Gotta love google! Found MS circuit

[kante1603]

Learn something new everyday.
Guess it works different than M-S Mic technique?

It is Mid/Side-technique-built with encoder and decoder circuits-so both directions!Please do a search on google/wikipedia.On microphones in ms you have a Cardioid/Omni-signal for the mid and a figure-eight for the side.After having them pre-amplified you can run them thru the decoder part to get the stereo signal.

Is it possible just to put level adjustments in between to do that?

Not on point "B" as it is just the de-balanced input before  the encoder-circiut.Best point is to level them after encoding and before decoding back to L&R.
I am doing this on my external gear with the i/o-controls or make-up (e.g. a compressor) inserted on the dedicated balanced sends and returns for level-matching.


Maybe I simply do not understand what excactly you want to achieve.
If you want spreading/widening of stereo-signals you might want  to read the following discussion:

http://www.groupdiy.com/index.php?topic=37144.0

Look at reply #6 from member ruairioflaherty,I absolutely agree with him!

Best regards from germany,

Udo.

 

[sr1200]

Good thread and yeah. I see the point. Ive only used ms micing for like acoustic guitar performance which basically just spread out the mono signal into stereo. I occasionally use a farchild plugin on mixes in the lat/vert mode (MS) to add a little distance if the main vocal or soloist is a little too up in your face (but why not just bring it down in the mix?... Sometimes using the ms plug just sounds better. And now that i think of it the plug IS compressing the M and/or the S.)  I do have to disagree with the statementbof the plug sounding better than doing it analog.  Our mastering eng. Uses an analog box to do it and i swear there is a genie in there or something.  Never heard a plug do what he does to it. But i guess thats the years if experience at work as well.

 

[abechap024]

digital is great until you have a real analog unit in the same room You know something about digital that is strange? Try to represent the number 1.5 in a computer. Its Impossible! You can get REALLY REALLY close but there is no way to represent the number 1.5 in binary. and that's just one example, there are many more numbers binary cannot represent exactly. (but for the same reason there are many binary numbers that can't exactly be represented in decimal)

Not that computers are bad or don't sound good, they can sound very good - but It just reminds me that analog is something more special indeed.

 

[Rochey]

Abe.

That is complete tosh.

If you've enjoyed any sound system playback in the last 10 years, the likelyhood is your listening to a product that has at some point passed through ADC's or DAC's, and has most likely, at some stage been stored in the digital domain.

What digital doesn't do is add sweet things that our ears like to hear like subtle harmonics. Such things look bad on the Scope, but sound wonderful to our ears.

In addition, most digital systems cope with accuracy by adding extra bits to deal with smaller and smaller values (a trick known as "double precision). To put it in perspective, research shows that most humans, at best have a dynamic hearing range of less than 130dB. (that's from the quietest sound we can hear, to near instant hearing damage). The old Vinyl records you may like have a dynamic range nearer to 70dB. CD's... nearer to 96dB.
I believe (and I may have missed something here) that actually makes a CD more accurate at "playing back" your value of 1.5, than an original vinyl LP, or tape for that matter.

However, just in the same way as analog, there is very badly designed digital, and very good designed digital. I've been privy to both, and believe me, when digital is done properly, it's a glorious sound.
Badly designed digital is more about software than about hardware. In digital, we all fear the mighty clipping, when there's no higher to go. (where as in analog, you saturate for a short time, and the analog circuitry will saturate gently, with some wonderful harmonics).
In *good* digital, you allow some extra bits in your databus for clipping, or you do some clever tricks to keep shifting your valuable data within the 24bit realm.
In badly designed digital, multiple clip's or data truncations occur (when you scale the audio lower and lower in the 24 bit word, and start to introduce data truncation). Well designed Digital will ensure that no data is lost, regardless of the amount of headroom you leave in the system.

Once you've gotten past that, then you need to start worrying about jitter in the clocks affecting high frequencies. You can definitely hear a jittery digital system as a harsher, slightly more distorted high frequency range.

Sorry to change topic flow a little hear... it's just that nothing frustrates me more than people taking stabs at the whole digital vs. analog discussion, and throwing a big old opinion out there, without having some understanding of whats going on under the hood.

A bad design, whether it's analog or digital, will make the sound sh*t.
There's just a few more parameters to worry about with digital. Something our friends in the audio interface, mixing software and musical instrument world still continue to learn on a daily basis.

 

[ruairioflaherty]

I do have to disagree with the statementbof the plug sounding better than doing it analog.  Our mastering eng. Uses an analog box to do it and i swear there is a genie in there or something.  Never heard a plug do what he does to it. But i guess thats the years if experience at work as well.

M = L+R
S = L-R

That's it.  No genies involved.  You nailed it when you said your guy years of experience.  I mastered an album today and used MS eq on perhaps 8 of the 11 tracks.  No widening or spreading, just gaining a little extra control by eqing the M channel differently to the S.  This was a folky album and quite a few of the tracks had little esonances or honks in the acoustic guitars which were panned mostly left and right.  Going MS allowed me to eq out those nastys and add a little sparkle to the acoustic without thinning the vocal or bringing out essy sounds.  The biggest eq gain or cut on the whole album was 2.5dB but mostly 0.5 to 1 dB cuts and boosts.

Cheers,
Ruairi


Using analog eq and maybe compressors in an MS loop can be very powerful and effective but it's not the analogue MS matrix that's adding the mojo, it's teh choice of processing (or not) in the loop.  Look at the great effort Wayne went to when designing his MS board to avoid any change in signal integrity or crosstalk.

 

[zebra50]

Try to represent the number 1.5 in a computer.

Isn't that one example that does have an exact solution? 0.5 = 2-1, so

1.5 (dec) = 1.1 (bin)

Of course there are many examples that don't.

http://www.exploringbinary.com/binary-converter/
http://en.wikipedia.org/wiki/Binary_numeral_system#Fractions_in_binary

Reminds me of the joke...
Why do mathematicians always confuse Halloween and Christmas?
Because 31 Oct = 25 Dec.

 

[kante1603]

M = L+R
S = L-R

That's it.  No genies involved.  You nailed it when you said your guy years of experience.  

Exactly!

@sr1200:

I think after having understood what MS really is it can be a powerfull tool.I grew up with these techniques and yes-it took some time to really get familiar with it.You can fool arround with digital and analog stuff-it´s up to you,but:Ms is not meant to be used as an effects-processor(allthough doing nuts things in the chain will make it be one ).

A pretty good way to understand the theory of MS is maybe creating a complete MS-chain in your DAW by yourself doing both the encoder and decoder.Then you can learn how parameters affect each other.This can be done right after the formulas stated by ruairi on the top of his reply (attention:it takes a certain amount of channels to do it).

Here are the overviews of how to handle it:

Encoder:

Decoder:




Anyway,my prefered way to use MS is the analogue way.
In general I simply take the best of both worlds (digital recording,analogue processing etc.).
I have made best experiences with Wayne´s wonderfull board and the awesome flexibility.Most if not all "jumper points" on board were created by Bob Katz himself and are the result of Wayne asking him how the board can be even improved.Mr. Katz gave him an answer with a lot of possibilities of what you can do additionally right on board and uses it in his analogue rack-not bad.

So-if you want to go this way:I again highly recommend Wayne´s board.


Best regards and a "happy learning" ,

Udo.

 

[emrr]

All the other variations one might want to explore are covered in depth in various posts at the other forum.

http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=112
http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=71
http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=66
http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=110
http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=15

 

[sr1200]

Sorry to beat the dead horse here, but is there anything wrong in putting level controls on the M and S channels to play with the balance if so desired?  There have been times when people have brought mixes in where
A) the main vocal was either WAY too low or loud
B) theres no more room to compress cause every down beat is hitting the red.
C) the mix sounds horribly narrow.

I guess this can be done with the gain on the eq/comp/etc that is in the chain, but was thinkin that a control right on the box might be nice as well.  Even thinking just a couple passive trim pots would work just to nudge down either of the 2 channels (might keep the signal a little more pure too no)? 

 

[abbey road d enfer]

IMO there should be level controls at least on the coder, for the simple reason that the basic equation M = L + R results in a level that is nearly 6dB above the individual L and R channels when they are highly correlated, as they are in most contemporary productions.
The German broadcasters had reasoned that the equation should be M = 0.707 (L+R), because it gives unity gain for separate sources (for the very same reason pan-pots were supposed to have 3dB attenuation at center); they had just forgotten the fact that L and R are almost never uncorrelated. In their infinite wisdom Neumann had internal level controls on their MS coders, in addition to the front panel controls.
On the decoder, it may not be really necessary since the outboard that's inserted has generally some form of level control.

 

[sr1200]

Anyone thats built one with the PS schematic that Wayne had to go with the unit... are these the right values for the caps and resistors (theres a couple of ? that i have NO idea what to put) and what transformer should go with this (running at the recommended 18v)

should i go higher than something like this (
http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=TE62064 or more like
http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=TE62085 (overkill?)
)

The top half has my guesses as to values.

Anyone know the REAL values?

 

[abbey road d enfer]

L1 & L2 are LED's. R7 & R8 can be anything from 1k to 10k. In fact they are not necessary for the operation; they are just here for visual monitoring of the rail voltages.

 

[sr1200]

So was i on on the rest of it? lol

 

[emrr]

You might want to talk about it at the other forum, since Wayne and Roger live over there. 

 

[sr1200]

Kinda deads-ville over there, and when i sent an email asking about it, i just got the original PS image as a response with no values again and a "this is all i have" note.

 

[emrr]

only deadsville if you don't post there.  you gotta try.  the creators certainly won't answer you here.