VNA Basics
Before diving into measurements, it helps to understand what a VNA actually does and what the numbers on screen mean. This page covers the core concepts without assuming prior RF knowledge.
What is a vector network analyzer?
Section titled “What is a vector network analyzer?”A VNA is a test instrument that measures how an electrical component or circuit responds to RF signals across a range of frequencies. The “vector” part means it captures both magnitude (how much signal) and phase (the timing relationship of the signal) — not just one or the other.
This is what distinguishes a VNA from simpler instruments:
- A scalar network analyzer or SWR meter measures only magnitude. It can tell you how much signal is reflected, but not the phase.
- A spectrum analyzer measures signal power at each frequency, but it does not inject a known stimulus.
- A VNA does both: it sends a known signal into the device under test and measures what comes back (or passes through), capturing the full complex response.
That complex response — magnitude and phase together — is what lets you compute impedance, plot on a Smith chart, and design matching networks. Without phase information, you know how much signal is reflected but not why, which severely limits what you can do with the data.
The word “network”
Section titled “The word “network””In RF engineering, “network” means an electrical circuit or component — not a computer network. A filter is a network. An antenna is a network. A length of coaxial cable is a network. The name “network analyzer” simply means “circuit analyzer.”
S-parameters
Section titled “S-parameters”S-parameters (scattering parameters) are the standard way to describe how an RF network behaves. They quantify how much signal is reflected and how much passes through at each frequency.
For a two-port network (a device with an input and an output), there are four S-parameters:
| Parameter | Direction | What it measures |
|---|---|---|
| S11 | Port 1 to Port 1 | Input reflection coefficient. How much of the signal sent into Port 1 bounces back. |
| S21 | Port 1 to Port 2 | Forward transmission coefficient. How much of the signal sent into Port 1 passes through to Port 2. |
| S12 | Port 2 to Port 1 | Reverse transmission coefficient. How much signal passes from Port 2 back to Port 1. |
| S22 | Port 2 to Port 2 | Output reflection coefficient. How much signal sent into Port 2 bounces back. |
The NanoVNA-F V3 directly measures S11 and S21. To measure S22 and S12, swap the DUT connections between the two ports and measure again.
Reading S-parameter notation
Section titled “Reading S-parameter notation”The two digits in “S21” tell you the signal path: the first digit is the receiving port, the second is the sending port. So S21 means “signal received at port 2, sent from port 1” — in other words, forward transmission.
S11: Reflection
Section titled “S11: Reflection”S11 tells you what fraction of the signal bounces back from the device under test. It is the primary measurement for:
- Antenna tuning. A well-matched antenna at its resonant frequency absorbs most of the transmitted power and reflects very little. S11 shows you where that resonance occurs and how well-matched it is.
- Impedance matching. Any point where impedance changes — a connector, a junction, a mismatch in a transmission line — creates a reflection. S11 quantifies that reflection.
- Return loss. When expressed in decibels (dB), S11 is called return loss. A return loss of -20 dB means only 1% of the power is reflected — a very good match.
S11 display formats
Section titled “S11 display formats”The same S11 data can be displayed in several formats, each useful for different purposes:
Plots the reflection magnitude in decibels versus frequency. This is the most common view for antenna work. The deeper the dip, the better the match at that frequency.
Typical values:
- -6 dB: acceptable for many applications (SWR ~3:1)
- -10 dB: good match (SWR ~2:1)
- -20 dB: excellent match (SWR ~1.2:1)
- -30 dB or better: nearly perfect match
Standing Wave Ratio — a dimensionless ratio derived from S11. Older ham radio convention, but still widely used.
Typical values:
- 1.0:1 — perfect match, no reflected power
- 1.5:1 — good match, ~4% reflected power
- 2.0:1 — acceptable for most transmitters, ~11% reflected power
- 3.0:1 — marginal, ~25% reflected power
A specialized circular plot that shows impedance (resistance + reactance) at every frequency on a single diagram. Indispensable for matching network design.
The center of the Smith chart represents 50 ohms (the reference impedance). Points above center are inductive; points below are capacitive. Points to the left have less resistance than 50 ohms; points to the right have more.
S21: Transmission
Section titled “S21: Transmission”S21 tells you how much signal passes through the device under test from Port 1 to Port 2. It is the primary measurement for:
- Filter characterization. S21 shows the passband shape, insertion loss, and out-of-band rejection of a filter.
- Amplifier gain. A positive S21 (in dB) means the device adds gain. A negative S21 means it attenuates.
- Cable loss. Connect one end of a cable to each port. S21 shows the attenuation at each frequency.
S21 display formats
Section titled “S21 display formats”The most common view for transmission measurements. Displays gain or loss in decibels versus frequency.
- 0 dB means the signal passes through unchanged.
- Positive values indicate gain (amplifier).
- Negative values indicate loss (cable, filter, attenuator).
Shows the phase shift introduced by the DUT at each frequency. Important for amplifier stability analysis and delay measurements.
Single-port vs. two-port measurements
Section titled “Single-port vs. two-port measurements”Some devices have only one RF port — an antenna, for example. These are single-port measurements. You connect the DUT to PORT1 and measure S11 only.
Devices with an input and an output — filters, amplifiers, cables — are two-port devices. You connect one side to PORT1 and the other to PORT2, then measure both S11 (input match) and S21 (transmission).
| Measurement type | What to connect | S-parameters measured |
|---|---|---|
| Single-port (antenna, load) | DUT on PORT1 only | S11 |
| Two-port (filter, cable, amp) | DUT between PORT1 and PORT2 | S11 and S21 |
| Reverse two-port | DUT swapped (output on PORT1, input on PORT2) | S22 and S12 |
Why calibration matters
Section titled “Why calibration matters”Every measurement system introduces errors. The cables have loss. The connectors have small reflections. The instrument’s internal circuits are not perfectly flat across the entire frequency range. These errors shift and distort your measurements.
Calibration characterizes those errors by measuring known standards — devices whose electrical behavior is precisely defined:
- Open — an unterminated connector (nearly total reflection, 0-degree phase)
- Short — a shorted connector (nearly total reflection, 180-degree phase)
- Load — a precision 50-ohm termination (minimal reflection)
- Thru — a direct connection between Port 1 and Port 2 (minimal loss, known delay)
By comparing the measured response of these standards against their known ideal values, the VNA builds a mathematical error model. That model is then subtracted from every subsequent measurement, leaving you with the response of the DUT alone.
The calibration reference plane
Section titled “The calibration reference plane”Calibration defines the reference plane — the physical point at which the measurement begins. When you calibrate at the ends of your cables, the reference plane is at those cable ends. Everything between the VNA ports and the reference plane is “calibrated out” and will not appear in the measurement.
This is why it is important to calibrate with the exact cables you will use for the measurement. If you calibrate with short cables and then add an extension, that extension becomes part of the measured network.
Common devices under test
Section titled “Common devices under test”| DUT | Primary measurement | What to look for |
|---|---|---|
| Antenna | S11 (return loss, SWR, impedance) | Resonant frequency, bandwidth, impedance at center frequency |
| Bandpass filter | S21 (insertion loss, shape) and S11 (input match) | Center frequency, 3 dB bandwidth, passband ripple, stopband rejection |
| Low-pass / high-pass filter | S21 and S11 | Cutoff frequency, rolloff rate, passband flatness |
| Coaxial cable | S21 (loss) and S11 (reflections) | Loss per length at frequency, connector quality, fault location (TDR) |
| Amplifier | S21 (gain) and S11 (input match) | Gain flatness, gain at specific frequencies, input return loss |
| Attenuator | S21 (attenuation) and S11 (match) | Attenuation accuracy, flatness, match quality |
| Crystal / ceramic filter | S21 (passband) | Center frequency, bandwidth, insertion loss, shape factor |
Decibels: a quick reference
Section titled “Decibels: a quick reference”S-parameters are almost always expressed in decibels (dB). Decibels are a logarithmic ratio, which compresses large ranges into manageable numbers.
| Power ratio | dB value | What it means |
|---|---|---|
| 1 (no change) | 0 dB | Signal is unchanged |
| 2 (doubled) | +3 dB | Signal doubled in power |
| 10 | +10 dB | 10x power increase |
| 100 | +20 dB | 100x power increase |
| 0.5 (halved) | -3 dB | Signal halved in power |
| 0.1 | -10 dB | 10x power reduction |
| 0.01 | -20 dB | 100x power reduction |
| 0.001 | -30 dB | 1000x power reduction |
For S11 (reflection), more negative is better — it means less signal is bouncing back. For S21 (transmission through a passive device), values closer to 0 dB mean less loss.
With these fundamentals in place, you are ready to start making measurements. Head to Your First S11 Measurement for a hands-on walkthrough.