OT: Screwball audio idea.

farss wrote on 7/12/2005, 1:25 AM
See here I was thinking about how to connect a balanced dynamic mic into a consummer camera without a balanced to unbalanced converter (to many boxes and connections confuse customers at times) and a thought just entered my head and I can blame Jay Rose for this one.
One thing he warns about are silly people who wire one side of the balanced mic to one channel and the other side to the other channel resulting in lovely dual mono with the two channels out of phase, fine until you run it through a mono TV, no sound at all.

But hang on me thinks, the signal we want is recorded onto two tracks nicely out of phase but the common mode noise that we don't want is in phase!
So if in post we were to invert one channel and add the two together (and maybe drop the level 6dB) then we've effectively emulated a balanced input. The common mode signal is now cancelled and the difference signals add.

Now there's a few caveats, if the camera has independant AGC on each channel things could get screwed up but I doubt they do, that'd maybe cost 10c more so I doubt they went to that expense. Also we don't have the RF isolation that good balanced inputs / tranformers give but it should be better than just a plain unbalanced connection.
Bob.

Comments

Chienworks wrote on 7/12/2005, 3:41 AM
Even if the two channels had independent AGC it shouldn't get too messed up. The signals are identical after all so whatever one AGC happens to do the other should follow extremely closely. I doubt the difference would be enough to notice during normal program material.

Prolly the only concern i would have is impedence mismatch. Most balanced mics are 600 ohms or less while most unbalanced inputs are expecting 3000 ohms or more. This might cause some nasty hum. However, (thinking cap isn't fully awake this morning yet, so bare with me), wouldn't the hum be in phase and therefore cancelled too? Or do i have that backwards?

Overall though, still seems like a dandy idea.
farss wrote on 7/12/2005, 4:24 AM
You're right, the hum would be (hopefully) in phase and therefore cancelled. In any case the impedance mismatch can easily be fixed with a resistor and that'd fit inside one of the connectors. As a general rule of thumb impedance mismatches have got to be more than an order of magnitude off before things get really nasty. Mics and mic inputs these days seem to be anything between 150 to 1000 ohms. If you're into signals going down long transmission lines then correct impedance matching is of course critical but that's another story, look up Maximum Power Transfer Theorem.
Bob.
Coursedesign wrote on 7/12/2005, 8:14 AM
What's with all these hum and noise problems?

Years ago, I shot training programs with two Sony TRV900s. I used a couple of different battery-powered phantom mikes, all balanced XLR with good quality balanced microphone cables to the camera, then a very short stub to connect to the 1/8" unbalanced jack, with no extra electronics.

It was hum and noise free even under hoards of fluorescents, only the noise of the camera built-in micpre. Definitely good enough for many applications.

For more demanding productions I recorded to DAT and synced in post, this gave a much "sweeter" sound of course, but created a lot more work.
farss wrote on 7/12/2005, 4:31 PM
Agreed, you should be fine with unbalanced mic leads for a few metres. Point was maybe there's an ingenious way to emulate having a 'balanced' input using what you've got.
Bob.
Hunter wrote on 7/12/2005, 6:13 PM
In my years in home and car audio field, I've think that some people are magnets for noise. They can do every thing right but still get noise, yet others can have wires all over the place and get nothing.


Hunter
ottowr wrote on 7/12/2005, 9:55 PM
Unbalanced devices should always be connected with balanced wiring for best results, connecting the shield and "cold" at one end, and leaving the shield unterminated at the other end (or terminated via a 0.1uF cap to ground at the "un-terminated" end.)

The idea of using a mic with balanced output to 2 x unbalanced inputs (but out of phase) may or may not work.... It probably wouldn't work if the mic has a output transformer with floating ground. I.E. The mic cable's shield may go to the mic case, but the mic signal is only on the "hot" and "cold" pins and not referenced to the ground, so the ground won't give you the "cold" side of the signal.

If you have a mic with an output transformer that has a centretap going to the shield of the cable it would work ok.

Don't you love it when someone's question fixes one of your own problems?

I just grabbed a Beyer M88 to check this, its been sitting on my bench for a year, has a brand new capsule, sounds thin.... hmmm seems I wired it between 1 & 3 instead of 2 & 3. At least that's sorted...

And while I had it open, it seems thet the transformer connects to 2 & 3, the shield goes only to the case....