Can someone possibly explain "impedance balanced"?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

strangeandbouncy

Well-known member
Joined
Aug 8, 2004
Messages
2,112
Location
West Sussex. UK
Hi, guys,

I have seen a couple of mentions recently ( in Tomwaterman's Calrec re-design thread for instance - great idea btw!) of "impedance balanced outputs" Coulsd someone possibly explain what this is? I couldn't find anything terribly extant with a search . . .


Thanks as ever,


ANdyP
 
Oops, there'll be lots of links but I can't remember one right now.

OK, let's try:
Impedance balanced means that for each of the two signal-lines the output-impedance is the same as 'presented' to the next piece of gear,
and it'll be in combination with only one line actually having signal on it.

The other line is then just simply a passive network connected to ground and that network is chosen as to mimic the output-impedance of the 'active' like as well as possible.

The thought behind al this is that you get the good things (disturbance-rejection) of balanced lines because of the identical sensitivity of both conductors for disturbances, so that's about impednaces of those lines.
Whether there are signals on those lines of opposite polarity OR just on one line doesn't matter ! The following sym.-input simply subtracts the nastyness on both lines and your disturbances are reduced.
You do get some improvement from both lines carrying signal but that's a bonus coming from twice the signal. But the essential part lies in the identical susceptibility, which is determined by the impedances. So make these as equal as can be.

So there's no signal of the opposite polarity present on that second line, something to keep in mind when you go to an unbalanced device
and want to use signal-swapping a few lines to get the other polarity.


Here's an example:

http://www.jensentransformers.com/as/as004.pdf

(C3 & R4 are mimicing C4 & R5.
The output impedance of the 'active part' is about zero, so little or nothing has to be added to C3 & R4)

Hope it's more clear now - it's a simple concept but hope I've written it down not too hasty.

More good info: http://www.jensentransformers.com/apps_wp.html

Bye,

Peter
 
Go to the neumann site and look for the PDF "Transformerless Studio Condenser Microphone" It is about the tlm170 design, 1983 AES.
 
Does this mean that when I come to wire my unbalanced desk outputs to a TT patchbay it would be advantageous to wire the 'sleeve' connector via a resistor to ground that matches the output impedence? :?
 
[quote author="emtee"]Does this mean that when I come to wire my unbalanced desk outputs to a TT patchbay it would be advantageous to wire the 'sleeve' connector via a resistor to ground that matches the output impedence? :?[/quote]

Things get sticky when the cabling itself is highly asymmetrical, such as the worst case of a simple single center conductor coax. The outer shield gets exposed to electric fields; the center conductor sees only the electric field of the shield, except for a bit of stuff that gets through depending on the type of shielding/screening material.

For magnetic fields the situation is rather different, and with equal impedance on shield and center conductor the mag fields will couple fairly equally at audio frequencies, allowing the common-mode rejection of the receiver to ignore the induced signals.

Then as well there may be conducted noise, based on potential differences between the source and destination.

Using a twisted pair under a shield often helps even when the source is unbalanced. See Bill Whitlock and Jensen for some good discussions of all of the issues going from unbalanced to balanced and vice versa. There is also his contribution "Grounding and Interfacing" to the 3rd edition of Handbook for Sound Engineers, Ed. Ballou.
 
[quote author="emtee"]Does this mean that when I come to wire my unbalanced desk outputs to a TT patchbay it would be advantageous to wire the 'sleeve' connector via a resistor to ground that matches the output impedence? :?[/quote]
Is your patchbay symmetrical ? And your feeding those unbalanced signals to gear with balanced inputs ?
It might be of use then (in TRS-speak, do it for the RING then, not the SLEEVE) but may not be worth your while/hardly noticable for short cables etc etc Hard to tell. If it's working fine now I wouldn't bother, but please don't believe me, ask the guys who have made those huge patchpanels.
And it's undesirable when you happen to use balanced sources (when getting another mixer some day)

Bye,

Peter
 
D'oh! :oops: I meant 'ring' connector, as in hot to tip, ground to sleeve :oops: and resistor (equal to the output impedance of the desk buss out) from ground to ring. Thing is, it ain't working at all at the moment, it's a ground up wiring job, as it were.

Basically sending unbalanced outputs of a TAC Matchless to TT patchbay that will be normalled to balanced inputs of several Motu interfaces.

I know that the Motu's work fine with a TS jack as input, just wondering if it's advantageous to wire the patchbay as 'impedance balanced' to send the signal down a 6-8metre twin & screen cable to the interfaces.

Hope that makes more sense than my last attempt... :oops:
 
impedance balanced was the old tube way to do things so to speak. In order to have the most power out of your tube equipment you had to have the same impedance from one unit to another.. this ment having 600 OHM line and back then was run at 0DBM operating level vs moderan +4DBU/M. It's why some guitar amp manufacutres have differnt output plugs of various ohm's to match with your speaker. However you can get some groovy sounds with mismatched impedance it's responsible for the bass sound on Taxman by the beatles.
 
"impedance balanced was the old tube way to do things so to speak."

Sorry, but that's incorrect. The way it was done then was transformer-coupled, symmetrical (what some today would call "balanced/floating").

As detailed in the posts above, "impedance balancing" is a fairly recent trick that attempts to bring some of the benefit of a balanced line to an unbalanced source.

"back then was run at 0DBM operating level vs moderan +4DBU/M. "

Also incorrect. +4dBM and +8dBM were the standard operating levels in the tube era.
 

Latest posts

Back
Top