LTspice tutorial - Simulating Magnetic Hysteresis

[@NIOC630]

Somebody should pay this guy... (besides patreon)

[@bigboss9817]

Whoever controls the money, controls what a person is doing. The big strength of this channel is its independence. I think the only way to keep it independent are donations. And if I had a lot of money, I would gladly give big amounts to support Fesz, while insisting that he keeps doing exactly what he wants to and not what I want just because I fund him.

[@vipinpnair]

Some time back I searched a lot about how the inductance increases before saturation, based on a measurement report. Now got the answer. Thanks for this video . 😊

[@johnconrad5487]

" inductance increases before saturation" It decreases not increases.

[@vipinpnair]

@@johnconrad5487 5:48

[@vipinpnair]

@@johnconrad5487 7:53

[@vipinpnair]

⁠@@johnconrad5487see the video 7:53

[@ghlscitel6714]

I am electrified. This is an outstandingly excellent presentation. The most interesting part however is postponed. I'm eagerly waiting to see the next part.

[@DottorFrankie]

Very useful video for LTspice users. Thanks a lot 😊

[@pieters286]

Thank you - very well received.

[@aleksandar4923]

amazing

[@mr1enrollment]

thank you! very good again. looking forward to the measurements strategy.

[@dominikgs]

Very nice done! Thank you for the work.

[@sanjikaneki6226]

very interesting. Will you continue this series with model vs practical toroid inductor and transformer comparisons? Also how do people find the correct/proper toroid for their inductor/transformer or E EE EP core in general?

[@patrikborjan5169]

A very interesting and informative video as usual. Could you do one in the future on how we could obtain the coercivity, remanence and all those other values needed for a simulation from a core whose datasheet is not available? Thanks.

[@tomasbergh]

Great video!

[@analogdesigner-Jay]

Excellent work, thanks!

[@sairamt1790]

Thanks for your video and sharing the simulation knowledge 😀

[@InductorMan]

Super useful! Man I didn't know LT spice could natively represent minor BH loops. Question, wouldn't it be easier to model nonideal transformers by simply using the built-in transformer model with very high inductance windings and coupling the nonlinear inductor to that, rather than building the coupled transconductance model? I'm not sure if perhaps this is just natively what LTSpice does under the hood when simulating k elements (maybe I missed that) but I would assume LTSpice has tricks to make coupled inductor simulation stable and fast. Probably good to leverage those, right? And we saw that you needed to add a small conductance to the transconductance model to stabilize it, seems like a megahenry ideal inductor isn't really different. The one thing I found when trying this is that the simulator doesn't like paralleled inductors, so it wants some nominal series resistance added to the ideal inductor winding that parallels the nonideal inductor. But even if adding 1uOhm or whatever of series resistance should be problematic, the nonideal inductor can be put on a tertiary ideal inductor winding coupled to the active windings, and the series resistance can be added to only that tertiary winding, which should allow the main windings to exhibit whatever resistance is desired.

[@FesZElectronics]

You cant use the nonlinear inductor in a k statement directly, but it should not be an issue to put an extra nonlinear inductor in parallel with a linear transformer. I tried using a 1mH inductor in parallel with a transformer with 1:4 H, the one I used in the examples, and it works as expected. I think you are right - this is an easier way of creating this simulation.

[@homer7011]

Where in the video are the minor BH loops? This guy shows that it is not fully possible -- kzbin.info/www/bejne/f5PadZmQnM1rlZI

[@InductorMan]

@@homer7011 Is that right, one can't do it? I didn't try, I just grabbed the example parameters for the example nonlinear inductor from the LTSpice IX for OS X Help file in order to test my suggestion of using coupled inductors rather than G sources, and happened to notice a figure with minor loops illustrated. Sorry if I was being sloppy and made a false assertion. Rereading, I see that figure is accompanied by the text "The initial magnetization curve is given by Bmag(H) = .5 · (Bup(H) + Bdn(H)) Minor loops are obtained by various translations of the above equations per the cited reference. The core's absolute and differential permeabilities are a function of H and the history of values of H. The plot below shows the path taken by an asymmetrical minor loop for a typical power ferrite(Hc=16 A-turns/m, Bs=.44T, Br= .10T)." So reading the help file at face value I'd have assumed that it does actually simulate minor loops, and I'm also not sure where the presenter in the video you linked asserted that LTSpice doesn't simulate minor loops, I only listened for a minute or two after the timestamp. But again, if I was wrong in that statement, I apologize. I have not tried it myself.

[@vasiliy5578]

Thanks for the interesting video. I missed one thing: why, at 5:45, when you check the inductance value, the initial value is near 0.6 mH, while according to the datasheet, the 5943000301 core has 470 nH/turn^2, so 20 turns will give only 0.176 mH?

[@FesZElectronics]

I have no clue; I did not really get a good correlation between the inductance measured with an LCR meter and the value from the simulation... I think the exact measurement (and datasheet) value is obtained as a average when a slightly larger current is applied.

[@dreamflying-ko6tn]

​@@FesZElectronics I have the same question. Through these equations, I derived another inductance value equation: L = dB/di *N*A. This result matches the curve at 08:01, and its maximum value would be near 3.6mH. However, this result is significantly different from the 0.176mH calculated using the AL value of 5943000301. What does the curve represent if it is not the real inductance in the application?

[@johnconrad5487]

Thanks for making this video. Very instructive. For a long time I have wanted to model the BH curve but did not know how to do it. What is happening at 15:55? We see very big currents in the inductor which progressively get smaller and show the saturation curves.

[@FesZElectronics]

That current is from the other transformer implementation - the one with the G-sources and the non-linear inductor; I'm not sure why it looks like that right at the beginning, probably something to do with the initial state of the simulation.

[@mostafanfs]

Thanks for everything. Very informative. Could you please upload the files somewhere?

[@md.zillurrahman6171]

Thank you very much, it helped me a lot. Thank you man.

[@homer7011]

How about piezoelectric hysteresis?

[@sekabone]

Why are you not using LTSpice version 24? It has some new features, and runs faster than previous version.

[@FesZElectronics]

I'm much more used to version 17.. for this particular example though it should not matter all that much

[@Stelios.Posantzis]

1:40 Why does the inductance rise at the two ends of the time period?

[@johnconrad5487]

The curve is almost vertical during those excursions. as it goes into saturation the inductance drops towards zero.

[@rupeshmohanr9850]

Which circuit simulator is best for learning and creating devices

[@FesZElectronics]

I use LTspice, its pretty good

[@MUSTAFABİRGÜN-b7l]

How can I eliminate the noise between chassis and gnd? How can I simulate this via ltspice? Can you help me with this?

[@May-r5h]

你是我的神

[@paulpaulzadeh6172]

You didn't put inrush current in model, is it ??

[@johnconrad5487]

Inrush current in a transformer is usually due to capacitance. Inrush can be very small in small transformers and very large in big ones.