|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
I got this off of a list I belong to, amd have not seen anything like it as to cable studies. Sounds interesting. Did not post the guys name who posted it.
Clayton
>Warning, long post (Summary of a study)
>
>About once a year this topic comes up and somesone makes an uninformed
>statement similar to the one below. I usually make one post pointin to an
>article appeared in the Winter Audio Ideas Guide (an objectivist
>publication) that gave the results of
>an extensive test of speaker cables that showed they do measure differently
>and that difference is audible. In the past I've dropped it there and assumed
>those interested would search out the reference. Seems the reference drops
>into
>an objectivist black hole so here is a summary.
>
>The article and study was done by James H. Haward. "Recently retired from
>Bell Canada Special Services Engineering and Operations Groups, Jim Hayward
>is a Ryerson
>Electronics Technology graduate whose professional specialty has been data
>communications and radio systems. He is also a pianist and longtime
>audiophile, and currently teaches at Radio College of Canada in Toronto."
>
>Paraphrasing:
>
>The laws of physics are the same regardles of which end of the power amp
>you observe. That said, there is one parameter that causes a profound
>difference in performance. It is the impedance interface.
>
>He procedes to outline aspect of cables, Zs,R and L. and follows with
>
>"a recent study of nyquist plots of the complex impedance curves of 21
>speaker systems wasn't good news for those audiophiles who believe the
>world would be a better place if all the audiophiles invested in
>transmissin lines with a characteristic impedance of 8 ohms. ... the
>minimum impedance being one ohm and the maximum 28 ohms. Phase deviations
>abounded between maximums of =56 and -67.5 degrees.
>As if that wasn't bad enough, a number of scientific reports indicate that
>certain dynamic signals...applied to speaker systems can cause a current
>flow that is up to two and a half times greater than that predicted from
>the speaker's minimum impedance."
>
>The power amp must be capable of delivering very high current and have an
>output
>impedance of nearly zero. A power amp must behave as much as possible like a
>perfect voltage source. In other words , its output voltage must not be
>influanced by changes in load impedance. ... Lets take an actual cable
>parameter and see how
>its impact changes withy the magnitude of the source and load impedance. 500pF
>is a typical value for parallel capacitance of 3 meter length of ... speaker
>cable. As capacitance reactance is lowest at high frequencies, lets analyze
>its
>impact at 20kHz where its nasty shunting behavior will be most detrimental.
>It's capacitive reactance is about 16 kohms at 20 kHz and this is in parallel
>with the load. If the source is a passive preamp with an output impedance
>of 16
>kohms also, the frequency response will be down 3dB at 20kHz. Also, a high
>impedance
>preamp must drive a very high input impedance power amp to avois excessive
>loading.
>so if in our example, the input impedance of the power amp was 250 kohms,
>there would be over fifteen times more current flowing through the cables
>capacitive reactance to
>ground than threouh the input of the power amp at 20 kHz.
>
>Discussion of features necessary for high performance speaker cables omittd
>for brevity.
>
>The following speaker cables were measured and studied
> Prism Time-Compensated
> Nordst Flatline
> Kimber 8TC
> Kimber 4TC
> Cardas Quadlink 5C
> van den Hul The Revelation
> van den hull Hybrid
> Amphenol RG8/U
> lamp cord (16 AWG)
>
>Impedance vs. Frequcecy (mohms)
> Below 2kHz all cables were bunched together below 50. Lamp cord ran about
>twice as much. The cables radically diverge above 2kHz - at 20 kHz they
>vary from 130 to 780. Flat impedance is better.
>
>Conclusions
>
>Risetime Tells Tells ... The cable with the shortest risetime is the one with
>the largerst bandwidth. It is also an excellent indicator ofuniform
>frequency and phase behavior within the audio spectrum. Low inductance
>cables have the flattest
>frequency response and shortest rise time. Cables with very low resistance are
>only contenders if they have low inductance also.
>
>Minimise power loss by using more cables. Bi-wire or bi amp. Not only does it
>reduce current in each cable but it reduces intermodulation distortion as
>well.
>(My aside-bi-wire with cheaper stiff for low/mids and quality stuff for hf.)
>
>Bi amping is expensive but the results are stunning.
>
>What speaker cables came oyt on top?
> 1. Kimber 8TC by a long shot
> 2. Kimber 4TC by a long shot.
>
>If anyone wants to discuss this, leave me out. This is presented for
>information
>only, not interested in debates. I someone can post information from extensive
>studies that directly contradict this I would be interested in seeing such.
Clayton,
You should try and get these two articles from KimberKable as they have it reprinted. Call them at 801-621-5530. It is an interesting technical article and eye opener. Just a couple of quotes from the article:
"It's puzzling that the fundamentals of cable design, analysis and application, so thoroughly documented by phone companies(ref.#3) and others in the early decades of this century, have not clarified this apparent confusion. Perhapstheir gems of wisdom have been forgotten, ignored, or worse, been taken out of technical context."
"Using the selection criteris that we developed earlier and reviewing our measurement data, predicting the best performing cable is straight forward."
He picked them by specs, a perfectly valid reson to do so.
http://kimber.com/8tc.htm#jameshayward
The post goes directly from a list of cables to a conclusion of
which are best. That's too large a jump for me.
You can tell the author is a data/RFI guy because for him life
starts at 20,000 Hz.!
I can't hear 20,000 Hz., can you?
So the "down 3 dB at 20 kHz. is not audible.
For 10 foot speaker cables resistance, inductance and
capacitance do matter but differences measured among
cables of a reasonable gauge (at least 16 gauge)
designed for audio use are small.
Debates on whether they are audible when using a ten foot
speaker cable will continue forever.
Characteristic impedance of cables is a data transmission
concern -- not an audio frequency concern for 10 foot cables.
I do like the comment "Bi-amping is expensive but the results
are stunning". That comment would not apply to bi-wiring which
simply adds more copper between your amp and speakers -- perhaps
"buy wire" would be a better spelling than "bi-wire" -- the
electrical circuit remains the same whether the wires from the
drivers inside your speakers meet at the speaker terminal or
at the amplifier terminal.
The comment on using "cheaper" wire for low/mid frequencies and
"quality" wire for high frequencies seems backwards to me
because most of the current flowing to our speakers is for bass
and mid-range frequencies. How do I know that?
Well, I just shut-off the tweeters on my "buy-wired" EPOS
ES11 speakers and could see no change on my amplifier output
meters while playing pink noise -- that must mean very little
power is required for tweeter frequencies. So if you want to use
"cheap" wire for some reason, I'd say don't use it for your
bass/mid-range driver. Of course we could debate forever on what
"cheap" wire is. I'd say it's any wire less expensive then the
30 cents per foot Home Depot 12 gauge speaker wire I use.
By the way, is Mtrycraft allowed on this website?
An audio website without his comments and the nasty
responses just doesn't seem very exciting!
Note: My comments on wires do not apply to 50 foot speaker
cables -- only to 10 foot or shorter cables.
I also tend to disagree with the author of the original article (Mr. Hayward), for many reasons, not the least of which are that he latches onto one single measurement factor as the ultimate metric, risetime, and that is a sure sign of having methodological blinders on. But your statements after that are not all factual or substantiated.
[ So the "down 3 dB at 20 kHz. is not audible. ]
This is right out of the ABX bible. Couldn't you be a little more imaginative? -3 dB implies - 1 dB at 10 Khz, and several fractions of a dB at lower frequencies. Nothing exists in a vacuum, including -3 dB points. In fact, examination of the inductance of typical zip cord cables, indicates that the rise from the DCR baseline starts at a relatively low frequency, one that gets lower and lower as the gauge does down (wire gets bigger). Most 12 gauge zip cords deviate from the DCR baseline by several hundred Hz. Yep, that big garden hose of a wire is no longer as low in impedance begining as low as 300 or 400 Hz!
The other factor invariably overlooked is that a -3 dB point implies a phase shift of at least 45 degrees, with corresponding amounts below that point, all the way down to the several hundred Hz point.
This means that the phase of the highs is no longer in step with the lows, and waveform distortion is the result.
[ That comment would not apply to bi-wiring which
simply adds more copper between your amp and speakers -- perhaps
"buy wire" would be a better spelling than "bi-wire" -- the
electrical circuit remains the same whether the wires from the
drivers inside your speakers meet at the speaker terminal or
at the amplifier terminal. ]
This is also right out of the ABX bible, nearly word for word.
Now I have posted frequently about how bi-wiring works. I find it hard to believe that you have not seen a one of them, or that you are not aware of the information at my website regarding current flow in bi-wire cables.
The electrical circuit IS NOT the same, nor is the resulting output signal from the speaker.
That makes your statement either supreme ignorance, or the desire to make unsubstantiated and misleading statements.
With the benefit of the doubt in mind, see this post and the related posts:
/audio/tweaks/messages/487.html
Jon Risch
A list of cables was presented in the post and then
we're told which are best with no explanation of how
the "best" were selected. That was my primary question
and I still don't know the answer.
The down 3dB at 20,000 Hz. seems to be a worst case
hypothetical scenario rather than an actual measurement
and not worth arguing about .. though I know you like to
argue!
I said can't hear 20,000 Hz. ... If the frequency response
was also down 1 dB at 10,000 Hz. (not stated in the post),
I'd probably enjoy that coloration, assuming I could actually
hear it while listening to music.
Concerning your wiring theories, I have read your theories that
audio frequencies travel MAINLY in a field outside the speaker
cables rather than primarily electron flow within the wires with
a weak magnetic field outside the cables.
A few electrical engineeis applicable to audio frequencies
-- perhaps there's a small possibility you are not correct.
You also write you can hear differences among speaker cable
insulator materials and rate each commercially available
material for audio use in your report.
I'll assume these insulator differences you claim to hear
were tested with blind or double-blind methodology before
you wrote your preferences and recommendations.
If this is true, you would be the only person I've heard about
in 33 years as an audiophile with such sensitive ears!
Perhaps your ears are too sensitive -- and your hearing is
not representative of a typical audiophile, who might have
difficulty hearing the difference between 10 feet of 12 gauge
copper speaker cable and 10 feet of 14 gauge copper speaker cable.
You certainly would have hearing sensitivity different than mine --
I have no idea what gauge of speaker wire is being used when I
listen to a stereo, much less what material the speaker wire
insulator is made of -- yet my hearing has been tested and it's
normal for my age (over 40).
Your buy-wire report used a unusual new test signal that you
invented, I suppose, rather than music or more common
test tones. Was the new test signal designed or chosen
specifically to emphasize a measurement that supports
your wiring theories?
You also do not mention the gauge and type of wires used
for the charts, except to say one is a single wire and
the other is a buy-wire. Does that mean you are comparing
one 14 gauge wire with two 14 gauge wires, for example?
Or would you be comparing perhaps one 11 gauge wire with
two 14 gauge wires in an attempt to have roughly the same
amount of copper in both circuits? You don't specify what
you are comparing but I certainly won't argue if you are
trying to prove two 14 gauge wires are better than one 14
gauge wire because I buy-wire my own speakers with two
12 gauge wires.
Your theory that an advantage of buy-wiring is the ability to
separate the magnetic fields of the woofer and tweeter wires
is based on your other theory that audio frequency current flow
is primarily a field outside the wire -- an unproven theory
theory for audio frequencies according to electrical engineers
I've talked with.
That you can cause a sound by hitting a speaker cable with a
hammer or with a pulse from a car battery does not seem to
prove your theory about separating buy-wire speaker cables
because the magnetic fields of two nearby cables would cause
audible distortion. On the other hand, it would make no sense
to tape those buy-wires together.
On a more positive note, the electrical engineers I talked
with support experimentation and agreed that if an audiophile
wanted to experiment with different cables and interconnects --
your do-it-yourself designs make a lot more sense than buying
expensive commercial products.
I've also passed out several copies of your DIY Acoustical
Treatments article. I'm not sure all spouses will like the burlap
covering -- speaker grille cloth can be used too, at greater
expense -- one audiophile i know built fiberglas acoustical panels
over ten years ago using a 1/4 inch thick charcoal grey foam speaker grille material -- you could see some of the pink fiberglas through it but
his wife could tolerate it -- they looked like two big speakers
from a distance rather than cheap 2 x 4's and fiberglass insulation.
[ A list of cables was presented in the post and then
we're told which are best with no explanation of how
the "best" were selected. That was my primary question
and I still don't know the answer. ]
It was based on the risetime of the cable. I have an unfair advantage, as I saw the original article some years ago, and remember the gist of it.
[ The down 3dB at 20,000 Hz. seems to be a worst case
hypothetical scenario rather than an actual measurement
and not worth arguing about .. though I know you like to
argue!
I said can't hear 20,000 Hz. ... If the frequency response
was also down 1 dB at 10,000 Hz. (not stated in the post),
I'd probably enjoy that coloration, assuming I could actually
hear it while listening to music. ]
There is more to what we hear than simple amplitude response.
[ Concerning your wiring theories, I have read your theories that
audio frequencies travel MAINLY in a field outside the speaker
cables rather than primarily electron flow within the wires with
a weak magnetic field outside the cables.
A few electrical engineeis applicable to audio frequencies
-- perhaps there's a small possibility you are not correct. ]
There is no doubt about the EM signal existing primarily in the space outside the wires, rather than in the wires. This is the accepted reality of what goes on, and is not my theory. An electromagnetic field and electron flow are two different things. The EM field travels primarily outside the wires even for 20 Hz.
The only reason that any of the field penetrates the conductors at any frequency, is the fact that they are less than perfect conductors. Superconductors completely exclude ALL magnetic and electric fields, hence, by definition, any EM field must travel outside such superconducting wires.
More mundane conductors have resistance, which allows some penetration of the magnetic fields. An EM field consists of an E field, and a B field, and even real world conductors pretty much exclude any E field potential differences INSIDE the conductor, therefore the EM field, by definition, can exist only outside the conductor.
[ You also write you can hear differences among speaker cable
insulator materials and rate each commercially available
material for audio use in your report.
I'll assume these insulator differences you claim to hear
were tested with blind or double-blind methodology before
you wrote your preferences and recommendations. ]
I posted a response to this before, yes, double-blind tests were conducted to ascertain said materials differences.
[ If this is true, you would be the only person I've heard about
in 33 years as an audiophile with such sensitive ears! ]
I personally know of at least a dozen people who can replicate my results, under blind conditions. There are many more that I do not know personally, and within the audio industry, professionals have long ago abandoned any faith in ABX methods to evaluate audio components. They also do not have the time or the inclination to tilt at the windmills that the ABX clique represents.
[ Perhaps your ears are too sensitive -- and your hearing is
not representative of a typical audiophile, who might have
difficulty hearing the difference between 10 feet of 12 gauge
copper speaker cable and 10 feet of 14 gauge copper speaker cable.
You certainly would have hearing sensitivity different than mine --
I have no idea what gauge of speaker wire is being used when I
listen to a stereo, much less what material the speaker wire
insulator is made of -- yet my hearing has been tested and it's
normal for my age (over 40). ]
My hearing is not particularly exceptional in any way, other than I have trained my hearing to discriminate minute differences. This can be taught (to a certain extent), and I have trained others in the past, and taught others to hear small differences via proper test methodology. It is mainly a matter of focusing attention, and knowing what to listen for, and how to listen to one aspect of the music at a time. This takes practice, and training for most others to reach a level close to my own level of discrimination.
[ Your buy-wire report used a unusual new test signal that you
invented, I suppose, rather than music or more common
test tones. Was the new test signal designed or chosen
specifically to emphasize a measurement that supports
your wiring theories? ]
The new test signal was developed as a result of investigations into the lack of current test signals to exhibit sufficient sensitivity to low level signal problems. It was honed and refined by testing a wide variety of audio components, primarily loudspeakers were used to develop the signals. There are about 6 main types of Phi Spectral multitone test signals, and they were presented at the 105th AES convention. The tests shown at my web page are the results of just one of those test signals.
[ You also do not mention the gauge and type of wires used
for the charts, except to say one is a single wire and
the other is a buy-wire. Does that mean you are comparing
one 14 gauge wire with two 14 gauge wires, for example?
The first tests were using 12 gauge zip cord compared to my cross-connected 89259 used as a pair of bi-wired cables. These tests are what were published in my AES preprint #4803. The CC 89259 is about 13 1/2 gauge, so the net copper difference was between 12 gauge total, and approx. 11 gauge total. Not very much difference in total resistance, and certainly not enough to explain any of the test results. Later tests used exactly equal amounts of copper, and other cables for the tests, including retail high end cables, and cables of intermediate sophistication. All results were similar, such that the distortion reduction effects were not significantly influenced by any gauge discrepancies.
[ Or would you be comparing perhaps one 11 gauge wire with
two 14 gauge wires in an attempt to have roughly the same
amount of copper in both circuits? You don't specify what
you are comparing but I certainly won't argue if you are
trying to prove two 14 gauge wires are better than one 14
gauge wire because I buy-wire my own speakers with two
12 gauge wires. ]
Converting the bi-wire cable runs to a single wired doubled up run will invariably raise the distortion levels as tested.
[ Your theory that an advantage of buy-wiring is the ability to
separate the magnetic fields of the woofer and tweeter wires
is based on your other theory that audio frequency current flow
is primarily a field outside the wire -- an unproven theory
theory for audio frequencies according to electrical engineers
I've talked with. ]
You are confusig different things here. As noted above, it is a FACT that the EM field travels primarily outside the conductor, even at audio frequencies. This not subject to debate or open to interpretation.
Then, you seem to be ignoring the current measurements I present, the ones where Fig. X shows that there is indeed a large difference in how much current is flowing in the different bi-wire cables. Current is directly proportional to how strong of a magnetic field is present, so the evidence for my theories about the magnetic field interaction between the HF and the LF seem to be quite compelling.
[ That you can cause a sound by hitting a speaker cable with a
hammer or with a pulse from a car battery does not seem to
prove your theory about separating buy-wire speaker cables
because the magnetic fields of two nearby cables would cause
audible distortion. On the other hand, it would make no sense
to tape those buy-wires together. ]
Not so. I have long proposed that distortion levels that seem too low to be problematic are in fact audible. Independant research has shown that under certain conditions, it is possible to hear down to 0.1% distortion. This is 60 dB down from nominal with steady state sine waves. If there is a more complex form of distortion that is smeared out in time, or causes some non-linear distortion to occur, it is easily possible that levels below -60 dB are problematic. Of course, I am speaking in terms of high performance SOTA playback, and not a Sears rack system.
The purpose of the magnetostriction demo is to show that ANY current flowing through the cables will cause motion. Any motion of the cables while they carry current WILL cause another signal to be generated, one that is delayed in time due to the mass and mechanical resonance characteristics of the cables, and will be added in with the original signal. How far down are these motion related distortions? They are at least in the realm of -60 to -80 dB. This makes them potential candidates for IM distortion, and when you factor in the single cable vs. bi-wire scenario, and realize that there is a fairly complete separation of the HF currents and the LF currents, you can see why I have theorized as I have. Keep the LF currents and associated magnetic fields from being able to affect the HF signals, and you WILL reduce any IM related effects, just as my measurements show.
Think about some of these things, instead of blindly accepting the ABX dogma and all of their propoganda without questioning it at least as hard as you question my own postings.
Jon Risch
bass frequencies can readily penetrate any protruding edges in a non-round wire, which I'm sure you know but some don't.
BTW, how do you keep the Capacitance down along your coaxial cable & between HF shield and LF core of opposing polarities.
[ BTW, how do you keep the Capacitance down along your coaxial cable & between HF shield and LF core of opposing polarities. ]
When connected in cross-connected fashion, Belden 89259 has approx. 48 pF per foot. Two coaxial 89259 cables with 17.3 pF per foot connected in parallel, but separated by some distance, will run twice the nominal, or about 35 pF per foot. The cross-connection actually increase the capacitance still further to the 48 pF value, but since it starts out so low, and remains at a modest value, the decrease in inductance that it enables is well worth the tradeoff.
Essentialy, it starts out very low, and never gets up that high even after cross-connecting the two coaxes.
Jon Risch
I have a pretty modest system consisting of a mid level
Denon 3200 DD receiver equiped with dual sets for the
fronts and center channels. My Paradigm Studio 20s and
CC450 are also so equiped.
For the first month I ran single leads of low grade
Vampire 12ga.. One knock on the Denon is that freqs. in the
sub 70-80Hz(?)range get cut off from the mains regardless of
what settings are used. I must say I noticed this. I
realize that the bookshelf size of my 20s don't lend
themselves to true lower levels but I should've been
getting more.
A week ago, I ran a second set of leads to the fronts using
the same wire and shit shack dual bananas. The difference
in bass output was immediate and unmistakable. It's as if
the speakers are breathing more freely. Maybe it's due to
"more copper", a lack of the jumpers, or one wire for each
driver, I don't really know. It has helped and the change
is not subtle in the least bit.
Chris Garrett
To the new and improved Mr. Greene:
So glad you finally decided to pay us a visit.
Welcome and enjoy your stay.
Of course mtry is allowed here, with a certain understanding.
We have asked him to consider the purpose of the founders.
I'll do my best to paraphrase what was so eloquently
stated by one of, and agreed to by the other who
provided the resources and volunteers support to open
the Asylum:
> > There are contrasts between a well versed audio objectivists,
those who are passionate about music, and rabble-rousers who
have no connection to the deep emotions that musical experiences
can evoke.
Some folks (take jj for example) are able to add an excellent
balance to our more radical tweakers without being a jerk.
Other, self-professed rabble-rousers are obviously not interested
in reasonable dialogue, and tend to post in insulting and sarcastic manners.
Some, even by their own admission, are not deeply into either
music or experiencing different reactions and emotions related
to audio; they only post for one reason - to preach negative
gospel and irritate audiophiles who dare state an opinion that
their science has not confirmed.
Some of the tweaks are pretty outrageous; some folks believe that
smoking cigars is a silly, obnoxious habit. Going to a cigar web
forum and posting insults and sarcasm to irritate people who enjoy
the habit/hobby is a waste of everyone's time.
As one who smokes cigars, and reads the reviews, I've found
it hard to taste flavors others are able to. At first I
thought them nuts, I mean "toasted hazelnuts" and "blackberries!"
After some time, I started to get hints of other flavors. I
realized that it's a question of sensitivity. Some of us
have it and others do not. Wine tasting is another good example.
Most of us can't tell a difference between vintages, let alone
Chateaus. Why is music any different? Measurements aren't the
end all be all.
This takes us to the "cable game." Amps/speakers/etc. all possess
many components which vary in quality, so discerning differences
between any two, isn't too far a stretch. Cables on the other hand,
are mainly composed of copper/silver and a few assorted insulators.
How could there be a difference from brand to brand?, the naysayers
argue. I'll leave the design to others who know, but I've read too many
intelligent posts labeling differences for me NOT to believe any exist.
I just want to let you know that I agree with your take on the matter.
Chris Garrett, Miami
This post is made possible by the generous support of people like you and our sponsors: