What wire gauge is it (AWG gauge number)? That's the main thing that makes most (virtually all) of the difference, if any, between different wire for speaker use. If it's AWG 16 or bigger (a smaller number), you're fine. Even AWG 18 or 20 would probably be okay.
Learn Ohm's Law. Just using DC resistance (so no complex numbers) is a good enough approximation for speaker cable. If the wire has one ohm resistance (both ways total, as they are effectively in series), which is more than any reasonable speaker cable would have, and you have an 8 ohm speaker, then the speaker gets 8/(1+8) or 8/9ths of the voltage from the amplifier, and the cable drops the other 1/9th. That's a fraction of a dB, and you're not likely to hear that being any different from a perfect cable. But feel free to compare different wire anyway.I'm burning through that thread right now. A lot of that is foreign talk to me. I'm stopping at each bullet point in the thread, and making sure that I understand what's being said.
I'm a mathematician/engineer, myself, but 22 years in a field that has absolutely no electrical application, leaves me a bit clueless on the subject.
If I ha a dollar for every one who ever pitched a fit, saying I'm violating code and gonna get sued by chopping the ends off extension cords, putting speakons on them and using them in my dj rig I'd be rich enough to afford some audiophile speaker cables. Or a few dozen pre-made double speakon cables at two dollars a foot. No one's ever gotten electrocuted as a result and no one ever will. Unless they're stupid/drunk enough to put the end in their mouth while the bass is thumping.
In my home audio rig, three strands of 16/2 lamp cord braided together. 12 foot run, rather difficult 4 ohm load (Zmin about 2.5 ohms). At least the cable is white not orange so it matches the baseboard.
It's great if the yunguns can get interested, nice work.
Cat 5e is 100 ohm per twisted pair. If you take all the solids together, and all the stripes together in one cable, you create a run with a characteristic impedance of 25 ohms. If you do this with 6 full cables all in parallel, you get 4 ohms. I did this to drive a 4 ohm load with a 10 microsecond rise time 40 volt square wave, worked perfectly.
Whether it sound different is debated hotly everywhere. However, the cost is capacitance and amp stability. If the amp is very high bandwidth, it may oscillate as a result. If you want a low z cable, put a zobel at the speaker that matches the cable z, that will protect the amp.
Jn
I think it's actually 14 AWG, but I'd have to measure it, to be sure. It was sold generically as "in-wall speaker wire" at Best Buy. Looks pretty. Like something I would have bought based on "dudeness" back when I was 21.
I couldn't agree more, and thank you.
I have read a bit on this subject. Mine is a Class D. Isn't that a characteristic of mainly tube amps?
The issue is the phase margin of the negative feedback.
If the amp has more than unity gain where the cable capacitance has reduced phase margin to zero, the amp will oscillate. I would think that a class D has already been designed so that gain at high frequency is limited so this condition won't occur. But ss amps certainly can be too "hot" for high capacitance cables.
Jn
If the amp has more than unity gain where the cable capacitance has reduced phase margin to zero, the amp will oscillate. I would think that a class D has already been designed so that gain at high frequency is limited so this condition won't occur. But ss amps certainly can be too "hot" for high capacitance cables.
Jn
Cobra/Polk speaker cable was responsibe for the death of many an amp in the latter half of the ‘70s.
dave
I've tried understanding this concept of capacitance in the cable, but it escapes me. A complete foreign subject.
.
Capacitance, as I understand it, is the ability of a an object to store an electrical charge. How do we obtain capacitance in a series of wires? Is it due to the geometry? (when woven/braided together) Or is it due to the small cross-sectional area of the wire(s) times the length?
.
Capacitance, as I understand it, is the ability of a an object to store an electrical charge. How do we obtain capacitance in a series of wires? Is it due to the geometry? (when woven/braided together) Or is it due to the small cross-sectional area of the wire(s) times the length?
It is geometry and materials.
For a wire pair, the product of C times L equals the pemittivity of the insulator times a constant.
For C in pf per ft, L in nH per foot, it is LC = 1034 times relative dielectric coefficient times a constant. Most solid plastics run about 3.
For twisted pairs, the constant is roughly two to 3. For coax, it is 1. The more pairs you use in parallel, the closer you get to the constant of 1.
Jn
For a wire pair, the product of C times L equals the pemittivity of the insulator times a constant.
For C in pf per ft, L in nH per foot, it is LC = 1034 times relative dielectric coefficient times a constant. Most solid plastics run about 3.
For twisted pairs, the constant is roughly two to 3. For coax, it is 1. The more pairs you use in parallel, the closer you get to the constant of 1.
Jn
In pure engineering terms the insulation for a speaker cable does not matter, as long as it is a reasonably good insulator and not too unusual in any sense. However, people who write recipes for audio cables have already decided that engineering is not their thing so they are free to invent good stories about insulation.
Yes, single objects can store an electrical charge on their surface, classic example is High School Physics study of spherical capacitors
An externally hosted image should be here but it was not working when we last tested it.
but "modern" capacitors (think from 1745 on) are made out of two metallic objects, separated some distance.
Textbook example is two parallel metal plates, simply because calculation is straightforward, but metal conductors in a speaker cable are *also* plates of a capacitor.
Calculation is more complicated but it may simply be *measured* and we get ab result in pF/meter .
Plain parallel cable is best (lowest capacitance) , and *some* "Audiophile" rated cable (tightly twisted or braided) is absolute worst, big time.
As mentioned above, some were even able to kill some amplifiers.
Another boring example of "you can´t actually beat Physics Laws" .
FEP has a dielectric coefficient of about 2, where the others are around 3. It does provide 2/3rds of the capacitance if used in coax, probably a smaller reduction with twisted pairs.
If you make a cable using 24 pairs of cat5 conductors, that could make a difference to an amplifier of high bandwidth. I suspect your class D won't care.
As to types of insulation for cat5e.. there's fire rating, bendability, temperature, various things that some insulations are better at than others. Also, Teflon has a tendency to creep, so if it stressed, over time it could short.
jn
I get that. I was being a bit rhetorical.
Physics tells us that when we gain one desirable property, we almost always trade it for some other. (there are no perfect designs, only reasonable compromises) So from a "purely engineering" perspective, I was just looking for the reason the insulation seems to matter. Because there is clearly some science to wrapping wires, and there were several recipes that I've read that indicate that material does matter. One write-up was conducted against many different types of wires and insulations, with actual empirical evidence - not just subjectivity - and the final determination was that the insulation type mattered. It just didn't say exactly why. But you have clearly indicated that the materials do possess inherent differences for such an application.
Notice that this application has nothing to do with audio speaker cables.
The 100 ohm and 25 ohm values are the Radio Frequency Characteristic Impedance values. These only kick in at (guess what?) radio frequencies.
But you can make speaker cables this way. Use two, three or more Cat3, Cat5 or Cat6 cables. Tape the cables together, connect all the solids to one terminal and the stripes to the other.
I get that a lot from those who seem only to, well I won't go there.
However...
The L, C, dielectric times permeability, the energy relationships..all are totally valid from DC to daylight. Every single analysis I've provided have never been disputed, just the standard "oh it's rf".
What is generally ignored are the conclusions. If you make a cable with an impedance of 4 or 8 ohms, my equations provide the exact capacitance, the exact inductance, exactly what the amplifier sees, exactly what the phase margin does, exactly what the bode plot looks like, exactly what the cable energy relationships look like, exactly how to compensate these issues by the use of a zobel.
In other words, I've provided exact information to the table. Engineering analysis, plain and simple. If you or anyone else feels the need to question my equations, analysis, do so with facts and rigor. Not generic unsupported statements.
If and when you and others are prepared to bring something other than floobydust, Google based "fluff" to the table, I will be happy to discuss. In other words, bring some valid engineering to the discussion. I have, and quite honestly, am hoping that others would..
"Rant off"
Speedy, long time no talk. How is it going? I've spent a while recovering, but am starting to get back into the swing of things, both job and personal. Sorry bout the vent, happy to dialog wit you.
Jn
Last edited: