BACKGROUND: There is always discussion on how to ground a shield. The answer lies in what the
worst-case noise situation is. It isn’t always the same answer. Do you leave one end open or
ground both
ends? If you ground only one ends, which end?
We can look at how the interference behaves to answer these questions and they follow understood electrical properties within swept frequency regions.
BODY: The first step is to understand what a shield is doing, and how. In the simplest terms, a shield creates two separate “electrical” environments, one on each side of the shield. One side is measured as a RATIO of the field’s intensity relative to the other side, in dB.
The shields we are working with are ONLY effective with an EM wave that is predominantly “E” field in nature versus “B” field, or magnetic. Magnetic fields are not shielded or blocked with conductive shields, but need a shield material that blocks magnetic flux lines or a low permeability material…think “material a magnet will stick to.” Those kinds of material allow magnetic flux lines an easier path than through air. We can capture and re-route the flux lines in low permeability materials. For this discussion, we will look at mostly electric field shields, stuff that conducts electricity.
What is an ideal electrical shield? It is a shield that 100% blocks electrical energy at the surface of the shield,
and
that has infinitely low resistance. Shields don’t have infinitely low resistance and they block electrical
energy at differing impedance based on the skin depth of the shield at a specific frequency.
We can measure the effectiveness of a shield across frequency with a transfer impedance plot. This is measured in milli-ohm/meter. It describes the resistance we can expect a shield to have, and thus the ratio of the fields energy in the shield based on the CURRENT the resistance causes to flow at that frequency…and it is not linear.
The graph below shows the frequency-dependent nature of a shield. A perfect shield would have NO RESISTANCE and both ends would be identical and thus seem like a SINGLE point of reference to a flow of current. Since we have zero resistance across the shield, we can’t have current flow caused by the shields. We CAN have current flow between the two points connected at the ends of the shield. In a “perfect” world the GROUND at both ends is the same potential and thus, an ideal shield has ZERO current flow, and is a measure of the POTENTIAL on one side of the shield relative to the other, in dB.
Since we don’t have a perfect world, Transfer Impedance describes what to expect at frequencies based on the shield’s impedance, and how that shield resistance creates a CURRENT flow and thus a voltage (shield resistance times the shield’s impedance = a voltage). When we have different resistances at each end of a shield we have current flow.
Shield Type
5 MHz
10MHz
50MHz
100MHz
500MHz
Bonded Foil +60% braid
20
15
11
20
50
Tri-Shield+60%Braid
3
2
0.8
2
12
Quad Shield 60% +40% Braid
2
0.8
0.2
0.2
10
Tri-Shield+80% Braid
1
0.6
0.1
0.2
2
Bonded Foil +95% Braid
1
0.5
0.08
0.09
1
The chart below is what JUST the shield impedance looks like for a set length of cable at lower frequencies. We see the same non-linear behavior of shield and frequencies.
A SEED (Shield Effectiveness Evaluation Device per IEC 61196-1) test shows the dB relationship to a Lower Shield Impedance. Series number 5 with a 95% coverage 45-degree braid and Duofoil tape is clearly superior.
Okay, we can see a shield is not perfect, and not linear. So how does this say what to do with each end of a shield? We have to weigh the CHOICE of HOW the shield WORKS to decide our fate.
– If you have ideal grounds and meet IEEE bus bar grounding (see the 568C.2 or later grounding specifications) limits it means BOTH ends can be grounded and the shield current will inductively couple less interference than the shield ATTENUATES through its material composition.
– If we have severe ground differential, we can induce a strong current in the shield that CAN, if the shield’s resistance is higher, induce noise into the core that is WORSE than if we disconnect one end. We convert our shield to an antenna, not a shield!
– An antenna does NOT create two separate environments between them with the ratio of one
measured to the other. One end of the antenna is infinite impedance (the open end) with the other end at ground.
The antenna “wire” is as close to zero impedance as possible in order to NOT
attenuate the antenna’s signal going to ground. The signal won’t go to the open end, but seeks the lowest potential in the circuit.
We trade the noise caused by a poor ground resistance potential between shield ends for the induced noise in an antenna’s wire parallel to the signal wires that induces a voltage based on the antenna resistance. In an antenna type ground, it is best to ground the SEND end, as the SIGNAL on the internal wires is as LARGE as possible relative to the antenna signal, improving the signal to noise between the two.
An often-ignored aspect of shields is HOW to ground one at lower frequency versus RF. There is a big difference and again, it is based on the shield characteristics at each frequency.
The charts below are derived from TWO slightly differing MODELS of RF shield inductive reactance resistance. I have this paper for those interested. But, the data is the same message in that as frequencies increase, the shield reactance goes UP. This necessitates a FULL 360 ground at the shield termination point in RF circuits. This is why good RF connectors are fully capturing the shield all the way around the cable. On your RF digital cables, use 360 degree grounds for the best true shielding.
SUMMARY – Most
systems will have proper GROUND differentials between them and thus have near
ZERO shield current. The shield relative to the signal wires will ATTENUATE outside interference. When you have poor grounds,
it may be beneficial to unhook the “receiver” end of the shield and hope
that the induced antenna current voltage is less severe than the induced voltage caused by differential
shield ground potentials. This should be a SECOND choice, not the first. A properly working shield, by
design, has a KNOWN shield dB rating that can be trusted in a proper electrical circuit.
An antenna ground’s induced voltage onto the cable is not fully described and is dependent on the GROUND proximity point and shield’s distance from the signal wires. In severe situations, it may be the best choice to mitigate noise to the lowest possible reference value as it is pretty hard to REMOVE a shield already on a cable. Some, such as coaxial cables, can’t be mitigated and need to be properly designed to EXCEED the ground differential by several orders of magnitude so as not to aggravate any ground differential.
ICONOCLAST will use double ground interconnect shields and proper DCR RCA grounds.
Power cables should also use grounds
at BOTH ends if you have a proper GROUND plane resistance such that ZERO current flows and thus you have ZERO induced voltage from differential current. An
antenna type ground CREATES a differential in each end of the “antenna” by design (one end is ideally infinity the other is ideally zero) and is thus a second choice if you have known ground issues.
One last note, those heavy 10 AWG power or more cables, may provide benefit as they induce less ground differential resistance than smaller power cords as the ground wires are larger. The circuit may not need the power delivery of a larger cable, but the lower ground resistance values may be of benefit on longer runs in marginal power grid situations. This will improve a shield’s current to nearer zero across frequencies. The dB isolation numbers values are for a proper shielded system with IEEE and TIA compliant shield differentials.
The Transfer Impedance numbers are between two-reference point probes on a shield, and DO NOT need the ground potential differences for characterizations.
NOTE: This paper was originally written prior to the introduction of Iconoclast Gen2 interconnects, so while some references are to the future, that future is now….
BACKGROUND: To possibly improve the performance of
the XLR, to maybe achieve even lower L and C,
we would need to revise the current design…and it will jump up the
electromagnetic complexity. The balanced of L and C would shift some but the
coherence will improve substantially.
Changing the “conductor” to a four insulated wire structure
will lower INDUCTANCE through signal phase cancellation. The star quad
arrangement will retain CMRR for NOISE reduction. Four smaller wires will
improve PHASE, and lower wire loop DCR to mitigate ground loops.
PROTOTYPE IMPROVED DESIGN:
Capacitance is the DISTANCE between
the plates (wires) and dielectric material(s).
Inductance is two-fold;
The electromagnetic field cancellation.
The loop area between the wires changes inductance.
For inductance dielectric doesn’t really matter,
inductance is DISTANCE.
We will have the same nearly loop
area in the design (C-C distance is the same) but each conductor in the new
design will further remove signal electromagnetic fields based on the
cancellation geometry. Inductance should drop compared to the single wire
conductor system.
The capacitance requires the same
meticulous attention paid to as the group dielectric. Since we are keeping the
cable the same size so the capacitance HAS TO go up as we have more wires
parallel to a dielectric (the center X-filler, beading and outer tube) , and
closer to the dielectrics. The conductor size and dielectric determine the final
size. The added wires and X-filler close to the wires are the main contributors
to the required capacitance increase. But, lower inductance improves PHASE
shift, and your ear is most sensitive to.
The current coherence, the main
objective of the design with minimal L and C changes, is based on the skin
depth penetration changes going from 1 x 0.018” wire to 4 x 0.010” wire for
each conductor.
4 wire “conductor”
Technically, four -wire per
conductor will increase capacitance some as we have more wires parallel to a
dielectric, but the current coherence
improves substantially, time aligning the low to high frequencies.
18214.4 μ inches = 18.2 mils @ one skin depth.
One skin depth is defined as when the surface current is 37% smaller going into the wire. If we had a wire that was 18.2 mils in size, the CENTER of the wire would have only 37% of the current measured on its surface.
Skin depth equation (below) is a
squared equation, so removing wire depth rapidly increases the inner current
magnitude. Dropping from 20 mils to 10 mils is a 4X improvement in current
coherence.
The very good,
and easier to make, current design does NOT use electromagnetic signal field
reduction technology I developed for the speaker cables in the series 1 signal
leads. The current XLR design relies on reduced loop area and uses AIR to
reduce the capacitance to a minimum for a given tighter spacing to achieve
inductance. The better the dielectric
the CLOSER I can physically locate the signal wires for a given capacitance,
thus lowering Inductance. The size of the wires determines the current
coherence, and with more uniform effect of the dielectric around each wire with
respect to frequency. The smaller the wire, the more uniform the velocity of
propagation from low to high frequencies.
An XLR cable’s
external noise utilizes CMRR based on all four noise signals being equal on
each wire and which cancels those noise signals in a star quad design through
electromagnetic field cancellation. If we look at the four wires, and using the
right hand rule (current out of the page). All the external noise currents in
the wire go CCW around each wire suspended in space. All the electromagnetic
fields cancel adjacent to any wire and across from any other wire. All the
fields superimposed onto one another forming a nearly ideal cancellation
circuit. Nearly perfect because stray magnetic fields would extends OUTSIDE the
four wires and reinforces the field. A first approximation says that this
doesn’t happen. The stronger fields are closets to the wire and cancel most
aggressively. Theoretical outer fields are weak, and don’t reinforce nearly as
much as the inner fields cancel.
TWO WIRE FIELD CANCELLATION ASSUMING FIELDS EXTEND PAST THE CENTER BOUNDARY. MAGNETIC FIELDS WILL CONCENTRATE BETWEEN THE TWO WIRES, HOWEVER,AND CANCEL:
OVERALL 1 X 4 WIRE CONDUCTORS ELECTROMAGNETIC SIGNAL FIELD CANCELLATION:
We DO NOT see
this nearly “perfect” rejection of signal magnetic fields to reduce the
inductance in the signal fields for series 1 RCA or XLR cable. We have a PLUS
and MINUS balanced signal current direction whose fields are only partially
cancelled. The partial field partial cancellation RAISES the inductance above
“zero” theoretically as we have a stronger field, and separated by the distance
needed to lower capacitance with any dielectric. The old design has ~36% higher
inductance, and thus worse PHASE shift than series II (0.015 uh/foot is reduced
to 0.11 uh/foot nominal). Lowering inductance directly lowers phase. See the
QED phase analysis measurements on a variety of cable;
QED – The Sound
of Science www.qed.co.uk/downloads/qed/soundofscience.pdf
OVERALL 4 x 1 wire XLR CMRR INTERNAL (signal energy) ELECTROMAGNETIC FIELD CANCELLATION:
The two MINUS
fields cancel between themselves.
The two PLUS
fields cancel between themselves.
But a MINUS to
PLUS field REINFORCES the overall magnetic field.
The reinforcement
makes the field stronger and the loop area effect worse.
BODY –To make improvements, we need to
reduce the signal electromagnetic field to ZERO, in theory, both from an
external interference view AND an internal electromagnetic conductor view. To
do this, we need to BALANCE the music signal by SPLITTING each of the four
SIGNAL wires into FOUR, or sixteen separate wires.
Making this
critical change will theoretically remove the signal field currents that
interact with the loop, creating inductance. It will also significantly improve
the dielectric group and Phase delay by forcing the dielectric to be seen more
uniformly across the 20-20KHz frequency range with smaller wires.
To keep
capacitance low for a given loop area, we need to use AIR around the wires, and
to make sure any plastics that touch the wire are super low dielectric constant
materials (FEP mini X-filler and external FEP bead wire). This is why the wires have to be BARE copper
with NO insulation around them. Only the
tangential surface of the FEP filler and FEP beading, the rest is air.
Capacitance is dielectric AND distance related where Inductance is distance and
electromagnetic field strength.
Each of the four
wire will be shorted together to make the typical four wires in a star quad.
The wires are 10-mil diameter 30 AWG for a total CMA of 4 x 102 =
400 CMA. I used 4 x 0.018” in iconoclast for a total 18 x 18 = 324 CMA for each
signal wire. 400 CMA is slightly lower DCR than the current design improving
attenuation and mitigated ground loop voltages.
The collateral
filler is foam FEP to manage capacitance. The power carrying signal braid
should also be as far away as possible from the internal signal wire QUAD
structure to lower the ground plane inside the cable, lowering capacitance.
This means making the outer belting thickness under the braid to the best fit
for an XLR connector, but not too big as the reduction in capacitance is a squared
law variable, once a threshold is reached, more is not too beneficial.
The X-filler is FEP, as would the
30-mil beading wrapped around the QUAD wire to lower the dielectric nearest the
wire where it is most critical. The material issues all control capacitance,
not inductance.
The overall belt is solid FEP, with
a 36 AWG BC braid and a drain wire. A final solid FEP jacket finished the
cable.
CORE
0.230” BELTING 0.030” BRAID 0.015” JACKET 0.030” TOTAL 0.305”
Does it really work on initial Capacitance and Inductance measurements? The final design using the ICONOCLAST™ all FEP design for ultimate performance appraisals measured as follows:
SHIELDED
CORE
Lab Rqst-177575 Sample ID – 60156Y (PDC2842)
Requestor – Galen Gareis Report
Generation Date – 22 June 2017
Capacitance @ 1 kHz
per ELP 423, Agilent E4980 Precision LCR Meter, Belden 4TP Cap/Ind Test Fixture
Meas: 18.23 pF/ft
Inductance @ 1 kHz
per ELP 424, Agilent E4980 Precision LCR Meter, Belden 4TP Cap/Ind Test Fixture
Meas: 0.10 µH/ft
JACKETED
SAMPLE
Lab Rqst – 177587 Sample
ID – PDC2842
Requestor – Galen
Gareis Report Generation Date – 29 June 2017
Capacitance @ 1 kHz
per ELP 423, Agilent E4980 Precision LCR Meter, Belden 4TP Cap/Ind Test Fixture
Cap @ 1 kHz
Spec: 10.5 pF/ft max
Meas: 17.48 pF/ft
Inductance @ 1 kHz
per ELP 423, Agilent E4980 Precision LCR Meter, Belden 4TP Cap/Ind Test Fixture
Meas: 0.10 µH/ft
Velocity of
Propagation (VOP) per ELP 392, HP8751A Network Analyzer, HP VEE Instrument
Control Software with Velocity of Propagation program and a GPIB card
installed.
Meas: 85.3%
4×4 Design
1×4 Ref. Design
231 pF/12.67′ = 18.23pF/ft
12.5 pF/ft
1.29 μH/12.67′ = 0.10 μH/ft
0.15 μH/ft
To really get better XLR
performance, both loop area and the field cancellation technology need to be
leveraged, with the latter being most critical. The capacitance is all about
materials and DISTANCE between them. Improving inductive field cancellation has
the added, and significant, benefit of improving signal coherence through four
smaller wires and phase with lower inductance while improving attenuation
performance.
A cable with smaller signal wires
and better coherence, low inductance (better phase) and slightly higher
capacitance will sound better sounding than a cable with larger signal wire and
less coherence, higher inductance and lower capacitance…as long as capacitance
isn’t too high!
The prototype run does indeed lower inductance with the
expected rise in capacitance.
0.15 uH/foot (X) + X = 0.1 uH/foot, X= 33% lower. Or the current design is 50% higher.
12.5 pF/foot (X) + X = 18.23 pF/foot = 45.8% higher Or the current design is 31.4% lower.
Since the original design is working from such low L and C
numbers, the percentages are not really illustrating the advantages of the
improved signal coherence with much smaller wires, and an advantage that should
play out in audible performance. The
-3dB first order filter frequency is still well above the audio band so first
order filter phase distortion is not going to be an issue. What must be the
major contributor is coherence with the smaller wires. Rs response, while
lower, is hard to quantify.
Rs (swept frequency
resistance) Values
The 4×4 XLR lowers swept Rs (proximity effect) values
significantly, and flattens the high-end linearity. Can you HEAR that
improvement, over the single the wire design? The truth is BOTH are
superimposed when the wire is used, and pushing the XLR designs to as near
perfection is certainly a better and better design. The lower DCR is evident in
the trace compared to the 1×4 25 AWG wire as is the flatter upper frequency
measurements.
The RCA interconnect has also been updated with the new 1×4
(ONE wire made with FOUR conductors) design. The reactive variables will track
with frequency like the single wire designs, but map to the altered L and C
values.
The following table shows the effects of changing the wire
size and number. The 4 x 4 has almost the same CMA as a single 22 AWG, but 1.82
times more total circumference, which shows up only at increased frequencies.
The lowest frequencies are essentially DCR.
The maximum Rs is lower with
the 4×4 design. Beta test feedback from customers on the 4×4 has been extremely
positive and, consistent with the numbers, shows this revision to be a
significant upgrade from the original Iconoclast design for analog
applications.
DCR INTERCONNECT LOOP CONSISTENCY
The interconnect cables of a given wire design (single to single and
verses quad to quad) have essentially the same loop DCR values.
From the Rs chart above at DC,
we see;
4×1 XLR and 1×1 RCA are 34.11
and 39.19 Milli-ohms/foot respectively.
4×4 XLR and 1×4 RCA are 27.53
and 28.79 Milli-ohms/foot respectively.
How was this done? The double
braid on the RCA was necessary to mitigate ground loop DCR variation between
sources, and the DCR was designed to be near a “free” return path for loop[
DCR. The loop resistance ios the braid plus the
conductor. But, the braid DCR is so low that the loop DCR is pretty much
the RCA center conductor. This is true for eother design.
The XLR DOUBLES the number of
conductors in each leg as a star quad. This reduces the DCR to one-half the
conductor’s value. Thus the two pairs in parallel are the same DCR as a single
conductor.
This was also done on purpose
to make sure that the RCA’s loop performance was as good as the XLR, and that
the RCA BRAID was essentially a ZERO DCR return path between grounds. If the
RCA braid was insufficient DCR, we would see more divergence between the two
singel ended and balanced design.
CONCLUSION
The measured XLR electricals are very good, and follow
design theory perfectly. ICONOCLAST once again shows that proper engineering
fundamentals are paramount to performance. Sound Design Creates Sound
Performance!
In a previous paper I
covered several issues that create signal distortion in audio cables. The most
demanding variables involve the TIME related distortions that the ear is most
sensitive to. Consideration must be made during cable design to mitigate the
TIME based issues through the audio band. The following paper is the journey
through the design process to arrive at a satisfactory RCA and XLR cable
design. I must stress ALL quality cable designers have to work with the exact
same known variables to solve problems at audio. Every cable is a compromise of
some sort as distortions can’t be eliminated.
ICONOCLAST has made the outlined design decisions to arrive at, what we
think, is an industry leading design based on real measurements.
SOUND
DESIGNS CREATE SOUND PERFORMANCE™
RCA DESIGN BRIEF
1.0 Conductors 1.1 Copper Size 2.0 Dielectric Material(s) 3.0 Dielectric geometry 4.0 Shield material and design considerations 5.0 Jacket design and material considerations
The design process will start with the RCA cable as
this provides the most pristine electromagnetic properties possible due to the
seemingly simplistic design. Once all is
said and done it is “simple” looking. The more complex XLR will have to, somehow,
match the RCA’s electromagnetic properties if it is to be an “equal” on
measured attributes. If the RCA isn’t any good, I may as well start over again!
“…granular building blocks in copper can never fit together perfectly, but are rotated causing an unexpected level of misalignment and surface roughness. This behavior, which was previously undetected, applies to many materials beyond copper and will have important implications for how materials are used and designed in the future…”
The
battle for material supremacy continues. However, what we tend to discount is
that while the overall design of the tire we put on the car is important, the
rest of the car has more to do with what that tire does than just the tire. We
over spec the tire and vastly under spec the car. I’m intent on building the
car, not the tire.
The
decision to use copper is based on several factors, none of which were price.
Copper offers the best material for affordable cables with a significant level
of performance in more ideal electromagnetic designs. Far more expensive
materials in lesser designs won’t work, and far more expensive materials in
superior designs won’t work…for most of us anyway.
Copper
is available in several process treatments and after process treatments;
ETPC
(as good as what used to be OF grade)
OFE (differing process, but far from vastly lower impurities content)
UP
OCC (what is often called long grain type, and again a differing process).
Cryo
treatments (used to improve copper’s PHYSICAL properties)
Grain
direction (music is AC. Which polarity do you like first and at what
frequency?)
I don’t use wire “quality factor” as
a design element since every contemporary draw science wire is of vastly better
quality than ever. Sure, some processes are more $$$ but there is scant
repeatable measurement that I can do other than conductivity, a passive
resistive measure that will influence R, L and C. The conductor type is an option
for the customer to listen to, only. There are differences. Belden just isn’t
in the position to create a pet project to define what isn’t yet scientifically
defined. That’s not our thing.
Belden
offers the three fundamental copper grades; ETPC, OF and UP OCC, as they DO
sound different in the exact same electromagnetic R, L and C referenced design.
No changes other than the copper, so we know what the culprit is. What we don’t
know, is WHY it is the culprit. Instead of making up a big old story, again, about
the material, we don’t. It is what it is in use and we leave it that way.
What
we don’t offer is what I can’t hear as a designer. Sorry, but I’ve yet to hear
CRYO treatments, intended to improve the wire’s PHYSICAL strength or grain
direction, change the sound. As far as grain direction goes, you can flip the
leads in any direction you want, as the wire’s grains all go the same way due
to the manufacturing process that we use. If you can hear the direction switch,
flip them any way you like. We won’t send you a bill for that!
Any
material used in a superior design SHOULD sound as good as it can, and cost
isn’t a direct line to better sound. I ignored cost when I designed
ICONOCLAST™, either high or low. If my system didn’t allow me to hear it, I
didn’t use it (materials) or do it (process / design).
This
isn’t a paper on conductors, although I may have some things to say about
alternatives to copper them later on in another paper based on some
measurements and calculations I’ve done. We’re talking copper in this paper as
it is the very best economical solution that we have right now.
Copper
has a very low DCR, a reasonably deep skin depth to manage current coherence,
is pretty high in tensile strength for processing, and in most applications
resists severe oxidation. The grain structure is clearly visible in form, but
that alone is NOT what makes the different grades sound different. It is a
trait of the draw science, but does not have as much effect on the sound as you would be lead to believe.
Use
solid or stranded wire? This, at least,
is easy. Is stranded better for the way the cable is used? Is stranded more, or
less, expensive? Is stranded easier or harder to process? Is the termination of
the cable better or worse with stranded wire versus solid? Are any gremlins
that I call tertiary variables (stuff there isn’t a measurement or calculation
for) removed if the truly measureable variables are accounted for between
stranded and solid?
ANSWER
– Solid wire wins hands down for this application. Every question is in solid
wire’s court. End use, costs less, processing cost, ease of termination and
lack of tertiary elements (all those diode effect “arguments” between strands
and more).
On
that, though, a note: the first generation of Iconoclast interconnects use
single solid wires for the signal-carrying conductors and that’s what’s
discussed in this paper. Our second
generation product (suitable for analog but not for digital due to impedance
issues) uses a star-quad arrangement of four separate wires, placed around a
separator, in place of each of these conductors for improved inductance; for
details see the fourth paper in this series.
Other than this change in the signal conductors, the “Gen 1” and
“Gen 2” interconnects are the same.
1.1 Copper Size / RCA
We
now have SOLID copper wire. The size selected sets the foundation for the whole
thing if we consider that the cable’s structure is supposed to allow a
conductor to be as near zero R, L and C measurement cable as we can design.
You
can’t use a conductor you can’t process. For the RCA cable, we want as small a
wire as we can process as this will force the best current coherence through
the wire (same current magnitude at all frequencies). The exact skin depth
calculation is a tool we use to gain the knowledge
to reduce the wire size in audio cables. At RF, we use it to tell us how much
copper to put over a STEEL support structure to maximize RF attenuation. Audio
is not RF, and the ENTIRE wire is used to move the signal and at ALL frequencies
concurrently, not the same issue at all in RF cable design.
RCA cables terminate into a theoretically infinite (47K-120K or there about) input resistor. We say impedance, but it is really as resistive as it can be made at the input op-amp level. Yes, purists will point out that input impedance DROPS some at higher frequencies.
If
the impedance is so high and the current is so low (it looks like an open
circuit) just use as small a wire as you can! Well, yes and no. It has to be
reliably terminated and secure in the end product, and it has to process evenly
under tension and not fracture from surface issues.
A
review of the end of process design backs into the initial design requirement.
Calculations and testing selected a 0.0176” diameter wire for ICONOCLAST. The
process has to handle less than 4-3/8 pound tension to avoid permanent wire
stretching. Wire was tested for the process requirement.
The
0.0176” diameter wire (0.0088” radius) is one half the diameters necessary for
one full 18-mil skin depth at audio, so we have significantly improved current
coherence through the wire @ 0.0176” diameter wire. Skin depth is FREQUENCY
driven for a given material. The smaller the wire the larger the inner current
magnitude will be relative to the surface current. We want as good a shot of
that as we can get.
The
RCA cable’s loop DCR will essentially be the center conductor in an RCA, if it
is made right, and ICONOCLAST is. The center wire governs attenuation. The
outer conductor is, in theory, infinitely low impedance so it nearly drops out
of the loop DCR calculation and leaves the center wire. The length of the cable relative to the input
impedance allows a SMALL wire at audio. At least attenuation works in our favor
at audio as it is a LOG relationship and gets really high very quickly as you
go up in frequency. For audio, we can relax a bit on attenuation as it is low
for the lengths we use and is in the right frequency range to stay low.
Attenuation is a passive “distortion” and is VERY hard to hear over TIME based
distortions.
Dielectric Material(s)
We’ve
already made a critical choice in our cable. The wire material and size. We’ve
used good engineering practice to KNOW what the decision will yield. Now, how
to RETAIN all that the material / size wire can provide? That’s easy, just
stick it in air and find an infinitely low ground potential for our unbalanced
/ single ended wire!
OK,
this IDEA is easy. The execution isn’t. I don’t care about speed of the process
and / or costs as I’ve used REASONABLY affordable material as my conductor. We
can always go back and break the bank on conductor materials. AIR is free, but
expensive to get. Air is by far the best dielectric to have, and especially
nearest the wire were the influences are the worst on group delay. The closer
to the wire the dielectric is, the more it influences the overall velocity of
the composite structure (wire / beading/ then plastic tube thickness / then
braid)
I
decided to go the tough route and use air. We can use RF as a HINT at what to
do overall. We have used designs called semi-solid core dielectric RF cables.
These partially suspend a wire in a tube with a spirally wrapped thread. The
problem is that the wire SIZE and the core tube properties aren’t suitable for
audio frequencies. Even the choice of materials isn’t as important at RF as we
can reach a set impedance vector (real + the reactive inductive and capacitive
parts all added together) by tweaking the thread and tube dimensions.
3.0 Dielectric geometry
The
audio signal is very sensitive to the dielectric effects of the plastics near
it. I chose a specially made beading thread to get the job done.
The
above picture beading around the wire is a glass thread coated in pure
TEFLON®. I use a ROUND beading shape
versus square, as it touches the wire at the tangent points for the very LEAST
effect nearest the wire. The electromagnetic field sees the entire cross
section of the plastics and material between the wire and the inner braid, so I
use GLASS thread inside the beading as it is a good dielectric, too. Why is the
glass there? A solid TEFLON® bead can’t be processes at this size and keep
consistent dimensional linearity. The glass is the true STRENGTH member in the
beading, not the plastic. The plastic is to set and hold the shape. The glass
lets me process the beading at production speeds.
Why
TEFLON®, really? OK, I’ll tell you. It has the lowest dielectric constant of
any solid plastic. It is TOUGH in thin walls for end product dynamic stability;
the bead should STAY round under side-wall pressure. This is a SMALL bead, so I
need that toughness. TEFLON® has high T and E’s (tensile and elongation)
properties for process toughness. We don’t have much process room, as I’ve
calculated backwards how big this bead would need to be in this design and wire
size.
How
big should the conductor be based on a tube ID? There is ONLY one optimum
asymptotic wire size driven MAX AIR volume (%) based on the tube ID. The ratio
of the tube ID with the 80% air void to the inner braid surface will determine
the capacitance. Maximizing the air content will improve the efficiency of the
dielectric so the smallest loop area for inductance will also yield the
smallest measured capacitance.
Here
is what happens when we CHANGE the wire size;
Tube ID (IN)
Wire Size (IN)
Air %
0.070
0.014
80
0.098
0.020
80
0.123
0.022
80
0.150
0.030
80
As
the wire gets bigger or smaller inside a given tube ID, it crowds out the air.
We COULD go drastically big in the ID of the tube and wire size (0.150” tube
ID)…but we want to hold INDUCTANCE and signal coherence in check. Inductance is
the loop area between the wire and the inner braid, and that needs to be infinitely
close, the opposite of capacitance. For a given tube ID size we want the
maximum amount of air void and the smallest possible wire to braid distance.
This means the conductor wire size has to be as small as you can process, and
with the desired capacitance. As the tube ID gets larger, cap will drop but
inductance will rise, and the opposite with a smaller tube ID. The design
target is 11.5 pF/foot on the bulk cable to assembly capacitance would be 12.5
pF/foot.
Using
too large a wire hurts frequency coherence so we pushed the wire size DOWN
until inductance was moving off spec relative to capacitance. A balance was
sought between wire size (coherence) and reactive variables (L and C).
I
can do a quick check to see how I’m doing by applying a test ground over a ten
foot sample. Using RF frequencies as a “constant” since the velocity has
stabilized to an asymptotic maximum, we measure really high VP values, ~ 87%.
This is good as it allows me to reference to end capacitance, too. I just treat
the cable like an RF cable and work the capacitance backwards from the open –
short Impedance; Z = 101670 / Cap * VP. This is about 104.6-ohms so capacitance
calculates to 11.2 pF/foot versus a measured value of 11.19 pF/foot.
We
know from the previous paper that Capacitance and Inductance are FLAT with
frequency, and are actually measured at 1 KHz. Our 11.19 pF/foot bulk cable
value is true at 20Hz-20KHz. Inductance is a low 0.15 uH/foot through the audio
band as well.
Capacitance @ 1 MHz per ELP 423, Agilent
E4980A Precision LCR Meter, Belden’s Cap/Ind Test Fixture
Spec for Cap @ 1 MHz: 12.5 +/- 1 pF/ft
PDB1610 B24 Cap @ 1 MHz: 11.1947 pF/ft
Characteristic Impedance per MIL-DTL-17H (ELP 142) using the included
equation:
Char. Imp per ELP 142: Imp = 101670/(C +VP)
Spec for Impedance: 100 +/- 5 Ohms
PDB1610 B24 Impedance: 104.631 Ohms
SEMI-SOLID PDB1610 finished RCA “assembly”
CAP 12.25
pf/foot IND 0.1450 μH/foot
Inductance
isn’t as critical in high impedance leads as current, which is ride time
limited by inductive reactance, which is near ZERO, but in my listening test,
cable with near zero on BOTH L and C attributes sounded best, and a BALANCE
needs to be considered. The cable isn’t big or small; it is what it needs to be
to WORK. The wire size we start with sets this all into motion.
The
FEP tube is critical to get right. Special processes are used to keep it
on-sized and ROUND over the beaded center wire.
Shield material and design considerations.
We
have a core tube and know the electricals, so now what? The braid is much more
important than people think, and for a different reason than people think. No,
it isn’t shielding, either. True, a double 90%+ braid have 90 dB RF shield
properties but, I sure hope your equipment isn’t THAT sensitive to RF. Foils
are much better and more economical for RF than a single 80% braid and the
shield reaches the 90 dB mark far more cheaply.
RF
cables are “shielded” to RF noise and IMMUNE to low frequency nose (outside their
pass band) because the shields have a low resistance to RF, measured as
transfer impedance. This is sort of like low DCR at audio frequencies, but
relates to how high frequencies work. Audio cables are not RF cables!
We
need to look at how unbalanced circuits work. They SHARE a ground…or do they?
They are SUPPOSED to SHARE a ground. They don’t. RCA unbalanced cables use the
CHASSIS as a ground to the wall outlet or it is floating in some cases but the
REFERENCE between the grounds is still there. In ALL cases, there is that pesky
WIRE thing called the SHIELD between the ground points on every piece of RCA
equipment you use. That wire has RESISTANCE and that resistance creates a
ground potential difference so current starts to flow between the two end grounds.
E=I*R, remember that? A VOLTAGE is impressed against the center wire and the
magnitude of that voltage is the current times the resistance. We can CONTROL
the “R” by using TWO 98% copper braids. This is $$$ to do, but it is the RIGHT
thing to do.
No,
those braids won’t shield MAGNETIC interference. The HUM you hear is more than
likely ground loop current through the braids resistance called SIN; Shield
Induced Noise. The lower the braid DCR is the better the SIN rejection. You
need low permeability shield to block low frequency magnetic waves (anything
below about 1 MHz starts to have a considerable B-field bent over E-field).
Good audio RCA cables ARE NOT going to shield B-fields. They will shield
E-fields and reduce SIN noise.
To
shield magnetic B-fields a MAGNET needs to be able to STICK to the shield. This
is an indicator that the material is “influencing” the magnetic field flux
lines INTO the metal and OUT OF the air. We can manage the SIN noise with a
good ground, but true extraneous magnetic noise is still tough with unbalanced
cables. Now you know why. It’s the ground system it uses.
Jacket design and material
considerations
ICONOCLAST
uses an FEP jacket for some good reasons. FEP is the most chemically inert
material there is, protecting your cables from chemicals and UV exposure
through those nice picture windows in your house. Lesser plastic material isn’t
as stable, or inherently flame retardant. Nor can many materials be used in
thinner walls.
Plasticizer
migration out of the cable, especially near heat, is a real issue in contact
with polyester or nylon carpet that would love to be the same color as your
cable laying on it! My previous cables were.
FEP does not have this issue and will look nice for decades to come.
Yes, it costs some more but these cables are an investment into the future and
can follow your system several steps above where you may be now. Based on
durability, stability and inertness to solvents, FEP is the best choice for the
long haul.
RCA SUMMARY – Knowing that RCA cables
aren’t as “shielded” at audio as we think, what can we do about that? If you
don’t have the problem, you’re good to go! RCA is a great sounding cable by
fundamental electromagnetic design. This is why it was created. It does have
magnetic noise immunity issues, though. There is no magic to good cables; it is
adherence to strict design rules that also encompass those “magic” tertiary
variables called wire science. The same
design adjusted for a new material’s skin depth properties can be made to the
same “ratio” and match the electricals with differing wire. The layers of the onion and their thickness
can be altered (L and C values) depending on what is most audible. Tests won’t
tell you that, this comes from design experience. This does NOT mean that either L or C can be
thrown to the wind. Both L and C cause
TIME based distortions and neither is welcome in good cable.
Then
there is the next cable I’m going to talk about that does exactly that, except
it is far, far harder to make as good as an RCA electrically. It is called the
XLR cable.
