Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I need help understanding how skin effect works

  1. Jan 27, 2014 #1
    I recently started working in low voltage electrical repair/upgrades in automobiles. I have a couple interns that i am working with and i truly want to be dam* sure im explaining things properly to them. Me and my boss have had heated discussions on the subject and i believe him to be incorrect on some understanding but as he is 50 years my senior and far more experienced i fear that perhaps it is I that does not fully understand how AC and DC power flows through circuits.

    I would like to lay out what i think to be true here in the hopes of being corrected by those who know far more than I and perhaps even be given some good documentation on how it all works. I have found some MIT youtube lectures which is where i have gotten much of my information to date.

    "I have come here in hopes of finding more scientific reasoning and rational as most of what i have found online thus far is conjecture and biased personal opinion from working in the industry with little to no documented backing to there opinions.

    I work with 12-16V DC current for power and 20hz to 35Khz with AC power. 20-150HZ operating over 16-8ga stranded wire, 150-1500hz on 16-18ga wire, and 1500-35,000HZ on 18-22ga wire.

    Solid core wire will have greater current capacity than stranded over long runs due to it having less resistance and thus less inductance to reduce the skin effect depth at higher frequencies. solid wire is also cheaper to manufacture and is more durable which is why its used predominantly in houses for electrical power and not so much in cars.

    DC current does not have frequency and thus skin effect has absolutely no place in the discussion because the electrons travel through the ENTIRE wire and not just on the surface. in that same breath AC current at low frequencies is not effected by skin effect as the skin effect depth is often greater than the diameter of the wire. case in point a 100hz signal traveling on an 8ga wire has a skin depth of nearly double the diameter of the wire and thus skin effect has absolutely no bearing on conductivity.

    as a wire heats up due to inductance the resistance increases and the usable skin depth lessens along the wire.

    solid core wire is smaller in diameter than stranded as due to the air pockets in stranded wire it must be larger than solid to retain the same current capacity.

    does stranded wire share a common skin effect or does each individual strand have its own skin effect occurring? (assuming this is not litz wire)

    Litz wire is advantageous at very high frequencies as it can carry a far greater load due to each strand having its own skin effect and there by increasing the usable surface area for signal transmission.

    when using CCA or CCS wire in high frequency transmission (<Ghz range) only the outside of the wire is used and thus there is little to no signal degradation due to the wire not being pure copper. however when dealing with DC currents having CCA or CCS wiring will lead to greater heat increase as the capacitance of the wire is greater and aluminum is 25% less conductive than copper so the CCA/CCS wire would need to be 25% larger than a pure copper wire to carry the same current. also with the increase in heat in the CCA/CCS wire you will loose capacity as your resistance will climb with the heat.

    OFC wiring has a negligible difference when compared to pure copper wire in both oxidation over time and over all capacity/resistance of the wiring.

    finally how does soldering a wire effect its specs? say i take a stranded wire and solder it to a stranded wire then i have created a solid wire in between two stranded wires. how does this effect the flow of electrons?

    does anyone know definitively how the various types of solder effect conductivity?
  2. jcsd
  3. Jan 28, 2014 #2


    User Avatar
    Science Advisor
    Gold Member

    Welcome to PF.
    A few points:
    Wire heats up due to Resistance and not Inductance.
    The use of solder joints need not affect the overall resistance as long as the joints are good (a good area of overlap and not corroded or 'dry'). Clean ends, fresh solder and the right temperature will ensure that there are no problems of that sort. The actual current path through the metal of the solder is very short.

    The increase in resistance as a wire heats up should not be relevant for cables that are being operated at their proper rating (the insulation will melt long before the resistance changes in a significant way).

    You have, presumably, seen the Wiki articles on skin depth and they tell you the skin depth of copper at various frequencies. Skin depth is not a line below which there is no current. It just represents the depth, above which most current is carried; it's a convenient indication, in the same way that Half Life is used to describe the rate of radioactive decay. For non-Litz, multi strand cable, the skin depth is more or less related to the outside of the whole cluster. I don't think the 'internal gaps' between the separate conductors do much because there is no significant field there.

    Litz wire has its advantages but it is a real pain to use and only exists (afaik) for thin conductors within equipment and components. Where conductivity is a problem (as with high power radio transmitters), the conductors carrying the RF power are usually just made thick and you'll find that tubing is often used, for cheapness.

    I understand that you are finding this interesting but, if you want to know what's best for any particular application, there is loads of information, available from manufacturers and regulatory bodies, to tell you what the normal requirements are and what is standard practice. There will be other factors (like the environment that the cable is in) that will be equally relevant to its viability for use in any particular circumstance. Just work tightly to the cable spec and you can't go wrong. Straying from the spec can land you with unexpected problems - unless you are very well qualified and know what you're doing. (Insurance companies can be very fussy, when it comes to a claim, for instance.)
  4. Jan 29, 2014 #3


    User Avatar
    Gold Member

    A few misconceptions stand out.

    less resistance and *thus less inductance* does not make sense to me. I think resistance and inductance are pretty much independent.

    Stranded wire is gauged by the same copper area, not by cross section.
    The nomenclature is also different:
    "For example, a 22 AWG 7/30 stranded wire is a 22 AWG wire made from seven strands of 30 AWG wire."
    The fusing current for equivelent copper cross sectional area should be similar.

    Stranded wire has less inductance for a given diamenter, but is similar for equivalent cross sectional area.

    no way. inductance has nothing to do with heating.

    correct, but it is also less flexible

    pretty much a common skin effect. Go to wikipedia and read about proximity effect.

    Not much, but if you are talking extremely high currents it could be noticable

    In summary:
    Solid is cheaper, less flexible, less surface area for corrosion.

    Stranded is more flexible, which is probably its main advantage.
  5. Oct 22, 2015 #4
    Conductors can heat up due to resistance or inductance.
  6. Oct 22, 2015 #5


    User Avatar
    Science Advisor
    Gold Member

    (Real) Inductors can get hot but it will only be due to the Resistive Component of their Impedance. Inductance (Reactance) cannot dissipate power, which is V.I (the dot product)
  7. Oct 22, 2015 #6


    User Avatar
    Science Advisor
    Gold Member

    This is all pretty much an incorrect statement, lots of mis-understandings
    Others have commented on the current capacity and the inductance issues

    Only true for DC, Not AC in power transmissions lines, homes or any RF situation ... have a look at any significant power transmission line
    they are all stranded

    No to your comment on skin effect on a solid conductor ..... stranded conductor has more surface area
    As the frequency goes up, the skin effect becomes more pronounced, and as a result most of the current is flowing in the surface region of the conductor
    You could make the conductor hollow and not increase losses. Therefore having a multi strand conductor results in a greater surface area and therefore
    a better current carrying capacity for AC

    no, it isn't more durable ... as others said, it's much less flexible, No, its not predominantly used in homes. You have to go back 60++ years to see single strand cable used in buildings .... Everything I have seen installed in at least the last 40 yrs is all stranded

  8. Oct 22, 2015 #7


    User Avatar
    Science Advisor
    Gold Member

    here's a few examples of HV AC stranded power cables .....



  9. Oct 22, 2015 #8


    User Avatar
    Science Advisor

    The top picture in post #7 shows a three phase Ultra High Voltage power transmission line, (maybe a megavolt). The nine near conductors in the image are all connected together as one phase, forming a cage. By using a cage the effective electrostatic diameter of the conductor is increased, which is done to reduce corona discharge from the wire. The current of one phase is shared amongst all nine conductors.

    The forces due to wind on a suspended wire are significantly greater than forces due to gravity. A cage offers a lower cross-section to the wind than would a hollow tube, so the towers can be lighter and the voltage higher.

    By using aluminium wires spun around a steel wire core, the tensile strength of the bundle can be greater and the weight less, so the towers with insulators for support can be further apart. Very little current flows in the steel core.

    The two legged gantry in the distance, with the three shorter pylons, supports the three phases, each on three long chains of insulators.
  10. Oct 22, 2015 #9


    User Avatar
    Gold Member

    So the 8 conductors arranged into a circle create one effective conductor. Is this voltage or current or frequency dependent?

  11. Oct 22, 2015 #10


    User Avatar
    Gold Member

    I have never seen in a domestic installation a stranded in-wall conductor (2mm ish or less diameter). However in all of the industrial installations I have seen stranded conductors used everywhere. In the more controlled areas the stranding is in the 2-3mm diameter per strand arrangement. In the outdoor situations the strandings is more like an extension cord. Probably hundreds or possibly thousands of strands.

    Do the regulations outside the US require stranded cabling in domestic in wall installations?

  12. Oct 22, 2015 #11


    User Avatar
    Science Advisor
    Gold Member

    Only lighting circuits would fall within that range, but not power sockets

    I don't know the regs well enough to know if it is a requirement ?

    All I know it is the practice in at least Australia and NZ

  13. Oct 23, 2015 #12


    User Avatar
    Science Advisor

    There are 9 conductors in the example. This is simply a high voltage effect. The air is ionised near tight radii which results in corona discharge.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: I need help understanding how skin effect works
  1. Skin Effect (Replies: 7)