Electronics tutorial - Inductor magnetic field emissions

#94 #electronics #inductor
In this video I look at the magnetic field emissions of various types of inductors. What I want to see is how the way in which the inductor is built influences the magnetic fields around it. I proceed the measure the magnetic emissions by using a magnetic field probe and since the signals are very small, this episode marks the return of the Cookie-Box!
Tested inductors:
SDR0805-100ML
PSDB1004NT100
DJNR8040-100
DE0704-10
HPI0624-100
Datasheet:
www.tme.eu/Document/7c752b00f...
Magnetic field probe: • EMC debugging - Near f...
Special Thanks to all my supporters on Patreon! Especially @afiskon and Xegomen!
If you liked this video be sure to check out my other videos and you can also subscribe to be up to date with all the new ones!
If you want to support the creation of more and better videos please consider checking out: / feszelectronics

[@WalidIssa]

Great video and like the in-depth investigation .... I generally like Power Electronics related videos more than communications and signals ....

[@FesZElectronics]

I try to keep a balance between multiple subjects.

[@ena6631]

Impressive, well done.

[@wi_zeus6798]

Absolutely awesome video!

[@Parirash123]

Thanks for making this video.

[@nbuzkila]

Another excellent video!

[@rabindrasharma]

7:45 the correct name for it is dumbells

[@NatFablife]

Now that’s suuuper cool the way you put your own spin on it!

[@LousyPainter]

Awsome video!

[@alexinal8514]

Great video, thank you so much for the great clarification )))

[@Arijit_VU3ICT]

Please sir make a video on FFT. What is its importance in signal analysis... Thank you...

[@CARlosDAN783]

Nuevo suscriptor amigo saludos desde mi amada Venezuela.

[@billwest257]

Another great video, thanks. You inspired me to start looking into the powder core inductors. In looking around I came across this document from Wurth on EM emissions and the mention of powder inductors shielding falling off over 1 MHz caught my eye. The document also has lots of other good information like connecting the start of the winding to the switch node. You are becoming my go to guy for EE information.

[@FesZElectronics]

Is it this app note by any chance?: www.mouser.com/pdfDocs/ANP047b_ENG_The_Behavior_of_Electro-Magnetic_Radiation_of_PI_in_Power_Management.pdf its quite a nice document with very useful info

[@billwest257]

@@FesZElectronics Yes that's the one, I thought I had put the link in my post, oh well. I'm thinking about building your test fixture to test candidate inductors.

[@FesZElectronics]

I guess one thing that I didn't cover, but should be mentioned is how the inductor construction affects the parasitic parallel capacitance - the metal powder core inductor has the highest parasitic parallel capacitance and although it is best at "low" frequency magnetic emissions - it has significantly higher emissions at high frequencies when used in switching converters. I guess the best way to test this is in an actual converter and checking not just the switching frequency and first few harmonics, but also up to ~300MHz all the ringing and stuff.

[@AfdhalAtiffTan]

Could you measure a toroidal inductor too?

[@corneliufilip5671]

Thank you for useful information.

[@falconhawker]

Disapointed that the camparison between inductor types did NOT include pot core inductor! ( orders of magnitude better ? or NOT)

[@sajidnaseeb5386]

This one is a nice video. I was working on simulating the conducted mode emission of a real buck converter using LTSPICE. After splitting the common mode and differential mode noise, I found that the conducted mode emission is way more than the EMC standard limits. So, I used the open source model of common mode choke filter. As a result the common mode noise were suppressed. But, now I want to make a simulation circuit of common mode noise filter in LTspice of my own which should give accurate results. Can you tell me how I can design it in ltspice. Any Book, Documents that could help me design it in LTspice?

[@sajidnaseeb5386]

If you have worked on this filter earlier, can you share with me your files that could help understand those circuits better.

[@p_mouse8676]

Since noise isn't correlated to the signal, you can improve the measurements by taking averages. This particular scope does have that option. Also, there are quite some more types of shielded inductors. Best known one is the toroidal inductor. But were the magic really happens, is when we also take the current vs inductance graph into consideration (saturation point). Or rather, were the rabbit hole starts, since it clearly shows that there isn't a perfect solution. A powered inductor may be great for shielding, but is pretty bad in keeping a constant inductance. Therefore it's totally unusable for audio or signal applications, because this means distortion is being introduced. Nice video again btw 👍🏻

[@FesZElectronics]

Well, of course there are other parameters involved in inductors, depending on the application, one or another is more important. I focused on these inductors, because these are most commonly used with high currents - in switching supplies. A low power signal inductor will not radiate that much because of the small currents. I did not test a toroid inductor mainly because its not that common - its a more expensive component because of its construction. Also to properly measure that one, you need to look at magnetic fields on multiple axes, since that inductor doesn't have a clear plane in which the windings sit.

[@p_mouse8676]

@@FesZElectronics I guess "not as common " is very relative. Toroidal inductors are actually very common in automotive applications as well as high current applications, mostly actually because of their price vs performance ratio. Their saturation isn't as steep and strong compared to ferrite. The downside is that their shielding isn't as good as well as a slightly higher DC resistance.

[@p_mouse8676]

I guess what I am trying to say, is that context is pretty important. Each type of inductor has its own unique compromises.

[@FesZElectronics]

You are right, its a perspective difference. My experience in power converters is mostly limited to 1-15W range where these types of inductors (that I analyzed in the video) are more common. I assume at higher power levels the toroid construction does become predominant. In what sort of converters did you see the use of toroid's? Also isn't the core material still ferrite?

[@oriole8789]

@@FesZElectronics I don't have much experience with automotive, but I've seen toroidal inductors pretty often in industrial and aerospace equipment. For aerospace, toroidal inductors are especially common if they share the circuit board with other digital/analog circuitry. Powdered core is the best value for money when it comes to power vs noise, so you see them everywhere in PC parts for instance (CPU/GPU regulation). Very cool comparison and explanation as always. That cylinder for the sense coil looks suspiciously similar to a cut portion of a toilet paper roll core haha. I've used those a few times as board supports when doing some BGA rework. Hey, whatever works~

[@axk1]

Thanks for the video! As always very clear and informative. Have you looked into magnetic coupling based in-wall mains wire tracing? Most of the detectors appear to be based on capacitive coupling and it doesn't always work because walls can be conductive and it would register everywhere along the wall not just where the wiring is. I'm trying to find some info online. Found a youtube video a guy using a 1mH inductor and an amplifier ("How To Locate Wires In Walls(EMF Detection)") but with the inductor's axis perpendicular to the wall it appears that he's getting capacitive coupling instead of inductive. I wonder if this type of tracing is feasible/practical with inductive coupling. I'm thinking about injecting some high-frequency current noise (10mA peak) with a simple switching load at say 100KHz and then sensing it with a 15mH inductor. Not sure if it's feasible though, it might be that the magnetic field ends up so small that it's overwhelmed by parasitic capacitive coupling even into a low impedance e.g. 75Ohm like in your video.

[@FesZElectronics]

The problem with generating magnetic fields is that you need a certain current to pass trough the wires. You could detect the magnetic emissions only if there is some sort of load added at the other end. The advantage with electrical field based detectors is that they do not require a load present, just an alternating voltage to be applied.

[@robertburget]

Hi, great videos. I am wondering how the result will change when you swap the polarity of those inductors. For example Coilcraft has power inductors with marking of the high dV/dt input for better noise shielding

[@FesZElectronics]

The polarity will help with electric field noise, not really with magnetic fields - in case the coil is connected in an SMPS, the terminal on the switching node sees more noise than the other, so if that terminal is inside the coil (rather than on the outside) the coil will generate less electric field noise;

[@josip1881]

Did you try to play with different core shapes/cross-sections, would be much of difference if one would make inductor on hexagonal core? triangular core?

[@FesZElectronics]

Honestly, I only tried commercially available components. But the geometry of the core should not have that much of an influence on how the field spreads out. A more important impact will come from the way the turns are arranged - if you have a single turn multi-layer inductor the filed will be more spread out and if you have a single layer multi turn, the field will be more concentrated in a specific direction.

[@kevy1yt]

I know this is an old video, but what if you took the best inductor shown and enclosed it in a metal shield itself and ground that metal shield to the circuit board ground. Wouldn’t that reduce emissions to nearly 0?

[@FesZElectronics]

That would also help, but perfect shielding will never occur... Vishay has a series of inductors called "IHLE" where they use powder core to get the best magnetic field enclosure but also a metal "cage" to reduce the remaining emissions. Maybe other manufacturers also offer similar technologies.