Transmission line behavior

Yes. Long cables are dominated by their resistance. At high frequencies, the active length is short enough that ohmic losses are small, and you can "see" the characteristic Z=sqrt(L/C) impedance of the cable.

That's like telephone twisted pairs: the characteristic impedance for a twisted pair, at high frequencies, or for short runs, is around 100 ohms, but the phone company treats them as 600 ohm cables, because they use long runs and low frequencies.

John

Phantom wrote in news: snipped-for-privacy@4ax.com:

Hmm....the average human can hear up to about 12 Khz, tops, and his audiometer chart beyond 10 years old looks like the Swiss alps in freq response. By the time he's 30, the DJ's pounding him with kilowatts has his freq rolloff down to 8 or 9 Khz.

Looking at the chart on the webpage, matching impedance of the line to the load, assuming the speakers HAVE an impedance, which is a joke across the spectrum of the chart, speaker impedance being measured at 1Khz, not across the bandwidth so wide, all this webpage is audiofool bullshit!

=======================================================

An audiofool moved in next door for a while. When my 1500 watt ham radio station hit the unshielded welding cables he was using for speaker cables, his 2KW audio amp lifted right off the floor from the RF feeding back to his input feedback amp.... I could even hear the horrible loud noise it made way over here! Until he figured it out, every time I got fed up listening to the rapcrap playing thru the walls into my living room, I'd go send a few packets on 20 meters, blast his living room with an amazingly loud buzzing noise, then all would be quiet for the rest of the night as he rolled around in pain on his living room floor.

After he figured it was me screwing up his stereo, he started pounding on my door and making threats. I, very calmly, offered to replace his triple-0 welding cables with some foil shielded #18 Belden mike cable to give it a Faraday shield from my LICENSED RF. He blew a fuse. The audio from #18 mike wire would only send audio below 800 Hz to his speakers, and other audiofool bullshit the cable hawkers feed them to sell the most expensive lampcord wire on the planet...professional speaker cable.

After a couple more days of my RF noise blowing his cones out, he was more reasonable and finally agreed to let me "give it a try" to see how awful it sounded. I brought my General Radio audio level meter when I swapped out the battery cables for shielded mic cable, leaving the foil shielding open at the speaker end and grounded to the stereo's chassis screw on that end. RF problems solved. The measurements with the audio level meter proved his audiofool brainwashing wrong as there was absolutely no difference, of course, in the audio bandwidth coming out of his speakers up to his highest test tone beyond what anyone could hear from his expensive speakers. He became a good neighbor soon after I solved this problem. I fixed a couple of things for him over the year that he lived there before buying his wife a nicer home.

Cables hooking up the speakers that don't melt at the loudest volume the neighbors will tolerate are just fine. The rest, like this webpage, is bullshit to sell $50 lamp cord with golden-colored ends. The cheapest zipcord (lampcord) you can find at Ace Hardware has the same audio produced as the most expensive "speaker cables" on the planet. The whole system sucks because of YOUR EARS and their horrible freq response. That's why you buy a nice multiband equalizer to compensate for your ears' awful frequency response curve. Speaker wires have nothing to do with anything unless the speaker current melts the insulation plastic.

There are 3 "bullet" points just before his figure 16. Just before that he says "Using a 50 ohm transmission line (purely for convenience), the following points are of interest .", so his example is using a 50 ohm line.

He then shows the setup in figure 16; a generator of 50 ohm impedance (putting out 1 volt into an open circuit), a 100 meter long cable with 1 ohm/meter DC resistance, and a 50 ohm load.

At low frequencies (his simulation goes down to 1 Hz) this is essentially solvable as a DC circuit: 1 volt driving a 50 ohm resistor, a 100 ohm resistor (the cable) and a 50 ohm load resistor. This is a series circuit of 200 ohms driven by 1 volt, so we would expect 5 mA in the 50 ohm load, which is what he shows at low frequencies.

But, at frequencies above 1 MHz his simulation shows 10 mA in the load. This is what we would expect if the cable were replaced by a short wire of zero resistance; in other words, a 100 ohm load on a 1 volt source. This is also what we would expect if the transmission line had zero attenuation at frequencies above 1 MHz.

It doesn't seem reasonable that a 100 meter transmission line with a resistance of 1 ohm/meter would have zero attenuation at any frequency (especially high frequencies) in a 50 ohm environment. This should be measurable. If what he says is true then wouldn't we expect that a length of lossy 50 ohm cable should show a rather sudden decrease in attenuation as we go above a certain frequency?

"Fred"

** Bollocks.

The average teenager can hear sound up to about 22kHz.

The average 60 year old can hear up to 14 or 15 Khz.

I am speaking of sine wave sound in isolation.

.... Phil

stance

igh

what he

should

quency?

Agreed, that does NOT seem reasonable.

I suspect the simulation is flawed due to confusion between the transient and steady state.

Because of the propagation delay, the initial transient is not the same as the steady state. The steady state occurs only after the signal reaches the end and is reflected back.

Think about how it would look on a TDR.

Initially the cable looks like it's Zo =3D 50 Ohms. I'm thinking that due to the distributed resistance of 1 Ohm per meter, there is a small distributed continuous mismatch that causes a small but continuous reflection, that over time creates a reflection that at the input, makes the line look like 100 Ohms.

I don't have the capability to simulate a lossy line right now. Can someone else do this?

Mark

sistance

high

If what he

e should

requency?

I found this applet

select lossy line lower left

you can play with this and see that the 1 ohm per meter actually changes the Zo of the line as a function of frequency for a fixed L and C per meter. A line with 1 ohm per meter resistance and an L and C the yields Zo=3D50 Ohms at 10 MHz actually has a very high Zo in the

1000's at audio frequencies. This is only due to the resistance. If you select R and G to =3D0 then of course the Zo is not a function of frequency. But with the lossy case the Zo changes as a function of freq.

Interesting

Mark

I wish I'd known about termination and its interference reducing effect

40 years ago. I lived 3km from a powerful Band 1 television transmitter. I could never get it out of the 'HiFi' system.

Sylvia.

The example on the web site you sited is bogus and would never exist in the real world. Speaker wire is not 50 ohms impedance, it would be somewhat higher (closer to 300 like twin lead) and that only applies in the Megahertz range, not audio, so it is irreverent. 1 ohm per meter speaker wire would be the worst speaker wire to use for a home or professional sound system, no one would use it. I also doubt that someone would use 100 meter long speaker wire, if you did, it would have to be thick (low resistance) wire so that you wouldn't loose your sound in the wire, maybe 14 gauge or 12 gauge. The frequency response shown in figure 17 shows that the characteristics of the cable would not change until above the audio range anyway, so again it is irreverent.

Don't worry about it dude, that article doesn't apply to real world situations.

Shaun

yes.... the point the article is trying to make is that speaker cables DO NOT make an audible difference. In that sense, it is correct.

some of the analysis along the way are interesting and surprising however...

Mark

I agree. Most audio applications are completely irreverent.

Bob

--
== All google group posts are automatically deleted due to spam ==

Wrong and wrong. Impedance is proportional to the distance between the conductors, and inversely proportional to the diameter of the conductors. Typical speaker wire would have a quite low impedance indeed. Admittedly, at audio frequencies, this impedance can be ignored, but the impedance is what it is, even at DC. When it's a consideration is when you start reaching 1/4 wavelength long.

Hope This Helps! Rich

Talking about that..

About 5 years ago, I was asked to look at the sound system in our company's recreation area. They have out door speakers and some they bring out when we have out door company events for the employees.

Any way, the problem was they had mostly bass and very little highs. The amp would over load and fault the power supply if they pushed it to hard, especially when their was lots of higher frequencies in the material.

So, I did the usual thing for testing shorts, opens, load impedance's etc., and found nothing wrong.. I Then started to take the system apart and found the problem... Because we are a wire & cable with a few other special things, facility, when they redid the grounds with new wire for the lighting and all, they also laid in the ground, power cable to be used for the speaker rungs. The conductors of this cable how ever, had a polyrad insulation and the components were tightly bundled which made for a very high capacitance between the pairs. At over 200 FT per circuit, there was a lot of cap their... WHich explained why there was no problem generating loads of Bass..

Jamie

Bullshit. Unless the speaker system impedance was so high that the voltage was hundreds or thousands of volts.

--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

OK, OK... My spell checker changed irrelevant to irreverent without me noticing what it was changing and I misspelled it too. Don't hang me for it. ;-))

Shaun

"Rich Grise is Always Wrong'"

** Generally, speaker twin leads (called figure 8 in some parts) have a characteristic impedance ( Zo) of about 70 to 120 ohms. However, you CAN get special cables for use with hi-fi speakers that have Zo values of 8 or even 4 ohms.

One way to make such a special cable is by using many strands of insulated wire for each conductor and weaving the pairs all along the length. A second way is using two flat strip conductors and having the space between filled with a very thin layer of dielectric - the latter type is rather expensive but goes easily underneath floor coverings.

** The issue with speaker cables that can *sometimes matter * is the series inductance of the whole run. What you are dealing with is a short transmission line ( way less than a 1/4 wave at most ) with a grossly low impedance load at the end - hence the cable's inductance per metre dominates the performance at high frequencies.

At the top of the audio band, this inductive ( series) impedance is way more than the cable's DC resistance over a typical run. Using large cross sections of copper conductors is no help.

With unusually long cable runs OR a speaker which has a low impedance (2 ohms or less ) at the top end of the audio band - the effect on response can be very significant.

** Only stupid radio hams pull that old war horse out every time cable performance comes up. Cos none of them understand the behaviour of a short transmission line OR the issues involved in using transmission lines (= any kind of signal cable really ) for purposes radio communication. ** You pissing off would help a lot.

..... Phil

Wrong!!. There is no Characteristic Impedance at DC. For audio at 20Khz (the highest audible frequency) the 1/4 wavelength is 3.75 kilometres. Nobody is EVER going to run a speaker line that long!! So it is irrelevant to home and professional sound system.

Shaun

"Shaun"

** Totally wrong.

See my other post a shot while ago.

You a radio ham by any chance ???

... Phil

Then phone lines have never carried audio? Or leased phone lines that carried radio network prgramming weren't audio?

--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

From: "Phil Allison" Subject: Date: January 29, 2011 7:39 PM

"Shaun"

** Totally wrong.

See my other post a shot while ago.

You a radio ham by any chance ???

... Phil

Yah I've read it now. Inductance huh; would the capacitance to two parallel conductors(the speaker wire) add to this effect too?

I have an advanced radio license but I don't use it. The only ham radio related things I do is Tesla Coils.

Shaun