Quote:



Really? I've read otherwise in the audio press. Also about the 75-ohm standard for S/PDIF cables. I can't judge it myself, I'm no expert. But many experts claim 50-ohm cables and RCA plugs/jacks to be problematic for digital connections. Moreover, what is the 50-ohm standard for RCA plugs worth for if it is worthless in your opinion?

In any event we're getting off topic. Actually I was just trying to counter your earlier statement...

Quote:



...a conception that doesn't apply to «20-ohm» speaker cables like the ones in the linked patent paper (read it!).
.

Intrigued and investigating speaker cable impedance further, I found this site which does some simulations:
Loudspeaker Cable Characteristic Impedance

Quote:



So you mean characteristic impedance, Leitungswellenwiderstand ? It is not the product (which means "multiplication") of capacitance and inductance. For a lossless line (and even if you have a really large mansion, your audio cables will probably fall under this), it is sqrt(L/C), where both are measured per some unit length. It is also purely resistive and independent of frequency. Again, you can convince yourself of the unimportance of the difference by using the "full" formula Z=sqrt((r+2*pi*f*j*L)/(g+2*pi*f*j*C)), where g is the conductance of the dielectric surrounding the wire. Audio cables of 8-ohm characteristic impedance are indeed easy to build - just pick one where inductance per meter is 64 times larger than capacitance

The reason there is a standard characteristic impedance for interconnects is for termination. Which most important application in case of audio is to keep shoddy amplifiers from oscillating

Quote:



The frequency at use in s/pdif cables is 100-300 times higher than that for audio. It is an entirely different world cables wise, and things from it don't really apply to audio frequencies. The impedance match is needed to eliminate reflections.

Quote:



The easier it is to build one of 20 ohm. I have no idea if a speaker connection would benefit from it, but I give the idea the benefit of the doubt. After all cables are generally said to make no sonic difference, whereas in fact, you know...
.

Quote:



Which audio press is this? I've been reading the audio press for going on 30 years now and I don't recall any mention of a 50 ohm "standard" for analogue interconnects.

Quote:



50 ohm cables and RCA plugs/jacks can be problematic if used on a 75 ohm high speed digital interface. Any impedance mismatch in such an interface can be problematic.

Quote:



RCA plugs weren't intentionally designed to have a 50 ohm characteristic impedance. They were designed because they were cheap to make. They were originally designed back in the 1930s for the purpose of allowing owners of RCA radios to plug in an RCA record player, which is why they are also referred to as "phono plugs."

They then ultimately became the ipso facto standard connector for consumer audio equipment. But it had nothing to do with their 50 ohm characteristic impedance.

Quote:



I read it.

The whole gist of it is that that cable's impedance should match the loudspeaker's impedance.

And even if you're talking 20 ohms, you're still talking about wasting about a third of the amplifier's power driving the cable.

k

Quote:



Huh? You stated yourself that the audio band is not affected by the cable impedance, so why does the cable eat up energy? Remember, it still has ~0 ohm ohmic resistance. And capacitance should only have an effect in the ultrasonic range, no?
.

Quote:



I didn't say any such thing.

What I said was that matching the cable's impedance to the loudspeaker's impedance was irrelevant at audio frequencies.

Quote:



Technically the reactive elements (i.e. inductance and capacitance) don't actually eat up the energy. Energy is only eaten up by resistance, which converts the energy into heat.

However because of the cable's impedance, it's still drawing power from the amplifier. And in the case of 20 ohms, about a third of the amplifier's power is going into the cable, and not the loudspeaker. Which means the amplifier has to provide a third more power with a 20 ohm cable in order to get the same amount of power delivered to the speaker if the cable's impedance was much higher.

Quote:



It's not the ohmic resistance we're concerned with. It's the impedance.

Quote:



Depends on how much capacitance there is. The greater the capacitance, the greater its effect on impedance at lower frequencies.

k

Quote:



How this?


Quote:



So the power is converted into heat as well, or what? I can't imagine anything else than capacitance acting as (sort of) short-circuit between the cable poles, but that would require heavy capacitance with effect deep down into the audible range. And this scenario makes it impossible to consider a fixed percentage (which you provide with your «third»), since the actual capacitance of a 20-ohm cable is dependent on the counterbalancing inductance. When I think about it: According to your theory every speaker cable draws considerable power from the amplifier because of its impedance, even normal speaker cable (let's say of 60 ohm impedance), just not as much as one with 20 or 8 ohm. Or did I get it wrong?
.

Quote:



Simple. Ohm's Law.

Quote:



Eventually, yes.

Quote:



Yes.

Quote:



The percentage is for wherever the cable impedance is 20 ohms and the load is 8 ohms.

Quote:



The lower the impedance of the cable, the more power is drawn from the amplifier by the cable and vice versa.

k

Quote:



You should also consider the "monster cables vs wire coathanger" test. Or what the fact that which one was the coathanger wire was not apparent 100% of the time implies about any short interconnects.

Quote:



You are confusing impedance and characteristic impedance. You need a rather long run of cable for its impedance to be 20 ohms. Even for a cable with characteristic impedance of 75 Ohms. Cables you usually meet in you home all measure (essentially) 0 Ohms and as a result consume essentially 0% of the output power of the amplifier.

Quote:



No, I'm not. I'm simply addressing the idiocy of Goertz's assertion that the cable's characteristic impedance should be close to that of the loudspeaker's impedance.

First, the cable's characteristic impedance doesn't revert to the square root of L over C until well above the audio range, at which point the speaker's nominal impedance is completely irrelevant.

Second, if the cable's impedance in the audio band where the speaker's nominal impedance would be relevant, you'd be wasting a lot of power just driving the cable, not the speaker.

k

Yeah, I must confess that I still doubt this scenario.


Quote:



Well, this test wasn't done with my ears, but somebody else's, and possibly it was a DBT. So it has no validity to me.
.

Quote:



So... if it doesn't match your preconception about things you will simply not accept it as evidence?