Monster Cables on Engadget

[MrQhuest]

It seems that Monster Cables have been put to task on Engadget...

http://www.engadget.com/2008/03/03/audiophiles-cant-tell-the-difference-between-monster-cable-and/

... against coat hangers!

MrQhuest

[XBoxWarrior]

Not a single one was then able to tell the difference between the Monster Cable and the hangers, and all agreed that the hangers sounded excellent.

Well, that settles it!

I am gonna use coat hangers for my stereo from now on!

By the way, where do you get metal coat hangers from in 2008?

[shadron]

Well, that settles it!

I am gonna use coat hangers for my stereo from now on!

By the way, where do you get metal coat hangers from in 2008?

An antique store, but it might be cheaper to go with Monster.
I got a big ol' roll of barbed wyre, just split it out, clip off them barbs and Listen Up.

[Grimoire]

Awesome. I always tell people to avoid Monster Cables like the plague. I knew they were a bunch of scammers when I noticed that the "See the Monster difference" setup had composite cables for the non-Monster TV and component cables for the Monster. That's like Shell saying "Shell gas is better than all the others. Look how much faster this red car (Ferrari) using Shell gas is than this other red car (Fiero) using non-Shell gas!"

[Mongrel]

By the way, where do you get metal coat hangers from in 2008?

In the UK getting your shirts dry cleaned will net you some wire coat hangers also a lot of charity shops seem to have a box of coat hangers for sale, a quick rummage will turn up loads of metal ones for a pittance.

[Round Robin]

Cheapo lamp cord from Radio Shack has always worked well for me. Thanks for the link!

[fuelair]

Beeeeeautifull!!!!!!!!!

[kellyb]

Well, I really would expect Monster (or it's much cheaper equivalents) to be better than a coat hanger. Copper really should sound better than...whatever metal it is that coat hangers are made from.
I wonder how long the coat hanger piece was?
If it was only an inch or two long, that might not be a difference that could be heard.
Or maybe audiophiles just don't know how to listen.

[noblecaboose]

Ooh! Just the thing to for my CREATION!!
I've got the bolts in his neck, the giant heavy boots on his feet and the staples along the top of his head. All I've needed is some really thick cables specifically designed for monster creation and -- what? That's not what these are for? What's all this about stereo equipment?
Talk about false advertising!

[sol invictus]

Well, I really would expect Monster (or it's much cheaper equivalents) to be better than a coat hanger. Copper really should sound better than...whatever metal it is that coat hangers are made from.
I wonder how long the coat hanger piece was?
If it was only an inch or two long, that might not be a difference that could be heard.
Or maybe audiophiles just don't know how to listen.

It's not like you hear the metal....

Speaker cables can be characterized by resistance, impedance, and capacitance - and none of those matter much at all as long as they are reasonably small. Many listening tests like this have been done in the past, and the only positive result I've ever heard of was something ridiculous (like 50 feet of 24 gauge lamp cord versus a five feet of fancy cable - and that test wasn't even volume leveled, making it pretty meaningless).

Expensive audio cables are a total ripoff.

[Paulhoff]

Well, I really would expect Monster (or it's much cheaper equivalents) to be better than a coat hanger. Copper really should sound better than...whatever metal it is that coat hangers are made from.
I wonder how long the coat hanger piece was?
If it was only an inch or two long, that might not be a difference that could be heard.
Or maybe audiophiles just don't know how to listen.
Oh please, metal of any kind does not change the signal, they have different resistances, but metal is linear, it neither adds and/or subtracts harmonics.

Paul

:) :) :)

[Gord_in_Toronto]

No report of any skeptics present during the test?

So the "Randi Effect" did not apply? :D

[kellyb]

Oh please, metal of any kind does not change the signal, they have different resistances, but metal is linear, it neither adds and/or subtracts harmonics.

Paul

:) :) :)
Resistance effects bandwidth, so there would be signal loss on the highest and lowest of the audible range.

[Paulhoff]

Resistance effects bandwidth, so there would be signal loss on the highest and lowest of the audible range.
No it does not, resistance effects all frequencies the same.

Paul

:) :) :)

[sol invictus]

Resistance effects bandwidth, so there would be signal loss on the highest and lowest of the audible range.

No it does not, resistance effects all frequencies the same.


You're both wrong... even if the cable is purely resistive it still acts as a voltage divider, and since speakers are not purely resistive loads it will attenuate different frequencies differently. In general the effect will not be a loss at the top and the bottom - it will be considerably more complicated.

However those effects are too small to be audible, even for a coat hanger.

[kellyb]

No it does not, resistance effects all frequencies the same.

Paul

:) :) :)

If you use an instrument cable instead of a speaker cable and apply considerable power, the resistance issue is quite substantial and manifests as loss in the upper and lower ends.

When you're in the line level voltage stage or preamp stage it won't matter, but once you're in the post amp AC voltage range, bandwidth will be effected by resistance when you're pushing it into a speaker.

Regarding using a coat hanger instead of a speaker cable, it's just a matter of how long the coat hanger is.

http://fohonline.com/index.php?option=com_content&task=view&id=1238&Itemid=69

[kellyb]

However those effects are too small to be audible, even for a coat hanger.

Would you agree that there would be an audible difference if the cables/wires were 100 feet long (to use an extreme example of what I'm thinking)?

[sol invictus]

Would you agree that there would be an audible difference if the cables/wires were 100 feet long (to use an extreme example of what I'm thinking)?

With or without volume leveling?

I have never seen or read of a volume-leveled blind test in which listeners successfully identified differences in cables. I have read one non-volume leveled test where they did, and IIRC that was 50 feet of 20 or 24 gauge lamp cord versus some short length of decent wire. The audibility in that case was almost certainly due to the overall signal attenuation - had they leveled the volumes I'll bet no one would be able to tell.

So, the answer is - it depends. But yes, at some extreme the differences will be audible.

[kellyb]

With or without volume leveling?

Are you talking about power application for the 3 db rule (3db loss per octive thing)?

[sol invictus]

Are you talking about power application for the 3 db rule (3db loss per octive thing)?

I'm not sure what you're trying to ask.

The primary effect of a resistive cable will be to lower the overall signal amplitude, and therefore reduce the SPL produced by the speaker. If the reduction is 1db or more it will be audible. However (since most audio systems are equipped with a volume knob) that's not very interesting - so to look for real effects, you want to control for volume first and THEN see if the differences are audible.

[Paulhoff]

If you use an instrument cable instead of a speaker cable and apply considerable power, the resistance issue is quite substantial and manifests as loss in the upper and lower ends.

When you're in the line level voltage stage or preamp stage it won't matter, but once you're in the post amp AC voltage range, bandwidth will be effected by resistance when you're pushing it into a speaker.

Regarding using a coat hanger instead of a speaker cable, it's just a matter of how long the coat hanger is.
You both miss the point, speaker wire people talk of the distortion of copper vs. silver vs. whatever metal they choose as if metal is nonlinear. You are talking about resistance and amps under very abnormal conditions. Good gage wire is all that is required besides clean contacts.

Paul

:) :) :)

[EvilBiker]

Meh, resistivity of copper is 15.80 nano-Ohm-meters, iron is 87.10 nano-Ohm-meters.

Resistivity = (Resistance*Cross-sectional area)/length = R.A/l

Considering a diameter of 3mm for both a copper cable and iron coathanger, we get A = 0.00942 m^2

So, for a 10m length of copper cable we have a resistance of 16 uOhms, and for the same length of iron, we have 92 uOhms.

(Please excuse the lack of Latex use, I'm in a hurry :-P)

Most commercial speakers have a coil resistance of either 4 or 8 Ohms - the contribution of the cable resistance is going to have almost zero effect.

If you want to talk about bandwidth limitations, forget it with these resistances. You would need a HUGE capacitance across the cables to get any effect below a few Gigahertz.

[kellyb]

I'm not sure what you're trying to ask.

The primary effect of a resistive cable will be to lower the overall signal amplitude, and therefore reduce the SPL produced by the speaker. If the reduction is 1db or more it will be audible. However (since most audio systems are equipped with a volume knob) that's not very interesting - so to look for real effects, you want to control for volume first and THEN see if the differences are audible.

The more I think about it, the more it seems that there really shouldn't be any difference between lampwire and speaker cable of equal gauge audio-wise, even with a BIG system pushed hard and really long cabels. Can you find that blinded test you were talking about?

It's the coat hanger I think would end up sucking pretty quick.

[nitpicky sidenote]
I don't think anyone could hear 1 db difference unless (maybe) you were placed in some high-tech silence chamber thingamajiggie that doesn't exist for the test.
[/nitpicky sidenote]

You both miss the point, speaker wire people talk of the distortion of copper vs. silver vs. whatever metal they choose as if metal is nonlinear. You are talking about resistance and amps under very abnormal conditions. Good gage wire is all that is required besides clean contacts.

Paul

Yes, those are abnormal conditions, but if you push a coat hanger (steel?) to the extremes you'd push a copper cable to (lampwire, speaker cable, etc) it will cease to perform at some point, and the determining factors are primarily wattage and length of the cable/wire. If the coat hanger is short and/or the power is small, there won't be an issue. Keep pushing it, and you'll eventually run into problems.

A piece of copper wire is going to be much more conductive than a piece of steel at the same gage because steel is part carbon which is non-conductive.
All metals are not created equal when it comes to conductivity, resistance, etc. At short distances it doesn't matter, but eventually it does.

[kellyb]

If you want to talk about bandwidth limitations, forget it with these resistances. You would need a HUGE capacitance across the cables to get any effect below a few Gigahertz.

What do you mean by huge capacitance? The limitations of the gauge of the copper cable? As in, it's time to step up from 12 gauge copper cable to 11 or 10?

(my perspective is commercial live audio, and Monster cables are mostly marketed to mainly consumer, home system folks...but they also market to small professional-rig owners. Monster is a scam, but I do think it's better than coat hangers when pushed to it's limitations.)

[Ivor the Engineer]

Lots on speaker cable and other audio woo topics here:

http://www.dself.dsl.pipex.com/ampins/pseudo/subjectv.htm

Why not use coat hanger wire for audio connections? It's too rigid.

The variation of speaker coil resistance has far more effect than any cable parameters. If the speaker cable is being pushed to the limit, the speaker coil is going to be fried long before the cable.

[EvilBiker]

What do you mean by huge capacitance? The limitations of the gauge of the copper cable? As in, it's time to step up from 12 gauge copper cable to 11 or 10?

(my perspective is commercial live audio, and Monster cables are mostly marketed to mainly consumer, home system folks...but they also market to small professional-rig owners. Monster is a scam, but I do think it's better than coat hangers when pushed to it's limitations.)

Well, if you are talking about any filtering effects, you would consider capacitance across the cables, which would then form a rudimentary lowpass filter (RC).

In fact, I would surmise that when considering any advantages of Monster cable over Coathanger Rigid Speaker Cable in terms of capacitance, the coathangers would be a better bet :D

As my engineering colleague Ivor points out, the only reason you would not use coathangers is because of the rigidity...

[Paulhoff]

Awesome. I always tell people to avoid Monster Cables like the plague. I knew they were a bunch of scammers when I noticed that the "See the Monster difference" setup had composite cables for the non-Monster TV and component cables for the Monster. That's like Shell saying "Shell gas is better than all the others. Look how much faster this red car (Ferrari) using Shell gas is than this other red car (Fiero) using non-Shell gas!"
BOSE speakers are another plague, their motto should be "Progress thru Advertisement".

Paul

:) :) :)

[sol invictus]

The more I think about it, the more it seems that there really shouldn't be any difference between lampwire and speaker cable of equal gauge audio-wise, even with a BIG system pushed hard and really long cabels. Can you find that blinded test you were talking about?

It was published in an audio mag (I forget which) maybe in the early 90s. You can find a scan of the article online if you look hard enough.

As for lampwire versus speaker wire (of equal gauge), I agree - wire is wire. The only difference is the price. That's not what this test was comparing, though.


It's the coat hanger I think would end up sucking pretty quick.

I don't see why.

[nitpicky sidenote]
I don't think anyone could hear 1 db difference unless (maybe) you were placed in some high-tech silence chamber thingamajiggie that doesn't exist for the test.
[/nitpicky sidenote]

I picked 1db because it's a round number that's pretty close to the threshold under decent listening circumstances. The actual value depends on the listener, the room, the sound being played, and which test you believe - I've seen as low as .3db reported.

[Paulhoff]

A good site, that doesn't do a lot of woo-woo.

http://www.audioholics.com/education/cables/speaker-cable-gauge/?searchterm=speaker%20AND%20cables*

http://www.audioholics.com/education/cables/skin-effect-relevance-in-speaker-cables

http://www.audioholics.com/education/cables/speaker-cable-length-differences-do-they-matter

http://www.audioholics.com/education/cables/debunking-the-myth-of-speaker-cable-resonance

http://www.audioholics.com/education/frequently-asked-questions/speaker-cable-gauge-distance

Paul

:) :) :)

[shadron]

The point about coat hangers is that there is no control over the distributed capacitance along the wires, so you have higher capacitance where they are close together and lower capacitance where they are farther apart. Capacitance between the two wires and between each wire and ground has the effect of preferentially, partially short circuiting higher frequencies so they never get to the speaker. However, as pointed out above, the capacitances and resistances are so low that the affected frequencies are far above the audible range. Therefore, you can reasonably use *any* conductors that: 1) don't increase the resistance (being very long or very thin, or having splices internally or at the terminals; iron has relatively high resistance, but clothes hanger thickness reduces that to negligible), and 2) don't increase the capacitance in a huge way (the conductors have large surface area that is very close together, one to the other or to ground). They said they soldered the terminals, so resistance through the iron coat hanger is likely to be measured in milliohms per foot, and the capacitance in picofarads per foot, and the appropriate formulas will show the roll off is at very high frequencies, much higher than the limit of human ability to respond, which is at most 25kHz. Even factoring in the effectively inductance of the speakers doesn't change that result in a meaningful, audible way.

The people who can "hear" these differences are not engineers, and they don't know what kind of impossibilities they utter. As in so many areas of consumer demand, money and marketing defeat ignorance.

[kallsop]

Have you heard the new Monster Triple Annealed Oxygen Free Double Helix Coathangers? They sound way better than cheap coathangers. Dude.

[Gr8wight]

Kellyb:

You keep saying you 'think' there should be a difference. You completely disregard those here with specific physics and mathematics knowledge who are telling you that the difference, while real, is far too insignificant for human ears to hear. You refuse to hear these words because you are still too caught up in your own personal biases on the issue. If you are so convinced that you can hear the difference, by all means, apply for the Million Dollar Challenge. Randi has repeatedly indicated his willingness to test audiophile claims. Send him and email and discuss the matter.

[Olowkow]

At 8 ohms impedance (or even 4 or 16), the capacitance of lamp cord is insignificant. I will go out on a limb and guess that perhaps 100 feet would be as much as 10,000 pf or .01 uf. I don't happen to have any to test though. The resistance is relatively small for copper, and will attenuate all frequencies equally. So even if you had 8 ohms of copper, you would lose just half the voltage or 6 db.. not something you would want to do.

So, the -3db point of a Butteworth low pass filter is:

f=1/2(pi)*RC

For 8 ohms and .01uf, I get just under 2 mHz (that's 2 megahertz). Allowing for ideal 20 kHz response being a bandwidth of 200 kHz, I would say you are pretty safe using lamp cord. I used to commonly put .01 caps across the speakers to prevent the CB lady from getting into my speakers.

As for the inductance...not a problem.

Now at higher impedances on the other hand, like line level 10 kohm, .01 uf is a problem. Resistance less so. Low capacitance shielded (coaxial) cable is used generally, or in broadcasting, two conductor foil shielded 600 ohm cabling is used, often with transformers.

[sol invictus]

The resistance is relatively small for copper, and will attenuate all frequencies equally.

As I tried to point out before, it's not true that even a purely resistive cable will attenuate all frequencies equally. That's because speakers are a rather complex load, and their impedance is generally quite far from purely resistive. So when you put a wire (with fixed resistance) in series with one, it acts as a frequency-dependent voltage divider and attenuates different frequencies by different amounts.

Of course for that to be audible you need to resistance to be pretty high - but 100 feet of cable might do it.

[kellyb]

Kellyb:

You keep saying you 'think' there should be a difference. You completely disregard those here with specific physics and mathematics knowledge who are telling you that the difference, while real, is far too insignificant for human ears to hear. You refuse to hear these words because you are still too caught up in your own personal biases on the issue. If you are so convinced that you can hear the difference, by all means, apply for the Million Dollar Challenge. Randi has repeatedly indicated his willingness to test audiophile claims. Send him and email and discuss the matter.

First of all, I'm not an audiophile; I work in live audio. My bias is that in live audio we push cables to their limits, and I'm completely convinced that a coat hanger of equal length and gauge (minumum 100ft, 12 gauge) under the conditions we put cables under, would perform less well than copper. That is my bias. I put gear through "extreme" challenges compared to recording and home stuff. We work with thousands of watts and hundreds of feet of cables at loads approaching 2 ohm and sometimes under. So my mind defaults to the effects seen under those kinds of conditions.

5% carbon steel has a resistance 10 times greater than copper. When (not if!) a steel cable will have an audible loss compared to copper depends on the length of the cable.

I don't think anyone here has disagreed with that.

That steel doesn't work as well as copper as a speaker cable is not an extraordinary claim to a professional live audio engineer. The math matters.
At very low power levels none of this makes a difference; we could sit in your living room and listen to coat hangers all night and it would probably be fine. Add some power and distance to it and you'll have problems.

[Ivor the Engineer]

Re: the calculations made by EvilBiker:

Meh, resistivity of copper is 15.80 nano-Ohm-meters, iron is 87.10 nano-Ohm-meters.

Resistivity = (Resistance*Cross-sectional area)/length = R.A/l

Considering a diameter of 3mm for both a copper cable and iron coathanger, we get A = 0.00942 m^2 0.00000707 m^2

So, for a 10m length of copper cable we have a resistance of 16 uOhms 22.4 mOhms, and for the same length of iron, we have 92 uOhms 123 mOhms.

(The area of a circle is given by pi*r^2 not pi*d.;))

For 100m of steel coat hanger wire the resistance would be 1.23 Ohms.

This would have a significant effect on the frequency response driving a 2 ohm load because, as sol has already pointed out, the speaker impedance varies with frequency. I'd imagine the most obvious effect would be boomy bass because of reduced damping.

KellyB's comments are correct for long cables driving low impedance speakers.

[EvilBiker]

Bah mutter - shoulda checked. For some reason I was thinking 2*pi*r instead of pi*r^2 -> pi*d. My bad. What a maroon!

Peer review in action, w00t!

[Paulhoff]

The whole point of the coat hanger thing was to point out that some so-called audiophiles believe that they can hear so well that they can hear things that are impossible to hear and when a double blind test was again preformed it showed that they couldn't, geeeeeeeeeeeeeeee.

Paul

:) :) :)

[Trantor]

BOSE speakers are another plague, their motto should be "Progress thru Advertisement".


While I don't consider myself an audiophile, I do own a very nice pair of older Infinity Kappa tower speakers that I purchased back in the 90's. Since my speaker selection was going to be a major purchase, I tested dozens of different types and brands of speakers before making my selection. I tested several Bose speaker systems including their top of the line 901 floor standing speakers and their two best Satellite systems. I was shocked by how these Bose speakers sounded in comparison to some other brands that I tested. The Bose were very expensive in relation to the actual sound quality they delivered. I even did some switching between a pair of cheap Pioneer speakers and some of the more expensive Bose, and I actually prefered the Pioneer speakers over the Bose that cost 10 times more.

[Gr8wight]

First of all, I'm not an audiophile; I work in live audio. My bias is that in live audio we push cables to their limits, and I'm completely convinced that a coat hanger of equal length and gauge (minumum 100ft, 12 gauge) under the conditions we put cables under, would perform less well than copper. That is my bias. I put gear through "extreme" challenges compared to recording and home stuff. We work with thousands of watts and hundreds of feet of cables at loads approaching 2 ohm and sometimes under. So my mind defaults to the effects seen under those kinds of conditions.

5% carbon steel has a resistance 10 times greater than copper. When (not if!) a steel cable will have an audible loss compared to copper depends on the length of the cable.

I don't think anyone here has disagreed with that.

That steel doesn't work as well as copper as a speaker cable is not an extraordinary claim to a professional live audio engineer. The math matters.
At very low power levels none of this makes a difference; we could sit in your living room and listen to coat hangers all night and it would probably be fine. Add some power and distance to it and you'll have problems.

Your claim is still the same. You believe this to be the case, but that belief is not based upon any actual evidence. You have not tested your hypothesis in even the most rudimentary manner. Come back when you have.

[Olowkow]

As I tried to point out before, it's not true that even a purely resistive cable will attenuate all frequencies equally.


What Sol says is of course true, but I was trying to keep it simple, since the losses are not detectable by humans. There is really no point in arguing about this stuff. I would recommend that someone who has the time...I don't...put 100 feet of lamp cord on a network analyzer with a speaker at the other end and plot it out...print it and display the results on this forum. If I had 1 hour in my day I would do it. Most folks don't use 100 feet anyway.

Now, what if there is a squiggly line that "proves" that there is a difference between lamp cord, and coat hangers? So what? If you can't hear it who cares? Coat hangers are ugly. Prove that you can hear a difference, and I will be a believer. The speaker is non linear in impedance at any cable length, so if the impedance at the output is not infinitely low, then all bets are off anyway.

I will say one thing along the "woo" lines, however. Having been a long time audiophile, my engineer buddy and I were totally convinced that we could hear the difference between "teflon silver plated steel coax", (the small RG174 type) and regular line level interconnects...the teflon was awful! I was convinced, but who knows, maybe my ears were off that night. I have no good explanation for this, but we gave up on using the stuff. Perhaps we just convinced each other? This stuff was never a religion with me, results were all that mattered. I spent years fiddling with it. Perfection....never happened. Then I got old, and bought a little CD boom box, and we listen to books now. Oh yes, of course I have 3 Ipods!!:)

One thing is for certain though. I found religion with Krell!! The Krell amp that my friend bought for $6000.00 was amazing. We just could not understand what they could have done so right, since we had built all kinds of class A 20 watt/channel and various other exotic amps...no question about it...Krell was unbelievable, just remarkable. I'll never forget listening A-B to Willie Nelson (sorry!) "Georgia on My Mind", and Carmina Burana on the Krell and our class A amps. The speakers were Swan copies, home built, with the most exotic drivers money could buy, bi amplified.

[Paulhoff]

One thing is for certain though. I found religion with Krell!! The Krell amp that my friend bought for $6000.00 was amazing. We just could not understand what they could have done so right, since we had built all kinds of class A 20 watt/channel and various other exotic amps...no question about it...Krell was unbelievable, just remarkable. I'll never forget listening A-B to Willie Nelson (sorry!) "Georgia on My Mind", and Carmina Burana on the Krell and our class A amps. The speakers were Swan copies, home built, with the most exotic drivers money could buy, bi amplified.
I like the Krell amps (Forbidden Planet named) that I have heard, there are differents in amps. I just got a new Yamaha receiver to replace an older one. I was not expecting any sound different, it was just an upgrade for new features, but there was a better sound to me. And without saying anything to my wife about it sounding better to me, I just turned it on and played music that she plays and is used to hearing and left the room without saying anything. When I returned and not saying anything she remarked that she like it and she heard an improvement in the sound. Later my son, who hears CRT TV's when on in the room, came from work and quickly said he heard an improvement to.

Paul

:) :) :)

Speaker wire on the other hand is BS.

[ktesibios]

What Sol says is of course true, but I was trying to keep it simple, since the losses are not detectable by humans. There is really no point in arguing about this stuff. I would recommend that someone who has the time...I don't...put 100 feet of lamp cord on a network analyzer with a speaker at the other end and plot it out...print it and display the results on this forum. If I had 1 hour in my day I would do it. Most folks don't use 100 feet anyway.


Okay, I was just wondering if anyone might be interested in some measurements. What I've done is to simulate a situation I encountered some years ago in a studio control room, where I found some very measurable frequency response effects caused by some idiot using around 20 feet of 24AWG shielded twisted pair to make cables running from a floor jack to a set of Yamaha NS-10M near-field monitors.

In this image:
http://forums.randi.org/imagehosting/37847d0c2fef041d.jpg

the red trace is taken at the terminals of an NS-10M with a 1 ohm resistor inserted in series with the amplifier output, simulating the resistance of 20 feet of 2-conductor 24AWG copper cable.

The blue trace is taken at the speaker terminals, but with no resistor, just the resistance of a 10 foot piece of 10AWG 2-conductor speaker wire.

The green trace is with the speaker and 10AWG cable, but the measurement is taken at the amplifier output terminals.

The effect of making a voltage divider out of a small (compared to the speaker's nominal impedance) resistance and a complex, frequency-dependent load like a loudspeaker can be seen pretty clearly. As for audibility, my experience is that a 1 dB hump or dip can be heard if it is wide enough- it seems to be an area-under-the-curve effect.

At any rate, the original incident which inspired me to re-create it on the bench started with the question "why do the NS-10s sound different here than they do in the similar room down the hall?"

ETA: these plots are all normalized so that 0 dB is referenced to the voltage measured at 1 kHz, which is why all three curves intersect there.

[Paulhoff]

Okay, I was just wondering if anyone might be interested in some measurements. What I've done is to simulate a situation I encountered some years ago in a studio control room, where I found some very measurable frequency response effects caused by some idiot using around 20 feet of 24AWG shielded twisted pair to make cables running from a floor jack to a set of Yamaha NS-10M near-field monitors.

No yes, inductance and capacitance, what an evil that they can weave.

Paul

:) :) :)

[ktesibios]

Perhaps, just for the sake of clarity, I should mention that the plot I posted is not from a circuit model. It's actual measurements taken with a real NS-10M, a Bryston 4B power amp and a Neutrik A2 audio analyzer controlled by a PC running Neutrik's A4 software. The only difference between this and the real-life incident I described is that I used a 1 ohm 5% carbon film resistor as a stand-in for the wiring resistance, rather than butcher up a piece of perfectly good shielded twisted-pair cable.

On the subject of distortion- metal conductors are indeed linear, provided that you don't push so much current through them that they heat up enough to change their resistance (thermal compression, the decrease in sensitivity caused by the increase in resistance of voice coil windings at operating temperatures that can easily exceed 125C, is an issue for sound reinforcement applications where speakers can spend hours at the edge of thermal failure), but metal-to-metal contacts, if they are contaminated or oxidized, can be very nonlinear. I've encountered a situation where a THD+N measurement taken from a tape machine in input monitor mode was nearly 0.3% if I connected the distortion analyzer to the machine via the console patchbay, but 0.004% if I connected the analyzer directly to the back of the machine. Reworking all the punchblock connections between the patchbay and the wall panel where the machine plugged in solved the problem.

Simply put, wiring is important, but not in the way that audio woowoos think. Clean, oxidation-free connections and keeping wiring resistance and distributed reactances low enough for the application matter; everything else is mere commentary.

[Paulhoff]

On the subject of distortion- metal conductors are indeed linear, provided that you don't push so much current through them that they heat up enough to change their resistance (thermal compression,....
Well let us hope so, if not I would have all kinds of problems with my 440 mhz amateur radio.

Paul

:) :) :)

[Olowkow]

The traces in post #43 are confusing, but likely due to interaction of the passive speaker crossover network, the speaker drivers, and the 1 ohm resistor for the red one.

What you are basically observing is the difference in current across the frequency spectrum, and you are measuring the current indirectly as a voltage drop across the speaker, when the resistor or high resistance cable is in line. Its sort of like using the speaker's impedance as a current shunt, then measuring the current through the resistor (with speaker in series). Clearly, using the heavy cable is preferable, but you would still see some variation at the speaker end, but less, just due to imperfections in crossover design.

We used to use a similar setup, with a series resistor, to find the lower resonance of the woofer, cutoff frequency, in a cabinet. No one is questioning the efficacy of heavy cables. #12 is good, #10 is very good, but #4 or 6 is getting carried away. Welding cable is cooler looking than Monster cables though.:)

One off thread note. For those who build speakers, check out Neutrik twist lock connectors. They even come for biamplification in 4 conductors. They are wonderful, and cheap. Nice metal shells and silver plated contacts, 20 amp very well made for about $6.00, from Mouser Electronics.

[Olowkow]

Well let us hope so, if not I would have all kinds of problems with my 440 mhz amateur radio.

Paul

:) :) :)

dit dit dit dit ..... dit dit!

[sol invictus]

One off thread note. For those who build speakers, check out Neutrik twist lock connectors. They even come for biamplification in 4 conductors. They are wonderful, and cheap. Nice metal shells and silver plated contacts, 20 amp very well made for about $6.00, from Mouser Electronics.

Seconded.

[Paulhoff]

OK, hold onto your hats, and go the the site below, and then, maybe then you will know where I am coming from when it comes to audio woo.

http://www.machinadynamica.com/

Paul

:) :) :)

[zeusbheld]

this is a bit off-topic as i'm referring to mic cables and not speaker wire, but it's still about wires and audio (more or less).

if you try recording sound and use an un-balanced line (NOT grounded) and an XLR (professional, grounded cable) you will likely notice the difference: non-grounded lines tend to have an audible hum.

[soulhill]

I just posted a bit about the "Audiophile" on my site: www.1000demons.com

[ktesibios]

The traces in post #43 are confusing, but likely due to interaction of the passive speaker crossover network, the speaker drivers, and the 1 ohm resistor for the red one.

I don't see what's confusing about it. What it illustrates is the fact that wiring resistance interacts with the speaker's impedance, which is a complex function of frequency, to form what's essentially an unplanned equalizer.

If the wiring resistance is Rw, the input impedance of the speaker is ZL, the voltage at the output of the amp is Vout and the voltage at the speaker terminals is VL then VL=Vout*ZL/(Rw+ZL). You know that, since you correctly identified the combination of wiring resistance and speaker impedance as a voltage divider and that equation is straight outta the voltage divider theorem.

The 1-ohm resistance I inserted in series with the amp output is just a hair less than the resistance of 40 feet of 24AWG copper wire, which is what you have if you make a 20 foot speaker lead out of 2-conductor 24 AWG cable. In this case, it results in the voltage at the speaker terminals varying over a 1 dB range compared to the voltage at the amp output, an effect which is not observed when the speaker is connected with 10 feet of 10 AWG cable, which would have a total resistance of 0.02 ohm.

Modern power amps are essentially constant-voltage sources; the source impedance seen looking back into the output of a decent solid-state power amp will usually be a handful of milliohms in series with a few microhenries (this from the parallel LR most amps have in series with the output to isolate capacitive loads at high frequencies). Properly designed speakers should produce constant on-axis response (and as close to constant power response, which is not the same thing, as possible) for a constant voltage at its input terminals. Consequently anything which alters the voltage-vs-frequency curve at the speaker terminals from the nice flat response at the amp output will cause a similar alteration in the speaker's sound pressure-vs-frequency curve.

What you are basically observing is the difference in current across the frequency spectrum, and you are measuring the current indirectly as a voltage drop across the speaker, when the resistor or high resistance cable is in line. Its sort of like using the speaker's impedance as a current shunt, then measuring the current through the resistor (with speaker in series). Clearly, using the heavy cable is preferable, but you would still see some variation at the speaker end, but less, just due to imperfections in crossover design.The wobbles in the curve at the low end are most likely due to the woofer/cabinet combination's motional impedance reflected to the voice coil terminals. The peak below 100 Hz is the resonance of the woofer, shifted up from it's free-air resonance by the compliance of the trapped air in the cabinet (the NS-10M is a closed-box design); the gradual rise from 200 Hz to around 1 kHz could be motional or due to the woofer voice coil's inductance (the NS-10M crossover is a primitive affair; a second-order LC circuit which doesn't include an RC network across the woofer to compensate for voice coil inductance, which is something you find in a lot of better passive crossovers nowadays). The dip in impedance that starts up around 3 kHz is probably due to the tweeter section of the crossover.

We used to use a similar setup, with a series resistor, to find the lower resonance of the woofer, cutoff frequency, in a cabinet. No one is questioning the efficacy of heavy cables. #12 is good, #10 is very good, but #4 or 6 is getting carried away. Welding cable is cooler looking than Monster cables though.:)

One off thread note. For those who build speakers, check out Neutrik twist lock connectors. They even come for biamplification in 4 conductors. They are wonderful, and cheap. Nice metal shells and silver plated contacts, 20 amp very well made for about $6.00, from Mouser Electronics.Speakons are OK; they've achieved enough acceptance in the trade that quite a few power amps made for the professional market now have Speakon receptacles as output connectors, in some cases in place of the 5-way binding posts that used to be the universal output connector.

Back when I did PA work, the cables I used for my low-freq cabinets (which were ported boxes with 2 JBL K151s in each) were made of two lengths of 10 AWG cable in parallel; for speaker-end connectors I used 30A twist-lock power connectors. My mid-high cabinets (each a horn-loaded 12" cone driver and a 2" throat compression driver on a constant-directivity horn) used Cannon EP-6 connectors and 12 AWG cable. For output connectors on the amp racks I used twist-lock receptacles.

[Olowkow]

I don't see what's confusing about it. What it illustrates is the fact that wiring resistance interacts with the speaker's impedance, which is a complex function of frequency, to form what's essentially an unplanned equalizer.

Perhaps "odd" is a better word. While I believe I stated that I am quite familiar with what it represents, I would have expected sharper peaks and vallies for the resonances. Of course that would depend on the sampling rate of the software. Also, the red trace starts up at 9kHz, while the blue trace starts coming down. Not sure why that should be the case.

I built a pair of Klipsch corner horns, (too many problems, like time alignment) and dozens of various speaker, active crossover, and amplifier designs. I even had a pair of 80 Hz Altec horns which I used with the Emilar drivers and Altec 500 Hz horns (damped) on the top end. 20 watts/ch class A sounded phenomenal, but for me the bottom line was always room acoustics, and this battle was never ending. 1/3 octave equalizers helped, but just for one "sweet spot". I finally decided that the "perfect" audio system does not exist, on my budget anyway.

When we remodeled the house, I discovered I couldn't get the 80 cycle horns out the door! I had to cut them up with a chain saw! :mad:

[Paulhoff]

When we remodeled the house, I discovered I couldn't get the 80 cycle horns out the door! I had to cut them up with a chain saw! :mad:
It is a good thing that there is no hell.

Paul

:) :) :)