Very Important For PCB Layout: Crossing planes explained by Eric Bogatin
HTML-код
- Опубликовано: 26 май 2024
- What is happening when signals travel over a gap in your PCB? Thank you very much Eric.
Links:
- Ansys free version: www.ansys.com/academic/studen...
- Ten Signal Integrity Rules of Thumb webinar: www.signalintegrityjournal.co...
- FEDEVEL courses: courses.fedevel.com/
Chapters:
00:00 The board and signal
00:44 How signal travels - solid plane
02:11 How signal travels - a gap in plane
03:31 Why noise is picked up by other tracks
06:17 Changing gap size and it's influence on the noise
08:30 How to fix the problem?
------------------------------------------------------
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
It would be really interesting to run a simulation in a GSSG configuration with ground pour on signal layers. I believe this configuration is quite commonly used on internal layers for low speed signals. Each signal trace is sandwiched between a continuous ground plane and a split ground plane (orthogonal signal layer with ground pour). I believe that a simulation could highlight effects similar to those shown in this video. Thank you!
Even at low speeds when you don’t think you really need a plane for signal integrity, a split plane like this is awful for both electromagnetic susceptibility and crosstalk, it becomes surprisingly awful even in the low kHz region. I’ve seen it in other people’s designs have been changing them to solid planes.
An interesting simulation would be signals crossing a split power plane with a ground plane under the power plane. I.e. Sig / Pwr / Gnd. The inter-plane capacitance between Pwr and Gnd should act as a much better capacitor at higher frequencies than a discrete component mounted on the top layer and connected using vias.
Awesome material. Thanks, Robert!
Awesome information. Thanks for the cool interview.
Very cool demonstration! I was wondering, what is the effect if you don't have a solid ground plane, but instead route return path as a trace directly under the signal traces? Does this make the crosstalk worse because there can be potential differences between nearby return paths or could it be a benefit kind of like a differential pair?
I've seen something like- this when there is split plane on one side of the signal and a solid plane on the other side. The split plane still affects it.
Isn't this an effect of, essentially, one side of a transmission line being a plane, and thus having a distributed pair charge to the line part of the line, and when it encounters the discontinuity, it's like a wave hitting a double (or in this case, a quadruple) slit? There are two pulses going back because there are two changes in charge direction, one when it encounters the beginning of the gap, and one when it encounters the far side. The impedance mismatch looks a lot like a partially reflective window/mirror, in fact you could maybe get a "thin film" effect cancelling out the noise if you match the width of the gap to the length of the pulse? Very cool simulation!
very cool!
Hi Robert, interesting video,
I have doubt let's say we have a digital signal on a "flex pcb", and generally we use "hatch pattern for return plane", Will it create the same problem because hatch pattern ll have cavities?
Thanks in advance
@robert ,could you please make a learning video on advanced power supply like fast charging and other power supply
Thanks Robert and Eric - but a break in return GND will occur whenever a signal travels through a via. Does Eric have another simulation of induced crosstalk when a high-speed signal travels from top to bottom layer (no stub).
like this? ruclips.net/video/2UKfWowTvOs/видео.html
Sorry, how do we know when making a pcb what type of component we will use and what component strength?
Experience, trial & error, measuring, reading & understanding datasheets, examples from others. Etc
And spice simulation
@Robert Feranec Hi Robert, I love your videos on PCB design and electronics, you are doing great job. I recently found out on wiki that you are from Slovakia, it's awesome I am from Czechia. Much respect for your work. Ať se daří PS: I just realized, that you sound kind of like indian tutors on youtube XD, no offense
Vdaka :)
Robert, I respect what you are doing but could you please work a little bit on your accent? Eg. capaaacitors, not capaciiitors. Please don't take it offensive is rather advice what you should improve. Thanks!
btw got funny situation on interview some years ago when I incorrectly put accent on "impedance". And guys had to hold they laugh when they heard "impotance". Got the job anyway...
I tried. There is no school in Slovakia where they would focus on my accent. I have tried two, it was useless so I just let it go.
I have no problems understanding him. Prefer english with an accent over subtitles
Use close caption and turn off speaker lol
Well, there is accent everywhere, even US accent is a type of accent. We just have to get use to it.