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

Why the efficiency of air core transformer changes (up and down)

  1. Jul 12, 2016 #1
    The graph below shows the efficiency of an AIR CORE transformer at different voltages (from MAINS). At low voltages, it has high efficiency and gets lowers because of high magnetising current causing heat loss. But why is it higher at some points again? I do not understand this pattern.

    Please give me a detailed explanation with scientific terminology.
    Thank you.

    Graph : https://postimg.org/image/vd50yniox/

    Table of current and voltage: https://postimg.org/image/cojojmvhd/

    Efficiency and voltage:

    The setup for air-core: the coil were put on top of each other with cardboard in between for support .(
    Coil used: http://int.frederiksen.eu/shop/product/coil-f--student-transformer--400-turns
    Last edited by a moderator: Jul 12, 2016
  2. jcsd
  3. Jul 12, 2016 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Looking at your graph it appears that at _one_ data point it has "high" efficiency (0.088%). Either side of that data point the efficiency appears more consistent with the rest of your graph. Is that data point correct?
  4. Jul 12, 2016 #3
    The graph looks like this if you zoom near the end parts. Link:https://postimg.org/image/t2y1eqhgh/
    I want to understand why it has a bouncing ball type graph (as it goes up and down and less up and down and so on)
    I am sorry but I do not understand what you are talking about.
    Thank you for your time
  5. Jul 12, 2016 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Ok I will try to explain.

    You said..
    So I looked at this graph..

    At low voltages (<1V) there are four data points.

    0.2V approx 0%
    0.4V approx. 0.09%
    0.6V approx. 0%
    1V approx. 0%

    Only one of these data points has "high" efficiency. Is it possible you made a mistake when recording the data?

    Is this...
    _all_ of the data from the experiment?

    If that is all of the data you have then the "bouncing ball type graph" is probably due to the curve fitting program you used. You do not appear to have enough data points to get a nice smooth curve that bounces up and down.
  6. Jul 12, 2016 #5
    I didn't make any mistakes. because i checked 3 times The problem is that I wanna understand why it has the "up-down-up-down" relationship and why the air core transformer behaves that way.

    The data i used to plot is the one below:
    I don't get why it rises again at the specific voltages and believe that people here can help me. I used the bouncing ball idea as it was easier to convey but it really does show a similar pattern although i agree that the data is a little low to truly call it that. But i am interested in the explanation of the rise-drop than the trend itself.
    Thank you for your time
  7. Jul 12, 2016 #6
    I apologise as I framed it wrongly. What i meant was that at specific voltages the there an increase in efficiency but as you from lower to higher voltages, the increase became smaller and smaller.

    Hope this helps you understand what i want meant.
  8. Jul 12, 2016 #7


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    What was the turns ratio of the transformer? I tried calculating it using the primary and secondary voltages and get either 3:1 (3 data points) or about 27-30:1 (8 data points).
  9. Jul 12, 2016 #8


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    How were you measuring the secondary current? What was the secondary load? A resistor?
  10. Jul 12, 2016 #9


    User Avatar
    2017 Award

    Staff: Mentor

    I don't think it is an effect of the transformer operation itself. I would expect some problems with data-taking, or maybe mechanical issues like a variable coil position (did you just put the other coil on top, or did you fix its location somehow?).

    You have three groups of efficiencies: 0.00033% (most values), 0.003% to 0.013% (three values) and then this one outlier of 0.088%. That could indicate three different setups, or some changes between them. The small differences within the low-efficiency group can come from measurement uncertainties, I wouldn't worry about those too much. The massive difference between the groups needs an explanation first.

    It would help to know more about your experiment, as CWatters already commented. A picture of the whole setup could be interesting as well.
    Which setup and which device(s) did you use to measure voltage and current? Did you change the range of the instruments?
  11. Jul 12, 2016 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    How did you determine efficiency? Power in vs power out ?

    How did you measure AC microamps to six significant figures the in presence of a line frequency magnetic field?

    What is physical arrangement of the coils? The meters and their leads?

    Be aware that in an air core transformer the flux is not constrained to an iron core, instead it's free to meander
    so as secondary current increases it will make flux find another path back to primary coil

    I think that a lot of what you measured is EMI.and you would have been better off to leave the secondary open circuited , using it as a flux detector, that way it wouldn't affect flux distribution


    When in a nice restaurant, if you shout demands at the servers you're apt to get soup spilled in your lap.

    old jim
    Last edited: Jul 12, 2016
  12. Jul 12, 2016 #11
    First of all, I would like to thank you all for commenting as i have been constantly bugged by this bizarre results and seeing people help me out makes my day. :)

    I totally agree sir but this is actually something i wrote for yahoo answers where people give minimal expression. I sincerely apologise if it came out rude or as shouting but i need the scientific explanation because i want to understand it. Also, I really like this how you made me understand sweetly.

    It was 400:400 which gives 1:1 ratio

    I used this coil for BOTH primary and secondary coil.

    I was using a bulb as a load (I think a load is an electrical appliance which uses up energy, please correct me if i am wrong).
    As for data collection, Digital ammeter and Voltmeter were used which had the option of A, mA and uA (the reason i could measure even small currents).

    Yes. It is efficiency so the ratio is made into a percentage.

    The coils have a square gap which you can see. So i put cardboard to hold it still.

    I believe so that i measured the EMI.

    Could you please elaborate on this?

    Actually, I am a grade 11 student working on this as my essay and i will credit the people who help me here. The problem is that if i give my images here then it will come as plagiarism. Therefore, I will give you a short summary and some other pictures- if any further problems then, I can provide everything i have (I sincerely apologize).

    The method went like this:
    • Powerpack attached to rheostat
    • rheostat attached to digital ammeter
    • Digital ammeter attached to Transformer coil (primary winding)
    • Transformer coil attached back to powerpack.
    • The digital voltmeter attached in parallel to the primary coil.
    • The secondary winding was attached to another Digital ammeter.
    • The digital ammeter attached to a bulb.
    • Bulb attached back to the secondary winding.
    • Another digital voltmeter attached in parallel to secondary winding.
    • All wires were copper wires.
    • All the voltage and current readings were taken from the respective devices and recorded.
    As for the setup of Air-core transformer
    The image below will help:
    Note: Coil were resting on the table and no gap between the coils.

    Interesting observation i forgot to mention: At higher voltages, the primary coil got heated to pretty high temperature. The time taken to increase in temperature decreased with Higher voltages. I didn't find significant heating in the secondary coil.

    Thank you for your time and guidance.
    Last edited: Jul 12, 2016
  13. Jul 12, 2016 #12

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    I too should apologize , you are likely not accustomed to local customs where i live
    so if you'll accept my apology and excuse this "grumpy old man" 's outburst we'll get along fine.

    That is impressive ! Glad to hear there are still practical courses in schools .

    I work in thought experiments... so,
    Fire up that young imagination !

    Your coil makes a "solenoid", try a search on magnetic field of a solenoid
    In yourimagination
    think of the magnetic field from your primary winding
    it might look like this
    of course reversing itself at line frequency

    Magnetic flux travels in closed loops just like current, so the analogy of "Kirchoff's Flux Law" applies.
    The picture just isn't big enough to show the bigger loops of flux
    but with no iron around flux does spread out into the surrounding space

    Now stack an identical coil on top of your solenoid but don't connect the lamp load yet .
    Some of the flux from your bottom solenoid, quite a bit of it in fact, will continue up and exit through the upper coil linking ts turns and inducing voltage there.
    I drew that flux in blue, please excuse my lack of draftsmanship. It's "MSPaint" after all...

    An open circuited coil in a magnetic field makes a simple flux detector - voltage is rate of change of flux through the coil , and whatever flux goes through it is changing at line frequency.

    Current is what pushes flux, and the bottom coil pushes some of its flux through the top one. Since the top coil has no current it doesn't push back against the flux from bottom one.

    Now connect a load to the top coil and let current flow in it
    Now the top coil can push flux of its own
    and it will push against the flux from bottonm coil because of something called "Lenz's Law"
    giving you in effect opposing electromagnets
    distorting the flux field
    so some of the flux from bottom coil won't go up into top coil it'll detour out into air, lowering the efficiency of your transformer.

    You could demonstrate that using paper iron filings and DC current through your coils
    i didnt draw the re-routed flux on this one , i think you're savvy enough to have figured it out by now ...
    observe opposing fields

    so any current flowing in your secondary pushes flux down against the upward flux from primary, lowering secondary voltage
    i think if you open circuit secondary you'd find secondary voltage linear with primary current
    but as soon as you allow current to flow it shape shifts your flux field .

    Now iron conducts magnetic flux hundreds or thousands of times better than does air
    so with ah iron core almost all the flux stays in the iron
    as shown in this picture from the datasheet for your coil that you linked
    http://int.frederiksen.eu/shop/product/coil-f--student-transformer--400-turns , click on pdf


    There's a small amount of flux that goes through the air
    indicated by the solitary black lines going there in that picture
    but a huge amount of flux going through the iron.

    The flux that shortcuts through air is a necessary evil in transformer design, it's called "Leakage Flux"
    and precision transformers arrange the windings carefully so as to minimize it.

    Your arrangement has a lot of it.

    Magnetics is rather fun
    i hope you get interested
    try searching on some of the terms
    and get a feel for the two basic units
    nowadays everybody uses SI
    The magnetism that flows around in loops is called Flux it's a diffuse continuous field but we draw it as lines to show intensity, SI unit is Weber
    and you'll find plenty of sites where intensity is shown by color
    and a few that still use old cgs units of Maxwells or Lines instead of Webers
    the force that pushes magnetic flux around loops is MagnetoMotive Force (MMF) , SI unit is Amp-Turns,

    most folks work in flux density Webers per square meter, called Teslas , that way their formulas apply to any size coil.

    I hope this gets you started.

    Now a practical thought
    When you connect a meter in the presence of an airborne magnetic field,
    any flux encircled by your meter leads creates voltage in them
    so the voltage arriving at your meter is in error by that induced voltage.
    It'll be small for your case because your 400 turn primary encircles a lot more flux than does the one turn loop made by your meter
    but i wonder how much it affected those feeble secondary current readings ?
    To reduce that induced test lead error you minimize the area they encircle by twisting them together. That's why you see so much "Twisted pair" wiring in audio work.

    It looks like you are in an exceptional school. I had the good fortune to be in a good one during my high school years.
    Have fun, learn a lot , think simple

    old jim
  14. Jul 12, 2016 #13


    User Avatar
    2017 Award

    Staff: Mentor

    Some current flows through the voltmeter. Knowing the internal resistance of the voltmeter would help, especially if you switched ranges in between (this changes the internal resistance and therefore influences different measurements in different ways).

    Quite sure jim meant electromagnetic interference. Noise from the environment. Then everything will influence the measured values, including the current location of the cables in the setup and even your own body position.
  15. Jul 12, 2016 #14
    I thought he meant electromotive force (EMF). My mistake.
    If you don't mind, can you elaborate on this
  16. Jul 12, 2016 #15
    I am sorry but the resistance of the voltmeter would be hard to obtain because my school is closed.
  17. Jul 12, 2016 #16
    It's my pleasure to be guided by smart and intelligent people by you.

    I have some question regarding your explanation but i will most likely post them later because it is 3:32AM here.

    I am very happy to have commented on this forum because not only everyone has helped me but also i have learned a lot of new things;So, do not feel annoyed if i ask a lot of questions.

    Thank you for your precious time everyone
  18. Jul 12, 2016 #17
    Your explanation is really great but still does not answer why there is a up-down-up-down graph than for eg. 1/x relationship
    Questions hoarding you later
  19. Jul 12, 2016 #18


    User Avatar
    2017 Award

    Staff: Mentor

    Alternating currents, e. g. from the power grid and all connected cables in the room, lead to changing electromagnetic fields around them. Your secondary circuit might pick those up, independently of the transformer coupling. It works a bit like a radio: you can receive radio transmission without connecting the radio to anything. You just need an antenna - and your circuit acts as one.
    Sensitive AC equipment never operates at line frequency for this reason (60 or 50 Hz depending on the region), unless it absolutely has to. Too much noise.
  20. Jul 12, 2016 #19

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    It'll take some more investigation
    Using your data

    could you plot secondary voltage on vertical versus primary amps horizontal ?
    The closer that is to a straight line the less your upsy-downsy was due to secondary coil's MMF pushing backward against primary coil's MMF
    Since the ratio of currents was like 10,000 to one i expect a fairly straight line.

    Primary voltage is divided between I X R drop in the winding and voltage required to push magnetic flux , called "induced voltage" or "counter EMF".

    Have you an idea about the resistance of your bulb load ? Plot secondary volts versus amps and see how straight is the line.
    Then you'll begin to get a feel for what your transformer and load were doing.

    One observation is worth a thousand expert opinions. See what you can glean from your observations.

    Have you had complex arithmetic yet ? Are you accustomed to polar and rectangular representation of volts and current ?
    Last edited: Jul 12, 2016
  21. Jul 13, 2016 #20
    A really simple to understand explanation. Thank you very much.
    But can elaborate a little about how my equipment could have read this at the 50 Hz region (Singapore)? Is it because of going very low for current readings or something like that?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted