0
\$\begingroup\$

Return current path follows least resistive path in DC signals; least inductive path in AC signals as you seen in Figure 1.

In Figure 1, there is no GND pour on the top layer, and only a poured GND on the bottom layer.

In Figure 2, there are GND pours on both the top and bottom layer.

What happens if I pour GND on the top layer and connect each GND polygon with vias as you seen in Figure 2? I mean, does the current path goes to its source on the top layer and bottom layer at the same time? If yes, how much of return current goes to top layer and bottom layer?

Figure 1: enter image description here

Figure 2: enter image description here

\$\endgroup\$
4
  • 9
    \$\begingroup\$ Current does not take the least resistive path. It takes all possible paths simultaneously, proportional to how conductive they are. \$\endgroup\$
    – floppydisk
    Feb 15 at 9:00
  • \$\begingroup\$ I may be wrong, but DC always takes the shortest path (least resistance). \$\endgroup\$
    – Gregory
    Feb 15 at 9:21
  • 5
    \$\begingroup\$ @Gregory you are wrong. A 1 ohm path and a parallel 2 ohm path doesn't have all the current flowing in the 1 ohm path does it. \$\endgroup\$
    – Andy aka
    Feb 15 at 9:29
  • 4
    \$\begingroup\$ @Gregory you are indeed wrong. DC current distributes itself over all available paths, the current density distribution being a function of conductivity \$\endgroup\$
    – Bart
    Feb 15 at 11:07

1 Answer 1

Reset to default
3
\$\begingroup\$

The DC case will be the same, both planes will be used for the return current.

As for the least inductive path: That depends on the trace-to-pour distance in the top layer and the layer-to-layer distance. But in general, one can expect that part of the current flows in the bottom plane, right below the trace, and a part of the current flows next to the trace in the top pour. Because both these paths will have probably similar inductances, whereas everything further away will have so much inductance that current there will be negligible.

\$\endgroup\$
2
  • \$\begingroup\$ I agree with you. For the resistive and inductive cases, how much of return current goes to top layer and bottom layer? For the inductive case, I think that almost all return current goes on the bottom layer. For the resistive case, I am not sure ratio of return current between top and bottom layer. \$\endgroup\$
    – yardi
    Feb 15 at 9:15
  • 1
    \$\begingroup\$ For the resistive case, the split will be around 50:50 given equal copper thicknesses. The inductive case depends also critically on the exact via positions and the distances I mentioned in my post. It can be everything in bottom all the way to everything in top... \$\endgroup\$
    – tobalt
    Feb 15 at 10:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.