Source: https://guthlabs.ai/docs/instruments/op-amp-noise-estimator-input-referred/mechanism/

instrument mechanism / verified-clean

# Op-Amp Noise Estimator (input-referred) mechanism

The inspectable method record for Op-Amp Noise Estimator (input-referred), kept separate from implementation and qualification.

Status: **VERIFIED_CLEAN**Lifecycle: **verified-clean**

## Model scope

The core white-noise formula is correct under the stated unit and spectral assumptions. Verdict is NEEDS_REVIEW because the specification advertises optional current-noise/source-resistance and 1/f-corner functionality without defining their formulas, required parameters, or integration model.

- Vn_rms = en × √BW is correct only for constant white voltage-noise density and a bandwidth expressed in Hz; it does not calculate total noise when 1/f noise is significant.

- The optional current-noise and source-resistance inputs have no stated formula. Their contribution would require at least (in)^2 × R_source^2 added to the mean-square voltage noise, with resistance in ohms and current-noise density in A/√Hz.

- The optional 1/f corner correction is unspecified: no 1/f noise model, integration limits, or formula are provided, so that feature cannot be verified.

- The specification says total input-referred noise while labeling the calculation a simplified white-noise-band estimate; this scope limitation must be explicit in the user-facing result.

## Units and conventions

- en is a one-sided white voltage-noise density in V/√Hz, commonly entered in nV/√Hz and converted using 1 nV = 1e-9 V.

- BW is the integrated noise bandwidth in Hz, not merely a nominal -3 dB bandwidth unless those are explicitly treated as equivalent.

- The output Vn_rms is in volts RMS; outputs may also be displayed in µV RMS or nV RMS after conversion.

- BW must be finite and nonnegative; en must be finite and nonnegative.

- If current-noise and source-resistance contributions are later supported, current noise must be in A/√Hz, source resistance in Ω, and temperature-dependent resistor thermal noise requires an explicitly specified temperature in kelvin.

## Inspectable vector

One of 4 source-bound review vectors:

````
{
  "inputs": {
    "voltage_noise_density": 8,
    "voltage_noise_density_unit": "nV/√Hz",
    "bandwidth": 20000,
    "bandwidth_unit": "Hz"
  },
  "expected_outputs": {
    "total_rms_noise_voltage": 0.00000113137085,
    "total_rms_noise_voltage_unit": "V RMS",
    "display_value": 1.13137085,
    "display_unit": "µV RMS"
  },
  "working": "Convert en: 8 nV/√Hz = 8 × 10^-9 V/√Hz. √BW = √20000 = 141.4213562 √Hz. Vn_rms = (8 × 10^-9) × 141.4213562 = 1.13137085 × 10^-6 V RMS = 1.13137085 µV RMS."
}
````

## Failure boundary

**BREAK_CLAIMED** · SELF_FLAGGED

A negative bandwidth produces an undefined real-valued square root and must be rejected. The signed-density probe also exposes a convention hazard: literal signed arithmetic can produce a negative RMS result unless the instrument enforces or applies a nonnegative-magnitude convention.

## Other records

[01**Overview**](<https://guthlabs.ai/docs/instruments/op-amp-noise-estimator-input-referred/>)[03**Implementation**](<https://guthlabs.ai/docs/instruments/op-amp-noise-estimator-input-referred/implementation/>)[04**Qualification**](<https://guthlabs.ai/docs/instruments/op-amp-noise-estimator-input-referred/qualification/>)

## Sources and evidence

These pointers bind the page to evidence. They do not grant authority or collapse verification into deployment.

- `VERIFIED_LEDGER.json`

- `evidence-tranche5/tranche5-results.json`

- `BOUNDARY_HAZARD_CATALOG.json`