Published on 10/13/2026
A DUT passes in the morning and fails in the afternoon. The instrument is unchanged, but a test cable was moved and an adapter replaced. A VNA measures the path presented to it; calibration cannot make an unstable physical setup repeatable.
Build a VNA bench with cables and interfaces qualified for the band, controlled reference planes and verified connection repeatability. Loss, phase stability and connector condition all contribute to measurement uncertainty. A cable that merely passes continuity is not a precision test-port cable.
This guide focuses on the installed measurement chain. It complements connector cleaning and torque guidance, while keeping calibration, verification and production acceptance as separate steps.
On This Page
Define frequency range, DUT interfaces, measurement power and expected uncertainty. A short fixture measurement and a long movable chamber connection have different cable demands. Match the cable and adapter system to the task, not only to the analyzer's maximum frequency.
| Requirement | Why it matters |
|---|---|
| Insertion loss | Affects available signal level and measurement margin |
| Return loss | Contributes to mismatch interactions |
| Phase versus flex | Movement can change electrical length |
| Connector repeatability | Repeated mating adds uncertainty |
| Temperature behavior | Bench drift can change phase and amplitude |
| Reference plane | Defines what the reported DUT result includes |
Calibration corrects systematic errors for the configured measurement path under its assumptions. Choose the reference plane deliberately. If an adapter remains beyond that plane, its behavior can be included in the result unless valid characterization moves the plane.
Do not disconnect test cables from analyzer ports or change adapters and then assume the earlier calibration still applies unchanged. Recalibrate or use a valid characterized method appropriate to the measurement. Keep records of the cable and adapter arrangement.
Port extension is not a universal replacement for full adapter characterization. A simple delay correction does not inherently remove every mismatch and loss effect. Use the instrument's supported method and verify it with suitable standards.
Key idea: Calibration establishes a corrected configuration; verification checks whether that configuration still behaves as expected.
Get the interface right before ordering
Send the equipment-port details, cable length and operating band to SigmaRF.
Specify a Custom AssemblyCable flex can change magnitude and phase, and its contribution can be important in uncertainty analysis. Arrange the DUT area before calibration, then minimize movement between calibration and measurement. A cable advertised as phase-stable still needs a stated tolerance and test conditions.
Inspect and clean connectors with approved methods. Use the selected interface's torque instruction, hold the appropriate backing surface and avoid rotating the cable body. Do not force a damaged pin or incompatible precision interface.
A port saver can reduce wear on an instrument connector, but it becomes part of the measurement path. Include it in calibration or characterization and inspect it regularly. More adapters create more repeatability and mechanical-load concerns.
Use suitable verification standards with a current calibration and expected acceptance limits. A verification failure can reflect connection problems, damaged cables, wrong setup or instrument issues. Follow a controlled diagnostic sequence rather than adjusting the DUT to make the trace look acceptable.
Characterize repeatability by controlled repeated mating or movement under a defined method. Store cable identities and results so an intermittent bench problem can be traced. A visual inspection is necessary but does not establish phase stability.
Read SigmaRF's connector cleaning, torque and cable test-report guides for handling and production acceptance. This article focuses on measurement uncertainty and stable bench configuration.
An assumed cable delay changes by 1 ps when moved. Calculate the phase change at 10 GHz. This is a small illustrative delay, not a cable specification.
A cable can look mechanically unchanged while producing meaningful microwave phase error. The calculation does not establish the full measurement uncertainty; calibration, connector repeatability and instrument contributions also matter.
What makes a good VNA test cable?
It must meet the required band, loss, match, phase and repeatability limits under the intended handling. Connector condition and a stable installation are as important as the cable label.
Can moving a VNA cable change the result?
Yes. Flex can alter phase and magnitude and contribute to uncertainty. Control movement and qualify the cable under realistic test conditions.
Do I need to recalibrate after changing an adapter?
Normally the measurement configuration has changed. Recalibrate or use a valid characterized correction method appropriate to the path and required uncertainty.
Does a port saver affect VNA measurements?
Yes. It is another RF component and interface. Include it in calibration or characterization and monitor its condition.
Is calibration the same as verification?
No. Calibration corrects systematic measurement errors for a configured path. Verification uses suitable standards to check expected performance afterward.
How much phase error does 1 ps cause at 10 GHz?
It corresponds to 3.6 degrees of phase change. The same delay change matters more as frequency rises, so specify phase stability at the relevant band.
A trustworthy VNA bench is a controlled measurement configuration. Choose appropriate cables, protect connectors, preserve reference planes and verify the setup before relying on the DUT result.
For a cable quote, include band, interfaces, movement, phase tolerance and measurement purpose. A test cable is more than a low-loss antenna lead with expensive connectors.
SigmaRF VNA-sweeps every assembly for VSWR and insertion loss before shipment. Its testing capability spans 100 MHz to 40 GHz; each assembly is specified for its own operating band. Builds are supported by the Vadodara volume facility and Cleveland-area precision facility. Lead times can be as short as 2 days, depending on the build and component availability.
Get the interface right before ordering
Send the equipment-port details, cable length and operating band to SigmaRF.
Standard AssembliesCustom AssembliesTalk to an Engineer