DIY AC POWER CORDS
by Jon M. Risch, -  Last Updated 3-24-2002
[email protected]

I was not able to hear any significant differences
due to highly specialized power cords when I listened
exclusively to a Class A SS amp, other than using the
heaviest gauge to minimize voltage drop due to the cord,
and if the cord has shielding. The heaviest gauge commonly
avaialble for a pre-molded IEC type cord is 14 Ga. The
shielding available is usually a foil type shield (although some
have a copper braid), and this can help with RFI problems
sometimes.

Once I started listening to a Class AB tube amp, then I
noticed more of an effect with power cords.

It is easy to underestimate how much drop a typical 18 gauge
AC power cord will have on a power amp, and even how much it
might affect a preamp or CD player. Sheer heavy gauge AC
power cords will minimize any dynamic fluctuations and shielding
will help with localized RF or EMI interference problems.

By way of further explanation, I provide this copy of a post on
AC power cords I have made before:

AC Power Cord Effects

So, how in the world can a power cord make any difference,
if it is properly rated for the current draw, and is UL listed?
Like many questions about audio systems, this one seems very
logical and reasonable. Just use Ohm's Law to calculate the
current draw, and viola, we have the voltage drop, and we have
the cold hard facts, Yes?

No.

Like many overly simplistic answers, this one does not take into
account all the facts. Let's look at a hypothetical power amp to
start. Lets say it is a moderately high power design, and is rated
to draw about 8 amps from the wall at full rated power. That's
about 960 watts from the wall. We won't even get into side
issues of switching power supplies, which play total havoc
with the simple Ohm's Law approach, we will stick to linear
power supplies.

So if the AC cord is UL listed for 10 amps, it is likely an 18
gauge cord. Many 18 gauge power cords are rated for 10 amps
of current. How is this rating determined? By how hot the cord
gets while carrying the rated current, NOT how much voltage
drop there is.

In our hypothetical amplifier, the RMS voltage drop in the 6 foot
cord would be approx. 2/3 of a volt according to Ohm's Law.
This does not take into account the wall outlet contact, or the IEC
connectors if present, nor does it take into account any other
factors, just the resistance of the 18 gauge wires in the AC cord.

On the face of it, this seems quite harmless. How could two thirds of
a volt make any difference? Well, because the voltage drop is NOT
2/3 of a volt! Has Ohm's Law been repealed? Are engineers all
insane?

No, just an overly simplified analysis that failed to take into account
ALL the factors. So what could possibly make that much difference.
Well, for one, it would help to know that linear power supplies
refresh their DC reservior from the AC line in bursts of current,
current peaks that are in time with the peaks of the AC line. That
is when the filter capacitors are refilled, when the AC line voltage
its it's peak value, and the output from the secondary reaches a
level above that which the PS capacitors have been drained to.
Instead of a nice steady drain of 8 amps, we have current peaks that
last only for a brief moment, and the 8 amps is an RMS amount.

What this means is that because the current is being drawn only for
a fraction of the AC voltage cycle, or for about 1 thousandth of a
second (one half cycle lasts for 0.0083 seconds). The peak currents
can easily reach ten times the RMS value. What is the voltage drop
for 80 amps? At this point, the resistance of the AC plug contacts,
and even the bond inside the cord between the plugs and the
internal wires become a significant factor. It is not hard for the voltage
drop to reach 5 volts or more. The AC line peaks never reach the full
value, because the line cord has lost some of it.
Power amps depend on receiving the full measure of AC line for
their full rated power, so such a drop will reduce the actual output
power in a seemingly disproportionate amount. A 200 W amp may
be reduced to 170 watts before clipping.

The above (still simplified) analysis assumes a steady signal, and a
steady current draw. Musical dynamics make it a much less
consistent thing, and the dynamic demands will cause dynamic
perturbations.

This is not the only effect on the power amp. These current peaks
can easily cause AC line distortion, and the heavy current draw can
generate harmonics on the line, the hash from the rectifier diodes
can increase, a whole series of events occurs that are not
immediately obvious just by thinking of the amp as a simple resistor
and using Ohm's law.

Most house wiring is either 12 gauge or 14 gauge, while many OEM
cords are 18 or at best, 16 gauge. Most OEMcords do not have
shielding or any provision for reducing radiated EM fields, do not
have premium AC outlet plugs or premium IEC plugs for better
electrical contact at these junctions.

So what happens with a bigger power cord? Replace that 18 gauge
cord with a 14 gauge cord, and the voltage drop will go down by a
factor of about 2 and a half.

Include shielding in that cord, and the possiblity of radiated EMI/RFI
goes down compared to an unshielded cord.
Shielding and radiated EM fields might not seem relevant since the
house wiring has neither advantage. However, the house wiring is
not laying right next to the other component's AC line cords, or right
next to the line level or speaker interconnects. Reduced levels of
induced RFI and radiated EMI/hum fields would not hurt SOTA
sound reproduction.

Since the AC power cord is usually laying right in there with all the
other AC cords, and probably the speaker cables, and the
interconnects (some people even bundle them all together for
neatness, OUCH!), it is quite possible that a premium AC cord
will help reduce interference in the system, and raise the amount
of power available before clipping, and smooth any AC line
distortions, etc.

This is all without even going into secondary effects, or other
more esoteric aspects. Just a more nearly correct way of
applying Ohm's Law to the real situation. Add in ferrite filters,
built-in filter components, shielding effects, and the esoteric aspects,
and it should be obvious that AC cords are not at all simple, nor are
they a no-brainer.

There is the issue of resonant situations. Certain power cords and
power supply transformer primaries might tend to resonate at RF or
high frequencies. This resonance might make RFI/EMI problems
worse, changing to a different cord will change the resonant
frequency, and change the RFI/EMI effects.

The plugs are not a trivial issue either, and may be more responsible
for sonic improvements than the other factors.  Hubbel and other
premium plugs and sockets will increase contact area and pressure,
reducing contact resistance and other contact related problems.
It has been claimed that poor AC plug contact can cause micro-arcing,
with it's attendant hash being injected directly into the audio component.

It is not as simple as just simple wire resistance. The connection at
each end of the cord adds resistance, the wall outlet socket adds
resistance, etc.
     For the raw wire, round trip:
     12 gauge, approx. 6 feet = 0.0206 ohms
     14 gauge, approx. 6 feet = 0.0328 ohms
     standard AC power cord
     18 gauge, approx. 6 feet = 0.0830 ohms

Measuring real AC power cords, I get around 0.128 ohms for an IEC
18 gauge power cord, and about 0.022 ohms for a 12 gauge IEC cord,
not including the AC wall socket connection.

Why is the 18 gauge resistance so much higher than just the wire
resistance? Ever tear one of those cheap cords apart? poorly crimped
or barely soldered connections are responsible for the bulk of the extra
resistance.

Preamps and CD players all have their special requirements:
CD players require shielding to help keep the digital hash that
back-feeds from the circuitry out of the rest of the equipment, preamps
need a nice steady voltage for minimum noise, and freedom from
RFI, etc.

Fancy AC power cord geometries might also reduce the inductance
of the 6-8 feet of line cord, raising the available voltages, but this
would be limited to the ratio between the length of the power cord vs.
the wall run. Such geometries often reduce the radiated energy,
and aid shielding of the cord.

Some power cords might have a built-in filtering action, like the
water jacketed ones, that have the conductors and insulation
surrounded by a conductive fluid. This fluid might short out and
reduce/damp any EM fields the cord would conduct to the component
besides the 60 Hz AC power signal.

It also helps to keep in mind that we are not supplying a Sears rack
system, that any system which aspires toward the SOTA is going to
be more sensitive to minute effects and minute improvements. How
much does a big fat shielded power cord help things? Probably
about as much as upgrading from an OEM interconnect or zip cord
speaker cable to some decent aftermarket cables, some systems
are more sensitive to AC cords, some are less sensitive to cords.

As always, the bottom line is: you have to listen for yourself, and see
if their is any benefit for you, on your system, with your listening habits.

Do AC cords have the potential to influence high end sound? Yes.
Does anyone who believes this also believe that high end cord
costs are justified? No.

Some of the cord/cable manufacturers get carried away with using
only the very finest materials and assembly techniques, carrying
over the technology and costs from their high end audio
interconnects and speaker cables.

Is this necessary? I don't know, I will not discount it out of hand.
Does it make the cords cost a lot? Yes.

Does this make them dishonest or imply that they are deliberately
trying bilk the customer? Not at all.

While AC cords made from certain models of speaker cable may
provide some extra benefit, I have not been able to hear distinct
improvements when comparing them to just plain heavy gauge and
shielded cords when listening to a Class A SS amp.  When I
started listening to a Class AB tube amp, then power cord
differences became more apparent.
Results in your system may vary.  I have not
tried Belden 89259 in the cross-connected speaker cable hook-
up as an AC cord, but plan to sometime.

Below, I have listed information on DIY heavy-duty AC cords, both
pre-molded and raw wire.

Commercial numbers for these types of cords are:

PREMOLDED
****************
Belden
17604 6 ft. (Around $10)
17605 10 ft.

Still a big heavy cord, but 16 Ga. (and still better than
most OEM cords) are:

17602/3

17608/9 (right angle into gear)

These are now available under the Volex brand name,
Belden has them make the cords, see:
http://www.carlton-bates.com

For info on the Volex cords,. see:
http://www.volexpowercords.com/

http://www.volexpowercords.com/power_cords/powercord2.cgi
?view=17604&catagory=3_conductor_plastic_detachable&
area=other&mfgloc=&partnumber=&userid=

(To get to the section with the Belden type cords, take the hard
line returns out)

Otherwise, Belden and other cable companies offer a wide
range of UL listed portable cordage which can be used to
make up a power cord using 20 amp or hospital grade plugs.
These run up to 12 Ga., with four conductors available.
Using four conductors, one could wire the cord in a
star-quad (cross-connected) arrangement and conceivably
buy some gains here, as this will lower inductance and
reduce the stray field and external field pickup.

RAW WIRE
*********************************

Some numbers are:

19217 14 Ga. shielded, 4 conductor

19216 12 Ga. shielded, 4 conductor

For grounded cords (3 conductor):

19364 14 Ga. shielded
19354 " "

In assembling AC power cords, it is imperative that they
be wired correctly, and with a good mechanical connection.
Use an existing cord to check the connections for the
plugs. Always check with a VOM to make sure they are
not shorted!

Since the pre-molded cords are so reasonable in price,
around $12, there is no reason not to experiment with
your equipment and see if there is any benefit with your
particular system. Even making your own AC cord from
scratch is not that expensive, as the hospital grade plugs
and the raw cable shouldn't run over $20-30 total, which
is still much less than the retail high-end AC cords
running hundreds of dollars.

Don't neglect the wall outlets, many have found the Pass &
Seymour 5262-I, 5252-I and the 5242-I
and  wall duplexes to be the best without going
to Hubbels, and some think they are better than the Hubbels.
There is the equivalent AC cord plug, too in that brand,
Wall plug Pass & Seymour 5266-X .

IEC plug for cords: Schurter 4300.0603 available from Allied Electronics, Allied Electronics PN 509-1215

Jon Risch

All applicable copyright laws apply, all rights reserved, except transmission by USENET and like facilities granted.  Any use or inclusion in print or other media are specifically prohibited.  The informational content is not warrantied in any way or form, and any use of said content are at the reader's own risk, the author shall not be held responsible in any way for any damages or injuries arising from the content of this web site.  Common safety practices are encouraged at all times.  Do not fold, spindle or mutilate.