transformer operation on small time scales.

Discussion in 'Physics' started by wes, Aug 29, 2012.

  1. wes

    Thread Starter Active Member

    Aug 24, 2007
    242
    2
    Hi, I have been thinking about this for a while now and well here it goes.

    Transformer will be large, like 10 feet between primary and secondary.

    What happens when the secondary of a transformer is connected to a load (resistor etc) or connected to a larger load within lets say 10 ns. Now obviously when a load is connected to the secondary, current will start to flow and we say that the increased power on the secondary causes a increase of current on the primary so they are equal. Even if the current flow is tiny given the timescale, it is still increasing, just at a very tiny rate.

    But what happens for those few nanoseconds where the secondary has started to pull current but the effect of it dong so has not yet reached the primary and caused the primary to start to pull more current?

    Do you see where I am confused and really just not sure, I have a few ideas of what happens, but just not sure.
     
  2. HumanTrojan

    New Member

    Aug 29, 2012
    1
    0
    Wouldn`t a drawing panel be better than a message panel so people could sketch an answer :) just a thought ...
     
  3. wes

    Thread Starter Active Member

    Aug 24, 2007
    242
    2
    Yea that would be pretty awesome, I assume you mean something like paint imbedded on the page instead of having to upload image's.
     
  4. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,535
    We already allow about 3-6 different methods to post images, but drawing is more work, so people don't generally do it. I seem to be an exception.

    As for the original problem, the inductance would prevent an initial surge, it would be large. It may even interfere with the proper operation of the transformer.
     
  5. cork_ie

    Member

    Oct 8, 2011
    348
    58
    Isn't the primary current limited by the back EMF induced in the secondary?

    I would assume that there is a certain amount of capacitance in the windings and that will have the largest effect for the first few nanoseconds assuming 50HZ or 60 HZ sine-wave supply.

    In (say) 10 nanoseconds the sinewave will only have moved approximately 20 millionths of a cycle.

    The hysteresis of the core and inductance of the windings is going to delay any sudden current rise at a rate far slower than that.
     
  6. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,535
    Yes, and no. Transformers usually depend on back emf, absolutely. You are talking about something special though, a truly massive unit. Base inductance and counter emf are two different things, though usually the base inductance is so small it is negligible. From your description it may not be so anymore. I don't really know.
     
  7. cork_ie

    Member

    Oct 8, 2011
    348
    58
    Of course, you are 100% correct in what you say.

    The point I am really trying to make, and I apologise if I didn't make myself clear, is that there is unlikely to be any current increase in the primary winding in such a short time.

    The increased load on the secondary is just going to result in a drop in the induced voltage in the secondary until such time (a lot lot more than 10 nS but dependant on transformer design) as the reduction in back EMF causes the primary current to increase.

    To further complicate matters, because of the enormous size of the transformer, there is going to be substantial capacitance in the secondary. This is likely to have a more than negligible effect in this particular instance and should be considered as part of any overall measurements.
     
  8. cork_ie

    Member

    Oct 8, 2011
    348
    58
    Isn't the primary current limited by the back EMF induced in the secondary?

    I would assume that there is a certain amount of capacitance in the windings and that will have the largest effect for the first few nanoseconds assuming 50HZ or 60 HZ sine-wave supply.

    In (say) 10 nanoseconds the sinewave will only have moved approximately 20 millionths of a cycle.

    The hysteresis of the core and inductance of the windings is going to delay any sudden current rise at a rate far slower than that.
     
  9. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    I'm guessing wes might want us to think almost in the purely abstract ideal world with the transformer parameters essentially ideal. For the purposes of discussion one might need to allow some non-ideal conditions - such as finite winding inductances.

    Presumably 10 nano seconds brings us to the domain of the spatial propagation of electromagnetic phenomena. Light travels about 3 meters [about 10 feet] in free space in about that time frame.

    I'd start by considering an open secondary winding substantially isolated from the primary flux source but still enclosing all the primary flux (a nonsense I know). This winding will have a standing emf across its terminals. A load resistance is then applied to the winding - what happens next?

    My guess is the current will rise with a certain time constant - determined by the winding self inductance and the load resistance. Presumably it will take a certain time for the secondary magnetic near field "disruption" to propagate through the "magical" magnetic circuit linking the primary and secondary.

    Here I get lost because I don't know what the nature of the propagation will be.

    For me this question is reminiscent of the concept of an EM wave launched onto an ideal transmission line of substantial length and terminated in an impedance different from the line characteristic impedance. It takes a "while" for the reflected wave to return and effect the conditions at the source. In the intervening period an observer at the source is none the wiser as to what could be at the other end of the line.
     
  10. wes

    Thread Starter Active Member

    Aug 24, 2007
    242
    2
    yea I was just thinking about this in a ideal type of way to make it easier to figure out what happens.

    what did you mean by this t_n_k?
    "I'd start by considering an open secondary winding substantially isolated from the primary flux source but still enclosing all the primary flux (a nonsense I know)."


    What I think happens when a load is attached is the secondary current starts to rise and at the same time reducing the flux in the core, at the same time the voltage on the secondary would be decreasing until it reached 0. If you think about it, as the secondary's current reaches a level where the magnetic field equals that of the primary, then how could the primary still induce a voltage on the secondary? If the secondary started to become more powerful then the primary's magnetic field, it's own back-emf should prevent the current from going any higher than that of the primary, actually it should prevent this from even happening I would assume.
     
  11. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    It's a nonsense because it's not practically achievable. However it might be a useful starting point for discussion.

    My approach is to first consider an isolated [open] coil aligned co-axially with a uniform alternating magnetic field extant in the entire region surrounding the coil. A standing voltage determined by Faraday's Law of induction would exist across the open coil terminals. If one applies the reasoning implied in your quote above, then : when load is connected across the coil terminals a current will start to build up. This current flow would generate a field opposing the uniform field initiating the initial increase in current - ultimately leading to zero induced emf [presumably due to total cancellation of the exciting field by the current flow induced field]. Such a circular argument doesn't hold true, since the claim that the induced emf has reduced to zero would of necessity lead to the current returning to zero which brings one back to the initial zero current condition. One must then assume a cyclic astable behavior ensues with the current rising and falling ad-infinitum.

    One must therefore argue that either the current would never rise or that a new condition would exist where (upon reaching steady state) there is a constant alternating current flowing in the closed circuit by virtue of the externally induced emf in the coil.

    Having decided on this point, one might then consider the proposed conceptual transformer case, where the exciting field is generated by another [primary] coil whose flux just happens to completely link the original [ now secondary] coil. In the context of your original question regarding the effect of a significant physical separation of the primary and secondary coils and very small time frames, one must then come up with a mechanism / explanation for the associated temporal-spatial mutual field behaviour - a matter concerning which I have no knowledge. This is where I defer to the physicists.
     
  12. wes

    Thread Starter Active Member

    Aug 24, 2007
    242
    2
    So are you saying that the voltage shouldn't decrease to zero since that would mean the current returning to zero?

    What did you mean exactly by this t_n_k:

    "new condition would exist where (upon reaching steady state) there is a constant alternating current flowing in the closed circuit by virtue of the externally induced emf in the coil."


    I think you are saying that the induced emf from the primary should still somehow exist and cause the current to keep rising or falling? If that is true then would the primary in a sense be separate from the operation of the secondary? Well obviously that can't be true but I mean like as the secondary pulls current, that information takes a finite time to get back to the primary and cause the primary to increase its current pull as well. But during that time the primary's voltage, current and thus its magnetic field were still changing at some at some rate, say 60 hz. This rate of change of the magnetic field is always propagating back towards the secondary just as the secondary's magnetic field rate of change propagates back toward the primary. Since the voltage induced on a coil relies on the rate of change of the magnetic flux through that coil, then even with almost no flux at all, shouldn't the primary still induce a voltage? Say the flux was at 0 for a moment, in the next moment there would be a change that was equal to the last rate of change that induced the voltage on the secondary in the first place. So basically it all comes down to the fact that the primary magnetic flux rate of change is always propagating out and thus always propagating through the secondary and because voltage induced relies on rate of change of the magnetic flux and not strength overall, the primary is always inducing a voltage on the secondary.

    So where did I mess up, lol.
     
  13. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    I'm not sure you "messed up". All this speculation is like handwaving - it doesn't provide a clear reasoning based on knowledge of the physics at work.

    You could ask yourself what's going on in a radio receiver several miles from the radio transmitter. A coil in a tuned circuit produces a voltage due an electromagnetic field. Current flows in the coil. Does the transmitter respond to this current in the receiver coil? So energy can be transmitted without the need for the primary source to adjust its operating conditions in response to a system receiving energy from that source.
     
  14. wes

    Thread Starter Active Member

    Aug 24, 2007
    242
    2
    Well I was just looking for a big picture of how it works, I have no idea how it works at like the very fundamental level.

    As for the transmitter, wouldn't the receiver be so far away from the transmitter, like several miles, that the near field effects would be practically nothing? I have read a bunch of stuff on near field and far field effects but still don't totally understand it. From what I do though is that transformers operate off the near field effects whereas antennas (transmitters and receivers ) operate on the far field. Well technically they both operate off both, it's just one is designed to take advantage of the near field and the other the far field. So in the case of the transmitter and receiver, the electromagnetic field, not the magnetic or electric field separately, which are dominant in the near field, since you can have a large magnetic field and tiny electric field or vice versa, plays the bigger role and is the reason for the long distance effect. So since the far field is the bigger player at these distances and since the far field is energy that has been stolen you could say or given and does not need to come back to the transmitter, when the receiver receives some of that energy, it shouldn't need to effect the transmitter to cause more power pull since the transmitter has already given that energy to the EM field the receiver is getting energy from. Now if the transmitter and Receiver where within range of the Near-Field effects, then yes the receiver should most defiantly effect the transmitter.



    I am sure you have read this before but here it is just in-case:

    wiki near field and far field:
    http://en.wikipedia.org/wiki/Near_and_far_field
     
  15. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    I'm OK on near & far fields. Thanks for the heads up.

    So you are Ok with an isolated loaded coil in a field not suffering voltage collapse. Which is fine.

    Which probably begs the question - Do you have any further questions?
     
  16. wes

    Thread Starter Active Member

    Aug 24, 2007
    242
    2
    umm, non that I can think of at the moment. If what I said was correct, then not really, maybe later I will have some question after I think about it more.

    Thanks a bunch for helping to at least get a big picture idea of what is happening.
     
  17. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    Sorry I didn't respond to this question ...

    Well consider an isolated coil placed in a homogeneous alternating magnetic field where the coil generates a steady-state open circuit emf of 10V rms due to the field. I then connect a load of say 10Ω to the coil terminals. After any transient behavior subsides, the voltage drops to [say] a new steady state value of 5V rms - meaning a steady state current of 0.5A [rms] is then flowing in the closed circuit.

    Of course I could short the coil terminals and still obtain steady state current but obviously no terminal emf.

    Again to re-iterate I mean by isolated the coil is not mutually [magnetically] coupled to any other coil. It's all alone in a "sea" of magnetic flux.
     
Loading...