Jneutron's fast rise time square waves contain what kind of frequencies? The idea is that if a cable can be made pure resistive and dispersion free into the megahertz it shouldn't have any inductive or capacitive effects (including non linear phase shift) at 20k. Whether that's overkill for audio is another question. But if you're determined to send square waves down it and don't want reflections it's what you've got to do.
Buy it at Blue Jeans Cable , read their info, no BS and no snake oil salesman.
Speaker Cable:
Speaker cable is a bit different from a lot of the interconnect cables we handle, in several respects. Because speakers are driven at low impedance (typically 4 or 8 ohms) and high current, speaker cables are, for all practical purposes, immune from interference from EMI or RFI, so shielding isn't required. The low impedance of the circuit, meanwhile, makes capacitance, which can be an issue in high-impedance line or microphone-level connections practically irrelevant. The biggest issue in speaker cables, from the point of view of sound quality, is simply conductivity; the lower the resistance of the cable, the lower the contribution of the speaker cable's resistance to the damping factor, and the flatter the frequency response will be. While one can spend thousands of dollars on exotic speaker cable, in the end analysis, it's the sheer conductivity of the cable, and (barring a really odd design, which may introduce various undesirable effects) little else that matters. The answer to keeping conductivity high is simple: the larger the wire, the lower the resistance, and the higher the conductivity. We offer a few alternatives in large-gauge speaker cable, either raw or terminated
Speaker Cable:
Speaker cable is a bit different from a lot of the interconnect cables we handle, in several respects. Because speakers are driven at low impedance (typically 4 or 8 ohms) and high current, speaker cables are, for all practical purposes, immune from interference from EMI or RFI, so shielding isn't required. The low impedance of the circuit, meanwhile, makes capacitance, which can be an issue in high-impedance line or microphone-level connections practically irrelevant. The biggest issue in speaker cables, from the point of view of sound quality, is simply conductivity; the lower the resistance of the cable, the lower the contribution of the speaker cable's resistance to the damping factor, and the flatter the frequency response will be. While one can spend thousands of dollars on exotic speaker cable, in the end analysis, it's the sheer conductivity of the cable, and (barring a really odd design, which may introduce various undesirable effects) little else that matters. The answer to keeping conductivity high is simple: the larger the wire, the lower the resistance, and the higher the conductivity. We offer a few alternatives in large-gauge speaker cable, either raw or terminated
I haven't read the whole thread so this point may have been made already but given that your system is active you may have another option. I can't recall when but years ago I asked about the relative merits of placing amps close to the speaker they were driving in an active system. I got a very forceful reply stating that it was much better to have very short speaker cables and comparatively long small signal (shielded) cables from the source/electronic cross-over to the Power amps. This may not be an option for you depending on the physical structure of the amps but it might be worth considering.......
Cheers Jonathan
Cheers Jonathan
Which actually has higher inductance than if you twist them. Inductance is about loop area. If it's short it won't matter. When you rack up a couple hundred uH you will for sure.
Voltage drop over the mains power grid is primarily due to inductance. At 60 Hz. That's what long wires spaced far apart does to you.
My analysis was done using step response in order to determine the fastest system response possible using t-lines of arbitrary characteristic impedance and arbitrary load resistance. At fast timing scales, it shows how the propagation speed comes into play, and when you back off to audio speeds, it duplicates EXACTLY the bog standard LRC lumped element model.
It shows that extreme mismatches common with zip cable and 4 ohm loads will have settling times well into the tens of microseconds. And it is verified using actual measurements anybody who chooses to can DUPLICATE, as well as various simulations which also confirm the accuracy. I make no claims as to audibility, but point out that humans localization sensitivity has been verified by researchers as measurable down to the sub two microsecond range.
My square wave example was an actual need to drive a 4 ohm resistive heater located in a cryostat at 4.5 kelvin (liquid helium), in a 9 tesla field(IIRC) to quench a superconducting sextupole. I designed a 4 ohm stripline half inch wide with 1 mil kapton insulation to get to the outside room temperature, a 50 ohm stripline to monitor the actual voltage in the cryostat, and drove the load using a 15 foot long cable made with six runs of cat5e paralleled stripe to stripe and solid to solid.
There were no reflections, no ringing, and rise time at the load was exactly what the amplifier was doing at the cable start, delayed by the propagation speed of the cable. This extreme case proved the accuracy of my model for applications I normally do every day. Oh, and I was using a Tigersaurus for the drive signal. (As a side note, all normal materials lose basically all heat capacity at 4.5 Kelvin, so the thermal propagation speed, or thermal dispersion speed, becomes fast enough that we needed to reach into the low microsecond realm to actually measure system response to a simulated mega joule radiation fan with 190 nS timescales.)
For normal humans (I make no claims about myself
..), my analysis shows what the amplifier sees if you go very low z cable (high capacitance) with a load that decouples below the amplifier's unity gain point. And, all speaker drivers unload at higher frequencies. Hence, the zobel.
Jn
It shows that extreme mismatches common with zip cable and 4 ohm loads will have settling times well into the tens of microseconds. And it is verified using actual measurements anybody who chooses to can DUPLICATE, as well as various simulations which also confirm the accuracy. I make no claims as to audibility, but point out that humans localization sensitivity has been verified by researchers as measurable down to the sub two microsecond range.
My square wave example was an actual need to drive a 4 ohm resistive heater located in a cryostat at 4.5 kelvin (liquid helium), in a 9 tesla field(IIRC) to quench a superconducting sextupole. I designed a 4 ohm stripline half inch wide with 1 mil kapton insulation to get to the outside room temperature, a 50 ohm stripline to monitor the actual voltage in the cryostat, and drove the load using a 15 foot long cable made with six runs of cat5e paralleled stripe to stripe and solid to solid.
There were no reflections, no ringing, and rise time at the load was exactly what the amplifier was doing at the cable start, delayed by the propagation speed of the cable. This extreme case proved the accuracy of my model for applications I normally do every day. Oh, and I was using a Tigersaurus for the drive signal. (As a side note, all normal materials lose basically all heat capacity at 4.5 Kelvin, so the thermal propagation speed, or thermal dispersion speed, becomes fast enough that we needed to reach into the low microsecond realm to actually measure system response to a simulated mega joule radiation fan with 190 nS timescales.)
For normal humans (I make no claims about myself
..), my analysis shows what the amplifier sees if you go very low z cable (high capacitance) with a load that decouples below the amplifier's unity gain point. And, all speaker drivers unload at higher frequencies. Hence, the zobel.Jn
Last edited:
How can one have an "argument" if one does not know what to disagree with??
You've constantly fallen back into the characteristic impedance of a cable at audio frequency, seemingly without following the analysis to it's completion.
Your statements about inclusion of R and G are accurate, but are unimportant to the high speed analysis. And you ignore the fact that the normal LCR model and the t-line model totally agree. Also, actual measurements of lumped and continuous on the bench in the lab also agree.
So, no, there is no argument..just a different POV.
jn
You've constantly fallen back into the characteristic impedance of a cable at audio frequency, seemingly without following the analysis to it's completion.
Your statements about inclusion of R and G are accurate, but are unimportant to the high speed analysis. And you ignore the fact that the normal LCR model and the t-line model totally agree. Also, actual measurements of lumped and continuous on the bench in the lab also agree.
So, no, there is no argument..just a different POV.
jn
Just saying hi, watching the usual fun. DF as jn says it would be good to consider that the exact characteristic impedance equation is not really relevant at tiny fractions of a wavelength. For short lines the energy stored is much larger than the energy lost. Furthermore I spent hours of my time demonstrating that t-line and RLC by SPICE simulation converge on exactly the same answer.
Last edited:
Easy. Jan Nordmark, 1972 is first one I can think of.
Greisinger also discussed it, don't recall his numbers but they were really down there as well. For some reason, when I think his name I think 11, 11, 11.. Maybe it was publication date, or date on his website?
I do recall he spoke about sitting in the audience of a concert with four instruments (string I think) side by side, and instead of just listening to the performance, was calculating the amount of interaural delay necessary to discern the right-left positioning of the instruments. Geeky. (sounds like something I would do if I could do trig).
This localization stuff is at least two computers ago, I have to look for it..
jn
Greisinger also discussed it, don't recall his numbers but they were really down there as well. For some reason, when I think his name I think 11, 11, 11.. Maybe it was publication date, or date on his website?
I do recall he spoke about sitting in the audience of a concert with four instruments (string I think) side by side, and instead of just listening to the performance, was calculating the amount of interaural delay necessary to discern the right-left positioning of the instruments. Geeky. (sounds like something I would do if I could do trig).
This localization stuff is at least two computers ago, I have to look for it..
jn
Last edited:
We're still talking speakers, right??
For absolute positioning, I absolutely agree.
For incremental relative positioning of specific sounds embedded in a stereo soundfield, where each specific sound has been assigned a specific horizontal location, no vice is needed.
jn
Last edited:
Which is exactly what would be expected. At first I thought jn was denying this, which surprised me, but then it became clear that he was not - that was when I became unable to pin down exactly what was the point at issue. Anyway, I am not going to re-run the discussion. Attempts to draw me in will fail.scott wurcer said:
In point of fact, I have always stated that they produce exactly the same results. You thinking otherwise was an error on your part. I suspect you were blinded by my ignoring R/G and sticking with LC only.
Using the calculation sqr(L/C) always produces the correct impedance that a zobel needs to properly terminate any speaker cable. But is only required if the cable capacitance is obnoxious.
Cool.
jn
The simple answer is to just use Canare 4S11. It's engineered well, and if something is wrong with it, something it probably wrong with your stereo. And everyone can afford it at $1.19/ft.
But of coarse you could listen to DF96 and assume all the properties are meaningless... capacitance, why measure it ever again? Propagation cannot be why teflon is so preferred for insulation... or it's low capacitance, or it's thermal properties. It's just fooey.
But of coarse you could listen to DF96 and assume all the properties are meaningless... capacitance, why measure it ever again? Propagation cannot be why teflon is so preferred for insulation... or it's low capacitance, or it's thermal properties. It's just fooey.
To actually reply to the OP's question:
I like (and use) Mogami W3103 for most ''critical'' applications.
For ''general'' use and around the shop,great quality at a very reasonable price,available in many gauges and from amazon:
Kord Speaker Wire
I like (and use) Mogami W3103 for most ''critical'' applications.
For ''general'' use and around the shop,great quality at a very reasonable price,available in many gauges and from amazon:
Kord Speaker Wire