• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Extension Cords Questions

Sunin

[H]ard|DCer of the Month - August 2008
Joined
Dec 27, 2005
Messages
3,421
Ok I bought 12gauge, heavy duty Extension cords 50 foot, rated at 15amps 1875watts... well putting about 1300 watts through them got them warm over the weekend. WHY??????

This really irritated me.. I bought the beefy one's in order to avoid any overloading/fires...

I subsequently bought 5 more cables and now each PSU with a Max load of 750w is on its own extension cord.

Thought I had a lot of wires before I just basically doubled it. :)

So what is the deal with the heat found coming from the extension cord.

Note: This is why I do these sorts of upgrades over the weekend so I can monitor them before burning down my condo!



 
Even though you were way under the rating of the cord .. still a lot of power to pull on that one cable.

I think you did the right thing. Multiple is always a good option.

I think i might have the same problem later on.
When I get my server room built, I was going to have one outlet power each rack (20 amp each) but I think I am going to run two outlets on each circuit.

That way each one is only running 10 amp and won't get anywhere near the max.

 
I guess that after a certain length, you need to derate it due to more resistance due to the extra length.

 
I guess that after a certain length, you need to derate it due to more resistance due to the extra length.


I imagine this is a big part of it. How hot are we talking about? Too hot to touch? Or just noticeably warm?

 
Watts generate heat. Whether it's in a processor, GPU or cable. Stranded cable (which extension cords are) put off more heat due to higher resistance than solid conductor (your circuit wiring to the outlets) so warm when pulling 70% of their rated load is not really surprising or that dangerous by itself. where the warm cords become a problem is in that it helps breakdown the insulation faster.

I stumbled into this awhile back and it's got WAY more info concerning extension cords/wire and loads than you'll ever want to know.

Code:
Generally, current delivered to static loads is lowered by
increased extension cord length. Resistance goes up; current
goes down (I = E / R).

Resistance increases with extension cord length, thus voltage drop
across the line lowers the voltage available at the end of the cord
for use by the appliance. A heavier extension cord (one with more
copper in it) has less resistance than a lighter cord, just as in
plumbing, a larger pipe has a lessened restriction to the flow of
water.

Extension cords are rated by an AWG (American Wire Gauge) number;
the higher the number, the smaller the cross-section (roughly:
diameter) of the (usually copper) conductors in the cord. Stranded
wires of the same AWG rating have higher resistance per linear foot
than do solid wires; thus solid copper (romex) drops less voltage
over longer runs. Extension cords are made of stranded copper (or
other metals, perhaps) to maintain flexibility, for temporary
applications that can tolerate some amount of voltage drop in order
to gain positional advantage (where it would be impractical to run
solid copper).

In electrical circuits, current (in Amperes) is shown in formulas
as represented by the letter I; voltage (in Volts) is represented
by the letter E; and resistance (in Ohms) takes the letter R as its
representation in a formula known as Ohm's Law:

E = I * R (Voltage equals Current times Resistance)

Power is calculated by this law:

P = E * I (Power equals Voltage times Current)

Power is expressed in Watts in electrical circuits.

Anything that uses power in an electrical circuit is generally
referred to as the load on the circuit.

The length of an extension cord, along with the wire gauge of its
conductors, influences the current by lowering it (I = E/R; R is
increased by extension cord length). For example, ten 60-watt
light bulbs connected directly to a 120 volt source draw 5 amperes
of current, and present a 24-ohm load to the voltage source and use
600 watts of electrical power between them.

However, if a 100-foot, 16-AWG extension cord is placed between
the 120 volt source and the light bulbs, a small resistance
(0.8344 ohms) is added in series with the 24-ohm load. This
affects the circuit in the following ways:

1. Total load is now 24.8344 ohms
(24 + 0.8344; R total = R load + R line)

2. Total current is now 4.832 amperes
(120 / 24.8344; I = E / R)

3. I-R drop of the load is now 115.97 volts
(4.832 * 24; E = I * R)

4. I-R drop of the line is now 4.032 volts
(4.832 * 0.8344; E = I * R)

5. Power dissipated by the load is now 560.36 watts
(115.97 * 4.832; P = E * I)

6. Power dissipated by the line is now 19.48 watts
(4.032 * 4.832; P = E * I)

7. Power dissipated by the system is now 579.84 watts
(560.36 + 19.48; P total = P load + P line)


Commentary:

1. The circuit gained resistance and will therefore draw less
current.

2. The current in the circuit dropped from 5 amps to 4.832 amps.

3. The voltage available to the light bulbs dropped
to 115.97 volts.

4. The extension cord dropped the line voltage at the far end
by 4.032 volts.

5. The light bulbs use 560.36 watts instead of 600 watts (and
are less bright).

6. The extension cord uses 19.48 watts of power (dissipated
as heat).

7. The system uses less electric power (579.84 watts
not 600 watts).


Now, increase the conductor diameter of the extension cord to
12 AWG and shorten it from 100 feet to 25 feet:

1. Total load is now 24.0825 ohms (24 + 0.0825)

2. Total current is now 4.983 amperes (120 / 24.0825)

3. I-R drop of the load is now 119.59 volts (4.983 * 24)

4. I-R drop of the line is now 0.411 volts (4.983 * 0.0825)

5. Power dissipated by the load is now 595.89 watts
(119.59 * 4.983)

6. Power dissipated by the line is now 2.048 watts
(0.411 * 4.983)

7. Power dissipated by the system is now 597.945 watts
(595.89 + 2.048)


Commentary (25 foot 12 AWG compared with 100 foot 16 AWG, same work-load):

1. The circuit with the heavier, shorter extension cord lost
resistance, and will therefore draw more current with the
same 600 watt bulb array (load) attached.

2. The current in the circuit increased a bit, to 4.983 amps.

3. The voltage available to the light bulbs increased to
119.59 volts, closer to nominal 120 volts (design voltage for
these bulbs) than the 115.97 volts supplied to the bulbs using
the longer, thinner cord.

4. The extension cord dropped the line voltage at the far end by
only 0.411 volts, a tenth the (4.032 volt) loss of the other
extension cord.

5. The light bulbs use 596 watts -- close to the 600 watts
nominal expected of a perfect circuit, and give off
more light.

6. The extension cord is heated by only 2.048 watts of power --
much less loss due to heat and provides a margin of safety as
well as a stability (resistance increases with temperature);
this is more than nine times less heat dissipated by this
extension cord as compared with the longer, thinner cord.
However, it is dissipated over a shorter run (25 feet) and so
the power dissipated per foot is 2.048 watts / 25 feet or
0.082 watts/foot as compared with the longer cord:
19.48 watts / 100 feet or 0.1948 watts/foot. The thinner cord
runs hotter, even though it is dissipating its power loss over
a longer run. Its resistance increases as it warms and this
creates even more heat loss through the cord, over time.

7. The system uses more power (597.945 watts versus 579.84 watts
with the other cord).

So, the more conservative cord (12 AWG and shortened to 25 feet)
will cost more to operate the ten bulbs at the same supply voltage;
we get more light, less heating of the extension cord, and if this
were a single-phase AC motor load (instead of light bulbs) there'd
be less risk to the motor due to brownout (lowered line voltage at
the motor). Motors heat rapidly during brownout, dramatically
shortening service life, due to longer acceleration times of the
rotor as it spins up to operating speed (the longer spin-up times
increase heating).

Wire Constants (varies by manufacture):

12 AWG stranded: 1.65 Ohms / 1000 feet (606.06 feet per Ohm)
14 AWG stranded: 2.624 Ohms / 1000 feet (381.1 feet per Ohm)
16 AWG stranded: 4.172 Ohms / 1000 feet (239.69 feet per Ohm)

12 AWG solid: 1.5883 Ohms / 1000 feet (629.6 feet per Ohm)
14 AWG solid: 2.5252 Ohms / 1000 feet (396.0 feet per Ohm)
16 AWG solid: 4.0160 Ohms / 1000 feet (249.0 feet per Ohm)

Formulas:

Power: P = E * I
Voltage: E = I * R
Resistance: R = E / I
Current: I = E / R

The algebraic sum of the voltages in a series circuit is zero,
thus

Source Voltage minus IR drop (line)
minus IR drop (load) equal zero.

Don't fall asleep reading it. ;)

 
about 85F I would say..

Real solution is just to run 8 seperate 20 amp circuits to the room!

Soon! Soon! Got to buy my condo first. Then install wood flooring, then handle power circuits.



 
I'm gonna plug(hehe, pun intended) my 220V circuit opinions again. If you have 220v available at your breaker box and you are running wire specifically for computers, you will have less line loss and less heat from your wiring per wattage delivered to the boxen. The only downside is that 220v UPS boxes are a little harder to come by in the US. They are available however.


 
A better idea is to have a 240V line between the main panel and a 60A sub-panel in the computer room. In that room, divide it in circuits whichever way you want. The advantage of this is that you will just have one wire to worry about routing. If you end up selling, you can easily pull off the extra circuit and patch the holes to avoir scaring buyers :p

 
Back
Top