Corrections: (a) In the video I mistakenly selected only two regions in the core to calculate the core loss. You should select the entire core. (b) At 18:00, the copper loss expression should be divided by two since the solution is in the frequency domain (sinusoidal current). (c) I forgot to multiply the total core loss (PL_total) by 2. Since only half the geometry was modelled the total loss should be multiplied by 2.
You have set your operating frequency to 60Hz which you also specified in Steinmetz equation, but wouldn't you need to put the frequency as 0.06 since the frequency in Steinmetz is given in kHz?
Why is the electrical conductivity of air equal to 1? Shouldn't it be equal to 0? Then again, I tried to put 0 and the model doesn't solve. Why is that?
Technically the conductivity of air is very close but not exactly equal to 0. Using 0 can cause numerical instability in a 3D model. So using a non-zero small number would fix the issue. That's why I used conductivity of 1. You can even use 2, 0.5, 0.1, etc.
thankyou very much for this detailed video, i request you to please make a tutorial video for measuring high frequency ferrite core based high voltage transform, parasitics elements like leakage inductance and parasitic capacitance on both primary and seconday side of windings. thank you 👍
I'm not sure about making such a tutorial any time soon since it is closely related. I'm currently focusing on uploading a variety of topics. Thanks for your interest!
Sir as this is closely related , so you can do it in lesser time, but it will be very much helpful for student and researchers working on electromagnetics design of power electronics. thank you.
@@AS-zs7jw I haven't done it before but I suppose the definition of capacitance can be used to calculate it. You can also check this Comsol link for help: www.comsol.com/model/computing-capacitance-12689
Thank you so much for the video sir. I am facing a problem regarding "coil". I am trying to simulate in 2D geometry. And below the "coil", subsection named "Geometry analysis" is not coming in 2D. So I am unable to define the input and output of the coil. While in 3D geometry the subsection below "coil" named "Geometry analysis" is coming so we are able to select input and output of the coil. Please help me with this. How do I specify input and output of coil in 2D geometry?
In 2d you don't have to define the input and output. The direction can only be into or out of the screen and Comsol knows that. So once you've selected the coil you don't have to do anything else. The coil geometry analysis study step is also not needed.
@@JafScience So sir for defining into or out of the screen what parameter should we change? Is it that somewhere negative sign we have to give to specify "into" the screen as current direction? Because generally it is taking out of the screen as current direction. How do I give into the screen current direction?
Sir I was facing some problem while validating a paper. It is regarding "laminar flow between parallel plates with a magnet placed near it". Can I get your email id to discuss this problem? I have taken the necessary physics required but still not getting the flow pattern in presence of magnet.
If you mean temperature rise by joule heating, then that's similar to the tutorial of water-cooled induction heating coil. You can use heat transfer (ht) and mf interfaces and the multiphysics coupling computes the heat energy dissipated.
@@JafScience The scaling you did at 1:22. I am basically trying to replicate the model but for an inductor with an E core. Is there something I need to be aware about? Im thinking the only thing i have to change is only have one Winding instead of two. Also would the model work for a coupled inductor to measure the leakage flux and mutual indictance? PS: I appreciate the videos about COMSOL and big like for fast reply :)
@@PlayboyHZ The scaling is a quick way to make the core, to give it a thickness. I'm not sure about modelling leakage flux but to model an inductor then yes you'll need one winding.
Bro, how did you get the animation? I couldn't animate it, didn't work. The fact is that I tried to change the primary coil voltage from 1[V] to 1*cos(2*pi*60*t)[V] and add a time dependant study but didn't work. I don't know how to define this variable "t"
The variable t is already defined in Comsol when you use a time-dependent solver. I generated the animation for a frequency-domain study. In your case you can loop over the time values once you obtain the solution.
@@JafScience i got it. But did you change the frequency or something like that? Because I did all the same steps like you but my animation didn't change, all the frames were the same
@@xololomejor No I didn't change the frequency. The animation plays through one cycle (0 to 2 pi). Check my video on modelling magnetic hysteresis to see how to create animation for time-dependent studies.
I'm facing a problem my version is 5.3.a the flow of flux at the core only matches you if I disable the secondary core and take out the primary coil region from the ferrite material but of course having deferent values of loses. my question is why if i took the ferrite the flux flow matches you and why I'm getting wrong flow when i keep it.
@@JafScience first i go with the video but the results are wrong and deferent from video even the flow of B is wrong arrows the flow will be in primary side only. So when i remove the coil region from ferrite block the B flow came out like video bur the losses values are wrong thats what i change the rest like the video. Sorry for prolongation.
@@porlanvors3762 I'm not sure if there's something different in the coil feature in version 5.3. If all steps are replicated the result should be the same. If the flux lines do not follow the shape of the core, that could indicate a low core permeability.
@@JafScience Thank you for your quick response. I am now aware of that the 3D fillet feature is only available with access to the Design Module. Thank you anyway!
The concept of a transformer can be applied from this video. Use separate coil feature for each segment of the transformer. The thermal gradient depends on what yo want to model. Is it heat generated from hysteretic losses or something else. What are the boundary conditions for the transformer.
The type of coil depends on the geometry of the coil and the application it is intended for. Assumptions can also play a role, if you are trying to simplify your model. So more information is needed to determine the coil type.
In that case the only option is to use line current. If you have many wires close to each other I would recommend replacing them with a surface and using the coil feature with homogenized multi-turn model.
@@JafScience you have the best videos on RU-vid and there is nothing useful on Udemy. I think you could get some reward for your skills and willingness to teach 😀
