ok so if there's no signal on half a balanced line can we call it balanced?
i'm not saying that it won't have the noise cancelling properties of a balanced line, i just tend to refer to it as sending a single ended signal across it, and what polarity is used becomes dependent on what termination is needed by the down stream gear or whether or not polarity would affect adjacent frequency bands in a multiway system like a 3 or 4 way PA...and to me without that 180 deg signal the loss of gain just ain't worth it...quasi balanced (1 half grounded) which we see more of these days is a work around to gain noise immunity, at half the gain it should have.
thought exercise if you have a source device with a balanced output and you connect it to a quasi balanced input that has a signal pin grounded will everything be ok...? and for how long?
i'm not saying that it won't have the noise cancelling properties of a balanced line, i just tend to refer to it as sending a single ended signal across it, and what polarity is used becomes dependent on what termination is needed by the down stream gear or whether or not polarity would affect adjacent frequency bands in a multiway system like a 3 or 4 way PA...and to me without that 180 deg signal the loss of gain just ain't worth it...quasi balanced (1 half grounded) which we see more of these days is a work around to gain noise immunity, at half the gain it should have.
thought exercise if you have a source device with a balanced output and you connect it to a quasi balanced input that has a signal pin grounded will everything be ok...? and for how long?
Last edited:
If there is no signal on half the balanced pair it is either open circuit or one side is earthed. Either way you will not have a balanced circuit. The noise cancelling works because balanced sends antiphase signal on each leg of the pair. Any induced noise will be common mode (equal phase on both sides) hence the cancellation by the input. Better the input balance the better the rejection.
well, quasi balanced, is it BS or not?
it's that polarity thing isn't it?
does it matter? noise would still be cancelled would it not? if the input has a high CMRR but one side see's 0 V (and yup i'll bring it up...again it's not ground just the signal return) does it not reject noise regardless of the presence of signal in the opposite phase???
it's that polarity thing isn't it?
does it matter? noise would still be cancelled would it not? if the input has a high CMRR but one side see's 0 V (and yup i'll bring it up...again it's not ground just the signal return) does it not reject noise regardless of the presence of signal in the opposite phase???
There seem to a good number of knowledgeable audiophiles who find this a technically valid anomaly and audible disturbance.
https://community.roonlabs.com/t/flip-the-absolute-phase-of-a-cd-album/37743/83
https://yabb.jriver.com/interact/index.php/topic,76164.0.html
and where one poster also refers to this https://www.gcaudio.com/tips-tricks/absolute-polarity-a-prerequisite-to-optimum-performance/
At least it seems that JRiver player developers may have done a fix for it. Maybe Roon has too by now.
No, it could be terminated into a source resistance equal to the source resistance of the active side.
In that case you do indeed have equal source impedances and a balanced circuit, even though there is no active signal on one side.
This does seem to be a confounding topic for many audio folks.
I suggest to read the excellent application notes from Bill Whitlock.
Dave.
I agree that you'll have a single-ended balanced connection and CMR. Zero volts doesn't mean ground or open circuit.
ok i'll weigh out as the concept of a floating reference is as confounding as the concept of a virtual ground is to most...these are the times i understand why i love scotch and my love of jesus (forgive them for they know not what they do!)
It’s a common misconception that balanced interconnection requires anti-phase signaling. The two concepts are often conflated. The term, balanced, refers to the impedance to signal-ground of the driven signal path, and of it’s return path. So, for example, 110 Ohm balanced interconnection means that there is a 110 Ohm impedance from the signal transmission drive path to signal-ground, and also 110 Ohms impedance from the signal return path to signal-ground. While it may not be readily apparent, it’s possible and valid to have single-ended drive via a balanced interconnection, and obtain every dB of common-mode noise rejection that an anti-phase drive signal would enjoy via that same interconnection.
The entire point of balanced interconnection, is to provide suppression of common-mode noise current (hum and such) conduction between connected devices via their signal interconnection path. Which typically stems from, and completes through, those device’s A.C. power mains connections. A balanced interface, essentailly, forms a Wheatstone bridge circuit. In overly simple explanation, presenting the common-mode noise as matched in amplitude to the receiving input circuit. Thereby, enabling that input circuit to reject common-mode noise by virtue of it’s Common-Mode-Rejection-Ratio (CMRR).
While all of the Jensen Transformer’s App. Notes are informative, AN003 is particularly relevant to what you are asking. Link below.
https://www.jensen-transformers.com/wp-content/uploads/2014/08/an003.pdf
The entire point of balanced interconnection, is to provide suppression of common-mode noise current (hum and such) conduction between connected devices via their signal interconnection path. Which typically stems from, and completes through, those device’s A.C. power mains connections. A balanced interface, essentailly, forms a Wheatstone bridge circuit. In overly simple explanation, presenting the common-mode noise as matched in amplitude to the receiving input circuit. Thereby, enabling that input circuit to reject common-mode noise by virtue of it’s Common-Mode-Rejection-Ratio (CMRR).
While all of the Jensen Transformer’s App. Notes are informative, AN003 is particularly relevant to what you are asking. Link below.
https://www.jensen-transformers.com/wp-content/uploads/2014/08/an003.pdf
Basic electrical theory deems you need a closed circuit for current to flow. Either a balanced loop between source and destination or one side of the circuit connected to a common earth. Connection to a virtual reference may allow the source to see equal impedance but you will not pass a signal. Yes a balanced input will work with one side referenced to earth and not carrying a signal but it will have no common mode rejection, essentially you create an unbalanced circuit.
@totally analogue Ken Newton has provided you with an excellent explanation, and provided a terrific reference as well.
But you're not convinced??
As I said, this is a confounding topic for many audio folks. 🙂
Take care.
Dave.
But you're not convinced??
As I said, this is a confounding topic for many audio folks. 🙂
Take care.
Dave.
If an adaptive configuration (assuming you mean at the receiving end) has issues with one pin being at ground, it is not a properly designed balanced receiver. A balanced receiver worth its name only reacts to the difference between the two pins. And any of those pins can be at any level within the common mode range of the receiver including ground. In fact, it is tricky to talk about ground in a balanced system as a balanced system has no particular relation or reference to ground.
But yes, if the receiver is incompetendly designed, all bets are off.
Jan
Nope, 50+ years in the professional audio industry and I'm still not convinced. The unbalanced/balanced diagrams on pages 1&2 in the Jensen 003 paper are conventional circuits. Diagram 2.4 is unbalanced over twisted pair and will not offer any common mode rejection against noise. The remaining use transformers which is the way it's been done for years. Sorry to make a fuss.
1) Are you saying that, in figure 2.4, the impedance between Pin 1 and Pin 2 does not equal the impedance between Pin 1 and Pin 3?
2) Completely separate question. Can a fully battery powered audio system have a balanced connection between a preamp and amp (with CMR)?
There is no connection to earth ground.
2) Completely separate question. Can a fully battery powered audio system have a balanced connection between a preamp and amp (with CMR)?
There is no connection to earth ground.
@totally analogue If you disagree with Bill Whitlock that's cool. At least you've taken a look at the Jensen reference. Thank you for doing that.
Dave.
Dave.
Yes. The receiving end is nominally free of ground but always has some high impedance bias point for the inputs, if only to keep those inputs within the CM range of that input. That bias may come from ground or from the supplies with some sort of resistive divider. In practice, the floating source with the battery and no ground connection with settle itself near the ground or reference potential of the receiver.
It does, but that's not the point. The point is that the source impedances seen at pin 2 and pin 3 are equal*, which is a prerequisite for balanced lines.
So then this is a real balanced line. The fact that one of those source is connected to the sender side ground is immaterial.
Jan
*Assuming that the opamp output impedance is zero, which it will not be, especially not at high frequencies.
The impedance may well match but the signal is only feeding one input of the receiving amplifier, the second being effectively grounded giving you and unbalanced circuit.
The answer to question 2 is yes. It is the same as using send and receive transformers with no earth reference. In that case the send coil will generate a voltage which excites the receiving coil thus producing an output. If we use op amps then the sending voltage will appear across the differential inputs of the of the receiver which are internally referenced to each other and again produce an output. That is one advantage of balanced circuits, you can connect a signal between two devices without connecting the earth. The other being the cancelation of induced noise over the connecting cable.