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

Why does the resistance of an inductor increase when an iron core is introduced?

  1. Mar 26, 2012 #1
    The inductor is initially an air cored one.When the Iron core is introduced,the resistance is found to increase considerably.Please explain.
     
  2. jcsd
  3. Mar 26, 2012 #2
    Resistance represents power losses; I e., Z(ω) = R + jωL. It is a real component of the inductor impedance.

    An iron core can have both eddy current (skin effect) and magnetic loop hysteresis losses.
     
  4. Mar 26, 2012 #3
    Assuming you are referring to Impedance and not Resistance, when the core is introduced, the varying magnetic field of the inductor induces eddy currents in the iron core. These eddy's have their own field which opposes the primary (inductor's) field, therefore increasing the impedance.
     
  5. Mar 26, 2012 #4
    Why is the resistance which is the real part increase?
     
  6. Mar 26, 2012 #5

    NascentOxygen

    User Avatar

    Staff: Mentor

    Losses.
     
  7. Mar 26, 2012 #6
    Like how? When you measure at DC, it is only the resistance of the wire and is constant no matter what. How is the induction of core change the DC resistance?
     
  8. Mar 27, 2012 #7

    NascentOxygen

    User Avatar

    Staff: Mentor

    DC? DC hasn't been mentioned.
     
  9. Mar 27, 2012 #8

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    DC resistance doesn't change.
    But if you measure AC impedance and phase angle and do a polar to rectangular calc you get a real component that is different from what a DC ohm-meter would report.

    That's because the core losses are energy losses which can only show up as resistance.
    And they're only present when excitation is AC.
     
  10. Mar 27, 2012 #9

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Yes exactly, sounds like the OP is confusing resistance and impedance and said resistance instead of the other

    Cheers
    Dave
     
  11. Mar 27, 2012 #10

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    DC was inferred as the OP spoke of resistance, not impedance

    D
     
  12. Mar 27, 2012 #11

    NascentOxygen

    User Avatar

    Staff: Mentor

    AC was inferred because the OP spoke of inductor, not solenoid.
     
  13. Mar 27, 2012 #12

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    I dont know whether he meant impedance or resistance,, but to notice a core he had to use AC.

    It's an interesting experiment. One should try diverse measurements for they always turn up some unexpected jewel of insight.

    AC resistance and inductance are both affected by temperature of core which at first is strange,
    because there's no temperature term in either L= N[itex]\Phi[/itex]/I or [itex]\Phi[/itex]=μNIA/[itex]\iota[/itex]ength

    It's the core's resistivity affecting eddy currents which both cancel flux and absorb energy.
    Inductance will go up slightly with core temperature because eddy currents go down..
    For Resistance you have competing effects between iron core and copper windings - i dont know which way it will go.

    It's most noticeable in un-laminated cores as were used in early PWR control rod position sensors.

    old jim
     
  14. Mar 27, 2012 #13

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Poor old OP.
    I bet he's scuttled off with his tail between his legs. It was his first post, too!

    Come back and tell us what you meant, Kal_Electri. You will get an answer once we really know what your question is.
     
  15. Mar 28, 2012 #14
    But R is the DC value only!!! Or are you implying the skin effect cause the R to go up and this has nothing to do with the inductance. It is pure resistance that increase with frequency!!! That actually makes sense!!! I think I answer my own question!!!
     
  16. Mar 28, 2012 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Yungman - check my thinking

    Iron losses in the core come out as heat so must cause an in-phase component of current. Else there'd be no net electrical energy transfer into core. P=VICos(Theta) and theta cant be 90 degrees if there's any watts heating the core.

    So they will appear to be another resistance in parallel
    which will show up as series in theveniin equivalent

    and the DC ohms will differ from the ohms in real component of complex Z.

    am i on track?

    old jim
     
  17. Mar 28, 2012 #16

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Skin effect and core loss are two different sources of additional Resistance. They can both be embarassing and will add to the DC resistance effect or even dominate.
     
  18. Mar 28, 2012 #17

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    What Sophie and Bob S said.
     
  19. Mar 28, 2012 #18
    The theory of eddy current losses in transformer laminations is given in http://www.elect.mrt.ac.lk/EE201_em_theory.pdf
    Page 7 shows that it increases as the square of frequency.
     
  20. Mar 28, 2012 #19
    I have no idea, till yesterday, the DC resistance was the only thing I consider. I am learning.
     
  21. Mar 28, 2012 #20

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

  22. Mar 29, 2012 #21
    To say very simply and plainly... The Iron has drift current (because of earth's temperature) more than air. So, naturally more magnetic flux into it causes more aligned electrons with respect to the magnetic field applied. Magnetic flux is induced into it because inductor has twisted loops (Remember Biot-Sarvat Law which explains about the motion of electrons in a circular loop). Oriented electrons, formed in this manner, are always in a way that they oppose the flow of current through it - This is the basic principle of the inductor. Now as the no of electrons increase the total opposing flux through it increases (Self- induction) and so more opposition force followed by more impedance.

    Remarkably when you place a similar coil near the same inductive coil, the same voltage is generated in it, which we call as a transformer. A transformer doesn't work in very low voltages just like an inductor, because the thermally drifted electrons are absent. This is as simple as that. Make it more generalized than specialized.

    Hope you get it... :)
     
    Last edited: Mar 29, 2012
  23. Mar 29, 2012 #22

    Averagesupernova

    User Avatar
    Science Advisor
    Gold Member

    I would agree that in order for heat to be generated the current would have to be in phase with the voltage. Would we look at this as a parallel resistance or a series one?
     
  24. Mar 29, 2012 #23

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    It seems to me intuitive to call it parallel with a lossless inductor but that's just my mental laziness tying to keep it simple . Magnetizing and loss currents would simply add at 90deg.

    But it could also be considered in series with an inductor of finite inductance

    One could make thevenin and norton equivalents for the transformer that would help visualize.

    To my tired old brain the polar to rectangular conversion of Z is the doorway to picturing it.
    Admittances in parallel add. I think of losses as an admittance that absorbs power. Picturing it that way it's easier for me to arrive at the formulas, but some people are gifted with a different thought path - they can start with the formula and arrive at the picture..

    When i can work something both directions is when i begin to feel confident about it.
    I had trouble with magnetics because my first course was taught in English units, gilberts et al,
    next one in CGS units, after that i picked up SI on my own. With my bad memory for names i still get confused.
    Repeat of a post someplace else - Jack M Janicke's book "Magnetic Measurements Handbook" is a great reference for the home experimenter. I built one of his fluxgate magnetometers.
     
  25. Mar 29, 2012 #24
    750px-Transformer_equivalent_circuit.svg.png

    in that (admittedly simplified) image, I think Rc are the core losses and Rp/Rs are the coil losses (DC resistance ofthe wire). its easy to see that its in parallel when you realise that its caused by the voltage rather than the current, as is the magnetising current (Xm).

    not only does the value of Rc vary with frequency but also with amplitude for an iron core, once you start getting close to saturation.
     
  26. Mar 29, 2012 #25

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Thanks EOW... that picture works perfectly for me and from it the formulas come naturally.

    Rc also varies with temperature because resistivity of iron affects eddy currents.
    But in a well laminated core they are small.

    One good picture is worth soooo many words !
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook