Free tool

Phase Noise to RMS Jitter Calculator

Enter a phase-noise profile and an integration band, and get RMS jitter — plus which decade of offset is producing it, which is the part that tells you what to fix.

Offset (Hz)L(f) (dBc/Hz)Slope to nextContribution
RMS jitter
Peak-to-peak
14σ estimate
Integrated phase
degrees RMS
Dominant decade
where the jitter comes from

Runs entirely in your browser. Nothing is uploaded, stored, or sent anywhere.

Phase noise and jitter are the same measurement

A phase-noise plot and a jitter number describe the same imperfection in different domains. L(f) is the noise power in a 1 Hz bandwidth at an offset from the carrier, in dBc/Hz. Jitter is what you get when you integrate that across the band your system cares about.

σφ = √( 2 · ∫ 10L(f)/10 df ) radians  ·  jitter = σφ / (2πfcarrier) seconds

The integration band is not a detail — it is the specification. The 12 kHz–20 MHz band shown by default is the telecom convention; PCI Express, Ethernet and JESD204 all specify different ones, and a part that passes in one band can fail in another with identical hardware. Always quote the band with the number.

Jitter scales inversely with carrier frequency

The same phase-noise profile produces half as much jitter at twice the carrier frequency, because a given phase error occupies a smaller fraction of a shorter period. This has two consequences that catch people out:

The per-decade table is the useful part

Jitter is an integral, so it is dominated by wherever the area is — and that is rarely where the plot looks worst. The table gives each segment's share, because knowing that 70% of your jitter comes from the 10–100 kHz decade tells you to look at the PLL loop bandwidth and its supply, while a far-out contribution points at the output buffer and the power rail feeding it.

Rough map: close-in offsets are the reference and the PLL's loop; the mid region is loop bandwidth and VCO; far out is the buffer noise floor. Supply noise appears as spurs, which this integral treats as broadband and therefore underestimates — a real spur needs adding separately as a deterministic term.

Peak-to-peak jitter is quoted here as 14σ, the usual convention for a 10−12 BER. That is a statistical construction, not a measurement — random jitter is unbounded, so peak-to-peak only means anything with a stated probability. The BER calculator covers that relationship properly.

More free tools

Each of these runs entirely in your browser. Nothing is uploaded, stored or sent anywhere, and none of them needs an email address.

Via stub resonance →
Stub length to its quarter-wave null, and whether the link needs backdrilling.
Resistor divider →
E-series values, worst-case output range with stacked tolerances, and dissipation.
Op-amp error budget →
Every DC error term referred to the input, ranked, and converted to ADC LSBs.
Signal chain noise →
Per-stage noise referred to the input, the stage that sets the floor, and ENOB.

See all 49 engineering tools →

Stay in touch

New tools and notes, once a month at most

We add tools here fairly often and write up the things worth writing up — a stackup that behaved oddly, a standard that turned out to be obsolete, a calculator that was quietly wrong. Join and you get told when something new lands.

One email a month at the very most, and usually less. No drip sequence, no sales cadence, no sharing your address with anyone. Unsubscribe whenever you like.

Join the list →

Talk to us about a board design review →