I mount the output inductor off board and try to space them as well as orient them to help minimise interaction.
Wire loops have inductance.
Wire pairs with low loop area have low inductance.
Twisting reducing interactions.
Self does indeed treat coil crosstalk in his book, and he does a pretty good job of it. In particular, he devotes quite a bit of time to coil orientation effects. I don't recall him addressing the issue of crosstalk reduction by the use of more compact coils. I also think that he measured crosstalk by passing a current through one coil and measuring the open-circuit induced voltage in the receiving coil. While this seems entirely reasonable for measuring relative crosstalk effects, I think the approach I described above gives more quantitative real-world results when the shunt R and load are included in the same way as they are in a real amplifier.
Cheers,
Bob
The key job of an author is not to invent, but rather present, explain and organize the material while providing the best possible insight and separating the wheat from the chaff. If you look deeply enough, you can find virtually all of this stuff, but often spread out in many places and often not glued together in the larger context for the application at hand.
Of course, it is nice when an author puts forth new concepts and some new circuits. It is also very helpful when an author looks at something from a new perspective or raises new questions and digs into the answers.
Very little is new under the sun - and that is one reason we provide references to preceding material. Not just to credit those who have gone before us, but to direct the reader to other work that has been done that is relevant, either to let the reader dig deeper or to expose the reader to other points of view. Whether we like to admit it or not, there is a lot of opinion buried in technical books, and different authors have different opinions about many things. The reader who wants to research some of the references can then make up his own mind, or even triangulate to a new insight.
Cheers,
Bob
I think there is something to be said for those who have advocated the use of CAT 5 for speaker cables, wherein all of the twisted pairs are connected in parallel at both ends. Of course, one must consider the resulting cable resistance.
Cheers,
Bob
I agree, yet some make speaker wires a cable loop rather than tightly bundled wire pairs
Cables stir controversy, BUT
While evaluating coils for speakers I have clearly seen a pattern in which types we prefer, and mechanical stability paired with a big cross section for low ohm'ic value always seem to com out on top. Air coils tend to sound better than Cored types.
For big induction values in bass-filters low ohm'ic value is more important than the absence of iron material in the core.
I believe most of these findings are valid for audio cables as well. One must also understand that any capacitative and inductive element that is vibrated generates a signal with the vibration this is the case in filters as well as in wires.
Bob,
I have seen quite a few people talking about using CAT5 cables for speaker cables. At the same time it never is just one cable run but multiples it seems. So do we then take those multiple cable runs and again twist or braid the cables together for maximum rejection of inductive coupling? Is there any advantage to using the CAT6 cables over something like a litz cable of large enough size?
I have seen quite a few people talking about using CAT5 cables for speaker cables. At the same time it never is just one cable run but multiples it seems. So do we then take those multiple cable runs and again twist or braid the cables together for maximum rejection of inductive coupling? Is there any advantage to using the CAT6 cables over something like a litz cable of large enough size?
Good point. I'm not sure what the round-trip resistance of a 3m length of CAT 5 with the four pairs in parallel is. I'm guessing it is higher than desired. Since the wires are already twisted inside, I'm guessing twisting or weaving several CAT 5 cables together might not yield any benefit as compared to just running the cables in parallel. I doubt CAT 6 would make any difference.
Cheers,
Bob
Hi Kindhornman,
That has always escaped me. A nice fine stranded 16 ga is normally more than enough, and going up to 12 ga should satisfy almost anyone. The fine stranded cables are very flexible too. They lay out really nicely. I ran mine under the floor - the ultimate hidden cable run. Also stops my wife from moving the stereo system around.
-Chris
That has always escaped me. A nice fine stranded 16 ga is normally more than enough, and going up to 12 ga should satisfy almost anyone. The fine stranded cables are very flexible too. They lay out really nicely. I ran mine under the floor - the ultimate hidden cable run. Also stops my wife from moving the stereo system around.
-Chris
Chris,
I've had some Kimber cable 8TC wires for years as I was given those for use at a CES show. Now whether there is a discernible difference with a nice soft 14 gauge wire is questionable. There is also the properties of the dialectic covering to consider. I'm not suggesting Teflon or anything like that but some of the cheap wires do seem to have much higher capacitance due to the insulation used.
I've had some Kimber cable 8TC wires for years as I was given those for use at a CES show. Now whether there is a discernible difference with a nice soft 14 gauge wire is questionable. There is also the properties of the dialectic covering to consider. I'm not suggesting Teflon or anything like that but some of the cheap wires do seem to have much higher capacitance due to the insulation used.
The special thing about ethernet cables is that they are cheap because they are mass-produced. The twists are actually tightly controlled so as to reject external fields well. Ordering such a cable custom would be very expensive. The only problem is the wires might be very small. And all this assumes you believe the advantages of ethernet cable for speakers will be audible.
Hi Kindhornman,
I have run a lot of CAT-5 and CAT-6 in my life, but it doesn't cut it for speaker wire. Will it work? Sure it will. But its difficult to work with a few cables going to each speaker. If you could run it under your floor you might remain married. I think the insulation might be Teflon for some high end brands, certainly a "good" kind of plastic.
My own experience has been that hearing a difference between wire only occurs in two situations. Firstly, the original wire was too thin and long for the power levels. Secondly, when the original wire had something wrong with it that created a high resistance connection. It's the connections at each end that make all the difference in the world. You can use the normal metal crimp connectors, and they are about the best. I take a ring terminal and cut a slot in so that the post will fit through. Then it hangs nicely for you as you tighten it up. You end up with roughly 270° of contact area, which is pretty good by anyone's definition. In fact, the ring will deform to fit the tiny surface irregularities. So, coupled with the superior surface to surface contact and the low cost, this is the way to go.
Cheap wire tends to have steel or other metals in it that are stiff. So a supple wire made like a "zip cord", maybe with a spacer in the middle, will give you an excellent connection with low-ish capacitance. This is very hard to beat as long as your gauge is thick enough to present a low impedance compared to the speaker over the length of the run. It doesn't have to cost a lot either.
-Chris
I have run a lot of CAT-5 and CAT-6 in my life, but it doesn't cut it for speaker wire. Will it work? Sure it will. But its difficult to work with a few cables going to each speaker. If you could run it under your floor you might remain married. I think the insulation might be Teflon for some high end brands, certainly a "good" kind of plastic.
My own experience has been that hearing a difference between wire only occurs in two situations. Firstly, the original wire was too thin and long for the power levels. Secondly, when the original wire had something wrong with it that created a high resistance connection. It's the connections at each end that make all the difference in the world. You can use the normal metal crimp connectors, and they are about the best. I take a ring terminal and cut a slot in so that the post will fit through. Then it hangs nicely for you as you tighten it up. You end up with roughly 270° of contact area, which is pretty good by anyone's definition. In fact, the ring will deform to fit the tiny surface irregularities. So, coupled with the superior surface to surface contact and the low cost, this is the way to go.
Cheap wire tends to have steel or other metals in it that are stiff. So a supple wire made like a "zip cord", maybe with a spacer in the middle, will give you an excellent connection with low-ish capacitance. This is very hard to beat as long as your gauge is thick enough to present a low impedance compared to the speaker over the length of the run. It doesn't have to cost a lot either.
-Chris
Last edited:
Hi keantoken,
But the blue and green pairs have a tighter twist than the other pairs (568B), with 568A it would be the blue and orange pairs with the tighter twist. Unless you are running an awful long way with your wire, capacitance shouldn't be a problem. FOr long runs you would normally use a 70 V line with a transformer at each end, maybe some in between to run audio through the entire house as background music.
-Chris
But the blue and green pairs have a tighter twist than the other pairs (568B), with 568A it would be the blue and orange pairs with the tighter twist. Unless you are running an awful long way with your wire, capacitance shouldn't be a problem. FOr long runs you would normally use a 70 V line with a transformer at each end, maybe some in between to run audio through the entire house as background music.
-Chris
Chris,
Some of the worst end terminals were on the Kimber Cables wires, They were a split connector that had some kind of polymer in the middle of two contact areas, no matter what you did you could never keep those things tight. I assume there was some real creep going on with the polymer that would let them loosen up over time. I just cut them up to make them work.
I always thought that the CAT5 cables would only work well at very short lengths if you only used one run per speaker, that wire is very small gauge, i know I've had to fix some cables in the past and they are a pain. So I just don't see how you would get away with less than a couple of pairs of wires to a speaker more than a meter away. I know some talk about the high capacitance of some of the dialectic material some manufacturers use on some cheap zip cord. Some soft Litz wire in a good dialectic material just seems the best thing to use if it isn't to expensive.
Some of the worst end terminals were on the Kimber Cables wires, They were a split connector that had some kind of polymer in the middle of two contact areas, no matter what you did you could never keep those things tight. I assume there was some real creep going on with the polymer that would let them loosen up over time. I just cut them up to make them work.
I always thought that the CAT5 cables would only work well at very short lengths if you only used one run per speaker, that wire is very small gauge, i know I've had to fix some cables in the past and they are a pain. So I just don't see how you would get away with less than a couple of pairs of wires to a speaker more than a meter away. I know some talk about the high capacitance of some of the dialectic material some manufacturers use on some cheap zip cord. Some soft Litz wire in a good dialectic material just seems the best thing to use if it isn't to expensive.
Hi Kindhornman,
Some times things are just too much of a pain to chase theoretical benefits. I just use "speaker wire" and am done with it. I did most of my experiments at a younger age, and now I can rest on what I learned. One question is always, will the difference be worth what the other solution costs?
If I had to connect a speaker and all I had was network cable, I'd use it. But. I would replace it later on with the right stuff.
-Chris
Some times things are just too much of a pain to chase theoretical benefits. I just use "speaker wire" and am done with it. I did most of my experiments at a younger age, and now I can rest on what I learned. One question is always, will the difference be worth what the other solution costs?
If I had to connect a speaker and all I had was network cable, I'd use it. But. I would replace it later on with the right stuff.
-Chris