mid/side splitter/combiner

Having read Kieth mention that he'd been sorting out a Fairchild 670 this week, i was reminded of the joys of using one nearly 20 years ago . . . I loved it not only for it's "sound", but also for the mid/side control. i believe that the 670 does it with transformers(?). I would LOVE to be able to do something similar. I have been a big fan of mide/side mic technique for the longest time. Does anyone have a circuit to split left/right into mid/side, so i could use a.n.y.other two channel compressor in the same way, and then re-combine into left/right? i'd love to build one with Melcors for example. Is this a "shuffler" circuit? Does the Pope wear a silly hat? . . . Do these shoes go with this bag . . .

v v excited right now,

Andy P

Here you go:

Active
http://www.uneeda-audio.com/ms-mat.htm

Hope that helps.....must be a diagram online for the transformer one.

Cheers Tom

Tom, thank you very much indeed! this shows me how to re-combine an m-s signal to x-y, but not how to "encode" x-y to m-s. i am afraid that my electronic knowledge procludes me from being able to work out how myself . . . .Anyone?

thanks,

Andy P

Me thinks you just need some summing amps for that but I may be wrong.

I think you sum L+R into one for M and need to get L-R for S

I guess you would do that by summing L + an out of phase R? I think that may be it.

Not sure tho.

Cheers Tom

Thanks Tom!

thinking cap is on . . .

Sowter have a schemo for the transformer version on their site:

It's here.

MS encoder and MS decoder should be the same thing, N'est-ce pas?

Keith

Yeah, I would figure a passive one should work backwards. Anyhow, here's Jensen's take, with only two transformers, but of course they specify their own. http://www.jensentransformers.com/as/as065.pdf

Do Cinemag or Lundahl have recomended circuits? (Mainly interested to find out what models to use.)

Bear

Hi!

Lundahl LL1588XL's are used by some mastering guys for this job.

Cheers Tom

There was a discussion about the (Neve?) passive approach over on Prosoundweb.com, maybe in Klaus Heyne's forum. Within the last six months, anyway.

Neat setup with balanced signals - basically a diamond-shaped bridge where you put in M and S on opposing corners and then get L and R on the other corners. And of course it works backwards.

My version:



Jakob E.

I've never actually tried this with transformers, only an electronic matrix. Presumaby to get good separation of the signals into their respective places, you need two absolutely identical transformers, otherwise the cancellation won't be complete. How good is it in practice? I mean, if you feed a signal into the M channel only, what's the actual difference in output from the L & R? Or is this easier to measure the other way around, by feeding an identical signal to L & R and observing whatever's present in the S channel? (assuming that there will be a residual output, of course...)

Have you tried this, Jakob? Are the measured results better or worse than the electronic version?

I do realise of course that there are other advantages to the transformer version - I'm just intrigued.

Absolute isolation is overrated - other than in theory - specially for MS-like stereo sources.

I've only used transformer-matrices, and they work very well indeed.

Note that you use the same matrix for conversion both ways..

Jakob E.

[quote author="gyraf"]Absolute isolation is overrated - other than in theory - specially for MS-like stereo sources.

I've only used transformer-matrices, and they work very well indeed.
[/quote]
Fair enough!

I was still intrigued, so I've just lashed together a couple of OEP A262A2E transformers (1+1:2+2) - that's all I have lying around that is vaguely suitable - and measured the rejection from both pairs of available secondaries. I got -71dB from one pair, and -73dB from the other. And as you suggest, for this application these sorts of figures are going to be more than adequate in real terms, I think. And I dare say that if I took a little more care, and laid these out sensibly, not hanging them together with croc clips, that a better result could be obtained anyway.

[quote author="Bear"]Yeah, I would figure a passive one should work backwards.[/quote]
Doesn't need to...

An M/S splitter works forwards as an M/S combiner... they're the same thing as I see it, active or passive.

Keith

Wow!

You guys will never cease to amaze me . . . So much valuable information in such a short time!

I think that i will try the passive approach first of all. Especially as I have a couple of Sowter 4652's already. They are 150/150:150/150 at 0.5% dist at 34dB(50Hz) - should cope, I reckon! I shall re-combine in the short term up 3 channels of the desk - mid panned up the middle, side up the other two, panned left/right, and rhs phase reversed, just to check it out!
Hmm, i wonder what it would sound like with 2@5@2@0's as a buffer, and Profile 2503's . . . . ? Better leave that for a rainy day . . .

Yeps! I love passive things, me . . .

Thaks, Guys!


Andy P :grin: :grin: :grin: :green:

Just wondering, what would be the advantage with such a separate box ?

Could imagine it gives better fidelity and more possibilities.

But for now I get decent results with this figure-of-eight mic and using the known method of an additional mixerchannel (driven by the first channel actually receiving the S-signal but with flipped polarity).

Or maybe in general: what's the trick/fun/benefit of the M/S-processing of a 670 i.s.o. L/R ?

Bye,

Peter

[quote author="clintrubber"]
Or maybe in general: what's the trick/fun/benefit of the M/S-processing of a 670 i.s.o. L/R ?
[/quote]
The obvious benefit with all M-S processing is that you get to alter the stereo width by altering the relative balance of the M & S signals. Depending on exactly what you are processing, this can be very powerful. For instance, if you record drum overheads as a stereo pair, it's much easier to control the apparent width of the kit this way. Well, I think it is, anyway... but since you are already processing MS, you'd know that already. Had you got something else in mind?

Dear Clintrubber,
Big, Big advantages in mastering. Historically the 670 could be used to limit the travel of the disc cutting head in both up/down(x) and left/right axes. In practice today, you can limit the mono(mid) componant whilst still allowing the l/r(side) componant to remain just as big,(and nowhere near clipping) resulting in a louder/wider fatter mix. There are other advantages. you can de-ess the middle, whilst leaving the side un-touched, for example, since vocals are mostly mono. This will leave the stereo info un-touched. or de-ess the side componant(remove those nasty over eq'ed cymbalsfor instance . . . Or you could eq the side differently to the mid, with no real phase collateral damage . . . also you can effectively narrow or widen the stereo image by adjusting the mid and/or side levels respectively. Original Maselec limiter had M/S de-esser funtion built in, very handy for my mixing inadequancy at the time! Thank you to Tony Cousins at Metropolis Mastering for teaching me SO much about mixing - with the benefit of HindSight . . .

- Oh! It's so exciting to think I can do this at the mix stage too, now!!!!!!


Thanks again, guys, I am ever in your debt!


Andy P

PS you could also use it for de-coding your m/s recorded tracks without needing three channels of your desk. You KNOW m/s makes sense, and if you don't, Check it out! you can effectively"move round" x/y mikes, whilst taking a perfect mono recording, not two off-axis recordings. i used to love m's in the good/bad old days of tape. Not every studio i used to use had a good pair of condensors, and m/s enabled mr to take great stereo with, say, a U87 and a 414(not my first choice, BTW!). Then at a later date of track shortage, if necessary record over the side componant, keeping a perfect mono recording of Pno/ Perc/ Gtr etc, without having to bounce! It's not THAT long ago . . . i only moved to Logic/Protools in 2001!

what would be the advantage with such a separate box ?

Click to expand...

If your using a small mixer you don't have to eat up a third channel for your fifure 8 mic. On a more ms aplication note Run an M/S on a stereo mix sometime. It will give you seperate control of Phantom center instruments Vs instruments panned left and right. You can use this to raise the level of a vocal which would be in phantom center of course all other instruments in phantom center would be raised as well. But with a slight push foward or slight push back it can really polish off a mix already recorded down to your 2 track. It's common in Mastering to do this.