Hi, I don’t have success to do 2 separated ground. I don’t find informations about it. Can someone lead me to a way to do it? On Kicad?
Thank you
Where does the problem start? Have you done the schematic, or is that the problem? How far are you, what have you tried/thought of? Can you e.g. show a screenshot of a project at some point where you are stuck?
Hi eelik, thanks for your interest…
I am making a “simple” project for my analog class. It’s a “voice keyed” for ham radio from the book “modern electronic communication”.
I made it on breadboard without any consideration of ground, wire length… to be sure the system works. It is. Now, when there is no reading the noise come and loud. So, I guess first the building test is off course noisy because of what I mentioned. Ok, fine.
I am planning to make 2 ground part in this circuit, 1 for digital area and one for analog (just for the output).
I can understand your curiosity of the schematic, building, ect, and sure, I can send you what you asked but this sounds outside of the topic. I just need a way to create 2 ground surfaces.
Later one, yes I may ask for good practice to make them, and yes, showing my pub will be crucial, but this is not my thead yet.
So pleaaase someone, could you tell me where I can find the way to fill 2 contiguous grounds ?
Thank you
ok, found it!
“You was just doing in wrong way silly boy.”
Guess why there was no that kind of question anywhere.
Kicad is very cool app!
If you have separated ground planes make sure no trace crosses the gab between them. Otherwise you just build a slot antenna.
If you must have separated ground and need to cross over it with a trace make you need to provide a direct path for the return current. This is done by placing a capacitor that connects your two ground planes as near as possible to the trace.
Hi Rene, thank you for your comment, that’s definitely a hook!
…Yes, there is 1 traces that crossing the gap.
I know that every case is different, but what value and type of capacitor you would advise?
After some research, it seems that a 1nF ceramic would be a minimal choice.
and 3 times minimum gap the width of the traces.
And while trying to create a shield, I found that bypassing the output capacitor the noise decreases over 90%. As beginner, I don’t really understand what is going on.
If someone has some technical experience, you have a thankful listener.
And, what about 2 pcb for the separate the 2 circuits? Is this way good?
We can not see what your pcb looks like. We can not give any guidance than some generic suggestions. Post screenshots.
While @Rene_Poschl’s advise is correct it applies mainly to high speed/frequency signals. Low frequency signals and DC can generally cross plane boundaries without issue.
what is this in reference to?
Again, you are articulating what your issue is or what you are trying to achieve, and you choose not to share your schematics etc. I’m not sure how you expect anyone to help you.
Thank yo for your intervention Rene_Poschl and 1.21Gigawatts. After starting reading guidance, and reading…
I discover the truth that without experiments(sss) or advices of those who had experienced errors all my “articulations” are sterile. Now, it’s clear. So please excuse my slow-understanding-walking-dead, eelik.
This my second attempt for my component arrangement, and now I understand I need a start point. But, if you have more suggestions to improve my design, I will be glade to receive them. I guess I will receive as well some slaps, but if it’s for the good cause, enjoy. It suppose to be 2 layers itched by hand.
I am planning to move the 9V connection close to the audio output in order to place my start point closer to the gap. Do you think that’s the best solution?
I don’t remember where I picked up 1uF reference, it seems to be false after other lectures. Sorry I will do better next statement.
THe bypass I mentioned is across the 250uF.
Thanks for your wise suggestions.
please ! up, up, up, up, up!
Yes baby, it is! I changed lot of your “work” and just fixed for you, And because someone asked where are good sources, Dr Archambeault loves you. And 1uF to 10uF was the best try.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.