The Long Overdue Introduction!: EMC For Everyone #1
ฝัง
- เผยแพร่เมื่อ 12 ธ.ค. 2024
- The Long Overdue Introduction!: EMC For Everyone #1
After what seems like literal years of me teasing this series, it is finally here! This is the first part of the EMC For Everyone Series. I go over the theme of the series and how it will be structured, as well as do some testing on a couple of boards.
-- Supplemental Video --
/ emc-for-everyone-58374344
-- Evaluation Board Series --
• Custom Eval PCBs
-- Support me on Patreon! --
/ microtypeengineering
-- Discord Server --
/ discord
-- About MicroType Engineering --
MicroType Engineering is a full service electronics and mechanical product design firm. We offer turnkey support no matter how far along in the design process you are. We have full schematic capture, PCB design, firmware, mechanical design, and prototype/small-batch assembly services. Please reach out to learn more! microtype.io/
Only 2 minutes in and not a single um, ah or er. You are a brilliant speaker, and I am looking forward to this series tremendously. Fantastic content, thank you.
0:42
@@mistadude you definitely need to get a life.
@@aarondcmedia9585 anything but that
@@aarondcmedia9585 he is still a really good speaker even though he said that. He gets his points across very well
Finally 🤩 I've been waiting for a great series on TH-cam about EMC! I've seen a lot on TH-cam but like u said, almost all EMC videos on TH-cam are very specific which is really hard for beginners in this topic
Awesome! thanks so much for sharing this with everyone.
Awesome looking forward to the videos!! Gonna learn a lot
Thank you for sharing.
I really appreciate the work you put on your videos keep it up!
Twenty minutes is about the sweet spot for a video, for me.
Great video!
There should be a required specified height between the DUT and the metal plane.
I believe that for non grounded DUT, that height does effect the result.
Thanks for sharing, you are helping the EE community better and better.
Yep, you're correct. It's fully spec'd in the various standards. For conducted emissions specifically, it normally doesn't make a huge difference. But yes, it ideally should mimic it as much as possible!
Awesome video, looking forward to the series.
Don't be deceived or disappointed by ferrite beads though - I used to also believe they were only useful 10MHz+ and started to put inductors in my power filters instead for a while. However, recently I bought a Bode 100 and basically debunked that. As long as you buy a decent one (e.g. 300-600R @ 100MHz) it will be just as effective, if not more, right down from 150kHz upwards.
Sometimes you need to filter in the 1kHz - 150kHz range, of course, and that's where heavy inductor filters come into play. But that needs much more care and affection and multi-stage filtering to achieve without massive resonance issues up the frequency chain.
Glad you enjoyed! Yes, I'll be going through various filters soon!
I'm looking forward to those video's, there is not much on TH-cam on EMC.
Thank you for this video!
Thank you very much
Looking forward to seeing these videos. It would be interesting to break out common mode and differential mode noise. I also hope you will test out multi-layer boards with different distances between the layers. 0.1 and 0.2mm are readily available and of course the 0.1mm layer should be lower inductance, but it would be interesting to see this in a real-life test...
Heh, no kidding, those are all on my list for this series!
The big thing is just making sure I have a design that will radiate enough so those changes will make an impact
Why 0.1mm layer height will have lower inductance?
@@flienky Because the magnetic fields generated by the signal line and return plane will be closer together and opposing, so the closer they are together, the closer they come to cancelling each other out...
Good demonstration!
I have the same dehumidifier! It's the best!
I appreciate this bud. I'm new to this phenomena and it's baby steps here to help us become familiar with the content. Thank you. 👋👈👀
Thank you
So how do you intend so fix that primary frequency spike a t 1MHz ?
pi filter ? or some other type of filter ? encasing the board in a metal coffin?
You'll have to wait to see the video on that 😉. Either the next EMC one, or the one after
@@MicroTypeEngineering would at least those ideas work ?
PI filter certainly, shielding not so much for conducted emissions. I talk about this quite a bit in the supplemental video
I think he was only measuring _conducted_ noise carried along the power input and output lines and not any _radiated_ noise that a shielded enclosure might address. From my own experiences, you can have a lot of noise be carried throughout a system on the power-supply lines unless you filter it out _at the source_ , and depending on what else is happening in the system--amplification, for instance--it can then _radiate_ beyond your control. I may be wrong, but I think shielding mostly protects you from other people's sins (excessive radiation of noise) and not so much your own sins within your own system. I can't wait to see what MTE does, but a very-low impedance to ground for 1MHz (say, 0.1uF) followed by a high impedance for 1MHz (say, 1mH), might knock that spike down enough to pass.
So what exactly were you measuring with that black box? Was it EM radiation or noise at the output?
It provides a known impedance to help measure conducted emissions. It gives a reproducible setup, while making sure you're only measuring noise from the device under test, and not the power supply.
@@MicroTypeEngineering This doesn't really answer my question. What are these "conducted emissions"?
The definition from google says
"Conducted emissions are the noise currents generated by the Device-Under-Test (DUT) that propagate through the power cord or harness to other components/systems or power grid."
So are you measuring the frequency content in the voltage at the input/output or voltage induced into the metal plates underneath or what?
Oh. Yes, it's purely the frequency v energy content on the wire going from the DUT -> PSU/LISN. The ground plane is primarily there to help prevent ambient interferance.
You measure radiated emissions in a completely different manner. I'll be going over that in future videos in the series. I also spoke on it in the supplemental video.
thanks
What is EMC? I see an undefined acronymn. Can't wait to…Run-EMC. :-) It's tricky.
What happened to the USB Hub series?
The chip shortage happened..
Ohh I see, that's unfortunate
@@MicroTypeEngineering RIP
Just to make sure I'm keeping up, where exactly does 60dB uV land in absolute terms? I hate that voltage-sized dB are different than power-sized dB. 0dBuV is 1 microvolt--RMS I presume. Every additional 20dB is a 10x. So 60dBuV is a millivolt. (Right?) And all these measurements assume what impedance? (In Ham Radio, we like to use power-sized dB for everything, especially when measuring spurs and the like, and treat everything as 50 ohms so that even voltage measurements can be expressed in equivalent dBm.)
Heh, yeah it becomes quite confusing keeping everything straight.
But yes, you're correct. The LISN I'm using is nominally right around 50R impedance depending on the frequency. It's actually a bit closer to 45R, but still within the CISPR 16 and 25 limits. Both of those specify a range of impedances, nominally 50R.
Great video
A teaser video to get you to the Patreon site. No real information except for sign up for the extended video///