Impedance

Posted: Mon May 26, 2008 1:18 pm

OK, I've spent a bit of time searching here and on wiki, etc. but I can't find anything to sum up the whole concept of impedance in audio gear.

I've come to a conclusion that the impedance of the output must be higher or equal to that of the input it's being wired to, or you get buzzing and noise etc.
eg My headphone jack is 300 ohm and I have no problems with 64 ohm HPs but putting this into a Line In at say, 10Kohm causes noise.

BUT I have no faith in that statement unless someone confirms or denies this.

Basically, what's the rule of impedance? Input = output, input < output, input > output or what?

Posted: Mon May 26, 2008 1:48 pm

You seem to be suffering from trying to convert Audio Engineering into Electrical Engineering.
Basically I don't get it either.

Posted: Mon May 26, 2008 3:03 pm

Impedance is a combination of resistance and reactance. On an X/Y graph, resistance is the X axis and reactance is the Y axis and 90 degrees out of phase with resistance. When you do the math for any given audio circuit, you get an impedance value on the X/Y graph, and it has a phase angle. Whether you want impedances between input and output to match or not depends on which end of the signal chain you're on. At the microphone end, you want the input of the preamp to have an impedance at least 10 times higher than the output impedance of the mic. This sets up a substantial gain stage. Resistance and voltage are kissin' cousins, so the voltage increase from the higher impedance of the preamp translates into higher voltage for a gain increase. At the monitor end, you want the amplifier impedance and monitor impedance to match so that you get maximum current and power transfer. Power (watts) and current (amperage) are also kissin' cousins, so by keeping the imepdances matched on this end, you get maximum power transfer from amp to speakers. Sidenote: reactive components (capacitors and coils) in circuits are frequency sensitive, this is why impedance causes a phase shift, and where most gear related translation issues arise from. If the net phase angle of a signal chain from one studio is drastically different than the net phase angle of another studio's chain, stuff will sound drastically different out of each set of monitors. Gear manufacturors have never gotten together to standardize this, unfortunately.

Posted: Mon May 26, 2008 4:46 pm

so now that, that is clear
we are all happy now

Smile

to begin to understand
it might be better to take one specific situation at a time
and discuss some of the concepts
with cause and effect scenarios

" I've come to a conclusion that the impedance of the output must be higher or equal to that of the input it's being wired to, "
completely wrong

although I get where Jon is coming from,
it might be easier for audio people to view impedance as resistance again frequency

in the world of electronics many consider audio to be near DC
this is incorrect but for people that work in Radio Frequencies in either analog or digital
you can see why they think this way

impedance(resistance) can change with frequency

SO
lets take one situation and look deeper

Codemonkey
we can talk about the headphones
but first
get a look at the RANE notes on driving headphones

Posted: Mon May 26, 2008 6:03 pm

"Resistance and voltage are kissin' cousins"

Is it just me or is it gettin' warm in here Smile

Posted: Mon May 26, 2008 7:30 pm

not without current nearby
so that would make it a threesome

Posted: Mon May 26, 2008 7:35 pm

Something between my ears is swelling badly. Fireworks will follow shortly.

I'll have a look at the rane thing.

I was hoping for a general guideline of say, plugging line outs into line ins on different devices although i suspect the whole -10dbV and +4dBu comes into it here?

Posted: Mon May 26, 2008 7:51 pm

yes it does

and there are times when the dbv makes sense
and times when dBu makes more sense

it's all about available current/power and the load presented

these specifications we like so much don't tell the whole story in isolation
and it requires many specs to begin to tell the story

ALL the spec in the world don't tell how it will sound
that is NOT to say that specs are useless

it's complicated BUT it can be simplified
and
always beware that a little knowledge is dangerous

knowledge ... persure you must

Posted: Mon May 26, 2008 9:27 pm

Quote:

Is it just me or is it gettin' warm in here?

Must be you, the thunderstorms just cooled it down up here in Nashville!

Codemonkey, didn't mean to make your head explode. Impedance is a very abstract concept, I'm still working on wrapping my noggin around it completely. My source on theory is Al Grundy, who was AES President for many years, and made my head spin on a regular basis (had him as a professor).

Posted: Tue May 27, 2008 3:38 am

Codemonkey wrote:

OK, I've spent a bit of time searching here and on wiki, etc. but I can't find anything to sum up the whole concept of impedance in audio gear.

I've come to a conclusion that the impedance of the output must be higher or equal to that of the input it's being wired to, or you get buzzing and noise etc.
eg My headphone jack is 300 ohm and I have no problems with 64 ohm HPs but putting this into a Line In at say, 10Kohm causes noise.

BUT I have no faith in that statement unless someone confirms or denies this.

Basically, what's the rule of impedance? Input = output, input < output, input > output or what?

There's some misunderstanding here.

Firstly, I think your headphone/line in noise difference has nothing to do with impedance. Headphones are inherently isolated (not grounded) whereas single-ended line inputs are referenced to ground. I suggest the noise is due to a ground problem.

Secondly, the matter of input and output impedance. Generally speaking, for signal transfer, the output impedance should be lower than the driven input impedance. There are several common situations:

(1) bridging: the output impedance is low and the input impedance high. This causes the lowest attenuation of the signal and also allows many inputs to be connected to one output without interaction or signal degradation.

(2) matched: the output (driving) and input (receiving) impedances should be equal. Maximum signal power is transferred in this case, half being lost in the driver and half in the receiver. An example of this case is 600 Ohm working. The signal amplitude is exactly halved at each interconnection, and the open-circuit gain of the equipment has an extra x2 to allow for this. It is important that the equipment is designed for this use, as putting a 600 Ohm load on outputs that expect higher impedance loads can cause clipping and distortion due to current limiting.

(3) power transfer: this is the case when for example you are driving loudspeakers from a power amplifier. The amplifier has a very low output impedance (often lower than the cables in use), and for current-limiting reasons is specified to drive loads of not lower than a certain impedance. You want as much power as possible to be transferred to the load.

In the case of your headphone jack, it would be unusual for it to have a 300 Ohm output impedance. It's more likely that it is specified to drive 300 Ohm phones, but that in itself would be unusual, as the majority of phones are nearer 30 Ohm.

Posted: Tue May 27, 2008 11:11 am

Ok. Im going to start freaking out!

Why can't we just keep this simple ????

Mic's need to go into 10x heir output impidence. Amps need o drive a speaker the same as or less than their output impedince etc....

Every i have tried to understand it any better I Confused

You see, I don't question it. For the same reason I dont question why 'But' is but, and 'Put' is put... gettting it ?

English and electronics are both funny languages....

I eed to have another beer....

Laughing

Posted: Tue May 27, 2008 3:27 pm

So basically my noise problem is not resistance, but ground issues, and plugging into onboard crap? I kinda let that slip past because I've had good results (relatively speaking) in the past with said line input.

Something tells me I should just plug it in and turn it up, and forget about the fine electrical details?

Posted: Tue May 27, 2008 4:57 pm

this is why I didn't want to give an explanation like Boswell did
take a specific situation and we can discuss the cause and effects

Boswell's power transfer description doesn't seem right to me but does make sense in the context that it followed the description for matched.

Matched is a bit historic and probably doesn't help younger people with new bridging interfaces who may never own a transformer coupled tubed circuit
It IS important to know this stuff if you get into the retro gear.

most of the interconnects need some degree of power transfer ... some more then others and power amplifier and speaker combination has much in common with a bridging transfer where
" the output impedance is low and the input impedance high "

Codemonkey ... read the RANE notes on driving headphones

Posted: Tue May 27, 2008 5:04 pm

Space wrote:

"Resistance and voltage are kissin' cousins"

Is it just me or is it gettin' warm in here Smile

Wait... so making out with your cousin is bad?

Posted: Tue May 27, 2008 5:28 pm

[unfounded joke]
Only in some states.
[/unfounded joke]

Remember, it ain't illegal unless you get caught!

This topic sponsored by: