Tools · Preview
Open the budget. Argue with it.
The nine terms of the published in-vacuum wavefront budget, opened as a worksheet. Change an estimate, a distribution, the criterion, or the coverage factor, add the term you think is missing, and watch the combined and expanded uncertainty and the guard-band limit move. Export the result as a document.
Everything runs in your browser and stores nothing on our servers. A worksheet is a structure for argument. It is not a certificate, an accredited result, or a claim about any instrument, ours or yours.
Saved on this device. Starts from the published budget.
| Term | Type | Estimate (nm) | Distribution | u (nm) | Share | Remove |
|---|---|---|---|---|---|---|
From the calibration certificate for the transmission element, which reports an expanded uncertainty at k=2. Divided by 2 to recover the standard uncertainty. This is the term that carries traceability into the measurement, and it is the one that stops mattering to a customer only when it is absent. | 3.000 | 20.2% | ||||
Standard deviation of the mean over ten consecutive acquisitions with the article undisturbed. Type A because it is evaluated from repeated observation, which is the only term here that is. | 1.200 | 3.2% | ||||
The measurement looks at the article through a vacuum window, and the window deforms under one atmosphere of differential pressure and carries stress birefringence. Rectangular over the bound established by a windowed-versus-unwindowed reference comparison. This is the largest term in the budget and it exists only because the measurement is made in vacuum. It is the price of asking the right question. | 4.041 | 36.6% | ||||
The fixture is part of the article because the fixture moves too. Bound taken from the mount's differential contraction across the temperature range, rectangular because any value inside the bound is as likely as any other. | 2.309 | 11.9% | ||||
The article is at a temperature; parts of it are at slightly different ones. Bound from the spread across the article's instrumented points during the acquisition window, converted to wavefront through the substrate's thermo-optic behavior. | 2.021 | 9.1% | ||||
Residual after tilt and defocus are removed in reduction. Rays that do not retrace their path sample the optics differently on the way back, and the error that leaves behind does not remove with the alignment terms. | 1.501 | 5.0% | ||||
Where the analysis mask is drawn changes the RMS it reports. Bound from re-reducing the same data set across the range of defensible mask placements, which is a term that only appears in a budget written by someone who has actually done the reduction. | 0.981 | 2.2% | ||||
Spread across independent setups by different operators following the same written procedure. Named separately rather than folded into repeatability, because a term you cannot see is a term you cannot reduce. | 1.500 | 5.0% | ||||
The terms that could not be separated from each other with the data available. Every real budget has this term and most published ones quietly omit it, which makes the total look better and the document worth less. Carrying it is the difference between a budget and an advertisement. | 1.732 | 6.7% |
- Combined standard uncertainty
- 6.683 nm
- Expanded uncertainty, k = 2
- 13.365 nm
- Guard-band limit
- 18.275 nm
- Largest contributor
- Chamber window contribution
Terms are treated as uncorrelated and sensitivity coefficients are unity, as in the published budget. Correlation, degrees of freedom, and non-unity sensitivities are not modeled here. A worksheet is a structure for argument, not an accredited result or a claim about any instrument.
Read first
The document this worksheet opens.
CUB-2026-01, the worked budget
Every term named with its type, distribution, and basis; the combination; the limits section that says what the budget does not establish.
CTM-2026-01, the test method
The measurement the budget belongs to: what is measured, under what conditions, against what criterion.
WIRES, power-beaming measurement
The program that will carry this structure into power-transfer measurements. In architecture.
Bring the budget you cannot close.
If a term in your own chain will not resolve, or the guard band leaves no room to accept anything, that is a measurement definition conversation. Describe the chain at a non-sensitive level.