An interesting interconnect experiment
Oct 13, 2001 at 2:04 AM Post #31 of 46
Quote:

Originally posted by Budgie
I know that is the standard take on the "if you need to go long" issue- longer interconnects are better then longer speaker wires, but I wonder if that should be considered to be a situation dependant rule. I think the chance to pick up noise and interferance is much greater in an interconnect. And one of the problems with unbalanced interconnection is that the shield is actually part of the signal path. (one of the reasons balanced connections are used in pro gear).
If you need to go longer for your speaker leads you can always use a heavier gauge cable.

kwkarth- You are right about the feedback issue. I attended a talk at the VSAC show, put on by Linn Olsen, regarding the amplifier/speaker interface, and noise, and back emf getting into the feedback loop is a real problem for high feedback designs, and may contribute significantly to the sound of an amp. One reason why high feedback designs may not sound so good, compared to the low/no fdeedback designs. It was a path I had not really thought about before. Interesting stuff.

Twisting will cause phase changes in the recived signal? I have a hard time visualizing that. Is it wave length dependent? Seems a long wave length would not see the twist and it would become more apparent as the freq gets higher. My understanding is that the reason for the twisting was to get equal noise pickup in both conductors, so the high cmrr of the differential input could ignore it. (the twisting converts the recieved nosie to common mode noise)


Lynn Olsen? I've known Lynn for many years. We both used to work at the same place. We weren't in the same department but we would often run into each other and talk about all things audio. What's he up to these days? I've lost track of him since we left Tektronix.

Sorry, back to cables...

We'll illustrate this using a powerful magnet, a pair of wires, and a very sensitive zero-centered mili-volt meter... The principle will hold true for electromagnetic energy of virtually any frequency. (within reason)

Visualize if you will, two parallel wires hooked to a voltmeter. Now push the magnet towards the wires, and as you do you'll notice the needle on the voltmeter swing in one direction, let's say to the right, as the invisible magnetic lines of force pass through the wire, inducing current in the wire.

If you reverse the direction/motion of the magnet, this time pulling it back away from the wires, thereby reversing the direction of the magnetic lines of force as they pass through the wire, the current induced in the wire will be of opposite polarity as before, and you'll observe the needle on the meter swing in the opposite direction, this time to the left.

Now, stay with me here...

Let's repeat the experiment again, this time reversing the position of the parallel wires, left for right and right for left, but leaving them hooked to the meter as before, such that the twist we've added to the wires by swapping left for right is just next to the meter connection at one end of the pair... Now when you move the magnet by the wires, the needle will move in the opposite direction as before, to the left as the magnet approaches the wires and to the right as we pull it back.

Now, let's move that single twist to the middle of the length of wires, right in front of our magnet. This time as we approach the wire pair with our magnet the current induced in one half of the pair will be of one polarity and the current induced into the other half will opposite, just as we've demonstrated with the entire length of the wire pair. What will the needle on the meter do? As the currents induced in each half of the wire pair, exactly equal but opposite in polarity add together, they sum to zero. Th plus and minus exactly cancelling one another.

Obviously for this to work, the number and strength of magnetic lines of force crossing each half-length of our pair of wires must be exactly equal in each half of the total length of the pair. If one side is stronger than the other, obviously that will show up one way or the other on our meter.

As hopefully you can see, if you have a reasonalbly high number of half twists in the wire, the current unduced in each small length of pair will cancel out as it meets the equal but opposite polarity current induced in the next twist. Simplistically, this is the principle by which a twisted pair of wires "reject" RFI.

I hope this helps. If you still have questions, feel free to ask!!
smily_headphones1.gif
 
Oct 13, 2001 at 2:19 AM Post #32 of 46
So it only works if you have an even number of twists, not an odd number, right? No, I am just kidding. I do see what your saying. Thanks for that. For some reason all I have ever ran into was the common mode noise angle. Now I see why faster twist rates may improve the band width of data cables. But for some reason it still feels weird. Maybe it's me.
 
Oct 13, 2001 at 2:36 AM Post #33 of 46
Ah Ha, here is what is bugging me. Go back to the two wires. As the magnet moves the field will induce a current flow. The direction of the current is dependent on the orientation of the field to the axis of the wire. Seems to me that the current flow will be in the same direction in both wire. There will be a very small differance (i.e. voltage) due to one wire being closer to the magnet then the other (loop area issue) that will result in a meter deflection in our " Meter of the Imagination" (gratuitouse Carl Sagan-ism). Okay, so now we add the twist to eliminate that difference in loop area , and Bam, the meter don't do no stinking deflection. Crap I just spent 15 minutes of hunt and peck typing to get the same results as you suggested. Oh well, now it will stick in me noggin, at least. Whew!
 
Oct 13, 2001 at 5:30 AM Post #34 of 46
Quote:

Originally posted by Budgie
So it only works if you have an even number of twists, not an odd number, right? No, I am just kidding. I do see what your saying. Thanks for that. For some reason all I have ever ran into was the common mode noise angle. Now I see why faster twist rates may improve the band width of data cables. But for some reason it still feels weird. Maybe it's me.


Data rate in cables both UTP and STP is a function of the cable's bandwidth. I suspect in your business you're familiar with TDR? (Time Domain Reflectometry) You no doubt use TDR in your business to characterise cables. A TDR will tell you how long a cable is until you get to an open or short or any change in characteristic impedance of the cable like a kink/crimp. How does it do that? Radar! The TDR send a pulse down the cable. When that pulse hits a change of impedance (think impedance mismatch in audio) all or a portion of the pulse is reflected back. The TDR measures the time it takes for the reflection to come back. The distance is calculated simply because you tell the TDR what the transfer function of the cable is before you begin. i.e. how fast in terms of a percentage of the speed of light signals travel in that particular cable. How do we know? Cable capacitance and resistance per ft. which directly translates to the characteristic impedance of the cable. High capacitance cable with lousy or no shielding? Slow data rates. A "faster" twist would increase the capacitance maybe, but reduce the noise intrusion. Depending upon the dialectric properties of the cable insulation (conductance and thickness) and condutor diameter you could still end up with a very low capacitance cable. A higher noise margin lets you run faster data rates because you have fewer retries from scrambled packets. Anyway, I think I'm getting dizzy from all this twisting!
smily_headphones1.gif
 
Oct 13, 2001 at 5:35 AM Post #35 of 46
Quote:

Originally posted by Budgie
Ah Ha, here is what is bugging me. Go back to the two wires. As the magnet moves the field will induce a current flow. The direction of the current is dependent on the orientation of the field to the axis of the wire. Seems to me that the current flow will be in the same direction in both wire.


Nope. Remember, it takes two to tango. "Current flow" requires a complete circuit. So current flows through the wire (one side then "through" the meter winding and back to the other side of the wire loop. I think my head hurts now.
smily_headphones1.gif
 
Oct 13, 2001 at 4:20 PM Post #36 of 46
Thanks for the time you put in. My understanding is much better now. I do know what a TDR is/does but I have avoided doing data cabling for networks. Too many people allready in that arena. (and it is boring as hell, too)
 
Oct 13, 2001 at 8:16 PM Post #37 of 46
Hello,

You say that:

C = Q / V

V = E * d

C = Q / (E * d)

Then smaller distance between wires will result in higher capacitance. Though note that electric field strength is as much of a factor in this matter. So if I were to make it that steady E field cannot develop then I can decrease the effective capacitance. Q is also a factor. So cables with smaller surface area will have lower capacitance. However, note that smaller surface area will result is lower total conductivity.

It is important to note that this only holds for parallel plate capacitors. Inevitably, parallel cylinders will have smaller capacitance.

What about inductance? Capacitance is one thing but inductance will slow signals and increase phase problems also. Longer wire will result in higher inductance. Also, steady fields should not develop. You have make sure that fields are compensated or even cancelled. I think Braiding is better than straight wire when it comes to that.

What I do are:

1. Shorter distance (lower cap, lower ind)
2. Teflon insulator (lower cap)
3. Braiding
(This is good idea since E field in one braid will interefere with the neighboring braid. So there is no steady field. So lower ind.)

What I want to do are:

1. Silver plated high purity Cu wires.
(electrons will flow mostly close to the surface.)
OR silver wire.
2. Cu sleeves over my cables.
(lower RF pickup)
3. Kick-ass connectors.
(Better contact)

Tomo
 
Oct 13, 2001 at 10:28 PM Post #38 of 46
Quote:

Originally posted by Tomo
Then smaller distance between wires will result in higher capacitance. Though note that electric field strength is as much of a factor in this matter. So if I were to make it that steady E field cannot develop then I can decrease the effective capacitance. Q is also a factor. So cables with smaller surface area will have lower capacitance. However, note that smaller surface area will result is lower total conductivity.

All very true.

It is important to note that this only holds for parallel plate capacitors. Inevitably, parallel cylinders will have smaller capacitance.

All very true.

What about inductance? Capacitance is one thing but inductance will slow signals and increase phase problems also. Longer wire will result in higher inductance. Also, steady fields should not develop. You have make sure that fields are compensated or even cancelled. I think Braiding is better than straight wire when it comes to that.

What I do are:

1. Shorter distance (lower cap, lower ind)
2. Teflon insulator (lower cap)
3. Braiding
(This is good idea since E field in one braid will interefere with the neighboring braid. So there is no steady field. So lower ind.)

What I want to do are:

1. Silver plated high purity Cu wires.
(electrons will flow mostly close to the surface.)
OR silver wire.
2. Cu sleeves over my cables.
(lower RF pickup)
3. Kick-ass connectors.
(Better contact)


Great suggestions!!
 
Oct 14, 2001 at 1:30 AM Post #39 of 46
The next big question (my opinion) is the subject of skin effect. (should probably be a new thread, but what the hell). I have read opinions from different sources, and found that there is disagreement as to the frequency that skin effect becomes noticable. I suck at math so I am totally reliant on smarter brains for this issue. So what your opinion /idea on skin effect.

(So far the idea I am at for interconnects is to use Litz cable in a two conductor twisted configuration. A pain to implement, but it may be worth it)
 
Oct 14, 2001 at 2:41 AM Post #40 of 46
I truly wish I had the time to test and discover all of this stuff for myself but much of what I "know" is from what I've read also. So with that in mind, let me share with you a clip from Robert Harleys book in which he gives a brief discussion about cable construction and behaviour as it relates to audio. Most of what's here I'm sure most everyone on this board is very familiar with, but for the benefit of everyone I thought I would share the section in its entirety rather than edit parts out that most will already know...

Some general rule of thumb info re cables Quote:

from Robert Harley's book "The Complete Guide to High End Audio"

Cables and interconnects are composed of three main elements: the signal conductors, the dielectric, and the terminations. The conductors carry the audio signal; the dielectric is an insulating material between and around the conductors; and the terminations provide connection to audio equipment. These elements are formed into a physical structure called the cable's geometry. Each of these elements-particularly geometry-can affect the cable's sonic characteristics.

Conductors
Conductors are usually made of copper or silver wire. In high-end cables, the copper's purity is important. Copper is sometimes specified as containing some percentage of "pure" copper, with the rest impurities. For example, a certain copper may be 99.997% pure, meaning it has three-thousandths of one percent impurities. These impurities are usually iron, sulfur, antimony, aluminum, and arsenic. Higher-purity copper-99.99997% pure-is called "six nines" copper. Many believe that the purer the copper, the better the sound. Some copper is referred to as OFC, or Oxygen-Free Copper. This is copper from which the oxygen molecules have been removed. It is more proper to call this "oxygen-reduced" copper because it is impossible to remove all the oxygen. In practice, OFC has about 50ppm (parts per million) of oxygen compared to 250ppm of oxygen for normal copper. Reducing the oxygen content retards the formation of copper oxides in the conductor, which can interrupt the copper's physical structure and degrade sound quality.
Another term associated with copper is LC, or Linear Crystal, which describes the copper's structure. Drawn copper has a grain structure that can be thought of as tiny discontinuities in the copper. The signal can be adversely affected by traversing these grains; the grain boundary can act as a tiny circuit, with capacitance, inductance, and a diode effect. Standard copper has about 1500 grains per foot; LC copper has about 70 grains per foot. Fig.11-5 shows the grain structure in copper having 400 grains per foot. Note that the copper isn't isotropic; it looks decidedly different in one direction than the other. All copper made into thin wires exhibits a chevron structure, shown in the photograph of Fig.11-5. This chevron structure may explain why some cables sound different when reversed.

Conductors are made by casting a thick rod, then drawing the copper into a smaller gauge. Another technique-which is rare and expensive-is called "as-cast." This method casts the copper into the final size without the need for drawing.

The highest-quality technique for drawing copper is called "Ohno Continuous Casting" or OCC. OCC copper has one grain in about 700 feet-far less than even LC copper. The audio signal travels through a continuous conductor instead of traversing grain boundaries. Because OCC is a process that can be performed on any purity of copper, not all OCC copper is equal.
The other primary-but less prevalent-conductor material is silver. Silver cables and interconnects are obviously much more expensive to manufacture than copper ones, but silver has some advantages. Although silver's conductivity is only slightly higher than that of copper, silver oxides are less of a problem for audio signals than are copper oxides. Silver conductors are made using the same drawing techniques used in making copper conductors.

The Dielectric
The dielectric is the material surrounding the conductors, and is what gives cables and interconnects some of their bulk. The dielectric material has a large effect on the cable's sound; comparisons of identical conductors and geometry, but with different dielectric materials, demonstrate the dielectric's importance.

Dielectric materials absorb energy, a phenomenon called dielectric absorption. A capacitor works in the same way: a dielectric material between two charged plates stores energy. But in a cable, dielectric absorption can degrade the signal. The energy absorbed by the dielectric is released back into the cable slightly delayed in time-an undesirable condition.

Dielectric materials are chosen to minimize dielectric absorption. Less expensive cables and interconnects use plastic or PVC for the dielectric. Better cables use polyethylene; the best cables are made with polypropylene or even Teflon dielectric. One manufacturer has developed a fibrous material that is mostly air (the best dielectric of all, except for a vacuum) to insulate the conductors within a cable. Other manufacturers inject air in the dielectric to create a foam with high air content. Just as different dielectric materials in capacitors sound different, so too do dielectrics in cables and interconnects.

Terminations
The terminations at the ends of cables and interconnects are part of the transmission path. High-quality terminations are essential to a good-sounding cable. We want a large surface contact between the cable's plug and the component's jack, and high contact pressure between them. RCA plugs will sometimes have a slit in the center pin to improve contact with the jack. This slit is effective only if the slit end of the plug is large enough to be compressed by insertion in the jack. Most high-quality RCA plugs are copper with some brass mixed in to add rigidity. This alloy is plated with nickel, then flashed with gold to prevent oxidation. On some plugs, gold is plated directly to the brass. Other materials for RCA plugs and plating include silver and rhodium.

RCA plugs and loudspeaker cable terminations are soldered or welded to the conductors. Most manufacturers use solder with some silver content. Although solder is poor conductor, the spade lugs are often crimped to the cable first, forming a "cold" weld that forms a gas-tight seal. In the best welding technique, resistance welding, a large current is pulsed through the point where the conductor meets the plug. The resistance causes a small spot to heat, melting the two metals. The melted metals merge into an alloy at the contact point, ensuring good signal transfer. With both welding and soldering, a strain relief inside the plug isolates the electrical contact from physical stress.

Geometry
How all of these elements are arranged constitutes the cable's geometry. Some designers maintain that geometry is the most important factor in cable design-even more important than the conductor material and type.

An example of how a cable's physical structure can affect its performance: simply twisting a pair of conductors around each other instead of running them side by side. Twisting the conductors greatly reduces capacitance and inductance in the cable. Think of the physical structure of two conductors running in parallel, and compare that to the schematic symbol for a capacitor, which is two parallel lines.

This is the grossest example; there are many fine points to cable design. I'll describe some of them here, with the understanding that I'm presenting certain opinions on cable construction, not endorsing a particular method.

Most designers agree that skin effect, and interaction between strands, are the greatest sources of sonic degradation in cables. In a cable with high skin effect, more high-frequency signal flows along the conductor's surface, less through the conductor's center. This occurs in both solid-core and stranded conductors (Fig.11-6). Skin effect changes the cable's characteristics at different depths, causing different frequency ranges of the audio signal to be affected by the cable differently. The musical consequences of skin effect include loss of detail, reduced top-octave air, and truncated soundstage depth.

A technique for battling skin effect is litz construction, which simply means that each strand in a bundle is coated with an insulating material to prevent it from electrically contacting the strands around it. Each small strand within a litz arrangement will have virtually identical electrical properties. Litz strands push skin-effect problems out of the audible range. Because litz strands are so small, many of them bundled together in a random arrangement are required to achieve a sufficient gauge to keep the resistance low.

A problem with stranded cable (if it isn't of litz construction) is a tendency for the signal to jump from strand to strand if the cable is twisted. One strand may be at the outside at a point in the cable, then be on the inside farther down the cable. Because of skin effect, the signal tends to stay toward the outside of the conductor, causing it to traverse strands. Each strand interface acts like a small circuit, with capacitance and a diode effect, much like the grain structure within copper. Individual strands within a conductor bundle can also interact magnetically. Whenever current flows down a conductor, a magnetic field is set up around that conductor. If the current is an alternating-current audio signal, the magnetic field will fluctuate identically. This alternating magnetic field can induce a signal in adjacent conductors (see Appendix B), and thus degrade the sound. Some cable geometries reduce magnetic interaction between strands by arranging them around a center dielectric, which keeps them farther apart. These are just a few of the techniques used by cable designers to make better-sounding cables.

Some cables and interconnects are not merely pieces of wire, but contain electronic components. These cables are identified by "boxes" on one or both cable ends housing the resistors, inductors, and capacitors that form an electrical "network." Pioneered by Music Interface Technologies (MIT), these products are often called terminated cables (Fig.11-7).

According to MIT, some of the audio signal's voltage is stored in the cable's capacitance, and some of the signal's current is stored in the cable's inductance. The amount of energy stored in the cable varies with frequency, which is manifested as tonal and dynamic emphases at those frequencies. Moreover, the energy stored in the cable is released over time, rather than being delivered to the loudspeaker along with the rest of the signal. This frequency-dependent, non-linear energy storage distorts the size and shape of the soundstage, according to MIT.

Terminated cables are designed to prevent such energy storage, and to deliver all the signal to the loudspeaker with the correct phase (timing) relationship. MIT claims that terminated cables are essential to realizing correct bass weight, smooth tonal balance, and a full-sized soundstage with precise image focus. Because terminated cables form a low-pass filter (i.e., they don't start filtering until about 1MHz, well above the audioband), they are useful for connecting very-wide-bandwidth electronics. The Spectral brand of electronics, for example, has bandwidths as high as 3MHz-not because the designer wants to pass those frequencies, but to make the circuits behave better in the audioband. Terminated cables filter those very high frequencies, ensuring that loudspeakers are never fed energy in the megahertz range, and making the system more stable.

There's a lot of hype and just plain misinformation about cables and interconnects. Manufacturers sometimes feel the need to invent technical reasons for why their cables sound better than the competition's. In reality, cable design is largely a black art, with good designs emerging from trial and error (and careful listening). Although certain conductors, dielectrics, and geometries have specific sonic signatures, successful cable designs just can't be described in technical terms. This is why cables should never be chosen on the basis of technical descriptions and specifications.
Nonetheless, some cable and interconnect specifications should be considered in some circumstances. The three relevant specifications are resistance, inductance, and capacitance. (These terms are explained in detail in Appendix B.)

A cable or interconnect's resistance, more properly called DC series resistance, is a measure of how much it opposes the flow of current through it. The unit of resistance is the Ohm. The lower the number of ohms, the lower the cable or interconnect's resistance to current flow. In practice, cable resistance is measured in tenths of ohms. Resistance isn't usually a factor in interconnect performance (except in some of the new non-metallic types), but can affect some loudspeaker cables-particularly thin ones-because of their higher current-carrying requirements.

The sounds of interconnects and loudspeaker cables can be affected by inductance. It is generally thought that the lower the inductance, the better, particularly in loudspeaker cables. Some power amplifiers, however, need to see some inductance to keep them stable; many have an output inductor connected to the loudspeaker binding post (inside the chassis). When considering how much inductance the power amplifier sees, you must add the cable inductance to the loudspeaker's inductance.

Capacitance is an important characteristic of interconnects, particularly when long runs are used, or if the source component has a high output impedance. Interconnect capacitance is specified in picofarads (pF) per foot. What's important isn't the interconnect's intrinsic capacitance, but the total capacitance attached to the source component. For example, 5' of 500pF-per-foot interconnect has the same capacitance as 50' of 50pF-per-foot interconnect. High interconnect capacitance can cause treble rolloff and restricted dynamics. (A full technical discussion of interconnect capacitance is included in Appendix B.)

The interface between a power amplifier and a loudspeaker through a cable is a critical point in a playback system. Unlike interconnects, which carry low-level signals, loudspeaker cables carry much higher voltages and currents. Loudspeaker cables thus react more with the
components they are connected to.

Damping factor is an amplifier's ability to control the woofer's motion after the drive signal has ceased. For example, if you drive a loudspeaker with a bass-drum whack, the woofer's inertia and resonance in the enclosure will cause it to continue moving after the signal has died away. This is a form of distortion that alters the music signal's dynamic envelope. Fortunately, the power amplifier can control the motion; the degree of this control, or damping factor, is expressed as a simple number.

Damping factor is related to the amplifier's output impedance. The lower the output impedance, the higher the damping factor. When you connect a power amplifier and loudspeaker with cable, the cable's resistance decreases the amplifier's effective damping factor. For example, an amplifier's damping factor of 100 may be reduced to 40 by 20' of moderately resistive loudspeaker cable. The result is reduced tightness and control in the bass. Loudspeaker cables should therefore have low resistance and be as short as possible.

One of the most important issues is cable dressing-the way cables are positioned in your system. In an attempt to make their systems look tidy, some audiophiles bundle all the cables together with tie-wraps. But grouping AC cords, interconnects, and digital cables right next to each other can degrade your system's musical performance.

To understand why, consider how a moving-magnet phono cartridge works. The tiny magnets, suspended between coils of wire, are moved back and forth by the modulations in the record groove. The relative motion between the magnetic field and the coils induces (creates) electrical current flow through the coils. When this electrical current flow is amplified and converted into sound by the loudspeakers, we hear it as music.

The same sort of magnetic field around a cable in your rack can induce an unwanted signal in an adjacent cable. Audio signals flowing through a conductor create a magnetic field around that conductor. This magnetic field expands and collapses around the conductor at the same frequency as the audio signal. The magnetic field's expansion and collapse provide the relative motion between the magnetic field and a nearby conductor, inducing an electrical signal in any conductor within that field. This means that AC power cords, which carry 60Hz from the wall to your components, will induce a 60Hz noise in any cable that happens to be nearby. Similarly, a digital interconnect from a CD transport to a digital processor will radiate high-frequency noise (in the megahertz range) into analog interconnects, and even into AC cords. Although you don't hear this contamination as audible noise, it overlays the music with a grainy patina.

That's why the first rule of system setup is to keep AC cords away from interconnects and loudspeaker cables. If they must meet, AC cords and signal conductors (interconnects and speaker cables) should cross at right angles, not run parallel to each other. Crossing cables at right angles minimizes the amount of induced noise.

A simple trick for further reducing interaction between cables is to place 1"-square blocks of Styrofoam between AC cords, interconnects, or speaker cables where they cross. Use the blocks at every cable junction to separate interconnects, speaker cables, and AC cords from each other.
Putting Styrofoam blocks between interconnects at the back of your rack is more of a challenge than laying out cables on the floor. A little glue on the interconnect or power cord will hold the lightweight Styrofoam, provided you're not constantly moving interconnects. Rubber bands can also hold the blocks in place.

Even a small physical separation has a large effect. The "inverse square law" states that the strength of a magnetic field diminishes in proportion to the square of the distance. Consequently, you don't need large distances between cables to isolate them from one another-that's why the 1" Styrofoam blocks are so effective.

Next, avoid stacking two components on one rack shelf. The components will be better isolated from each other electrically, and get better ventilation, when each is given its own shelf. Also pay attention to the positioning of components within the rack. Don't put phono preamplifiers near digital sources (CD players, transports, digital processors), or close to a power amplifier. You can also avoid running a digital interconnect next to analog interconnects by thoughtful component positioning. Take special care with the interconnects that run from your turntable to your preamplifier's phono input; they carry extremely tiny signals that are easily contaminated by radiated noise.

Here are several other tricks to help you keep interconnects and cables from degrading your system's sound:
Because all wire degrades the signal passing through it, the less wire you have in your system, the better. Keep interconnects and loudspeaker cables short.

Keep left and right loudspeaker cables, and left and right interconnects, the same length.

If you have excess cable or interconnect, don't wind it into a neat loop behind the loudspeaker or equipment rack. This will make the cable more inductive and change its characteristics. Instead, drape the cable so it crosses other loops at right angles.

Periodically disconnect all interconnects and loudspeaker cables for cleaning. Oxide builds up on jacks and plugs, interfering with the electrical transfer. Use a contact cleaner (available at most high-end stores). It works. In fact, switching interconnects sometimes cleans the jacks, making the system sound better even though the interconnect may not be intrinsically better.

When connecting and disconnecting RCA plugs, always grip the plug, never the cable. Remember to push the tab when disconnecting XLR plugs.

Ensure tight connection of all RCA plugs, and particularly spade lugs on power amplifiers and loudspeakers. Get lots of contact surface area between the spade lug and post, then tighten down the binding posts.

Avoid sharp bends in cables and interconnects.


 
Oct 14, 2001 at 11:57 PM Post #41 of 46
Kwarth,
Now that is one of the best descriptions of the in's and out's of cables I have ever read. It is unbiased unlike the propaganda used by cable companies to sell their products.
I have often heard this and that about certain types of wire, geometry, etc, but this is all inclusive.
I basicly believe in using pure silver when possible. Some state that silver is "Bright". I think this is simply a reaction from those who are only used to copper conductors and therefore have not listened to music played through silver, which if nothing else, it is a better conductor than copper. So, in my opinion I think many of these listeners are actually hearing the music as it should sound (with silver)without the degradation the a lesser conductor (copper) causes.
Arguements will never cease though, as some prefer certain sounds either from conditioning or because of the economics involved.
This article also explains why silver plated wire can be better than copper alone. It explains so many things.
Thanks,
Dan
 
Oct 15, 2001 at 10:18 PM Post #42 of 46
Did you type all that out? Sheesh! You got a beer (or whatever) coming when we finally meet face to face. Thanks for that info. (The reason I bothered measuring cable capacitance in the first place was my growing belief that cable geometry is the real issue and other factors are less significant.)
 
Oct 15, 2001 at 11:08 PM Post #43 of 46
"Twisting the conductors greatly reduces capacitance
and inductance in the cable. "

Am I missing something? Is this a typo? I
suppose if two wires were perfectly adjacent but
untwisted, the capcitance would not go up, but

"reduce" it!!
 
Oct 16, 2001 at 12:12 AM Post #45 of 46
Quote:

Originally posted by Budgie
That part does not agree with what I experienced when I tried it.


I have to agree with both of you. It may be a typo. Although keep in mind that if you're using an insulator with a very high dialectric constant, at reasonable lengths, the capacitance will be negligable.
 

Users who are viewing this thread

Back
Top