Analog Stuff : Microphone Compressor

On a sunny day (Sun, 3 Jan 2016 00:48:14 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :

JFETs will be around. I remember somebody using a JFET to silence music while speaking in the mike (disco). You can make an AGC too with one.

But 'compression' is slightly different.

here I describe how you can use software (Linux sox) to compress audio and replace the old NE570 chip. That NE570 was used in satellite phone systems (compander on tx side, expander on rx site), to reduce noise (was not digital in those days). The advantage of that chip over digital signal processing is that there is no latency, say annoying delay, although in a sat link there always is a huge delay, but for earthling projects the delay would not be good.

Maybe there is a replacement (modern) compander / expander chip?

If you do not care about distortion and that speaker as generator .. sorry mike... two diodes in anti-parallel ;-) LOL

A compander chip on the TV output would help a lot to get rid of that silly trick where the movie sound is too low and then the commercials blast in at close to clipping... I thought they made that illegal, but its still happening at least one German TV.

Yeh... what kind of mic?

And, you mean to shunt or load the transducer element, to shift its power range before clipping?

The common active capsule (electret + FET) won't do diddly, of course.

An electric (HV bias + high-Z buffer) needs a shunt capacitor to control its dynamic range, because its output impedance is capacitive.

A dynamic mic (akin to your speaker example) will work, but only up to the range afforded by its DCR. You have to shunt it with a negative resistance to go towards zero. So, you get maybe 3dB of extension, hardly worth it.

I think electric mics are most popular among professionals because they have stupidly high range to begin with (mechanical clipping >120dB?), and the gain can also be controlled by the bias voltage (which almost serves the same purpose, but won't actually change the clipping threshold).

Tim

Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

I am looking for appropriate devices to shunt the input of a microphone to be able to compress it, and to possibly damp its diaphragm's motion when the broad screams into it on karaoke night.

I think a JFET would do for sure, but can we count on them being around long enough, this is new design. And if there is a better such device out there, I figure this is the place to ask.

Eventually I will have to make this demonstrably better, but I am sure I can. Look, take a nice 15" woofer on the bench and push on the cone. Now short across the terminals with a nice stout wire or whatever and push on it again. Feel the same ? NOPE.

While this idea is part of it, it is not at the core of it. And I am sure it has been done somewhere. But if you cannot find it, do it and be findable. If they come and sue on patents, make sure you got something to sue for.

Y'know, I walked by a TV the other day and there is a show on about UFOs and people sitting there saying "30 people calling in can't be all wrong" and shit like that. Know what ? I think there are mad scientists working out in heated/airconditioned barns all the time figuring out new shit, but do not want to give it to the world yet.

So anyway, the concept of damping the mic might have been done, but I want to try it anyway, and this is only part of the whole thing.

And none of this involves any high power, I can use a wallwart. Fuck you UL ! LOL.

Anyway, I need a fast electronically variable resistor, like a JFET of the past. What is out there that can do that ?

Thanx.

Why don't you buy a mic compressor off the shelf that's designed for precisely that purpose - vocal leveling for live sound? There are only about twelve thousand to choose from...

--


----Android NewsGroup Reader---- 
http://usenet.sinaapp.com/

to be able to compress it, and to possibly damp its diaphragm's motion whe n the broad screams into it on karaoke night.

long enough, this is new design. And if there is a better such device out there, I figure this is the place to ask.

I can. Look, take a nice 15" woofer on the bench and push on the cone. Now short across the terminals with a nice stout wire or whatever and push on i t again. Feel the same ? NOPE.

re it has been done somewhere. But if you cannot find it, do it and be find able. If they come and sue on patents, make sure you got something to sue f or.

s and people sitting there saying "30 people calling in can't be all wrong" and shit like that. Know what ? I think there are mad scientists working o ut in heated/airconditioned barns all the time figuring out new shit, but d o not want to give it to the world yet.

ant to try it anyway, and this is only part of the whole thing.

u UL ! LOL.

the past. What is out there that can do that ?

he probably wants a limiter not a compressor

-Lasse

Thats a good deal.

There's little difference between a limiter and a compressor, it's just a compressor that's working with an infinity -> 1 ratio. Many pieces I've seen can do that as well.

--


----Android NewsGroup Reader---- 
http://usenet.sinaapp.com/

e y

the standard vocal microphone is is still a dynamic the SM58

It'll never clip, I think Shure say the theoretical max SPL is something like 180dB

-Lasse

o be able to compress it, and to possibly damp its diaphragm's motion when the broad screams into it on karaoke night.

ong enough, this is new design. And if there is a better such device out th ere, I figure this is the place to ask.

can. Look, take a nice 15" woofer on the bench and push on the cone. Now sh ort across the terminals with a nice stout wire or whatever and push on it again. Feel the same ? NOPE.

it has been done somewhere. But if you cannot find it, do it and be findab le. If they come and sue on patents, make sure you got something to sue for .

and people sitting there saying "30 people calling in can't be all wrong" a nd shit like that. Know what ? I think there are mad scientists working out in heated/airconditioned barns all the time figuring out new shit, but do not want to give it to the world yet.

t to try it anyway, and this is only part of the whole thing.

UL ! LOL.

e past. What is out there that can do that ?

In radio we used to use the LA4 on the mic.

google VOGAD, take your pick

POTS phones used a stack of metal diodes & a filament lamp.

I put a tiny filament lamp in series with the speaker on my last TV. Worked quite well.

NT

Just buy an Alesis 3630 and use a limiter in the insert circuit of the channel strip. The preamp will overload but it's less likely to damage things.

If you want to get clever, run the same mic to two strips and set the gain staging so that when the 3630 ducks the one strip, the other comes through clean.

Or even better, use the two halves of the 3630 to manage the dual gain staging for you.

Much less soldering and R&D that way.

There's one for $59 plus shipping at Music Go Round.

--
Les Cargill

Confirmed - but with the standard 4 opamp input on most gear these days,

*that* will clip.

--
Les Cargill

There's one of those for $6xx on Reverb.com .

--
Les Cargill

er

l

the

dly

nd

s

Agreed. Easiest solution is use a decent *dynamic* mic like the SM58 (or al most any dynamic mic marketed for "vocal" and "live sound" applications). I doubt your vocalist(?) will exceed that mic's max SPL rating. Dynamic mic s generally impart "color" to the audio, though (response curves aren't fla t), but you may like the sound anyway.

If you want a flatter response, then there are *capacitance* ("condenser") mics made with high SPL rating - you just have to look for this spec. I've used industrial versions with 140dB-SPL rating.

Next concern is to prevent overloading the audio system (clipping, etc.). L oop in a comp/limiter (comp = dynamics compressor) on that mic's input ch annel. You can set the comp/limiter to function only as a limiter, but you may like the sound effect of compression as well (e.g., sound like all the pop recordings).

Some "live sound" mixing consoles have built in comp/limiters. Check that o ut first.

-Rich S.

This is probably too fiddly, but you can put a 600 ohm resistor in shunt across pins 1&2 of a Shure SM57/SM58 and it'll tame the presence peak @4k.

These were built initially to drive 600 ohm transformer inputs, so the shunt resistor tames out the peak caused by driving a voltage-coupled input.

But the little amps inside those will distort and overload at higher SPL.

--
Les Cargill

I have read and appreciate all the responses, but don't have responses to t he responses yet.

They are not hitting the spot. Good info, but I specifically want to do som ething a certain way to see if I can actually measure any difference. The d ifference may be minor but it makes for great advertising copy. A graph of distortion or whatever with the circuit and without. Plus I do not want exp ensive mics. In quite a few applications they are going to be handled by pe ople who are, to say the least, inebriated.

This market is so limited that this is only viable doing it nationwide and that is what these new associates of mine do, in fact worldwide.

How much difference does this particular circuit make ? Probably not enough for all these drunks to hear. But it is a feature. The rest of the process ing this thing will do will carry it. It's just that it gives the impressio n of exclusivity.

But I want to literally take a loudspeaker, for example the midrange out of one of my Boston A-150s which have extremely low distortion, and keep the level to that constant at say 1 KHz, and move it closer and farther. The ro om reflections and their effects on the response will have little to do wit h a pure sine wave. And it will be pure enough electronically, my analyser floor is 0.0016 %. Distortion caused by the mic will change as I change the distance between the speaker and the mic. Get it ?

I believe about the mics that can handle 180 dB. However, aren't they towar d the high end ? Cost is a factor because I want to make money. Remember th at sometimes they are imperiled by imbibers.

As such, I would like to have these things EQed and set or an exact model o f mic that is going to be available forever or we can get enough of them to last the product life cycle, and WE sell the replacements at a profit of c ourse. Even tell people, YES, you can use other mics on it but they will no t be properly calibrated. So we get those sales.

The rest of the details on this thing are not important for this discussion . I want to have like a shunt as a limiter. If y'all are SURE it will not m ake any positive difference at all say so. I just find it hard to believe t hat it has absolutely no merit. Thing is, it is not expensive to implement as long as those FETs are available. Maybe I just use what Digikey has and give it a whirl. If any of you have a good part number I would appreciate i t. It just has to be really linear.

** The Shure SM58 has long been the reference standard for vocal mics used for live sound and there are many imitators.

Like any dynamic mic, diaphragm excursion it much greater at low frequencie s that at higher ones and so there is a physical limit which sets the maxim um SPL that can be handled cleanly. There is also a small transformer insid e that overloads due to core saturation a little above this level.

Expect a typical example to handle 140dB SPL at 40 Hz and maybe 160dB SPL i n the mid range. 180dB SPL is impossible as air itself becomes non-linear a nd distorts heavily at such levels.

... Phil

** No it won't.

Electrical damping only works if there is an impedance peak produced by back emf at some frequency.

The impedance of an SM58 is 280 ohms between 500Hz and 5kHz rising smoothly to 560 ohms at 20khz, due to voice coil inductance. So adding a load resistor in parallel rolls of the high end by a few dB.

BYW: The relevant signal pins are 2 & 3.

.... Phil

Sounds like what you really need is an on-off switch.

Sylvia.

The idea is to provide a 600 ohm overall load for the mic to damp it, rather than say, a 2k ohm load.

If there's a fly in the ointment, it might be worth taking up with rec.audio.pro. I'm sure there'd be interest.

Indeed they are.

--
Les Cargill

On a sunny day (Sun, 3 Jan 2016 20:02:40 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :

If you want to use a JFET you will probably find that you need an opamp and a peak rectifier to get enough gate voltage, that is why you AFAIK do not find it in the first stage, And mind the time constant,

To use a JFET as variable resisor is something like this:

-------------------------------- + | R1 | | [ ] | === | C1 | -- --- Cd |- ||---->| | | R2 | | -- /// [ ] [ ] R3 | | | | | Vc | Ain ---||----------------------A out Cin

C1 decouples audio for DC R1 and R2 for you to figure out, R3 high C1 couples some audo to the gate Vc is the control voltage that now is negative relative to the drain The circuit refers to te h+, and that must be prop[erluy decoupled with Cd. Your wordwrap sucks, Get a real newsreader

It is linear within reason.

Some singers like to eat mikes, I was wondering last night if the cheaper solution would be to add some bad flavor to the mike surface. Or sugar if the signal is too low.

Why not patent it? Maybe little dispensor bitter / sweet. Sink outside the box ..

Wow an that a 8:90 in the morning, oops, 8:30, sleep in the eyes...

Hopefully you will get over it.