KiCad 5.1.2.
First experiments with shove.
When I try to move track D15 (hot-kay D) then suddenly track D8 jumps much more then I would suppose.
Is it typical/expected behaviour or I should report it as a bug?
When I come back to the left D8 jumps back to its origin position.
My Router settings:
The shove capability of the Interactive Router is a wonderfull thing, but it does take some time to get used to it, and use it most effectively.
One of the most important things is that you have to fully acknowledge that it moves your tracks around a bit and therefore you loose some control about the locations of your track positions.
In this particular case, you can very easily make the connections, without paying any attention to laying out your tracks in a “neat” pattern, and then at the end grab one of the outer tracks and shove them into the others to line things up.
I’ve used the shove mode to make room for via’s in ways I would not have thought possible because of (local) board density. It just shoves 5 or 6 tracks aside without me having to think about where I should put them.
Because the shove option can so easily shove (and line up) multiple tracks, it has become more efficient to not care at all about the intermediate track positions, and simply make your connections, and then do a short cleanup pass later if needed.
Also: Long parralel tracks may look nice visually, but it’s a bad practice for signal integrity because it increases unwanted coupling between (non related) signals.
If you uncheck “Optimize Pad Connections”, it will disable this behavior.
Sometimes betwean tracks I add extra track ended with vias to GND. Till now I had to connect all such vias but I believe in KiCad V5 I need only to put vias - we will see.
Today I started routing. I noticed some problems in following situation. Under TQFP 64 I wonted to go with VCC 1mm width tracks as close as possible to VCC pins and then connect them with 0.4mm track. When routing these 1mm I see the track where I wont it, but when click to finish it is not there.
Will check tommorow. Now I have the KiCad computer switched off and I’m 30 minutes late related to my plans to finish work at 20:00.
I can confirm the mentioned behavior.
Just have tried to reproduce similar situation, and in my case [5.1.2] disabling the “Optimize pad connection” did not help. Sometimes the interactive router breaks my “parallel tracks”.
When experimenting with the shove to verify it, found another quirky behavior:
I think It can be the effect of:
After showing that situation I have moved that resistor set max to the left - no room left for shove experiments.
So I get back one day and there repeated the situation.
Switching off “Optimize pad connections” don’t change that behaviour.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
