6

For an application we are considering pulsing a current through a conductor.

The conductor have an ampacity rating of 12.5A for continuous operation.

During the pulse we get currents around 25A and are therefore limiting the on-time to <50% of the period giving us a mean current of 12.5A.

Is this reasoning correct?

Edit:

Period = 1 second

Size AWG kcmil: 14

Material: Copper

Temperature rating of conductor @ 90C = 25A

Number of current carrying conductors = 10-20 => Percent of values adjusted for ambient temperature = 50% => 12.5A

Shape of pulse is rectangular

user1029697
  • 71
  • 4
  • Can you provide the data sheet or part number for the cable? There doesn't seem to be enough information to confidently answer the question. – travisbartley Aug 21 '13 at 08:17
  • 4
    It's also worth mentioning the time period of the pulses. – PeterJ Aug 21 '13 at 08:49
  • 1
    And the shape of the pulses. – Stephen Collings Aug 21 '13 at 11:23
  • And what kind of switching period we're talking about. There's a big heating difference between millisecond pulses and ten minute pulses! – Stephen Collings Aug 21 '13 at 11:33
  • 1
    Not posting as an answer as I don't have one, but both the existing answers raise red flags to my mind: Surge power averaging out equally isn't a good measure, as temperature rise is affected by heat dissipation *rate* (surface area, insulation, cross-section, other factors). Also, current alone isn't a good measure (RMS or any other method), since at one point a given metal will melt or vaporize if sufficient heat is generated within the active part of the duty cycle - The conductor metal is not specified, and that's a factor. There are way too many factors to allow a definitive answer. – Anindo Ghosh Aug 21 '13 at 11:45
  • @user1029697, you may wish to consider accepting one of the answers. It's an important part of the process. – Stephen Collings Jan 08 '14 at 16:27

2 Answers2

3

This is a question of resistive heating, so it's about RMS, not average current. You need to maintain an RMS current of 12.5 amps. That depends on the shape of the pulses and the duty cycle. If we assume your pulses are perfectly rectangular:

$$ I_{RMS}=I_{PK}\sqrt{D} $$

Plug in 12.5A RMS and 25A peak, and your duty cycle is .25.

Now, this is only really useful if your pulses are short and close together. If you're on for ten minutes, off for thirty, you may have a 12.5A RMS, but it's an RMS over a much longer period than the thermal time constant of the wire.

The thermal time constant of the wire will be on the order of 30 minutes to a few hours. Welders use a 10 minute cycle for "short enough to use RMS".

Community
  • 1
Stephen Collings
  • 17,373
  • 17
  • 92
  • 180
  • 2
    "Perfectly rectangular" pulses have infinite energy and would annihilate the universe. – travisbartley Aug 22 '13 at 08:11
  • 2
    Ha! That's funny. Reminds me of that Batman comic where Batman drives off a tiger by generating a sound at 20,000 dB. (Yes, dB.) It would indeed drive off the tiger. All the tigers. Everywhere. – Stephen Collings Aug 22 '13 at 10:18
  • If 340 dB is the equivalent of a 14 gigaton bomb, 20,000dB is probably enough to vaporize most of the galaxy. http://www.makeitlouder.com/Decibel%20Level%20Chart.txt – travisbartley Aug 22 '13 at 10:42
1

No, it's not correct. The power is proportional to the square of the current, so doubling the current will lead to 4 times the power. You'll have to limit the duty cycle to 25%.

Johan.A
  • 1,806
  • 1
  • 15
  • 27
  • 3
    I don't trust this. Deriving the max continuous power from the max continuous current can not tell you the max surge current and duty cycle for said current... – travisbartley Aug 21 '13 at 08:13