10 MHz OCXO Error Budget for Temperature Drift, Aging, Calibration, and the Measurement Reference
Acceptance criteria for a 10 MHz OCXO require more than choosing the smallest ppb value. Temperature drift, aging, initial calibration offset, and the measurement reference represent separate sources of error. The static budget below shows how to establish verifiable limits while tracking short-term stability and phase noise separately.
1. Calculate relative and absolute frequency error
The relative frequency offset is y = (f − f₀)/f₀. At 10 MHz, 1 ppb corresponds to 0.01 Hz and 10 ppb to 0.1 Hz. An integer-hertz display cannot establish whether a 0.1 Hz requirement is met by itself. Use a suitable measurement method, gate time, and frequency reference.
For temperature effects, record the temperature range, reference point, and whether the value is defined as a maximum or peak-to-peak. For aging, record when the interval starts, any preconditioning time, and the units of time. Preserve each definition before converting values and report values measured under different conditions separately.
2. A conservative static budget
Suppose the project allows an absolute frequency offset of ±50 ppb over the specified interval. Assume up to 10 ppb residual initial calibration error, 15 ppb temperature effect, 5 ppb supply effect, and 10 ppb aging over that interval. If the limits are distinct and all must be covered, their magnitudes add up to 40 ppb and leave a 10 ppb margin.
These are maximum limits, so a root-sum-square calculation for statistically independent random variables would not justify a smaller result. If a datasheet already includes certain components in total frequency error, count each once. Choose the budgeting method based on the system requirements and the definitions of the individual errors.
3. Estimate annual aging using applicable specifications
Daily, first-year, and ten-year aging figures may have different preconditions and guarantees. Devices with substantial early drift may age at a nonuniform rate even over a short interval. Base the calibration interval on the guaranteed device limits and the system maintenance strategy. A single day of observed aging cannot establish the annual value.
When verifying aging, log powered-on time, environmental conditions, the measurement reference, and any calibration. Mark power interruptions, reference changes, or frequency readjustments in the data so they can be distinguished from device aging.
4. Track ADEV, phase noise, and warm-up separately
State the averaging time τ for Allan deviation (ADEV) and the carrier and offset frequencies for phase noise. These metrics describe behavior on different timescales or in the frequency domain. A phase-noise reading at one offset does not establish long-term frequency error. Define warm-up as the time from specified power-on conditions to reaching a stated frequency-error window.
Check the measurement system itself. Reference accuracy, the counter method, connections, and the environment can affect the results. Report the device readings and measurement uncertainty separately. A small displayed value alone cannot establish the device's actual error.
Calculation and record table
| Budget item (all hypothetical) | Limit in ppb | Equivalent offset at 10 MHz |
|---|---|---|
| Residual calibration error | 10 | 0.10Hz |
| Temperature effect | 15 | 0.15Hz |
| Supply effect | 5 | 0.05Hz |
| Aging over the specified interval | 10 | 0.10Hz |
| Sum of conservative limits | 40 | 0.40Hz |
| Project limit | 50 | 0.50Hz |
References and applicability
This technical analysis uses public documentation and was prepared with AI assistance. All worked examples are hypothetical. They do not establish product specifications, measured results, or compatibility. For a specific device, use the exact ordering configuration, the latest manufacturer datasheet, and board-level validation.
See the manufacturer reference Epson Crystal Device technical terms and applicable conditions.
Related selection guides
For help documenting these parameters in a BOM or RFQ, see MERLIN’s OCXO stability and selection documentation. Confirm specific part parameters against the applicable manufacturer datasheet and project validation.
