Finding ground loops in mixed analog and digital designs

[Voltranger]

I've been debugging my latest board, and the analog signals are noisy, the digital lines are twitchy, and I suspect there's a ground loop involved.
My scope is definitely saying 'uh-oh.' How do you usually track down these sneaky loops without tearing everything apart?

What tricks have actually worked for you?

 

[Ohmwardbound]

The first thing I do is lift the grounds carefully and keep an eye on how the scope behaves. A quick test with the star ground, using temporary jumper wires and isolating the analog return paths, usually reveals the issue pretty fast

 

[Calypso]

I begin by cutting off and jumpering the analog and digital returns.
If the noise decreases, then I know we’re onto something.
It's also worth checking for high-current paths that might be sneaking beneath sensitive traces,
those hidden loops can cause some serious trouble.

 

[LEDZepp]

I typically start by measuring the ground paths and mapping out the return currents, not just the signals.
Then I’ll try a single-point ground approach using short jumpers. If that stops the flickering, you’ve likely pinpointed the culprit behind your issues.

 

[barfle]

Although it’s never happened to me (because I was warned abiut it as a student), closed turns (a complete circle) in your ground traces can pick up AC mains hum. It’s also worth making sure your bypass capacitors are all doing their job.