A NanoVNA is one of the most useful low-cost RF tools for testing filters. With a proper calibration and a simple two-port setup, it can show whether a band-pass filter passes the correct frequency range, whether a low-pass filter rolls off where expected, whether a high-pass filter blocks low frequencies, and whether a notch filter rejects the target frequency deeply enough.
The most important measurement for RF filter testing is usually S21. S21 shows how much signal passes through the filter from port 1 to port 2. It is used to check insertion loss, passband shape, cutoff frequency, roll-off, stop-band rejection, and notch depth. S11 is also useful because it shows how well the filter input is matched, especially inside the passband.
This guide explains how to test RF filters with a NanoVNA, including band-pass, low-pass, high-pass, and notch filters. It covers calibration, S21 and S11 traces, marker placement, insertion loss, return loss, cutoff points, stop-band rejection, common mistakes, and recommended RF accessories.
Browse the NanoVNA-H4 portable vector network analyzer, spectrum analyzers and RF analysis tools, RF dummy loads, RF power meters, RF test and measurement equipment, and request a formal quote from SDRstore.eu.
To test an RF filter with a NanoVNA, connect the filter between Port 1 and Port 2, calibrate across the frequency range you want to measure, display S21 LOGMAG to see the filter transmission response, and use markers to read insertion loss, cutoff frequency, passband ripple, and stop-band rejection. Add S11 LOGMAG if you also want to check input return loss.
| Filter type | Main NanoVNA trace | What to check |
|---|---|---|
| Band-pass filter | S21 LOGMAG | Passband, insertion loss, center frequency, bandwidth, roll-off, stop-band rejection. |
| Low-pass filter | S21 LOGMAG | Low-frequency passband, cutoff frequency, high-frequency attenuation, roll-off. |
| High-pass filter | S21 LOGMAG | Low-frequency rejection, 3 dB cutoff point, high-frequency passband, insertion loss. |
| Notch filter | S21 LOGMAG | Notch frequency, notch depth, notch width, passband loss outside the notch. |
| Any filter input match | S11 LOGMAG | Return loss, impedance match, passband matching, reflections. |
The simple rule: S21 tells you what gets through the filter. S11 tells you how well the filter input is matched.
S21 is the most important filter trace. It shows how much signal leaves Port 2 when the NanoVNA sends a signal into Port 1.
Use S21 to measure:
For most passive filters, the S21 trace will be near 0 dB in the passband and much lower in the stopband. For example, a good receive filter may show -1 dB to -3 dB insertion loss in the passband and -30 dB, -40 dB, or more rejection in the stopband depending on design.
S11 shows how much energy is reflected back from the filter input. A well-matched filter should usually have better return loss in the passband than in the stopband.
Use S11 to check:
For simple SDR receiving, S21 is usually the first trace to inspect. For RF design, lab reports, or product testing, inspect both S21 and S11.
Keep cables short, use good connectors, and avoid moving the cables after calibration. Cable movement can change the measurement, especially at higher frequencies.
Set the NanoVNA start and stop frequency around the filter you want to test. Do not always scan the entire NanoVNA range. A narrower sweep gives better detail around the filter response.
| Filter example | Suggested sweep | Why |
|---|---|---|
| 433 MHz band-pass filter | 350–500 MHz | Shows passband, roll-off, and nearby rejection. |
| 868 MHz band-pass filter | 750–950 MHz | Shows LoRa/ISM passband and surrounding rejection. |
| 1090 MHz ADS-B filter | 900–1300 MHz | Shows ADS-B passband and out-of-band rejection. |
| FM broadcast notch or block filter | 70–130 MHz | Shows rejection around 88–108 MHz and loss outside the notch. |
| HF high-pass filter | 100 kHz–30 MHz | Shows AM broadcast rejection and HF passband behavior. |
For a clear filter measurement, use these traces:
For most users, S21 LOGMAG alone is enough to see whether the filter passes or blocks the correct frequencies.
Calibration is the most important step. Calibrate after setting the exact start and stop frequency you will use for the measurement.
For a two-port filter measurement, perform:
After calibration, connect the filter between Port 1 and Port 2 without changing the cables more than necessary.
Most passive filters have an input and output. Some are symmetrical, but not all. If the filter has labels, connect NanoVNA Port 1 to the filter input and Port 2 to the filter output.
Basic connection:
If the filter has N-type, BNC, U.FL, MCX, or other connectors, use the minimum number of adapters possible. Extra adapters add loss and mismatch.
Place markers at important points:
Use the marker values to compare the filter against its datasheet or expected behavior.
A band-pass filter passes a chosen frequency range and rejects frequencies below and above it. This is common for ADS-B, AIS, LoRa, GNSS, ham radio, satellite reception, and SDR front-end protection.
| Marker | Where to place it | What it tells you |
|---|---|---|
| Marker 1 | Center frequency | Insertion loss at the wanted frequency. |
| Marker 2 | Lower 3 dB point | Lower edge of useful passband. |
| Marker 3 | Upper 3 dB point | Upper edge of useful passband. |
| Marker 4 | Known strong interferer below passband | How much the filter rejects that signal. |
| Marker 5 | Known strong interferer above passband | How much the filter rejects that signal. |
A good band-pass filter should show low insertion loss in the wanted band and strong rejection outside it. For SDR receiving, a filter with slightly higher insertion loss may still be useful if it prevents receiver overload from strong out-of-band signals.
A low-pass filter passes lower frequencies and rejects higher frequencies. It is often used to reduce harmonics or block unwanted high-frequency signals.
For example, if you are checking a low-pass filter designed to pass up to 30 MHz, sweep from 1 MHz to 100 MHz or wider if your NanoVNA and filter specification allow it.
A high-pass filter rejects lower frequencies and passes higher frequencies. SDR users often use high-pass filters to reduce strong AM broadcast signals, especially when using direct-sampling HF modes.
A practical example is an AM broadcast reject high-pass filter. The expected result is strong rejection through the AM broadcast band and much lower loss above the filter cutoff.
| Marker | Example placement | What it tells you |
|---|---|---|
| Marker 1 | 1 MHz | Rejection in the AM broadcast area. |
| Marker 2 | 2.6 MHz | Approximate cutoff area for some AM reject high-pass filters. |
| Marker 3 | 7 MHz | Passband insertion loss in the HF range. |
| Marker 4 | 14 MHz | Higher HF passband behavior. |
If the high-pass filter is designed to reject 500 kHz to 1.7 MHz, the S21 trace should be far below the passband level in that region and rise after the cutoff point.
A notch filter rejects one narrow frequency range while passing frequencies below and above it. This is useful for reducing strong local interference such as FM broadcast, paging transmitters, or a known unwanted carrier.
For example, an FM broadcast notch filter should show strong rejection between 88 and 108 MHz while keeping lower and higher frequencies less affected. The exact trace depends on filter design.
Insertion loss is the loss added by the filter in the passband. On a NanoVNA S21 LOGMAG trace, it appears as a negative dB value.
| S21 value | Meaning | Typical interpretation |
|---|---|---|
| 0 dB | No loss | Ideal value, rarely perfect in real passive filters. |
| -0.5 dB | Very low loss | Excellent for many receive filters. |
| -1 dB to -3 dB | Moderate loss | Often acceptable if rejection is useful. |
| -6 dB | Significant loss | May still be useful in some cases but reduces signal noticeably. |
| -20 dB or lower | Strong attenuation | Usually stopband or notch region. |
| -40 dB or lower | Very strong rejection | Good stopband result, if measurement noise floor allows confidence. |
Do not judge a filter only by insertion loss. A filter with 2 dB passband loss may still improve real SDR reception if it removes a much stronger out-of-band signal that was overloading the receiver.
S11 LOGMAG shows how much signal is reflected from the filter input. More negative is generally better for return loss.
| S11 value | Meaning | Interpretation |
|---|---|---|
| -6 dB | Poor match | May still work, but mismatch is visible. |
| -10 dB | Usable match | Common minimum target in many practical RF setups. |
| -15 dB | Good match | Generally good for many receive filter setups. |
| -20 dB or lower | Very good match | Strong result if measured correctly. |
For passive filters, S11 may look worse in the stopband. That can be normal because the filter is rejecting energy instead of passing it into a matched load.
The 3 dB point is where the S21 trace has dropped by about 3 dB from the passband level. It is commonly used to define cutoff frequency or passband edge.
For example, if a filter passband is around -1 dB and the response drops to -4 dB, that point is approximately 3 dB below the passband level.
For a band-pass filter, bandwidth is usually measured between the lower and upper 3 dB points. Place markers at both points and subtract the lower frequency from the upper frequency.
Stop-band rejection is how much the filter blocks unwanted frequencies. On the S21 trace, compare the passband value to the stopband value.
Example:
Be careful when measuring very deep rejection. A NanoVNA has a limited noise floor and dynamic range, so very high rejection values may require a better VNA or spectrum analyzer setup for confirmation.
Use this measurement to see whether the filter passes ADS-B while reducing nearby mobile-network or other out-of-band energy.
This is useful for Sub-GHz sensors, remotes, telemetry devices, and receive-side monitoring around 433 MHz.
This is useful for LoRaWAN EU868, industrial sensors, and Sub-GHz monitoring.
This helps determine whether an FM filter can reduce strong FM broadcast overload without hurting nearby wanted signals too much.
This is useful when strong AM broadcast stations overload an SDR receiver during HF direct-sampling work.
NanoVNA and TinySA Ultra are both useful RF tools, but they answer different questions.
| Question | Best tool | Why |
|---|---|---|
| What is the filter insertion loss? | NanoVNA | S21 shows how much signal passes through the filter. |
| Where is the 3 dB cutoff? | NanoVNA | Markers on S21 show cutoff points clearly. |
| Is the filter input matched? | NanoVNA | S11 shows return loss and impedance behavior. |
| What signals are currently in the air? | TinySA Ultra or spectrum analyzer | A spectrum analyzer shows live RF energy. |
| Does the filter reduce real interference? | Both | NanoVNA tests the filter response; TinySA shows real spectrum before and after. |
| Does the antenna match the filter and cable setup? | NanoVNA | NanoVNA can check antenna match and cable behavior. |
For the best workflow, measure the filter with NanoVNA first, then use TinySA Ultra or an SDR to confirm how it affects real-world reception.
Calibration is valid for the frequency range used during calibration. If you change the sweep significantly, calibrate again.
Long cables add loss and mismatch. Moving cables after calibration can change the result. Use short, stable cables and do not move them unnecessarily.
For S21 filter testing, through calibration is important. Without it, cable and adapter loss may be included in the filter measurement.
Active filters, powered filters, LNAs with filtering, and amplified modules can overload the NanoVNA or create misleading results. Check power, bias, input limits, and test setup first.
NanoVNA is excellent for learning and practical RF work, but very deep stopband rejection may require a higher-dynamic-range VNA or professional analyzer.
Some filters are directional. If the filter has input and output labels, test it in the correct direction and optionally repeat the measurement reversed.
Cheap or damaged adapters can add ripple, loss, or mismatch. Use good adapters and keep the adapter chain short.
A filter may have a good-looking S21 passband but poor input match. For serious work, check both S21 and S11.
A good filter test report should include enough information for someone else to repeat the measurement.
Best for: beginners, SDR users, ham radio operators, students, and first RF filter measurements.
Best for: SDR users who want to test filters on the bench and then confirm real reception improvements.
Best for: defensive RF monitoring, rogue wireless detection, Sub-GHz audits, drone RF awareness, and facility interference troubleshooting.
Best for: IoT product developers, RF labs, universities, and product-security teams testing filters before integration.
NanoVNA-H4 is required for RF filter testing, including S21 insertion-loss measurement, S11 return-loss measurement, cutoff frequency checks, passband and stopband validation, antenna matching, and RF lab training.
TinySA Ultra is required to confirm real RF spectrum behavior before and after filtering, investigate interference, check band occupancy, and support field troubleshooting alongside NanoVNA measurements.
Short SMA cables, adapters, calibration standards, attenuators, and RF accessories are required to create repeatable NanoVNA filter measurements and reduce errors caused by poor cabling or connector mismatch.
A higher-grade vector network analyzer is required when filter measurements need higher dynamic range, wider frequency coverage, better calibration confidence, group delay analysis, production validation, or customer-facing reports.
Universities, RF labs, SDR users, ham radio clubs, IoT developers, cybersecurity teams, telecom labs, and product-testing teams can request a formal quotation directly from SDRstore.eu.
Use the Add to Quote button on product pages or the document icon on product cards. Add NanoVNA-H4, TinySA Ultra, filters, attenuators, dummy loads, RF power meters, antennas, cables, adapters, SDR receivers, and project notes to one quote request.
A quote request is useful when you need:
Read the SDRstore.eu quote-request guide.
Use NanoVNA to test RF filters before installing them in an SDR, receiver, transmitter chain, or RF product. Start with S21 LOGMAG to see insertion loss, cutoff, passband, stopband, and notch depth. Add S11 LOGMAG when you also need to check return loss and impedance matching.
For basic SDR use, NanoVNA-H4, short SMA cables, a calibration kit, and a few known filters are enough to learn filter testing. For better real-world troubleshooting, add TinySA Ultra to compare the RF spectrum before and after filtering.
The best filter test is not just a screenshot. It is a calibrated measurement with known cables, correct markers, documented settings, and a clear answer: does this filter pass the signal we want and reject the signal we do not want?
Yes. NanoVNA can test RF filters by measuring S21 transmission and S11 reflection. S21 shows insertion loss, passband, cutoff, stopband rejection, and notch depth. S11 shows input match and return loss.
Use S21 LOGMAG as the main trace for filter testing. Add S11 LOGMAG if you also want to check return loss and input matching.
Connect the band-pass filter between Port 1 and Port 2, calibrate across the target range, display S21 LOGMAG, and place markers at the center frequency, lower 3 dB point, upper 3 dB point, and stopband frequencies.
Set a sweep from below the passband to well above the cutoff, calibrate, connect the filter, and inspect S21 LOGMAG. Check passband insertion loss, 3 dB cutoff, roll-off, and high-frequency rejection.
Set a sweep from below the rejected range to above the wanted passband, calibrate, connect the filter, and inspect S21 LOGMAG. Check low-frequency rejection, cutoff point, and passband insertion loss.
Set the sweep around the expected notch, calibrate, connect the filter, and inspect S21 LOGMAG. Place a marker at the deepest point of the notch and markers outside the notch to check normal passband loss.
Insertion loss is the loss through the filter in the passband. On a NanoVNA, it appears as a negative S21 LOGMAG value such as -0.8 dB, -1.5 dB, or -3 dB.
Return loss is related to how much signal is reflected from the filter input. On NanoVNA, S11 LOGMAG shows this behavior. More negative values usually mean a better match.
Common causes include poor calibration, wrong sweep range, missing through calibration, bad cables, too many adapters, moved cables after calibration, wrong filter direction, receiver dynamic range limits, or testing an active device like a passive filter.
Yes. Use the Add to Quote button on product pages or the document icon on product cards. Add NanoVNA-H4, TinySA Ultra, filters, cables, adapters, attenuators, dummy loads, RF power meters, and project notes so the full filter testing setup can be quoted together.
No posts found
Write a review