What is the reason for plane to pad

I’m new to electronics, and this is my first PCB
Looking at planes, I can see that when I fill a plane, kicad makes this kind of 4 legs to conncet the plane to the pad, instead of having them fully connected:

But, when I connect a tick trace, I get a full connection between the trace and the pad instead:

Why is that?

The 4 legs or spokes are thermal reliefs which serve the purpouse of decoupling the pad thermally from the plane in order to make hand soldering easier.


Thank you! I would have never guess that! :slight_smile:

Does this means that it is better to use smaller traces when possible? In my case I was using traces as big as I can. In the example above, those pads will have 3.3v and few mA, so I could go as thin as I want


It’s a balance between solderability (power of soldering iron needed) and current carrying capability, possibly also mechanical robustness of the pad if you happen to solder/desolder it more than once.

1 Like

For a current of 1A, you need a width of just 0.3mm (Although wider tracks do lower voltage drop and that may be beneficial).

Resistance of any sheet material (such as copper on a PCB) is also counted in “squares”.
If you take a PCB track and double it width, it halves the resistance, but if you also double the length, then the total track resistance doubles again, and is the same as what you started with. (It also has the same number of “squares” as the original track).
This also means that making short sections of tracks thicker or thinner does not change much over the overall performance, as long as it’s not too extreme. Sometimes very thin PCB traces are explicitly designed to act as fuses.

Yet another part is that adding (subtracting?) thermal reliefs makes it easier to hand-solder a PCB, but the copper on a PCB is also regularly used as a heat sink. For example the 1117 voltage regulators and TO252 cases are explicitly designed to use some PCB area as a heat sink.

There are a lot of factors in designing a PCB. The difference between a proper designed PCB and a badly designed PCB can be as big as a properly working circuit, and the circuit not working at all, even when part placement is the same and all connections beep the same with a DMM.
PCB tracks can even be used as inductors or as (low value and not very accurate) shunt resistors.

In your case it may be best to not connect the GND pad on the red layer at all, but only let the GND plane on the blue layer handle all the GND connections. But it’s just a guess. For a better answer, you have to analyze how the currents flow through the various tracks of the PCB.


Too quick and too far-reaching conclusion.
Imagine a plane connected to pad by 4 legs 0.25 mm width each. So the total connection is 1 mm width and such pad is enough easy to hand solder.
So what track width (if you have only track connected to that pad) will dissipate comparable amount of heat. Your first thought can be - 1 mm track. But it is not true as plane dissipates head mach better than single track. May be the comparable in sense of ease of soldering will be 3 mm track.
So the conclusion can be that to make soldering easy you should avoid using wider than 3 mm tracks if you need not to use them. Using thinner tracks from that reason gives you a little so if you have the other reason (voltage drop because of track resistance) to use tracks up to say 2 mm just do it.
The true reason to use thinner tracks is to have a possibility to make more connections at the same area. For PCBs ordered from manufacturer (not destined to be home made) I use 0.25 mm tracks for signal lines. If really needed I use 0.2 mm. The thinnest track my PCB manufacturer accepts is 0.1 mm but I have never tested his technological limits.

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.