Skip to content

PCB Grounding & Return Paths

A Grounding Review Checklist for a Finished Layout

Fourteen checks you can run in your PCB editor before the files go out.

  • Lesson 10 of 10
  • Intermediate

5 min read

This is a review you run on a finished layout, in your PCB editor, before the fabrication files go out. It takes under an hour on a medium board and it finds the problems that a design rule check cannot, because none of them are rule violations.

Nothing here requires a simulator. It requires looking at the board in ways the default view does not encourage.

Set up the view first

Three settings make the rest of the review possible:

  • Display plane layers as filled copper, not outlines. Most defects in a

plane are invisible in outline mode.

  • Learn the shortcut that isolates a single layer. You will use it

constantly.

  • Have a list of your fast nets. Clocks, serial buses above a few

megahertz, anything with an edge under about 2 ns, switching nodes.

The checklist

1. Look at each plane layer completely alone

Hide everything else. Scan for slots, necks, islands and unexpected voids. Mark anything that looks like a channel rather than a sheet.

2. Follow every fast net over its reference

For each net on your list, trace it from driver to receiver with the adjacent plane layer visible. You are looking for any point where there is no copper directly beneath the trace. This is the single highest-value check in the list.

3. Check every layer transition on those nets

For each via on a fast net, is there a ground via within a few millimetres? If the signal changes from a ground-referenced layer to a power-referenced layer, is there a stitching capacitor nearby, and could a stackup change have avoided the transition entirely?

4. Inspect every deliberate split

If there is a moat, list every conductor crossing it. For each one, say out loud where its return current goes. Any answer involving the words "around" or "the long way" is a finding.

5. Inspect the accidental splits

Connector footprints, dense BGA escapes, long thermal relief chains, and any trace routed on a plane layer. These are the splits nobody decided to make.

6. Check the plane around every connector

Is the plane continuous up to the footprint? Are the plane layers stitched together near it? Do fast signals approach across the row of anti-pads or from a continuous direction?

7. Review connector pin assignment

For each fast signal leaving the board, which pin is its return and how far away is it? Is one ground pin serving many signals?

8. Find the switching regulator loops

For each switching converter, identify the loop that carries the switched current: input capacitor, switch, diode or synchronous device, and back. Is it small, is it on one layer, and does its return current have any reason to flow under sensitive circuitry?

9. Check decoupling return loops

For each decoupling capacitor on a fast device, how far is it from the pin, and how does its return reach the device ground? A capacitor with a long via path to the plane has an inductance you did not budget for.

10. Verify mixed-signal partitioning by looking, not by trusting

If the board has an analog section, display the digital nets and look for any that enter the analog region. One is enough to matter. This cannot be checked by rule, only by eye.

11. Check converter and reference grounding

Are the converter's ground pins connected as the manufacturer's layout guidance for that part specifies? Is the reference close, with its ground connection next to the converter's analog ground rather than across the board?

12. Trace the high-current paths

Motor drivers, power stages, anything above a few hundred milliamps. Follow the return from load back to source and note what it passes under. High current flowing under a sensitive section is a problem regardless of plane continuity.

13. Look for stubs and unstitched copper

Pours connected at one point, isolated regions, copper under a shield can that is tied at two corners. Either stitch it properly or remove it.

14. Check the board edge

On a multi-layer board, are the plane layers tied together near the perimeter? This matters more as the highest significant frequency rises.

Recording what you find

Treat findings like any other review output rather than fixing them as you go. Note the net, the location and what you expect to happen. Three reasons:

  • Some findings turn out to be acceptable once considered, and you want that

judgement recorded rather than repeated next revision.

  • Some are not worth fixing on this spin and should be on the list for the

next one.

  • Fixing while reviewing is how you lose your place and skip half the board.

If your project keeps a record of design decisions, this is good material for it, and there is a guide on documenting design rationale that covers how. The reasoning behind an accepted finding is exactly the thing nobody can reconstruct two years later, and it is the sort of context that belongs with the revision rather than in somebody's memory.

What this review does not cover

Honesty about scope matters more than a long list.

This review finds geometric problems in return paths. It does not verify impedance, it does not predict emissions quantitatively, and it does not replace a signal integrity simulation on a board that needs one.

What it does is remove the class of problems that are free to fix at layout time and expensive to fix after fabrication, which on most boards is the majority of grounding problems.

Where this course ends

Ten lessons, one idea: current flows in loops, and the loop is the circuit. Every rule in this course is that statement applied to a situation, and when a new situation appears and no rule covers it, the way to answer it is to trace the loop.

The grounding advice you will meet elsewhere is mostly true, in the context where it was first said. Now you have the means to work out what that context was, and whether it is yours.

Sources and further reading

Where to go next

¿Te ha resultado útil?

Reddit LinkedIn

The XKEMA newsletter

Hardware engineering worth keeping

New courses, engineering notes and hardware guides, whenever there is something genuinely worth reading. Nothing else.

We will email you to confirm. You can unsubscribe from any issue.

XKEMA, in one sentence

Engineering knowledge management software for companies that develop electronic products: the product structure, its bills of materials, its documentation and its revisions, all in one place.