Every HW Engineer should know this: Measuring EMC - Conducted Emissions (with Arturo Mediano)

[@Gengh13]

Only a few minutes in and this guy is really good at explaining. Thanks for always finding interesting people to interview and learn.

[@JamesLebihan]

This is the type of information that can save thousands of dollars in lab retesting fees. Having access to Robert and Arturos expertise for free just blows my mind. I'll be using what I've learned here in my next conducted emissions tests for sure. Thank you for making this video! Radiated emissions next?

[@RobertFeranec]

Thank you James PS: Radiated emissions are on my todo list. Possibly also near field measurement ....

[@XeveL2]

@@RobertFeranec I second near field and radiated EMC measurements ! Your videos are very clear and the interaction, with you asking questions, helps a lot !

[@dario_fresu]

Robert, awesome video as always, excellent topic! Something nice would be if your guests could share the presentation slides, or at least where we could have a look at them. Thanks for sharing, and all the best. Dario

[@nadeeragunarathna2410]

I have watched several videos and learned a lot. Many Thanks for this video series. Please do a session on FPGA programming

[@jayvadodariya207]

Truly thank you for making these kind of videos.... 🙏

[@jayvadodariya207]

Could you please make a videos on like how FCC/CE testing are perform. Like what are the tests in conducted/Radiated emission are performs in labs for certification. e.g ( surge,ringwave,ESD,EFT, Radiated etc..) It will be very helpful for us.

[@mariusdarau9997]

Very good explanation, do you have a link with the splitter used for the common mode noise and differential mode noise, please?

[@christophermatthews638]

I have seen some great videos on KZbin from Wurt where they used a SPICE simulation where they model the LISN as well as the circuit under test. It shows whether the noise is DM or CM. It also allows for having the test limits on the output graph. Very useful and illuminating.

[@Pumoneon]

Hey, thanks for the info. Could you share the link to the video you refer to? Regards

[@dario_fresu]

Hi, would you mind sharing the source?

[@christophermatthews638]

@@dario_fresu I have the youtube video url here: kzbin.info/www/bejne/hZfRqqSaYt6fb9E. I hope this works. I made a mistake with the company name its Wurth Electronik and the video title is Anticipate EMC with LTSPICE

[@christophermatthews638]

@@Pumoneon the link is - kzbin.info/www/bejne/hZfRqqSaYt6fb9E . Anticipate EMC with LTSPICE and the company is Wurth Elektronikl

[@sergeyivanov3453]

Excellent video for nerds. Very interesting

[@jebinsatheeshkumar]

Hi Robert, Thanks for putting lots of effort to create this video. Jebin

[@cirotremola]

Excelent Video Robert. Thanks a lot. Is it possible make one video about Uncertainty calculation a management in EMC testing and second video about Reberveration Chamber in Automotive EMC testing? Thanks

[@RicardoPenders]

Thanks for sharing all this knowledge here on KZbin Robert. I do have one suggestion for you maybe, it's great that you show all these very professional stuff but you know that this is pretty much all unobtainable to the normal guy, the hobbyist so to speak. Maybe it would be interesting to show what you can do when you're doing this for a hobby, what can you do as a DIY guy and you don't have access to the professional equipment, let's say you only have a cheap oscilloscope that has an FFT function like the one I have for example which is the Hantek DSO5072P upgraded from 70MHz to 200MHz and you don't have an arbitrary function generator but only a simple DIY function generator but still capable to inject some simple signals up to about 1MHz. Or maybe if the answer is that with my equipment it's impossible to do any of the tests or measurements, well than you can just tell me that with a comment that I can't do any significant or usable with the cheap equipment that I have that would be good too, an honest answer is always helpful even though it's something that I don't like to hear.

[@AlbertRei3424]

Why would a hobbyist perform emc test?

[@RicardoPenders]

@@AlbertRei3424 I guess I'm a crazy hobbyist, well sometimes you make something that really interferes with other stuff and then it gets a bit more important to do an emc test just to see if it's the thing you built that's interfering with your other stuff but nothing more than that just to confirm it's the thing you built interfering.

[@asierdanjou3905]

Great video! thank you very much. I have a question. If the EUT does not have ground terminal, why is necessary to measure common mode noise ? Is this noise radiated from ground plane instead of conducted ? thanks in advance.

[@bartek153]

another brilliant vid Robert. Thanks

[@forresterickson6225]

I am only sixteen minuets into the video but making notes. The linear frequency scale is helpful for showing that the interference is periodic and the amplitude of the harmonics are nearly all the same. This signal is a COMB in the frequency domain. The Fourier transform of a COMB back to the time domain is also a COMB. A COMB has very narrow on time. This is a clue that there are periodic very narrow pulses at the root of the interference. These are likely to be current pulses. This is a clue that your interference is conducted common mode emissions coincident with the switching supply edges. You could probably see them with an oscilloscope in the time domain by triggering the scope from the switching supply by connecting the oscilloscope to the LISN. Read about the COMB signal at Wikipedia: en.wikipedia.org/wiki/Dirac_comb Hope this helps

[@forresterickson6225]

A about 1 Minuet Robert asked about input protection built into the Spectrum Analyzer. I am sure for many CB (test agency companies) they would rather fail an external protection device and throw it away than fail something inside the Analyzer which then is an expensive fix by the manufacturer anyway. There will always be a way to fail the protection devices. The consequences of the inevitable failure is what you as the spectrum analyzer user / owner is you want to get to choose. Hope this helps.

[@forresterickson6225]

For people wanting to start measuring radiated emissions the RTL SDR, (Software Defined Radio) can become a spectrum analyzer with which you can start looking for the significant emitters in your projects. Hope this helps.

[@seanscon]

HI Nice video. If possible, please make a video on decoupling capacitors an how to determine their correct value, say on an circuit carrying an ARM CPU. thank you.

[@RobertFeranec]

Thank you Sean. I wish it was so simple, just ... I am not sure, have you seen this? That series started with the same question as you have: kzbin.info/aero/PLXvLToQzgzdcuyFu3KruvkR_TjpnF1Z7f

[@seanscon]

@@RobertFeranec i missed them, Robert. Thank you for pointing them out to me. I will watch them.

[@siddharthmali5841]

Excellent video. Very useful information. 😊

[@arpanrathore6826]

Thanks a lot both of you✌️

[@salvador3063]

Great lecture. Also wonderful content Robert.

[@ЯленьОгель]

Thank you very much, guys

[@cherryq91]

Excelent video! Thank You.

[@ppionke]

I guess there was no explanation regarding extracting (CM , DM ) information about mathematical operations on waveforms using an oscilloscope instead of D.M.R.N`s. The system was even assembled in 1:33:08 after all GJ!

[@jkhasahgl]

Thank you a lot for this video!

[@Graham_Wideman]

1:20:02 Robert mentions that the scope math function could be important. I don't think Arturo ever got back to that point, but I assume it's useful to perform the additions and subtractions that can separate out the CM and DM signals, without needing a hardware component to do those operations as shown at 1:42:30.

[@RobertFeranec]

yes, you are right. if people are interested, we can do more detailed measurements. We did experiment with the oscilloscope measurements & math functions, but I decided do not include them in the video as it would be better to use something what would create nicer and more visible results (a device with more clear DM and CM noise).

[@sepphuber7434]

Yes, because first of all a scope has more than one input (you need to measure both LISN outputs at the same time to separate DM from CM) and then you can calculate the DM and CM components.

[@jimjjewett]

I wonder if this would be more digestible if each chapter had been a separate, shorter video.

[@vginside]

Hello, Could you please share the tools model you used such as oscilloscope and spectrum analyzer and buy link too if possible? I have used Power Integration LNK305 chip for AC to DC and i have lot of EMI and i got no clue how to measure it.. this video will really help me...

[@andradecaio_]

Excellent!

[@edouardmalot51]

Thanks for this content

[@kacperbardecki3671]

Would I perform an emissions test on a PCB that isn't a power supply or converter?

[@Konecny_M]

Was mandatory AC coupling of the LISN measurement port mentioned? Because someone might be quite nastily surprised omitting that from a setup with SA otherwise.

[@Hardware_Team]

Thank you. May I ask what software do you usually use to simulate EMC, SP, SI ?

[@RobertFeranec]

Thank you Nguyen PS: We discuss simulators here, it may help: designhelp.fedevel.com/forum/main-forum/altium/17829-si-analysis-tool?p=18203#post18203

[@Graham_Wideman]

In connection with use of a scope FFT, it would be useful to hear why one would bother with a Spectrum Analyzer if the scope can do the job. I'm assuming that the Spectrum Analyzer does a better job in some dimensions -- ease of use, speed, frequency range and bandwidth, sensitivity, functions specific to comparing spectrum to limit graphs and so on. But then again, I don't know how good are the spectrum features on scopes at what price range. Like do you need a $10k scope to get one with spectrum features of a $2k spectrum analyzer? Comments?

[@ahbushnell1]

That spectrum analyzer goes down to -100 dbm. A scope has 8 bits or maybe 10.

[@theondono]

The trick is realizing *what* the device is doing internally and it becomes clearer. Oscilloscopes and SAs work in very different ways. When using a scope FFT, you are picking a “slice of time” and running maths on top. This means that FFT measurements that get close to the bandwidth limits, you’ll get poor results. In some sense, the scope can only measure the range from ~0Hz to F, but a SA can measure from f1 to f2, with f2-f1 in the kHz range, even if they are frequencies in the GHz. Since noise ~~ bandwith, the SA can give you way more sensitivity.

[@RixtronixLAB]

Nice video, thanks, Pi Filter will help us :)

[@liron28]

How do I lower 2kHz noise in my board?

[@UTBHolyGift]

Thank you for your good video. Could I ask you something? I have a Spectrum analyzer and I would like to EMI test(150KHz~30MHz) my SMPS PCB. What more do i need? AC Conducted Emission Receiver or Measurements?

[@avi-crakhome2524]

You were looking for the noise problem in simple circuits - no result. They looked for a problem in complex chains, and got even more confused. The reasons for the failure are in the tools used. Your attention is completely focused on the shape of the voltage, because there are many tools and techniques available for this. During the discussion, the word "impedance" was heard. Which is very rarely used for power electronics. But if energy values are used in calculations and stimulations, then the error will be seen immediately.

[@Graham_Wideman]

Who are "you" and "they"? What "no result" and "failure" and "error" are you talking about? And you remark that "impedance" is rarely used in power electronics. Well clearly the word "impedance" is appropriate in this video, as it centers around measuring noise using a LISN, which has "Impedance" right in the name.

[@larryandrews696]

According to the standard on how to setup and test for conducted emissions, none of this is correct accept for maybe the correct thickness of the ground plane he used. Please don't repeat any of this people! It will be wrong and you would never be 17025 certified by an inspector if he saw this as your way of testing.

[@nameredacted1242]

You can pre-test your device before going to an expen$$$ive off$$$ite te$$$t facility to make sure your product is designed properly, but I don't think you can EMC certify "yourself". You don't know the standard, you don't have certified and calibrated equipment, who will trust you?

[@deniakb]

Thanks, good job!