Must Know This to Understand High Speed PCB Layout Simulation | S-Parameters Explained, Eric Bogatin
HTML-код
- Опубликовано: 9 июн 2024
- How the model of PCB used in high speed board simulations is created. Explained by Eric Bogatin. Thank you Eric.
Links:
- Eric's and The EEcosystem class - Transmission 101 (use FEDEVEL coupon): electronicsmasterclass.regfox...
- Signal Integrity Academy (use FEDSI - 3 months free): www.bethesignal.com/bogatin/3...
- FEDEVEL courses: courses.fedevel.com/
Chapters:
00:00 What is this video about
00:29 What are s-Parameters, Why we need them
05:51 How S-Parameters models are created
08:02 Including components in simulations with S-Parameters
14:44 What is in S-Parameters file?
17:10 Opening and explaining S-Parameters file
23:04 S-Parameters ports explained - what they are
26:09 Floating ports
30:23 S-Parameters numbers explained
30:59 What ports to use when using S-Parameters model
------------------------------------------------------
Would you like to support me? It's simple:
- Sign up for online courses hosted on our platform: marketplace.fedevel.education/
- Sign up for my Hardware design and PCB Layout online courses: academy.fedevel.com/
- You can also support me through Patreon: / robertferanec
- Or sign up for my Udemy course: www.udemy.com/learn-to-design...
It is much appreciated. Thank you,
- Robert 
Наука
I always get excited to see Eric Bogatin as the guest speaker! LiteVNA (6GHz) or NanoVNA (3GHz) are two very affordable tools to measure S-parameters, with some additional interconnects and RF demo kit well under $200 total, anyone can measure empirical S parameters. I've been using a liteNVA for characteristic impedance of microstrip transmission lines with great success. If anyone was interested in learning or verification of PCB designs I highly suggest either of the affordable VNA's available today, it's quite amazing how accurate they are for less than the price of a calibration kit for one of the professional offerings. I went from "what's a smith chart?" to building my own antenna's for less than a text book on the subject.
why would we want to measure S-parameters on the PCB, is the simulation not good enough?
when you say network analyzer, it sounds like it is something used to analyze internet traffic.
where would you put the probe to measure the S-parameters of a PCB track, especially since there are so many points for there to be emissions from and absorption like cross talk, and things outside of the PCB that are radiating.
@@user-ww2lc1yo9c Unfortunately the theoretical world doesn't always match up with the real world so we measure the actual PCB to verify the manufactured design matches the theoretical parameters. As you know with RF design the physical dimensions dominate and any variance in the manufacturing process can make simulation a pointless exercise. I personally don't have access to simulation software but even if I did, as a measure of quality assurance, the empirical data is more valuable. VNA (vector network analyzer) is the tool used to measure S-parameters, it's connected in a very specific manor and the setup is super important. I used solder to attach a pigtail directly to the IC RF pad after I calibrated the VNA using the exact same interconnects. The concerns you mentioned are the reason for proper PCB layout and design, if you follow the guidelines and verify the S-parameters as I mentioned, the only other factor would be outside influence which we cannot control.
If you would like more information pertaining to the design of PCB RF layout there are plenty of resources online, in particular the Altium channel industry expert videos. They cover everything from layer stack up, microstrips, via spacing and much more. Another resource would be to look into the manufacture suggested layout reference designs for whatever Rf IC you are using, although that can sometimes be misleading since applications engineers aren't always the best but it's a solid starting point.
I hope that helps you out, realistically the answer to your question comes with experience, in order to isolate questionable issues do every measurement possible to confirm it's not something simple. After chasing my tail for a few weeks trying to troubleshoot a wireless range shortcoming I learned that empirical data is far more valuable than simulating, not that simulations aren't valuable learning tools but in the real world they are pretty worthless in most applications. If you have any aspirations to learn RF design a VNA is an invaluable tool and for the price of the liteVNA and RF demo kit you too can understand the importance of empirical measurements.
Eric Bogatin, explaining RF. Now that's great content. Thanks Robert. You got me to join.
Thank you very much Sean
Excellent call Robert and Erik! I got a bit overwhelmed with information on this one, but your calls I tend to come back to over and over until I have the ‘aha’ moment, then somehow it all just makes sense. I think I will need to take a deeper dive into the derivation of Maxwell’s equations. 😊
I often hear/notice new things in the videos when re-watching them second or third time.
@@RobertFeranec Absolutely!
If you were to count how many pieces of new information/useful information you are receiving from these videos it sort of makes sense that you have to watch them a few times, or be very focused the first time and listen to everything word for word, but even then your brain needs some processing time. Just the way the brain works I guess 🧠
Thanks for the video, the questions you asked might sound a bit weird, but they really bring out the best, simplest explanations from Eric! As I was thinking I'm starting to understand something, I heard a question and answer, then I realized I didn't really understand the concept. It's like learning by doing but someone already asked those weird questions I would have when doing something, not just reading or watching a video about it. :D
Great lesson. It is a little bit ironic that PCB may be most advanced part of next project;)
fantastic thanks
OK lets see, so the video was kind of rushed near the end but I guess that is just because the discussion was agreed to be for 30min.
The discussion was quite clear a the start to describe what S-paramters are. It was quite clear about the distinction between the "topology" moden and the S-parameter model. It also described how they are created. However, what it did not do was, how these are generated from an actual PCB. An actual PCB will have a lot of PCB tracks and vias and pads. A track itself could have a continuos ground plane or splits in it. A track itself could have have the same width throughout or taper to smaller widths near the pads. The track could could have magnetic and electric field coupling with other tracks or even signal sources that are outside of the PCB. Now, how is an S-parameter model created from an actual PCB and then used in simulation, this topic was not covered at all in this video. Because of this, the video does a good job of describing what S-parameters are, but not exactly how a model is used to solve a real world problem.
Do you intend to make a follow up video on this? I guess Eric must be charging hundreds of pounds per hour for his service.
Hi Robert thanks for your videos.
Can you please do one session on how to calculate rise time using slew rate data ..
many data sheets shows rise time and fall time in slew rate units ..
Very nice video again ! I know the focus of this video is mainly on simulation. However, I wonder when s-parameters are actually measured on an actual PCB in the PCB design flow. It seems that the measurements are never made on the actual product but on some preliminary validation board. Would the goal of this validation board be to test connectors, via design, PCB material, etc. before finalizing the design of an actual product? Is that correct?
This man is shrewed, this presentation was more about him advertising himself so people take his course than anything.
Idea for antoher video - comparison of simulation tools, 2D/3D field solvers.
Would like to use more of simulation in my work. Would apprentice at least and overview of what tools exists and what they are best for.
謝謝！
Thank you very much Joanne
I love your videos. I should not nitpick but I really want the audio for the video calls to improve. It would be really awesome if the guest you're interviewing records the audio on this computer rather than you recording it on your resulting in lower bitrate and sample rate for the audio.
I don't mind audio, but also other voice call app that uses better bitrate would do the job.
Thank you. I keep improving the setup, I may come up with something new in future.
There are losses in every transmission line. The goal is to get enough information at the end of the transmission to make a good enough estimation of the signal that was transmitted. I think it's easy enough to understand the spoken words in these videos, even with the low bitrate and compression. The hard part is to actually understand the abstract concepts and practical meanings of the transmitted information and apply that to my work.
Pls Can any tell me the conclusion!
_in the simulation output we can see wave forms and we can check the Vomin ,Vimax,overshoot and undershoot time speed to assure that signal is in our limits thats fine but the output of s_parameter is different we can get only losses and crosstalk ,by using this how can I tell that this signal through a connector is fine ? How we can tell
Pls help me out .
This video will help you a lot - Eric is explaining the graphs there: ruclips.net/video/p6vx34jhTtI/видео.html
@@RobertFeranec thank you so much sir 👍
Robert when are NASA going to get you out of space, you've been up there for a long time floating around.