Energy stored in Inductor - Transformer core

Thread Starter

Skeebopstop

Joined Jan 9, 2009
358
Hi all,

sometimes we introduce an air gap into a transformer so as to not saturate the core as quickly when trying to store energy up in the primary (i.e. flyback converters).

Can somebody help clarify to me what happens to the stored up H field in the primary coil during the off cycle? How does this H field persist such that it pushes the flux through to create the required voltage emf in the secondaries to create electrical power.

I realize what I just stated seems like the answer, however I do not have a 'clear' visual image of this.

Any help would be greatly appreciated. See page 5 of attached app note for a clearer picture of what I am after.

Thanks
 

Attachments

b.shahvir

Joined Jan 6, 2009
457
Hi all,

sometimes we introduce an air gap into a transformer so as to not saturate the core as quickly when trying to store energy up in the primary (i.e. flyback converters).

Can somebody help clarify to me what happens to the stored up H field in the primary coil during the off cycle? How does this H field persist such that it pushes the flux through to create the required voltage emf in the secondaries to create electrical power.

I realize what I just stated seems like the answer, however I do not have a 'clear' visual image of this.
:) Hi, the phenomenon you are after is somewhat abstract in nature and hence a bit difficult to visualize. I myself have been dabbling with it for quite sometime now! A strong ‘H’ in the magnetic core tends to saturate it, as a result a finite reluctance in the form of an air-gap is introduced in the core circuit so that now a much higher ‘H’ is required to saturate the core. It is somewhat akin to a magnetic de-rating process.

Coming to your query on storage of ‘H’; It is helpful to remember that electrons in motion are infinitesimally small mechanical particles. Having said that, the mechanical particles undergo moment of inertia. When the voltage wave is at off cycle, the electrons still remain in motion as per the laws of particle mechanics since they are unable to cope with the speed at which the voltage wave passes thru zero.

However, it is important to note that this inertia is mainly caused by the changing magnetic flux associated with the electrons in motion (that vary their flow rate and direction of motion after each half cycle). This changing magnetizing force (‘H’) in turn ‘opposes’ or ‘assists’ the action of the electrical current, i.e., the changing magnetic field delays the fall or rise of the electric current thru the coil…..as each time the magnitude of coil current changes, the strength of ‘H’ also changes. As a result a self-induced alternating voltage appears in the coil (due to Faraday’s laws of electromagnetic induction) which ‘opposes’ or ‘sustains’ the flow of electrons at each minute step in the primary coil. This phenomenon is sometimes known as Magnetic Inertia.

Please note, my explanation may not be scientifically accurate (as the physics of charges in motion is quite complex), but this is how I have come to understand this abstract concept….awaiting a better and scientifically accurate reply myself ! ;)

Kind Regards,
Shahvir
 

Wendy

Joined Mar 24, 2008
23,415
Do you understand counter EMF? Basically the same function that makes motors efficient also works in transformers.
 

Nanophotonics

Joined Apr 2, 2009
383
Hi
It's really to do with the world of Physics. Some laws are fundamental and are considered to have no further core explanation other than to just accept it as it is, until someone else finds something with a deeper explanation later on. The idea of flux and the concept of primary-secondary transformer coils are explained by the law of electromagnetic induction. From an electrical and electronic point of view this should adequately make sense, however, a deeper analysis and understanding will require a very good and deep knowledge of Physics and it becomes really complex. As for visual image, for example, we know many theories about the electrons and its motion, but personally so far, I've never seen an electron as such, but instead we tend to represent it graphically for better understanding.
 

Wendy

Joined Mar 24, 2008
23,415
You can also overcomplicate things, to the point where they become harder than they need to be. The fundimental mechanism of Counter EMF is use in a lot coil type functions, in such things as solernoids, transformers, and motors.

If you want to design one from scratch then you have to dig deeper, but that rarely comes up.
 

Thread Starter

Skeebopstop

Joined Jan 9, 2009
358
Thank you all for your replies.

The concept of induction is clear. Faraday's law deals with 'flux densities', and as we all know, flux is not representative of H fields unless in non ferromagnetic materials. So Transformers certainly are a bit more convoluted.

Let me simplify the question for now, while I re-read Shahvir's comment to try and see if it matches my mental image.

H=uB can be thought analogous to Ohm's law. H is like V, u like R and the flux density is like current. So ignoring saturation and such for now, as long as B isn't saturated, and there is energy stored up in H or wherever, it can push B through to the secondaries in the off stage of a flyback converter. This change of B in the secondaries is what is inducing the voltage/current in the off stage as it is a rate of change of flux as H in the primary is used up.

If we can clarify this point up just to make sure my fundamentals are correct, I'll proceed with the more detailed analysis thereafter.

Cheers!
 

b.shahvir

Joined Jan 6, 2009
457
Let me simplify the question for now, while I re-read Shahvir's comment to try and see if it matches my mental image.

H=uB can be thought analogous to Ohm's law. H is like V, u like R and the flux density is like current. So ignoring saturation and such for now, as long as B isn't saturated, and there is energy stored up in H or wherever, it can push B through to the secondaries in the off stage of a flyback converter. This change of B in the secondaries is what is inducing the voltage/current in the off stage as it is a rate of change of flux as H in the primary is used up.

If we can clarify this point up just to make sure my fundamentals are correct, I'll proceed with the more detailed analysis thereafter.

Cheers!

Your analogy with Ohm’s law is correct. ;) There is a remnant flux ‘B’ which is still changing even during the off stage of the converter (Magnetic Inertia). It is responsible for the induced voltage/current in the secondary by Faraday’s laws of EM induction.

I am not good with equations though….so it will be difficult for me to analyze this phenomenon mathematically. :(

Cheers!
Shahvir
 

Thread Starter

Skeebopstop

Joined Jan 9, 2009
358
So, if I do saturate my core, my apparent primary inductance will decrease, however the amount of energy I can store up in its H field is essentially infinite dictated by 0.5*L*I^2. Naturally with a much smaller L now that I've saturated.

I suppose even air probably has a saturation limit, however the coil would generally burn out long before this I assume.

The question becomes, if I do saturate my core and my over current protection does not react fast enough on the other end of the switching FET, excessive Vds spikes can occur in the primary due to the energy store in the primary exceeding that which can safely be dumped through the flux to the secondaries (due to flux saturation), and as such appears back on the primary as a massive voltage until it can discharge it through something arcing or breaking down.

Does this sound about correct?
 

b.shahvir

Joined Jan 6, 2009
457
So, if I do saturate my core, my apparent primary inductance will decrease, however the amount of energy I can store up in its H field is essentially infinite dictated by 0.5*L*I^2. Naturally with a much smaller L now that I've saturated.

I suppose even air probably has a saturation limit, however the coil would generally burn out long before this I assume.
Your assumptions are correct.

The question becomes, if I do saturate my core and my over current protection does not react fast enough on the other end of the switching FET, excessive Vds spikes can occur in the primary due to the energy store in the primary exceeding that which can safely be dumped through the flux to the secondaries (due to flux saturation), and as such appears back on the primary as a massive voltage until it can discharge it through something arcing or breaking down.

Does this sound about correct?
Well i can't predict about the intensity of the spike voltage (so as to result in an insulation failure or otherwise!) but the resulting transients/spikes will definetly be quite prominent. There might be a possiblity of sustained harmonics too. :eek:...but i can't prove it to you mathematically though!
 

Thread Starter

Skeebopstop

Joined Jan 9, 2009
358
Ok, thanks for your responses thus far.

Now the question loops back a bit to the design phase of such a transformer. In following the Fairchild application note, the process of design is as follows:

1. Calculate primary inductance required to utilize duty cycle. This is independant of air gap/windings at this point, only dictated by the saturation flux density Bmax, voltage, power and frequency requirements.

2. Calculate the primary turns required to accomodate this inductance. This also is independant of air gap and is dependant on the primary inductance, maximum current, Bmax and effective core area.

Now what I don't understand is this. Inductance seems heavily related to the core material and its dimensions. So how can I introduce an air gap, thus shiftining my flux density saturationi curve, without impacting the primary inductance?

Is it more an approximation that the inductance introduced by the air gap will be considered negligible in relation to a non-saturated core?

I swear, that this will be the end of this thread once I get to the bottom of this final bit. The rest was a build up to here.
 

b.shahvir

Joined Jan 6, 2009
457
Is it more an approximation that the inductance introduced by the air gap will be considered negligible in relation to a non-saturated core?
Well, negligible would be too harsh a word i must say, it would be more like 'considerably small'. Of course air core cannot practically compete with non-saturated magnetic core if the magnitude of inductance is considered, but the reluctance of air-gap is considerable so as to modify the saturation curves of the magnetic core between which it is introduced.

Keeping this in mind, appropriate amount of air-gap (reluctance) is introduced to attain desired saturation curve for the desired application.
Hope this helps Matey. :)
 

Thread Starter

Skeebopstop

Joined Jan 9, 2009
358
Well, negligible would be too harsh a word i must say, it would be more like 'considerably small'. Of course air core cannot practically compete with non-saturated magnetic core if the magnitude of inductance is considered, but the reluctance of air-gap is considerable so as to modify the saturation curves of the magnetic core between which it is introduced.

Keeping this in mind, appropriate amount of air-gap (reluctance) is introduced to attain desired saturation curve for the desired application.
Hope this helps Matey. :)
So than, after introducing the air gap to avoid saturation, why do non of the app. notes go back and re-iterate their number of turns etc.. to accomodate for this 'new' inductance?

Cheers
 

b.shahvir

Joined Jan 6, 2009
457
So than, after introducing the air gap to avoid saturation, why do non of the app. notes go back and re-iterate their number of turns etc.. to accomodate for this 'new' inductance?
Cheers
Good observation! ;) practically they should....but as i'm not a design engineer so this reply is intuitive. Sometimes the note writer might have skipped this fact assuming that the reader will self compensate out of common sense. It may also be that the changes in the magnetic core due to the introduced air-gap might be considered quite small to be practically compensated for! But this is all said intuitively.

Cheers!
 
Top