USB Chargers, PSU and Why CM/DM Noise Feels Worse in the USB-C PD Era

Modern phone and USB-C laptop chargers are marvels: tiny, efficient, and able to deliver anything from a gentle 5 V trickle up to laptop-class power. But in RF- and audio-sensitive setups, they can also be surprisingly “noisy.”

That noise is usually not about the DC voltage being “wrong” ... it’s about high-frequency switching energy escaping the charger, riding on the cable, and turning your power lead into an antenna.

What people mean by “noisy” chargers

Two different things get lumped together as “noise”:

• Audible noise (coil whine / buzzing): magnetics or ceramics vibrating, often when the supply runs burst/skip modes at light load.
• Electrical noise / EMI (the RF problem): broadband switching energy that shows up as HF hash, SDR spurs, raised noise floors, audio hiss/clicks, or unstable behavior in sensitive modules.

This article focuses on the electrical/RF side ... especially how noise “escapes” as conducted energy on cables.

DM vs CM noise ... the fast definition

Noise on a cable travels in two main “modes.” They behave differently, and they need different mitigation.

Why CM is the “RF killer”

With DM current, the outgoing/return currents tend to cancel their fields (especially on a tight pair). With CM current, the fields add ... and your cable becomes an efficient radiator. That’s why CM often looks like: “My charger wipes out 40 m” or “My SDR waterfall turns into snow when USB is plugged in.”

Why modern compact chargers can be noisier in a shack

They’re high-density switching converters

Almost all modern wall chargers are switching supplies. Smaller + higher power forces higher switching frequencies (or faster edges), higher dV/dt and dI/dt, and less physical room for magnetics, shielding, and filtering. Regulatory compliance is real ... but “meets EMI limits” does not automatically mean “quiet next to an HF receiver.”

Two-prong “floating” supplies often have tricky CM paths

Many phone chargers are Class II (no earth). The output is floating, so the design must control EMI without a solid earth reference. In practice, parasitic capacitances across the isolation barrier (transformer, switch node, layout) can inject high-frequency CM energy onto the output. The result: USB cables happily carry that CM current into your station environment.

Efficiency tricks can create wideband “junk”

To meet efficiency and standby targets, chargers often use burst/skip modes at light load, variable frequency control, synchronous rectification, and fast edge transitions. That combination can create “messy” spectral content ... not a single tone, but bursts and harmonics spread across HF/VHF and beyond.

GaN often means faster edges (and more EMI headroom needed)

GaN enables smaller, efficient chargers, but the faster edge rates can increase ringing and high-frequency spurs if the layout/filtering is marginal. In RF terms: more energy at tens or hundreds of MHz is easier to accidentally couple into everything.

What changed from the “old 15 W” era to USB-C PD?

Much higher possible power

USB-C Power Delivery expanded the ecosystem beyond “fixed 5 V.” Modern PD can negotiate multiple voltages and currents, and newer PD revisions extend power delivery up to 240 W (EPR). Even if you never use that, many PD supplies are designed for a wider operating envelope than a basic 5 V brick.

Negotiation and dynamic operating points

USB-C uses the CC pin for role detection and current advertisement, and USB-PD can negotiate contracts over CC (BMC signaling). The PD signaling itself is usually not the main RF-noise culprit on VBUS, but the charger’s internal power architecture now supports multiple operating points, which can change switching behavior and the “noise signature” depending on load and negotiated voltage/current.

Why the USB cable makes the noise feel worse than it “should”

The cable is part of the EMI system:

• USB power leads are often long enough to radiate efficiently at HF and above when CM current is present.
• Your device and bench create stray capacitances to chassis, coax shields, ground straps, instrument cases ... giving CM current a return path.
• Small changes (cable swap, ferrite clamp, moving the charger 1–2 m away) can produce huge differences.

Filtering ... “we have filters” isn’t one thing

CM and DM require different suppression techniques:

• DM filtering helps when the rail itself is dirty (VBUS-to-GND ripple, switching residue). Think LC/π filtering, feedthrough caps, local decoupling.
• CM filtering helps when the cable is radiating (same-direction current on both conductors). Think common-mode chokes, clamp ferrites, routing and referencing strategy.

In many ham/SDR/audio setups, CM is the bigger villain ... because it turns a “power cable” into an antenna.

Practical checklist for quieter USB power in RF/audio setups

• Separate “power cleanliness” from “RF cleanliness”
  Acceptable DC ripple does not guarantee low CM noise on the cable.
• Shorten and improve the cable
  Shorter is usually better. Avoid running USB power parallel to coax, antenna feedlines, or sensitive audio leads.
• Use ferrites as a fast CM test
  A clamp ferrite near the device is a quick “does CM matter here?” experiment.
• Filter near the device
  Prevent the cable from carrying (and radiating) noise throughout the sensitive area.
• Use a known-good supply when possible
  Official supplies (e.g., Raspberry Pi PSUs) tend to be predictable. Still, even “good” supplies can be annoying when placed close to RF front-ends.

Quick diagnosis: is your problem mostly CM or mostly DM?

Two clean ways to power RF-sensitive USB devices

Set expectations (the honest part)

• Inline filters and “quiet supplies” mainly target conducted noise on the power lead (and by reducing CM current, they also reduce cable radiation).
• If a charger is radiating directly from its body and it’s right next to your receiver front end, distance still matters.
• Stay within current ratings for the supply, cable, and inline filter, especially at higher 5 V currents.

Bottom line

Modern USB-C PD chargers aren’t “bad” ... they’re just more demanding designs: higher power, smaller size, faster switching, and trickier CM paths. In RF/audio environments that combination makes cable-borne noise easier to notice.

Once you look at the problem through the DM/CM lens, the fixes become clearer:

• DM noise ... clean up the rail (LC/π, decoupling, feedthroughs)
• CM noise ... stop the cable current (common-mode choke, ferrites, better referencing and routing)

That’s exactly why inline CM/DM filtering and purpose-built low-noise USB supplies can be such effective “drop-in” mitigations in the USB-C PD era.

Interested in more technical content? Subscribe to our updates for deep-dive RF articles and lab notes.

Questions or experiences to share? Feel free to contact RF.Guru support and engineering.