Sorry, i misunderstood the motives / point of your question RW.
If you don't think that "skin effect" comes into play, try finding some cabling that is identical other than the conductor type i.e. solid vs stranded with the same gauge and geometry from the same manufacturer. Since surface areas, dielectrics and grade of conductive materials should be near identical, the only variance would be "skin effect" due to "strand jumping". The biggest difference will be in treble clarity, which is where most of the smearing takes place. Sean
> |
Sean wrote:
>> Not to single Eldartford out, but he was of similar belief to you (and many others) on the subject of speaker cable audibility <<
I never said there wasn't an audible difference between speaker cables. I'm using Goertz MI-2s in my system because they "sound" good to me. I'm postive that there is a difference in the sound of speaker cables.
What I *did* say was that "skin" effect was grossly over-rated as a reason for these audible differences. I also realize that the equations used to buttress my argument may not be 100% accurate in a real-world environment. But, those equations are *not* off by an order of magnitude. And they'd have to be off by that much (> .2db) before you even *approached* audibility of skin effect.
*Something*, or some things, *do* make an audible difference - skin effect ain't one of 'em.
Nice chattin' with ya!
-RW- |
Rlwainwright: We've been down that road before several times here. Depending on the test methodology, results vary quite a bit in terms of "skin effect" / inductive reactance. When factored into actual use tests i.e. by combining the electrical characteristics of the cable with the load of a typical loudspeaker, bandwidth is further reduced. As such, taking steps to reduce inductive losses in the cable itself can pay off and does become audible under real world conditions.
The other variable here is that inductive losses & skin effect are also part of the impedance equation. Taking steps to reduce both simultaneously also lowers the nominal impedance of the cable. This improves transient and power response, which also factor into the audiblity equation.
Go back and read some of the various debates here discussing such things. Not to single Eldartford out, but he was of similar belief to you ( and many others ) on the subject of speaker cable audibility. As you may be aware of, Eldartford is also an EE and "probably" based his beliefs on the logical deductions that come with the basic formulas that we are taught. After making a direct comparison in his system between two different types of speaker cables i.e. heavy gauge zip cord and an "esoteric" brand, El publicly stated that he EASILY heard major differences in presentation. He also went so far as to put his money where his mouth is and purchased all new cabling, based on his own first hand auditory experiences.
You'll also find that i've offered to lend speaker cables to others that are VERY vocal proponents of "cables iz cables". The person that was the most vocal on the subject in recent memory ( can't even remember who it was ) refused to even consider performing such a test. This is true even though the testing would be performed within the confines of his own system and it wouldn't cost him a thing. At the same time, they continued to "preach" their message, even though they weren't willing to put their own beliefs to the test.
As such, those that are willing to experiment can end up reaping the sonic benefits, along with the knowledge and experience gained. Those that cling to formulas and "theory" become limited by those constraints, missing out on the potential their system and listening enjoyment may hold for them.
Below is a excerpt from a post i made a week or so ago in a power cord thread. It applies here too. Sean
>
"PS... Those laughing about cabling ( speaker, interconnect, power, etc... ) making / not making an electrical and / or audible difference lack actual experience in the field. If they had the actual experience, and therefore the knowledge that comes with it, they would no longer be laughing.
Instead, they would be asking "why" and working on ways to further their limited experience and education on the subject. That is, if they were truly intelligent and truly interested in such things. Graduating from a school and / or having a limited background in one specific area of electronics doesn't make you intelligent and / or "well versed" in all areas of electronics.
Most technological breakthrough's come about because someone couldn't understand how something that wasn't supposed to be happening was happening, hence their taking the time and making the effort to understand how / why the impossible was possible. That is, the "impossible" in terms of the level of theory that they / we understand at the time.
Most of the time, things are going exactly as theory would predict. This is why we can do what we can do with repeatable and predictable results. The problem is that most people don't take all the various parts of theory into consideration. Instead, they try to keep things simple, therefore neglecting several of the more complex variables that enter into any multi-faceted equation. This is why looking at the big picture typically explains what is going on in the little picture.
Unfortunately, even the smallest things, like an atom, are phenomenally complex". |
>> Skin effect / impedance related high frequency losses definitely come into play in the audio region. This is well documented and has been known since the 70's. You can find independent measurements and test results all over the place if you look around. Sean << Well, as suggested by Sean, I *did* look around a bit. I still stand by my previous assertion. The losses that were demonstrated were on the order of .02db at 25KHz, and approx. half that value at 10KHz. I challenge ANY human to prove he can hear a .02db loss at ANY frequency. It would appear that skin effect is of no practical consequence at audio frequencies. This is not intended as a slam to Sean, he's a great guy and very well-informed. However, in this case, I feel that the skin effect is grossly mis-represented and of no real effect upon a cable carrying audio frequency signals. Go to this link: http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/skineffect/page2.html to read the following: "For the example chosen, at low frequencies the bulk resistance of the copper wire causes a power loss of around .008dB. At 10kHz the loss rises to .009dB if the internal impedance were absent, and .016dB with the internal impedance taken into account. At 25kHz these values rise to .012dB and .028dB respectively. Hence the change in relative signal level from near-d.c. to 25kHz, with internal effects taken into account is around .020dB." -RW- |
Sean, I had a couple of ideas on the twisting, but, its not really that important. I admire your ambition, and hope you share your results. |
Clio09: I have NO idea what Alpha-Core or their dealers are charging for their products nowadays. I haven't purchased any cabling in quite some time.
Other than that, i don't know who "Dave" is, but if he's chosen to carry Goertz speaker cabling as a product line, give him a pat on the back for me, will ya??? : ) Sean
> |
Thanks again Sean. Dave is 40 miles down the road from me. Maybe I'll give him a call and chat with him to see if I can get a pair on a trial basis.
BTW - Did the prices drop on the AC MI-2 cables. I was looking to get a 6 ft. pair (silver spades) as a result of reconfiguring my system and the cost came out to a whopping $209 with shipping. |
Unsound: If you look at how Alpha does their "purling", the conductors are "stacked together" and THEN twisted. As such, their impedance remains consistent and they are not configured as a standard "twisted pair" configuration. Maybe i misinterpreted what you were saying i.e. the twisted pair configuration and ended up putting words in your mouth, but that's what came to mind. It would be hard to keep "hollow" cables "stacked together" when using some type of core, hence my "logical deduction".
Clio09: As far as the Magnan's go, they might well be worth a shot. The reference might have some slight theoretical advantages over the Goertz and it would be an interesting comparison. The foil would have even less skin effect than the "strapping" that Goertz uses, although i see the potential for a far less consistent impedance in that specific configuration. That is, the Goertz conductors tend to cling together relatively tightly in the way that they are bound. I don't think that the Magnan's are built in similar fashion.
On top of that, it would be easier for the foil to kink, bend, crinkle and / or seperate from one another, potentially introducing other electrical side effects. All of this is besides the previously discussed comments pertaining to how the foils themselves are terminated.
As far as speaker cables go, there's a comparison that i wanted to make for quite some time. I purchased the necessary parts quite some time ago, but never got around to it. Once i could put these cables together, the end result would be a conductor that was nearly equivalent to the Goertz MI-2's in gauge with a similar, but slightly higher nominal impedance. Due to the materials used, skin effect should be slightly reduced and treble speed may be slightly increased.
This is all theory though, so until i can get around to doing this and performing the associated tests, it's all a moot point. As we all know, there can sometimes be major divurgences between theory and reality : ) Sean
>
PS... When making comparisons, try not to confuse "different" for "better". It's an easy thing to do because of the "newness factor" : ) |
Sean,
Thanks for the clarification. I have rather enjoyed the AC MI-2 cables over the last year or so. However, I have been interested in doing A/B comparisons against the AC MI-2 just for the heck of it. The Magnan's were on the list, but now I think I'm going to give the Supra Swords a listen instead. |
Sean, thank you for your reply. At the risk of appearing argumentative, wouldn't it be possible to twist these tubes in such a manner so that the same side of each coductor would maintain contact? The only reason I suggest this arrangement is because Alpha-core suggests that "purling" has some advantages. Perhaps we could even abandon the idea of twisting. Would hollow oval tubes have an advantage over solid oval conductors? |
Unsound: The Reference speaker cable is stacked ala Goertz and the Signature is placed side by side ala Nordost. I guess that David is trying to play both sides of the street simultaneously. The end result is that these two cables would measure and perform quite differently, even though they might use extremely similar materials. The difference in geometry i.e. stacked or side by side makes for a huge difference electrically. Stacked gives you a MUCH lower impedance( what we want ) whereas side by side will put you up over 80+ ohms ( at least ).
For the longest time, i thought that the Magnan's were "side by side", but when i looked at the info on the Reference last night, it stated "stacked". As such, i chalked it up to memory failure and "ass-u-me'd" that the Signature's were built like this too. They aren't and i'm glad that you pointed this out. As a point of reference, i've used and liked some of Magnan's interconnects, but never their speaker cables.
As to their terminations, i too think that they are the weak point of the design. It almost looks like they are using a "standard" solid or stranded lead attached to the foils. To my mind, this pretty much negates many of the benefits of the foil. On top of that, there's got to be a very measurable impedance bump in this region, creating inconsistencies in the signal path.
As to the AP cabling, they've used two different geometries over the years. Their copper cable used one design geometry and their silver cabling used another. I was involved in several different MAJOR debates about this cabling. My question to proponents of this cabling, and the manufacturer of this cabling, was that if one design is electrically superior, which is what they stated was demonstrated in their computer modeling, why would they use anything but that geometry regardless of the type of conductor used? The fact that they talk so much about skin effect yet neglect to mention "strand jumping" was also a HUGE falling point with me. After hearing these cables, i could not get rid of them fast enough...
As to your idea about using hollow cables and configuring them into a twisted pair, that would be an inconsistent mess. Due to the oval shape and twisting them, you would run into massive differences in contact area between the conductors. This would result in an ever changing impedance and inconsistent Td ( Time delay ) through the cabling. One would hear this as a smeared sound with a lack of focus and solidity. Compared to many cables though, it would be no better or worse.
Clio09: Most any Solid State ( SS ) amp should be fine here. Output impedance for most SS amps is quite low and are therefore quite suitable for use with Goertz flat series of speaker cabling. On the other hand, many tubed amps have an output impedance that is anywhere from one to a dozen ohms or so, introducing further variations into the equation. Using a low impedance cable like that of the Goertz design may cause poorer sonics rather than improvements. This would be due to the lack of "buffering" that most higher impedance speaker cables would present as part of the load that the amp sees. The less "buffering" that a high output impedance amp sees, the less linear and consistent the sonics will be.
As a side note, Nelson Pass ( and a select few others ) are making special amps with VERY high output impedances on purpose. The lack of linearity is being purposely taken advantage of for use with other non-linear devices i.e. limited bandwidth full range speakers utilizing a single driver. The lack of "buffer" or "damping" introduces a bass peak, which helps to balance out the sound of these otherwise "lean" sounding speakers. Kind of like using two negatives and trying to make a positive. Only problem is, the results can vary SOOOOO drastically, that a LOT of trial and error will be required in order to obtain the desired results. It is my opinion to start off with the highest level of consistency possible and build from there. Sean
> |
I hope this isn't taking this thread to far off course; upon reading Duke's post I went to the Magnan site and noticed that Magnan and Alpha-core while similar have a d different thoughts re; orientation. Magnan suggests that the positive and negative legs should be kept side by side and Alpha-core suggests top over bottom. Wireworld is marketing similar video cables and suggests top over bottom. When I spoke to Alpha-core re: using seperate legs for postive and negative (in order to facilitate simpler termination amongst other reasons) they said that I would be reducing their cable to "expensive jumpers". IMHO, the Magnan terminations seem a bit inelegant and might make cable dressign quite a nuisance. Analysis Plus concept of hollow cables would to my ignorant mind seem like a good thing. I can't help but wonder if an oval configuation of the Cogan Hall cables wouldn't be a bad idea. Any thoughts of about using twin thin hollow oval silver tubes individually wrapped in teflon and twisted that could be integrally termintated as speaker cables? |
Sean -
I'm a little curious about the comments you made on output impedance of the amp and impedance ratings of the speaker cables and speaker. I use AC MI-2 and from what I can tell from specs published on the website they are 2.5 ohms impedance. My speakers are Spendor 1/2e rated at 8ohms nominal impedance (they dip but stay pretty flat overall). I have not perceived any issues with my amp, which is custom built and while I know the input impedance, sensitivity and gain, I don't know the output impedance. Ideally if I have the AC MI-2 specs correct what should be a good output impedance range for my amp? |
Skin effect / impedance related high frequency losses definitely come into play in the audio region. This is well documented and has been known since the 70's. You can find independent measurements and test results all over the place if you look around. Sean
> |
I see this talk about skin effect and wonder what you are talking about? In my electrical engineering schooling, we were taught that "skin effect" only occurs at much higher frequencies than what is present in an audio signal - we're talking MHz and GHz, not KHz. Do y'all have some new physics that you use to explain this?
-RW- |
Duke: I'm surprised at your recommendation of AP speaker cables. We typically share similar results / observations ( as we both like Magnan's products ), but we are 180* out of step on this one. On top of that, low skin effect on any stranded cable, especially one that uses bare conductors in a braid woven design, is a contradiction in itself. You literally have one bare wire crossing over another bare wire, resulting in what is referred to as "strand jumping". To my ears and those of my brother and ( now Ex ) girlfriend on several different systems, the AP 9's were horrible sounding.
Other than that, the Alpha-Core's "might" work well with the Soundlab's. Most of this would be dependent on what the output impedance of the amplifier being used was. With a low impedance SS amp, no problem. With a high impedance tubed amp, i would look for something else.
The reason for this has to do with power transfer characteristics and the nominal impedances involved. For best results, the nominal impedance of the speaker cables being used should be somewhere between the output impedance of the amplifier and the load impedance of the speakers. That is, IF the amp is stable and capable of driving the speakers without a problem. Otherwise, using speaker cables of a higher nominal impedance and / or a more restricted bandwidth may work better. Sean
> |
To my ears, Alpha core is a little bright. I agree with the Magnan cables. They are the next speaker cables that I will buy. All of my interconnects are Magnan.
Jeff |
Yes, Alpha-Core cables work quite well with SoundLabs. That's one of the cable lines I considered carrying, but ended up with something else (I'm a SoundLab dealer).
A cable that takes the wide, flat, thin, low-skin-effect geometry even further is Magnan Signature. David Magnan designed them using stacked original Quads as his reference. Magnan Signature speaker cables are what I use in my system, and I have sold them to quite a few SoundLab owners. One of those sales was to an AG2 owner.
In a less expensive cable, I have found Analysis Plus Oval 9 to be very good on electrostats. Their low-skin-effect geometry is well suited to this application.
Duke |
