable between each signer and the mixer. Some signers are gonna be at 10 feet, other at 50 ft and so on. XLR cable a expansive, and my guess is that they might be overkill anyway if the carry a signal from a crappy headset. I've worked a lot with network cable in a previous life. Is it crazy to use good'ol cat 5 cable for this purpose? I'ts cheap and available everywhere. I would then create a physical femake RJ45 to female TRRS 3.5mm adapter for the signer end, and another kind of adapter for the mixer side.
My setup so far:
1. Normal cheap headset (for each signer)
2. Custom box adapter (for each signer) - TRRS input, potentiometer for the headphone volume, battery for the microphone (a diy version of that?), and each 4 wires (L, R, Ground, Mic) to a pair of wires in the ethernet jack.
3. Regular ethernet cable, various length (for each signer)
4. Custom mixer - Ethernet "inputs", all microphones inputs combined (with resistances?) to an XLR output. All L/R outputs combined to XLR inputs.
5. Some sort of amplifier or regular mixer connected to the custom mixer to amplify the microphone and send it back to the headphones.
So, am I completely crazy? I know it's a very long shot. I'd like to start with a prototype for my choir, and try it with 4 or 8 people. If it works, I'll look into a way to expand it and to put it in the hand of other choirs.
I'm not a audio / electronics pro. Just a hobbyist
Thanks for your help!
Felix
*For many experts, normally signing in a choir will not be possible before 2021 : Choirs may have to remain silent long after society reopens - Portland Press Herald
I'v been signing in a choir for +10 years. With the pandemic, no choir can rehearse right now. While many things can be safely accomplished at a 6 feet distance, signing in a choir requires closeness and most certainly increases the risk of spitting the virus further.*
Software for online rehearsal do exist, and we are trying them, but they all have latency problems (of course) and are very hard to use for older and/or non-tech people.
I'd like to come up with a solution that will allow a choir to rehearse in the same location but at a safe distance from each other. If money wasn't an issue, the setup would look like something you would see in a studio, extended to let's say 50 people : 50 microphones, 50 pairs of headphone, a 50+ inputs mixer, a 50+ output headphone mixer, 50 microphone stands, etc.
My perfect solution has a few criteria, but I'll consider anything in between.
- Somewhat affordable
- Easy to build for someone with experience with audio, computers, technology, soldering
- Easy to install and operate at each rehearsal
A few considerations / ideas :
Microphone and headphones
Best case scenario : using the most basic smartphone headset. Yeah, I know, they sound bad. But if you look for "virtual choir" videos that were made recently, they are all made with that type of microphone, and they sound more than OK for this purpose. The target audience here is not audiophiles.
Mixer
I think this is the most complicated part. I know that the signal from the headset microphone isn't compatible with a mixer, and that it needs a different voltage. Also, it's a choir, so signers are supposed to be their own potentiometer. So there is no need for a volume adjustment feature. Is there a product out there that can achieve that? If not, I can solder. I've seen many DIY passive mixer tutorials out there. Would something like that be feasible, if combine with an amplifier?
Long cable
There has to be a long cable between each signer and the mixer. Some signers are gonna be at 10 feet, other at 50 ft and so on. XLR cable a expansive, and my guess is that they might be overkill anyway if the carry a signal from a crappy headset. I've worked a lot with network cable in a previous life. Is it crazy to use good'ol cat 5 cable for this purpose? I'ts cheap and available everywhere. I would then create a physical femake RJ45 to female TRRS 3.5mm adapter for the signer end, and another kind of adapter for the mixer side.
My setup so far:
1. Normal cheap headset (for each signer)
2. Custom box adapter (for each signer) - TRRS input, potentiometer for the headphone volume, battery for the microphone (a diy version of that?), and each 4 wires (L, R, Ground, Mic) to a pair of wires in the ethernet jack.
3. Regular ethernet cable, various length (for each signer)
4. Custom mixer - Ethernet "inputs", all microphones inputs combined (with resistances?) to an XLR output. All L/R outputs combined to XLR inputs.
5. Some sort of amplifier or regular mixer connected to the custom mixer to amplify the microphone and send it back to the headphones.
So, am I completely crazy? I know it's a very long shot. I'd like to start with a prototype for my choir, and try it with 4 or 8 people. If it works, I'll look into a way to expand it and to put it in the hand of other choirs.
I'm not a audio / electronics pro. Just a hobbyist
Thanks for your help!
Felix
*For many experts, normally signing in a choir will not be possible before 2021 : Choirs may have to remain silent long after society reopens - Portland Press Herald
Those headsets usually have an electret condenser microphone with built-in JFET common-source stage. Normally the sound card biases the drain by connecting it via a 2.2 kohm resistor to a supply of typically 2 V to 5 V.
With n microphones and no need for individual volume control, you should be able to just connect all n of them in parallel and use one resistor of 2.2 kohm/n to one 5 V supply. The trick is to wire it all up without ground loops.
Is your Ethernet cable shielded? If not, you will probably get hum. Using the same cable for the headphone signal could also cause oscillations.
On the headphone side, an amplifier meant to drive an 8 ohm loudspeaker should be able to drive 50 headphones with 390 ohm series resistors. So give each headphone a fixed 390 ohm series resistor and a 10 kohm or 25 kohm potmeter wired as a variable series resistor, and connect the whole bunch in parallel to a normal audio power amplifier.
With n microphones and no need for individual volume control, you should be able to just connect all n of them in parallel and use one resistor of 2.2 kohm/n to one 5 V supply. The trick is to wire it all up without ground loops.
Is your Ethernet cable shielded? If not, you will probably get hum. Using the same cable for the headphone signal could also cause oscillations.
On the headphone side, an amplifier meant to drive an 8 ohm loudspeaker should be able to drive 50 headphones with 390 ohm series resistors. So give each headphone a fixed 390 ohm series resistor and a 10 kohm or 25 kohm potmeter wired as a variable series resistor, and connect the whole bunch in parallel to a normal audio power amplifier.
Last edited:
I sympathize with your situation. One of the things that helps me stay sane is a weekly music jam, and we had to shut that down because of the pandemic. That has taken away one of the biggest joys in my life, and that too at a time when we're all dealing with far more stress than normal.
So, how many singers are you dealing with? Your costs go up sharply with the number of people in the choir, so something that's practical for five singers may be quite impractical for thirty.
The necessary circuit is simple, and the signal from these microphones is relatively large, so you may be able to get away with a DIY passive mixer - basically, combining the signal from a few microphones just using resistors. Again, this will probably work well with five microphones, but if you're talking about twenty or thirty people, this is not going to work, and you'll need a more complex solution.
Of more concern to me - how are you going to get the monitor signal back to people so they can hear themselves (and the sound of the choir)? Can you just use a few foldback loudspeakers at the front of the hall, rather than try and drive one pair of headphones per choir member? Because driving thirty pairs of headphones at the far end of long cables won't be a trivial problem.
By the way - there was a product called "Jamhub" that I think would do exactly what you wanted for a small number of people. Unfortunately the company is now out of business. But if you can find a suitable Jamhub in the second-hand market, it might solve your headaches.
Here's a 2010 review article on the Jamhub: JamHub
If you can't find (or don't like) a Jamhub, it might still give you ideas you can steal for your own project.
-Gnobuddy
So, how many singers are you dealing with? Your costs go up sharply with the number of people in the choir, so something that's practical for five singers may be quite impractical for thirty.
One thing to watch out for: trained singers are a lot louder than conversational speech. Many of these headset microphones will overload and distort badly if placed in front of a singer with a powerful voice.
Computer / smartphone headsets usually have an electret condenser microphone in them. These need to be connected to a small positive DC voltage (3V, say), through an external load resistor.
The necessary circuit is simple, and the signal from these microphones is relatively large, so you may be able to get away with a DIY passive mixer - basically, combining the signal from a few microphones just using resistors. Again, this will probably work well with five microphones, but if you're talking about twenty or thirty people, this is not going to work, and you'll need a more complex solution.
Of more concern to me - how are you going to get the monitor signal back to people so they can hear themselves (and the sound of the choir)? Can you just use a few foldback loudspeakers at the front of the hall, rather than try and drive one pair of headphones per choir member? Because driving thirty pairs of headphones at the far end of long cables won't be a trivial problem.
It's a creative idea! I have no idea if it will work, but you can answer your own question easily with a quick experiment. Build one Cat 5 cable, try it out, see if it works! (I suspect mains hum will be a major problem, but I don't know for sure.)
If you also put one transistor in each box, acting as a buffer to lower the impedance of the microphone signal, that might help a lot with the hum problem, and allow the Cat 5 cable idea to work.
Aha, that's what I had been looking for. Four people, sure, you can hack something together. Eight people, gulp, it will cost twice as much. Twenty people...well, you get the idea!
By the way - there was a product called "Jamhub" that I think would do exactly what you wanted for a small number of people. Unfortunately the company is now out of business. But if you can find a suitable Jamhub in the second-hand market, it might solve your headaches.
Here's a 2010 review article on the Jamhub: JamHub
If you can't find (or don't like) a Jamhub, it might still give you ideas you can steal for your own project.
-Gnobuddy
It's not that temporary. A choir shut down for five months or a year is unlikely to survive - there will be nothing left to put together. Singing skills (and singing voices) fade quickly if not exercised regularly.
The time-scale of this pandemic is dictated heavily by the two to three week incubation period of the virus - that's one "time step" in the mathematics of the disease, and it takes many, many time-steps for the numbers (of people infected, or recovering, or dying) to change significantly. It takes some five months for one peak in the infection to form, peak, and die away. We can expect a second peak when the current one finally dies down. Barring a miracle or a vaccine, we're looking at a time scale of at least a year. Heck, if a vaccine is developed, it will probably take a year to manufacture 6 billion doses and get them to 6 billion people, which is roughly what it would take to stop the pandemic (the last 2 billion people, if sufficiently dispersed over the planet, won't be enough hosts to keep the pandemic going.)
We have economic pressures forcing political leaders to pretend things are better than they are right now, and we should all get back to work. But the virus doesn't know or care about human politics and human economics.
-Gnobuddy
I think you are too pessimistic. This microphone capsule https://cdn-shop.adafruit.com/datasheets/CMA-4544PF-W.pdf as well as the far better ones used in this thread: Need Advice for a Weird Project all have an output noise of about 55 nV/sqrt(Hz) when loaded by the default 2.2 kohm. Connect 50 of them in parallel and use a 44 ohm resistor and you get 55/sqrt(50) nV/sqrt(Hz), or about 7.8 nV/sqrt(Hz). A good microphone preamplifier has a far lower noise floor, so you can get away with a simple parallel connection without losing SNR - provided you use a good, clean supply and a reasonable microphone preamplifier.
Why 50 mikes ?
You don't use 50 mikes for recording an orchestra I think. Perhaps a couple for the piano and drums.
Borrow a few condensers (more sensitivity) and try to put them suspended above the heads (church balcony perhaps) and try to capture a few singers and then You will have a picture of how many do You need.
Perhaps Your local Radio / TV station can help. Have You asked them.
Any Recording / Sound engineer online who can help ?
You don't use 50 mikes for recording an orchestra I think. Perhaps a couple for the piano and drums.
Borrow a few condensers (more sensitivity) and try to put them suspended above the heads (church balcony perhaps) and try to capture a few singers and then You will have a picture of how many do You need.
Perhaps Your local Radio / TV station can help. Have You asked them.
Any Recording / Sound engineer online who can help ?
You have a lot to consider so I will just hit a few points
- singers often have tonal issues if they cannot hear themselves at all, the instead of stereo headphones consider sending the singers voice through one channel (unmixed) and the choir through the other.
- cat 5 cable should work fine but the mixing issue is not trivial, good luck
- if the goal is to practice as a choir, sound like a choir I think individual volume control might complicate the end result. Maybe allow volume control over the channel of the individuals voice and centrally control the mixed output channel?
- singers often have tonal issues if they cannot hear themselves at all, the instead of stereo headphones consider sending the singers voice through one channel (unmixed) and the choir through the other.
- cat 5 cable should work fine but the mixing issue is not trivial, good luck
- if the goal is to practice as a choir, sound like a choir I think individual volume control might complicate the end result. Maybe allow volume control over the channel of the individuals voice and centrally control the mixed output channel?
As someone who worked in radio as onair personality, I need to hear my own voice, so I did what MANY do, slide one side of my headphones off my ear. I could hear the mix or cue in one ear, and my voice in the room in the other.
As to singing as a group, having a mixer running voices you only hear in the cans denies you the chance to learn to sing as a choir. You learn to sing as an individual, someone else makes you into a choir...hopefully.
As to singing as a group, having a mixer running voices you only hear in the cans denies you the chance to learn to sing as a choir. You learn to sing as an individual, someone else makes you into a choir...hopefully.
A bit off topic, but the term is physical distancing, we are supposed to try to stay socially close.
Many in the USA have been slow to take up the “correct” term.
dave
6 ft/2m is insuffiicent separation for people singing as droplets go much further when people are singing, particualrily if it is done loudly. Experiements show that even twice that is likely insufficient.
dave
And what do you do for those rehearsals when you only need ten people, or just the sopranos, or just the soloists? If you unplug 45 mics out of the 50, the 44-ohm idea will not work at all.
The OP can tell us more about his requirements, but IMO the circuit needs to work whether you have just one mic, or the maximum number of mics, or any number in between those limits, plugged in. A practical compromise would be to use small groups of four or five mics with individual load resistors, but mix the signals using individual mix resistors. Thermal noise goes up, certainly, but the circuit becomes more usable; you don't have a situation where the thing only works if you plug in fifty mics at once.
-Gnobuddy
There was already plenty of social distrust in North America, particularly within the USA's borders. One of my concerns is that the present pandemic will teach many of us (perhaps subconsciously) to be even more distrusting of our fellow human beings. Not good.
-Gnobuddy
COVID SAYS:
Short answer, FORGET about Choirs rehearsing/performing together, period.
There´s TONS of everyday activities, pleasures, JOBS, CLASSES, SCHOOLS, CHURCH ACTIVITIES, GYMS, etc. being temporarily being given up, sorry but trying to keep a CHOIR running under current circumstances is MADNESS, there is no other way to say it.
IF you can find a way to PRACTICE WHILE EVERYBODY STAYS AT HOME, cool, go for it.
Any other not meeting that constraint is MADNESS.
Somebody had to say it.
Short answer, FORGET about Choirs rehearsing/performing together, period.
There´s TONS of everyday activities, pleasures, JOBS, CLASSES, SCHOOLS, CHURCH ACTIVITIES, GYMS, etc. being temporarily being given up, sorry but trying to keep a CHOIR running under current circumstances is MADNESS, there is no other way to say it.
IF you can find a way to PRACTICE WHILE EVERYBODY STAYS AT HOME, cool, go for it.
Any other not meeting that constraint is MADNESS.
Somebody had to say it.
Actually I see no reason why that would not work. The drain bias voltages of the common-source stages will increase a bit, but that's all.
The hardware to talk to a cell phone headset is pretty simple but not quite as simple as just putting them all in parallel. And you would need some sort of mixing to make the sound in the headphone meaningful. however one dual opamp per mike could do it. And possibly provide a solo + a mix. The high impedance of the mike will be pretty noise sensitive (I have done those experiments) you do need shielding. Still you could possibly come up with a small PCB run off of a single 9V battery with a shared buss connection to all the other modules. From JLBPCB I suspect 100 PCB's could be incredibly cheap, even assembled.
The issue of mike overload is real and you would like to have the the mikes not overload but that may be manageable. The actual level near a mouth speaking loud is quite high and those mikes do need to account for it. Still an operatic soloist would need something different.
For $15 on eBay you can get over ear headsets with boom mikes which could be a fine simple solution to work with.
The bigger challenge will be Plex isolation "booths" that would contain any potential virus but still permit working together. It would be more like working in a recording studio where you need separation to permit later mixing but everyone needs to hear each other as well as themselves. Those would be more expensive but could be built from a mix of celotex and plex with a light wood frame.
From what I see this will be our way of life for some time into the future. It is better to adapt than to rage against it.
The issue of mike overload is real and you would like to have the the mikes not overload but that may be manageable. The actual level near a mouth speaking loud is quite high and those mikes do need to account for it. Still an operatic soloist would need something different.
For $15 on eBay you can get over ear headsets with boom mikes which could be a fine simple solution to work with.
The bigger challenge will be Plex isolation "booths" that would contain any potential virus but still permit working together. It would be more like working in a recording studio where you need separation to permit later mixing but everyone needs to hear each other as well as themselves. Those would be more expensive but could be built from a mix of celotex and plex with a light wood frame.
From what I see this will be our way of life for some time into the future. It is better to adapt than to rage against it.
Yeah, I'd hate to advocate an electronic solution that would encourage several dozen people to congregate in an enclosed space. It's already been shown that a single individual infected 85% of a choir under ordinary circumstances - what percentage would be acceptable, other than absolute zero?
Really have to think out of the box here, to allow members to accomplish it and stay at home. Something like everyone calls in on their cell phone - ladies one number, gents another - then a radio station broadcasts the (stereo) mix back to FM radios at everyone's homes for monitoring. Some churches around these parts have radio towers... Pricey? Yep. Feedback? Yep.
Death of any yet uninfected member due to C-19? Not on your watch!
Really have to think out of the box here, to allow members to accomplish it and stay at home. Something like everyone calls in on their cell phone - ladies one number, gents another - then a radio station broadcasts the (stereo) mix back to FM radios at everyone's homes for monitoring. Some churches around these parts have radio towers... Pricey? Yep. Feedback? Yep.
Death of any yet uninfected member due to C-19? Not on your watch!
Last edited:
According to the article Felix referred to in the opening post: "At the micron level, the aerosolized virus could travel as far as 16 feet, a pair of experts told a national audience of musical educators on Tuesday night." So you either need a very large rehearsal room or screens of some kind.
Off topic: does anyone understand why the distancing rules are very different depending on what country you are in? Here in the Netherlands the official number is 1.5 m, wherever Felix may be located it is 6 ft = 1.8288 m, for Dave it is 2 m and in France a mere 1 m will do.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.