Skip to content
RFrftools.io
RF

Phase Noise Under Vibrations Calculator

Calculate vibration-induced phase noise degradation for oscillators on defense, aerospace, and mobile platforms using acceleration sensitivity (Gamma) and vibration profiles

Loading calculator...

Formula

Lvib(f)=20log10 ⁣(Γaf0fvib)3 dBL_{vib}(f) = 20\log_{10}\!\left(\frac{\Gamma \cdot a \cdot f_0}{f_{vib}}\right) - 3\text{ dB}

Reference: Vig, "Quartz Crystal Resonators and Oscillators"; MIL-PRF-55310; IEEE 1139

L_vibVibration-induced phase noise (dBc/Hz)
ΓAcceleration sensitivity (Gamma) (1/g)
aVibration acceleration (g rms)
f₀Carrier frequency (Hz)
f_vibVibration / offset frequency (Hz)

How It Works

Phase noise under vibration is one of the most critical yet often overlooked performance limiters in defense, aerospace, and mobile RF systems. When an oscillator is subjected to mechanical vibration, the resulting micro-deformations of the quartz crystal lattice cause its resonant frequency to shift in proportion to the instantaneous acceleration. This relationship is characterized by the acceleration sensitivity vector, commonly denoted as Gamma, which is an intrinsic property of each individual crystal resonator. The physics behind vibration-induced phase noise stems from the piezoelectric nature of quartz crystals. When an external force (acceleration) acts on the crystal, it produces mechanical stress that alters the crystal's elastic constants and dimensions. This changes the resonant frequency according to the relationship: delta-f/f0 = Gamma * a(t), where delta-f is the instantaneous frequency shift, f0 is the nominal carrier frequency, Gamma is the acceleration sensitivity (typically expressed in parts per billion per g, or ppb/g), and a(t) is the time-varying acceleration. For sinusoidal vibration at frequency f_vib with amplitude a0 (in g), the oscillator output develops FM sidebands at offsets of plus and minus f_vib from the carrier. The single-sideband phase noise at the vibration frequency offset is given by: L_vib(f_vib) = 20*log10(Gamma * a0 * f0 / f_vib) - 3 dB, where the -3 dB factor converts from peak to RMS for a sinusoidal signal. This equation reveals a crucial insight: vibration-induced phase noise scales directly with carrier frequency. An oscillator that performs adequately at 1 GHz may be completely unusable at 10 GHz under the same vibration environment, because the phase noise increases by 20 dB for every decade increase in carrier frequency. Random vibration, which is more representative of real-world platforms like aircraft, ships, and ground vehicles, requires a different treatment. Instead of a single sinusoidal tone, the vibration is described by a power spectral density (PSD) in units of g-squared per Hz. For a flat random vibration PSD of W(f) [g^2/Hz], the phase noise at offset frequency f is: L_rand(f) = 20*log10(Gamma * sqrt(W(f)) * f0). MIL-STD-810 defines standard vibration profiles for various military platforms, and these profiles are essential inputs to this calculation. Typical acceleration sensitivity values vary dramatically by oscillator type. Premium oven-controlled crystal oscillators (OCXOs) achieve Gamma values as low as 0.1 ppb/g through careful crystal cut selection and mounting. Standard OCXOs typically range from 0.5 to 2 ppb/g. Temperature-compensated crystal oscillators (TCXOs) are generally worse, at 2 to 10 ppb/g, while basic crystal oscillators can have Gamma values of 10 to 50 ppb/g. When reading oscillator datasheets, look for the acceleration sensitivity specification, which may be listed under vibration performance or environmental specifications. The practical impact of vibration-induced phase noise is enormous. In Doppler radar systems, poor phase noise masks slow-moving targets and reduces detection sensitivity. In coherent communication systems, phase noise causes constellation rotation and increases bit error rates. In navigation systems (GPS receivers, INS), vibration-induced phase noise degrades position accuracy. Engineers must account for the total phase noise budget, which is the sum (in power) of the quiescent phase noise and the vibration-induced phase noise. In many mobile platforms, vibration-induced phase noise dominates the total budget by 40 to 80 dB over the quiescent level, making oscillator selection and vibration isolation critical design decisions.

Worked Example

An X-band (10 GHz) radar on an aircraft uses an OCXO with Gamma = 1 ppb/g. The aircraft vibration profile is 1g rms sinusoidal at 100 Hz. 1. Convert units: f0 = 10 GHz = 10e9 Hz, Gamma = 1 ppb/g = 1e-9 /g 2. Calculate numerator: Gamma * a * f0 = 1e-9 * 1 * 10e9 = 10 3. Sinusoidal phase noise: L_vib = 20*log10(10 / 100) - 3 = 20*log10(0.1) - 3 = -20 - 3 = -23 dBc/Hz 4. This is very poor. A good OCXO might have quiescent phase noise of -120 dBc/Hz at 100 Hz offset. 5. Degradation: -23 - (-120) = 97 dB -- vibration completely dominates the phase noise budget. 6. Peak frequency deviation: 1e-9 * 1 * 10e9 = 10 Hz Solution: Use a premium OCXO with Gamma = 0.1 ppb/g to gain 20 dB improvement (L_vib = -43 dBc/Hz), and add anti-vibration mounts with 20 dB isolation at 100 Hz to bring the effective vibration-induced phase noise to -63 dBc/Hz. This is still above the quiescent level, so further vibration isolation or a lower-Gamma oscillator may be needed for demanding Doppler radar applications.

Practical Tips

  • Always verify the oscillator's Gamma specification from the datasheet -- do not assume typical values, as individual units can vary by a factor of 2-3 even within the same model
  • Anti-vibration mounts (isolators) are most effective above their resonant frequency; select mounts with resonant frequency well below your critical vibration band to maximize isolation
  • For frequency multiplied or synthesized signals, vibration-induced phase noise scales with the multiplication factor N: add 20*log10(N) dB to the reference oscillator's vibration phase noise
  • In systems with multiple oscillators (e.g., PLL reference and VCO), analyze each contributor separately and combine as power sum -- often one oscillator dominates
  • Consider the vibration axis: Gamma is a vector quantity, and the sensitivity varies with direction. Worst-case analysis should use the maximum Gamma across all three axes

Common Mistakes

  • Confusing peak and RMS vibration levels: MIL-STD-810 specifies g-rms for random vibration but peak (0-to-peak) for sinusoidal. The -3 dB correction in the formula accounts for peak-to-RMS conversion for sinusoidal vibration only
  • Ignoring vibration-induced phase noise and specifying oscillators based solely on quiescent phase noise: in mobile platforms, vibration often dominates by 40-80 dB over quiescent performance
  • Using ppb/g values directly in dB calculations without converting to 1/g first: Gamma in ppb/g must be multiplied by 1e-9 before use in the phase noise formula
  • Assuming vibration isolation mounts eliminate all vibration: mounts have finite isolation (typically 20-40 dB above resonance) and can amplify vibration near their resonant frequency

Frequently Asked Questions

MIL-STD-810 is a U.S. Department of Defense test standard that defines environmental test methods, including vibration profiles for various military platforms (ground vehicles, fixed-wing aircraft, helicopters, ships, missiles). It specifies the vibration PSD (power spectral density) in g^2/Hz that equipment must withstand. These profiles are the standard inputs for calculating vibration-induced phase noise in defense systems. For example, Method 514.8 defines composite wheeled vehicle vibration at roughly 0.01-0.04 g^2/Hz across 10-500 Hz.
OCXOs (oven-controlled crystal oscillators) typically achieve acceleration sensitivity (Gamma) values of 0.1 to 2 ppb/g, while TCXOs are usually 2 to 10 ppb/g. This 10-20 dB improvement in Gamma translates directly to 10-20 dB lower vibration-induced phase noise. Additionally, premium OCXOs use SC-cut crystals and stress-compensated mounting that further minimize vibration sensitivity, while TCXOs generally use AT-cut crystals with less sophisticated mounting.
Vibration isolation mounts (elastomeric, wire rope, or pneumatic) attenuate vibration transmitted to the oscillator. A well-designed mount provides 20-40 dB of isolation above its resonant frequency, which directly reduces the effective acceleration level seen by the oscillator. However, mounts amplify vibration near their resonant frequency (typically by 6-12 dB), have limited low-frequency isolation, and add weight, size, and cost. The isolator must be selected so its resonant frequency is well below the critical vibration band.
Testing follows IEEE Standard 1139, which defines methods for measuring oscillator phase noise under vibration. The oscillator is mounted on an electrodynamic shaker, and phase noise is measured using a phase noise analyzer or spectrum analyzer while the shaker applies a controlled vibration profile (sinusoidal sweep or random). The measured phase noise versus quiescent phase noise reveals the vibration contribution. The acceleration sensitivity Gamma can be extracted by comparing the measured sideband levels to the known vibration amplitude at each frequency.

Shop Components

Affiliate links — we may earn a commission at no cost to you.

SMA Connectors

Standard SMA RF connectors for board-to-cable connections

RF Cables

Coaxial cable assemblies for RF signal routing

Spectrum Analyzer

Compact spectrum analyzer modules for RF measurement

Related Calculators