EC-22.6 Measuring at Radio Frequency: Calibration, and What Goes Wrong
You can calibrate a network analyser and explain what each calibration standard removes, use a spectrum analyser without misreading its own artefacts, and recognise the handling errors that quietly invalidate high-frequency measurements.
At these frequencies the cable, the connector and the fixture are part of the circuit, so an uncalibrated measurement describes the whole arrangement rather than the component. Calibration is the procedure that mathematically removes everything except what you meant to measure, and it is why a network analyser measurement is trustworthy in a way that most laboratory readings are not. The second half of the topic is the list of ways it still goes wrong.
Work through these
Why the fixture is part of the measurement
Connectors, cables and the piece of board a component sits on all have length, loss and reflections of their own at these frequencies. Without correction, an instrument reports the sum of all of them.
NPTEL: Passive Microwave Circuits, Devices and Measurements (IIT Bhubaneswar) · CourseCalibration standards, and what each one lets the instrument solve for
Measuring known terminations gives the instrument enough equations to determine its own errors and subtract them. Each standard in the set removes a specific term, which is why a partial calibration is worse than none.
NPTEL: Passive Microwave Circuits, Devices and Measurements (IIT Bhubaneswar) · CourseDe-embedding, and moving the reference plane to where the part actually is
Once the instrument is calibrated to the end of its cables, the remaining board between there and the component can be modelled and removed. Getting this wrong shifts phase and makes a good part look badly matched.
NPTEL: Passive Microwave Circuits, Devices and Measurements (IIT Bhubaneswar) · CourseThe spectrum analyser: resolution bandwidth, noise floor and sweep time
The three settings interact, and a narrow filter that lowers the noise floor also slows the sweep enough to miss transient signals. Most misleading spectrum measurements come from one of those three being wrong.
NPTEL: RF Transceiver Design (IIT Roorkee) · CourseArtefacts: images, harmonics and the instrument's own distortion
An analyser overloaded at its input generates products indistinguishable from the ones you are looking for. Stepping the input attenuation and seeing whether the peak moves is the standard test for whether a signal is real.
Handling, torque and connector care
A connector tightened by hand rather than to its torque figure gives a different answer each time, and a damaged one damages every connector it subsequently meets. This is unglamorous and is a genuine cause of wasted weeks.
Sign in to keep your progress.
Free resources
We haven't checked most of these for screen reader use yet.
Links last checked 4 Sept 2026.
Stuck here?
Ask a mentor. A real person answers, and they can see exactly which topic you're on. Usually within a couple of working days.
Checking your session…
Topics shown in module order.