As I understand the definition of 'balun', yes. Some do, some don't and if they don't, it's likely to hum if the signal travels long distance, so it's easier to just use something with a transformer. I had looked at the link- they come in many forms- that's much smaller than the ones I have used, but as long as it works.....

The description includes 'fully shielded, Balanced-Line for best hum rejection.'.

But using a high Z output does work, as long as grounding isn't an issue, which it was when I used Cat5e for the audio from my main system to the garage system. The electrician (turd) didn't wire it to code, so the sub-panel in the garage wasn't bonded to the main panel and I had hum. From the main stereo to the basement system, it was fine as long as I disconnected the garage or used an isolation transformer.

At some point the cable capacitance will roll off the highs, regardless of grounding. High Z suffers more than low Z in this respect.

Ernie and others

The term balanced relates to the the "impedance" that each wire presents. Weather it's referenced to ground or not is of no concern. A balanced line will present equal impedance on each wire. Unbalanced lines present unequal impedance; one of the wires is at ground typically.

Your telco copper line are balanced with a typical impedance of 600 Ohms, but not to ground. If either telco line touches ground you would get a bad hum in your telco set. Ethernet is also the same.

For an expanded discussion:
[Link: en.wikipedia.org]

I know- that's the reason for using some kind of balun or line driver.

If there is a ground potential difference between different "ground" points, a lower resistance/impedance along the length of the ground conductor will result in lower injection of noise. I've seen a demonstration (Middle Atlantic seminar with Bob Whitlock) of cheap-ass audio cables with very low shield resistance being quiet, while a super expensive audio cable with 5 ohms resistance long the ground conductor just hummed along!


You also have an unsafe electrical system. "How did he die?" "He made the mistake of touching the ground of something plugged into Panel B while touching the ground of Panel A."

I'd think your threat to report him to the state contractor's licensing agency would get him to take care of that. And you should probably report him whether he fixes it or not.


That definitely shows there's a potential difference ( = danger) between the two grounds.

This is not correct. Any statement of impedance must include impedance TO _______ or FROM _______ TO _______.

The beginning of the article that buzz refers us to:


I added the highlights that name the TO and the TO FROM. g007 does not include the detail that it's impedance along a line and specifically states that impedance to ground is not relevant.

At this level of physics, a proper definition survives an extreme example that conforms to the definition. What happens if we have a stupidly large capacitance from each conductor to ground? We'll say the source and input impedances are 600 ohms. Let's say capacitance from each conductor to ground is 1000 microfarads, which has an impedance of about 8 ohms at 20 Hz. This would drop audio at 20 Hz by more than 30dB. Impedance to ground MUST be taken into account.

And:

Unbalanced lines present unequal impedance to ground since one of the wires is typically at ground. Unbalanced lines may or may not have equal impedances along their lengths. For instance, RG6 often has a copper center and almost always aluminum braid and foil. These are almost guaranteed to have different impedances along their length. And copper doesn't have the same  impedance at different frequencies as copper clad steel. But unequal impedance is not a requirement for unbalanced lines, so it can't be part of the definition of unbalanced lines.

That camera that I wired with a pair from a CAT5E cable had two conductors of equal resistance and impedance along their length, but it was unbalanced. There's nothing about an unbalanced line that requires the two conductors to be of unequal impedance. They MAY BE, but that's not part of their definition.

Ernie

The very definition of unbalanced means unequal impedance, "not balanced".

How you hookup the wire does not soly determine if it's balanced or not. It's the presentation of the impedances in the circuit that's sets the outcome and therefore the definition, balanced or unbalanced. I can use twisted pair cable either way, balanced or unbalanced.

When you say unbalanced wires present unequal impedances, you miss stating from where to where. A pair from a CAT5 can be an unbalanced pair while the wires have equal impedances along their lengths. The impedance to ground will be unbalanced. What the impedance is referenced to is critical to whether it's balanced or not.

Your definition is similar to stating that decibels are ratings of loudness or ratings of power. They're not. Decibels are an indication of ratios of loudness or power referred to some specific value. If you omit the reference, the value is meaningless. Or, to put it another way, if you leave out the reference, it's not possible for a person to create the same circumstance.

Ernie

It is the circuit as a whole we are talking about, not the wire.

You can use dB in relative or absolute mode. If you are specifying power out you must state the impedance, like 1mW across 600 Ohms = .7745 volts. If you are specifying the voltage gain of some amplifying device then relative dB is implied. This amp has a voltage gain of 20dB, here we are stating a ratio of out/in.

You're right, but. Sorry.

But when you talk about the gain of an amp, then we understand that it's output versus input; those are the TO and FROM points. When you define decibels are, it's a poor definition that omits the points of reference. I mean, what's 87 dB? Have you seen people talk of S/N of -20 dB instead of S/N of 20 dB, because they're confused about the reference points and how to express the direction in which they're measuring?

This is about definitions. Conversations among people who understand all the concepts, where no learning is going on, usually involve knowing the reference points, so they don't have to be expressed. That's not true with definitions.

You erred in specifically stating that impedance to ground does not matter in a balanced situation. If you have a balanced 600 ohm line where each conductor measures 8 ohms impedance to ground, then you have a crap cable, although balanced.
(Preferably you'd want at least 10 times the characteristic impedance of the system at the top end of the audio spectrum from each conductor to ground, with infinite impedance at 0 Hz.)

Ernie

87dB IS ratio of 22,387.211 to 1 in terms of voltage or current, or the inverse
1/22,387.211

Ernie

As to the S/N comment the minus sign is a measure of how far below max output the noise sits, a positive number would indicate dynamic range. It could also indicate the gain setting of a particular amp section, + for gain - for loss.

Sure, that's the ratio, but what IS it?
Is it the efficiency of a speaker?
Signal to noise? re what level?
dBm?
dBmV?
dBu?
The ratio is the same, but that doesn't identify anything. This whole thing between us started when you gave a definition and purposefully left out references that would make the definition meaningful and correct. You didn't identify, that is.

Also, it's generally assumed that a number without a + or - in front of it is a positive number, so the ratio is indeed 22,387.211 to 1 in terms of voltage or current. But it's not 1/22,387.211 in the same terms. That would be either -87dB or "a loss or reduction of 87 dB" or some other label that tells us we're talking about a reduction in voltage or current.