Sean: It's good to see that you're trying to understand people's views and learn from them. I guess I just interpreted your threads as being very defensive. Perhaps a statement saying something like : "good point, I never thought about it that way" is something we all need to say more.
In regard to your statement of "velocity of propagation", that is only ONE factor of wave propagation, and a simple one at that. I think you are not realizing that to do a complete wave propagation analysis of a wire, it cannot be a static measurement, but a time slice based series of measurements. This, in conjunction with the varying phase vector of the load, it's various characteristic interactions, the associated every changing signal (both in amplitude, frequency and time) all add up to some unbelievable complex calculations (we are talking about having to use a Cray computer here).
To elaborate: since there is not a definitive point in time or reference signal or load (recorded music is the opposite of a static single frequency, or time and amplitude), you then need to capture every millisecond of data, and chart it against the next millisecond. This plot would then have to be repeated for a whole series of loads with associated phase vectors, along with a whole series of dynamic signals, which compose a plethora of frequencies and amplitudes. When you add up all the permutations, you end up with over 100,000,000,000,000 measurement possibilities (and I'm being very conservative with my calculations!). This is why a measurement is so hard, because it isn't just one measurement. The world we live in is dynamic not static, this one of the reasons why Calculus was developed: to have the ability to describe non-linear functions at a specific slice in time.
Many things will be able to be measured and quantified in audio components and wires within 25 years. By then, the world should have access to computers that have processors that can do trillions of calculations per second. With the harnessing or this processing power, many things in dynamic environments will be able to be broken down and analyzed, with an excellent understanding of what is the cause and effect, regarding design iterations of electron carrying and transfer devices. |
Anytime there is an "exchange of ideas", there is "bickering" taking place. Each party is "arguing" or "presenting" their side of the story. This is NOT to say that understanding or learning can't come out of a "heated discussion".
With that in mind, i do try to learn as much as possible. I try to better understand where various points of view are coming from so that i can more easily pick them apart : )
As such, i probably do come across as "confrontational" in some / many of the posts that i make. This probably limits the amount of input that specific threads would receive, so i probably need to shut up a LOT more often than i do. I'll keep this in mind. I know that you didn't say that, but i'm sure that it is true.
As to your comments about velocity of propagation and the waveshaping that takes place in various components, etc... i agree wholeheartedly. Those are the things that are measurable ( with the right equipment ), so i have a hard time understanding why it hasn't already been done. Sean
> |
Just for the record: I don't prescribe to the notion that a cables' sound can be characterized by electrical equations. You are very right that you CANNOT do an experiment with resistors, capacitors and inductors to mimic a cable. One major reason is because the electrical wave propagation characteristic through these passive devices are much different than through any sort of cable. Anybody on this forum that has taken 4 or more years of high level calculus, should be able to tell you that the math involved to understand electrical wave propagation is extremely complex. This is what I was referring to when I wrote the words "calculations".
BTW: Sean you may think there is "bickering" going on here, but I think people are trying to open your mind to new ideas and concepts. Every post should be treated as a measure of learning and only a personal attack only when people don't back up their comments, or make malicious, demeaning statements about you. I've read many of your posts, and you make some extremely good comments, helpful suggestions and provide a great overview on many topics. But please try to learn from others also. This is what I think myself, Karls, bomarc and Bishopwill have been trying to do with our comments regarding your statements. We are not "bickering", we are expanding on your ideas, hypothesizes and assumptions. |
Ehider, you are touching on MANY of the subjects that Jon is trying to cover over at AA in the cable forum. I do NOT attribute all of the differences amongst cables strictly to "skin effect". Then again, if what we were hearing was merely a matter of simplistic electrical equations taking place, we could "duplicate" the sonics of a cable by lumping electrical components together ( resistors, capacitors, inductors, etc...) and duplicating the end impedance via Thevenin's Theory. If you have ever tried doing such an "experiment", such AIN'T the case. There are obviously other things that we aren't measuring that obviously DO affect the interaction between components and various cables.
Outside of all of the "bickering" that goes on here, some of you might want to check out this post. It has TONS of references in it regarding wires and cables, etc... Sean
>
http://www.AudioAsylum.com/audio/cables/messages/40899.html |
I and others have done identical comparisons with cables which we designed that measure the same but have different size conductors. Yes you are right Sean, there is a DEFINITE difference in sound. Here is where you are incorrect though; You are assuming this sonic difference must be associated with "skin effect". I now ask you, why does this difference have to be attributed with "skin effect"? There are MANY other things that could be going on here other than skin effect when you change the amount of conductors and/or conductor size in a cable. The transmisity differential between conductors, the inherent motor effects of different conductors beside each other, and the ratio of conductor to insulator materials, are just three things (off the top of my head) that have changed, in your wire change scenario listed in the above thread.
"Skin effect" is, has been, and always will be a marketing driven explanation for those who don't fully understand the minutiae of electrical theory, regarding electrical wave propagation, within a particular wire design. We are talking about tons of high level math here! I have over 6 years of graduate study math under my belt, and I am still not qualified to understand the calculations of the effects of wire topology differences.
Many years ago everyone assumed that "negative feedback" was a definate bad sonic culprit, so that's what some the high end manufacturers used to "hook" the audiophiles. Many audiophiles bought into this "low feedback" design concept, not realizing that this was most likey not the reason that the particular amplifier sounded different another non low feedback design. Just because I ate carrots every day this year, and was diagnosed with cancer last week, does not mean that carrots cause cancer! Please watch out where you learn your audio theory from. If it's from the audio manufacturers, look out! It took me almost a decade to "unlearn' what I thought I knew about the sonic "cause and effects", and the associated electronic theory to substantiate these cause and effects. |
Sean, interesting test which I haven't done. But I will say that you're assuming way too much here, namely (1) that gauge can't possibly make any difference since the currents are so small (not necessarily so), and (2) that gauge and surface area are the only things being varied (certainly not so). In an earlier post, I postulated that capacitance, inductance, and insulation dielectric properties are all potentially important, and it is a virtual certainty that they are all varying as well in your hypothetical test setup. If you were willing to take the time to measure them all and manage to somehow hold all of them constant throughout the test regardless of wire gauge, then we would be getting somewhere. But that still doesn't rule out gauge as a possible influence. To do that, you would have to start comparing different conductor geometries, while still holding all the above constant (extremely difficult), or compare solid to stranded cable (which introduces a whole new can of worms which would be virtually impossible to neutralize). I don't have the time or the equipment to run such a test, so I can't say. What I can say, based on good sound engineering judgment, is that skin effect is probably very near the bottom of the list when it comes to the relative importance of factors which influence the "sound" of a cable. The Goertz cables, as an obvious example, are designed that way not for skin effect considerations, but to tune the impedance and capacitance characteristics to the designer's preference. |
If "skin effect" has no effect at audio frequencies, try building some IDENTICAL geometry interconnects using identical materials BUT with the only difference being wire gauge. For the sake of comparison, you would have to use solid wire since stranded wire does not come in "micro" gauges.
For specifics, i'm talking about using some teflon jacketed solid wire of various gauges arranged in a twisted pair configuration. Start off small ( for instance, 26 gauge ) and then make a pair that is several gauges heavier ( 20 gauge ) and one other pair that is several gauges heavier than that ( 16 gauge ).
Given the fact that the materials are the same or very similar, the geometries are near identical, etc.. the electrical parameters should be "ballpark". Not exactly the same, but ballpark. The one that i know that you will mention is probably series resistance, but that should be less than 2 ohms total so long as the cable is not more than a few feet. Since the load that the source will see is in effect multiple thousands of ohms, this should have little to no bearing on the results.
Now hook one set up at a time and listen to them. Then swap to the next heaviest, and repeat the swap for the last set. You should instantly notice a difference in tonal balance, high frequency extension, the clarity and quality of high frequencies, etc... between these three cables. Overall, there should be a MARKED difference in treble performance.
If anyone HAS done "testing" like this, PLEASE post your results. Then try to explain why you think you hear the differences that you do. Keep in mind that "current limiting" has NOTHING to do with an interconnect, as we are talking the lowest levels possible. Of course, EE's and anybody else with a "theory" is welcome to join in. Just so long as you know and understand that it is a "theory" that we are discussing. Sean
> |
Correction to my last post: I meant to say: "personal friends with the designer of Kimber's first cable" |
Sean: Karls, Bomarc and Bishopwill have some very good points regarding "skin effect" and other electrical claims. As an Electrical Engineer myself who is personal friends of Kimber's first cable (not Kimber), along being involved with some realistically priced proprietary designs that smoke most of the ultra expensive cables. I can say unequivioquly that you should NEVER believe any of the "electrical theory" which manufacturers write about cable design. Many (but not all) cable manufacturers feel the need to write mumbo jumbo claims, based on "electrical theory" such as "skin effect", just so audiophiles will think their particular design has a unique scientific basis.
There are quite a few decent cables out there, with quite a few more stinkers (one of the worst starts with a "T"). Just remember, that "sizzle sells".
As an industry insider I've learned to be very wary of the so called "electrical theories" of circuits, cables and the like. Once you get to personally know the designers, you realize that most of their marketing material that you read is just that, marketing material! |
I've read so many conflicting claims from manufacturers and dealers and afiles about the supposed merits of solid v stranded, round v oval, thin v fat, (you-fill-in) that I've become convinced--sorry, Sean, and others--that none of you really knows what they hell you are talking about.
Use the ears test. If it sounds good to your ears and you can afford it, buy it. If it sounds good and more expensive stuff sounds better, save up. If more expensive stuff doesn't sound better, spend the cash on more software. Or something.
And have fun.
Will |
Karls, the SHAPE and SURFACE AREA of the wire is what determines if there will be a problem with skin effect. That is why Tara uses an oval wire, as it has more surface area than a round wire. Goertz takes this WAY further and makes the entire conductor "surface area". Since the conductor has no depth to it, there is no time delay or smear as frequency is varied. You get the best of all worlds i.e. large gauge for low series resistance, high current capacity, no smear or time delay from skin effect, etc...
Nordost takes a similar approach but in a different manner with different electrical characteristics. They too use flat conductors to minimize skin effect and increase surface area, etc... Instead of one big conductor though, they use several smaller conductors arranged in a different geometric pattern. This is done to alter the electrical characteristics to meet their specific design goals. Sean
> |
Sean, actually "skin effect" works the opposite way from what you think, if I'm reading your post correctly (and I may not be). That is, at higher and higher frequencies, a solid core wire of a certain gauge will transmit current (i.e., electrons) worse than a bunch of paralleled smaller wires that make up the same gauge. This is because skin effect causes the electrons to travel on or near the surface only, and stranded wire has much more surface area than solid wire of the same net gauge. Again, this really doesn't come into play until WAY above the audio range, at which point all kinds of things are different and solid-core wire has other benefits as well. That said, I'm a big fan of solid-core wire for audio use as well. The problem is that gauge is one of the most critical properties determining wire performance and "sound", and 10 gauge solid core wire is stiff as hell. Thus the preference for stranded conductors or solid-wire Litz windings in most people's setups. |
Bomarc, as to "doing my science homework on on cable manufacturers websites", let me know when scientists know and can explain everything. Until then, i can stop learning all of this "dis-information" and get the REAL "edjamikashun" when all is said and done. Sean
> |
I also agree a high pressure contact withstands corrosion tremenduosly. I have taken apart crimped copper connectors, together for fifty years, and they were bright. The two contacting metals must be of the same atomic structure. If they are not the same, they will start to corrode upon contact. (Here he goes again!)
~~~~Basically metals of dissimilar atomic structures, no matter how minute, have differences of electrical potential by construction. If you ever had an electrical device spark/arc/shortcircuit, there was always a pitting or deposition of metal from one surface to another. Along with the pitting is some oxidization. Warning! Warning! Stagnant portion of brain now being accessed!
~~~~This has been brought to you by the makers of "Space Food Sticks" the snacks of the astronauts.(Anybody remember those things?) My typing finger is tired, my brain is smoking, time to butt out. <(+)><(+)> Thanks AG for puttin up with me!............................ ............... () .....................................
............... ~~
|
Sean: Dont' do your science homework on cable purveyors' Web sites. Karls is dead right about skin effect--it's inconsequential at audio frequencies. If you like solid wire, use solid wire. Just spare us the "scientific" justifications.
As for corrosion, I have my doubts about how important corrosion is on exposed surfaces, which are, by definition, not points of contact. But I agree with other posters who state that a good crimp is critical (and undoubtedly better than a so-so solder job). |
~~~~Crimp, solder, crimp/solder. Differential coeficient of expansion due to minute differences of atomic structures. Iffin day ain't da saym, day mighta bees growin bigguh ata diffint rayt wen da hots is applyd. Dare gos da metle ta metle corntac.(Taken from the smart-alecky manual of thingamabobs)
~~~~Seriously, I will have to try these suggestions. It would be an inexpensive tweek. I love this web site! |
Karls, stranded wire is a poorer conductor at ANY frequency. This has to due to with skin effect, stray capacitance, strand jumping, etc... You will find that the best audio cables and high frequency RF cables all use solid wire for very specific reasons. The ONLY benefit that stranding has is that the cable is more flexible and less brittle.
Having said that, please keep in mind that i do own and use some stranded wire in several of my systems. Less "technically correct" or not, they seem to work better in those specific situations than some of the others that i've tried.
As to trying to insulate the excess exposed cables from corroding, you can easily slide shrink wrap, electrical tape, wire sealant, liquid tape, etc... up over the cable and receptacle end of the spade, banana, pin, etc... This might not make it airtight, but it will surely slow down the process. Sean
> |
As the 3 posts above mention, you are going to have some exposed copper somewhere which is going to oxidize. I asked Ray Kimber about bare wire connections at Audioasylum (I use 8TC) and his response was that a number of the Kimber cables were actually designed to be used with bare wire connections (the 8TC is a a little fat for that but I'm able to do it at my speakers and would do it at my amp if I could). He suggested just to tighten down the connection and not worry about it. After all, if you tighten it down and get a solid connection, it's not going to oxidize where the connection is. |
Sean, first of all, it would be tough to tin all the way back under the insulation so that no copper is exposed directly to the air, plus it kind of makes a mess, plus then you just end up with corroded solder rather than corroded copper, and a solid conductor to boot, and there goes your skin effect again. Second, "skin effect" is one of those things that is thrown around a lot but few people understand the physics behind it. Without going too deep, fundamentally it is a phenomenon that is frequency-dependent and only comes into play at frequencies far above the audible range (we're talking MHz range). Some of the worst cables I've ever heard were the original Monster Cable with its extremely finely stranded braided conductors, so clearly there are other factors at work here. I would place far more importance on other factors, such as good solid connections, and cable capacitance, inductance, insulation dielectric properties, and gauge. |
Karls, you state that all of the exposed copper is oxidized EXCEPT for where the actual crimp is making connection. Since "skin effect" is a common term used in wire / cable conversations, don't you think it would be better to treat and seal the entire stripped and exposed area than to have corroded copper /good connection at the crimp / corroded copper at the end with all of that being in the signal path ? After all, once something begins to corrode, that corrosion tends to "wick" into other areas. Kind of like "rust" on a metal surface. It starts small and gradually works its way into everything. Sean
> |
I don't like using bare wire because copper oxidizes immediately upon exposure to air, and continues to oxidize further as time goes on. The same is true of silver, and of all solders. Best to crimp to gold-plated spades which gives a true high-pressure metal-to-metal interface. (If using a proper dedicated crimping tool, that is. A pair of pliers from your toolbox isn't worth squat.) Gold is not as good a conductor as copper but is vastly more corrosion-resistant, so in the real world it is a lot better. I don't believe that solder is necessary if the crimp is done properly-- it's like wearing both a belt and suspenders:). I have disassembled good crimped connectors that were many years old and the copper in the crimped area is as bright as the day it was made, while everything else is dark brown. But on the other hand, soldering after the crimp is done probably doesn't hurt anything either, and if it lets you sleep better at night, then do it. |
Fbi, sorry, I can't remember whether I cut off the tinned ends and started with bare wire to solder on the spades. But would it matter? It got plenty hot and any tinning that might have been there would have been melted into the new solder, wouldn't it? |
Good post Sean. I'm always intrigued by the concept that soldering connectors (on interconnects or speaker wires) is going to degrade the sound. After all there are a gazillion solder joints on the PCB of each component, including the power transistors in the amp, and the speaker binding posts. Fusing the metals together is obviously better, but beyond the means of most of us.
PS I also keep some electrical cleaner handy to wipe off flux residue (brown gunk) which is messy and can be damaging over the longer term. |
Bare wire is SUPPOSEDLY best. This takes into consideration that the metals being used ( wire material and speaker terminal i.e. binding post, spring terminal ) don't react to each other. Dissimilar metals tend to corrode quickly. To help protect the wire from corroding, you might treat the wire and the terminal with something like Caig Pro Gold Liquid. This supposedly cleans and treats the metals, minimizing oxidation.
As to using some type of connector, a "real" crimp is supposedly superior to soldering the connector on. This would be "preferred method #2". Soldering the connector on is third in terms of a good connection. Crimping and soldering is supposedly fourth. I take this from information presented by Jon Risch on AA and others that are pretty well versed in things of this nature.
Having said that, i prefer to crimp and then solder any type of connection. You have to make sure that all the connecting surfaces have been properly prepped and cleaned, crimp the connection using a REAL crimping tool, use good quality solder, allow the solder to completely flow into the joint, let the joint set up and cool ON ITS' OWN ( no blowing on it to help it along ) before moving or disturbing it, etc...
Here's why i do this: crimping makes a metal to metal bond between the connector and wire. This assures low resistance. Soldering the connection seals the joint from oxidation. It also fills in any gaps and increases the surface area that is making contact. All bases should be covered if done properly.
The argument against soldering is that most solder is not very "pure" or a great conductor on its' own. That is why i crimp first, as this makes the basic connection and does not rely on the solder itself. Besides that, i see NO "theoretical" losses since there are HUNDREDS or THOUSANDS of solder joints in the signal chain. If someone was truly worried about the poor conductivity of solder, i would HIGHLY suggest re-flowing all of the connections on the circuit boards of each component using a high grade solder.
If you are not used to soldering, always use a THIN solder. You can always feed more as it melts and is needed. On the other hand, using a large diameter solder can tend to cool the solder joint / solder tip too quickly. This can end up in a high resistance / low strength connection that look poor and is not secure. Some good solders that i have used are Wonder Solder and Cardas. I find that Wonder melts easier and flows better. The Cardas requires higher heat and a little more attention. Supposedly, Jena Labs states that Cardas is the best sounding solder ( in their opinion ). Silver Bearing solder is MUCH harder to work with and not for beginners by any means. Common "generic" solder such as Radio Shack will work, but tend to degrade much more drastically over time. It becomes brittle and can become resistive as it gets older. Kester is a good alternative to this without costing an arm and a leg. The "no residue" solders that i have tried worked like junk, so skip them.
I think that much of the differences in the various formulations are not so much the metals used to make the solder, but the actual rosin. Keep in mind that you should never use Acid core for electronics unless you want a big mess with potential damage to components. Hope this helps. If you have further questions, try doing a search over at AA. Sean
> |
I really haven't compared tinned (soldered) wire versus bananas versus spades for sound quality. I have had difficulty in fitting the tinned wires through the holes in the speaker terminals, though.
Sugarbrie ... to be clear I wasn't vouching for the audio quality of soldered wire ... just letting you know how to solder large pieces of wire (where getting the wire hot enough can be a challenge, especially with a small soldering iron).
With 5 cats and an infant roaming the house mechanical integrety rates high for me, so if the wire's too fat to go through the holes in the speaker terminals then I put spades on (Soldered just like Bob said to do) and do them up really tight. |
First crimp your spades onto bare clean copper. Then if you wish, apply some solder to additionally seal the crimp connection from oxidizing, but don't solder onto the lugs' flat connecting surface where it meets the binding post. |
Tom,Spade lugs on the bare wire or on the tinned ends? |
I was advised by the person who upgraded my Qud ESL 63's to use bare speaker cable wires, tinned. Then I was advised by another expert to use spade lugs instead, which I did. The spades made the system sound enormously better. I'm sure this is just one man's experience, and may not carry over, but for what it's worth, I thought I'd throw it in. I used good solder too, and then mediocre spade lugs at first. |
First spread some flux on the wood panel that you will be using for soldering (some small plywood piece should work)
with hot iron melt the flux and dip(or thread through) your bare wire through the flux equally spreading it over the wire surface with the hot iron.
Than(if you're using a RadioShack silver solder) cut approximately 1 ring of solder and melt it inside the spreaded flux and thread through the bare wire equally spreading solder over the bare wire surface.
Define for yourself how often you need to clean soldered ends(and sometimes binding posts as well) simply by checking it's surface once in a while starting from once per week. |
Put some solder on the iron first and use the solder blob to heat the wire (better heat transfer ... faster results). When the solder begins to flow on the wire add more solder until the wire is completely coated. You'll know when the wire's hot enough because surface tension will draw the solder onto the wire.
To clean I would just wipe with electrical cleaner, or rarely just resolder with some fresh solder. I wouldn't cut the ends off .... but then I'm a cheapskate. |
I am not a solder expert. I should heat the wire with the iron and apply the solder to the heated wire, correct??
Does cleaning them often just mean cleaning the soldered ends, or are you guys cutting them off and re-soldering?
|
|
Yes Sugar, indeed they can and that's what I'm using. I soldered them with RadioShack silver solder and threaded them through the binding posts. Still you will have to clean it often but with much better convenience. |
When using bare wire ends, I "tin" the ends using a lead-free, silver solder (this solder is available from Audio Advisor, as well as electronics supply outlets and Radio Shack). |
