Verifying an Attenuator

Attenuators are fundamental RF building blocks — pads for level reduction, impedance matching improvement, and VSWR reduction between stages. The NanoVNA-H can verify the three key specs that matter: attenuation value, frequency flatness, and impedance match.

What You Will Learn

• Measuring attenuation (S21) and comparing to rated value
• Evaluating frequency flatness across the rated band
• Checking input/output match (S11) with a proper termination
• Testing reversibility of a resistive pad
• Identifying common problems with homebrew and surplus attenuators

Equipment Needed

• NanoVNA-H (firmware 1.2.40 or later recommended)
• Attenuator under test
• 50 ohm SMA load (for S11 measurements)
• SMA adapters if the attenuator uses N, BNC, or other connectors

Tip

If the attenuator has SMA connectors matching your NanoVNA-H, connect it directly between the ports. Fewer adapters means less measurement uncertainty.

Calibration

Perform a full SOLT calibration before measuring. The Through standard establishes your 0 dB reference line — any loss you see after inserting the attenuator is the attenuation itself.

1. Set the frequency range

   Cover the attenuator’s rated frequency range with some margin on each side.

   For a general-purpose DC-1 GHz attenuator:

   • START: 50k (50 kHz)
   • STOP: 1.2G (1.2 GHz)

2. Configure traces

   • Trace 1: S21 LOGMAG (attenuation)
   • Trace 2: S11 LOGMAG (return loss)

3. Run full SOLT calibration

   OPEN, SHORT, LOAD on Port 1, then THRU between Port 1 and Port 2.

   See Full Calibration for the detailed procedure.

4. Verify the cal

   With the THRU still connected, S21 should read 0 dB across the band and S11 should be well below -40 dB. If not, recalibrate.

Measuring Attenuation Value (S21)

1. Insert the attenuator

   Connect the attenuator between Port 1 (CH0) and Port 2 (CH1).

2. Read S21

   The S21 trace shows the attenuation in dB. Place a marker at the center of the frequency range.

   Example: a 10 dB attenuator should read approximately -10 dB.

3. Compare to rated value

   General-purpose attenuators are typically specified with the following tolerances:

   Grade	Tolerance
   Precision (Weinschel, HP/Keysight, Aeroflex)	±0.2 dB
   Good quality (Mini-Circuits, etc.)	±0.5 dB
   General purpose / surplus	±1.0 dB
   Homebrew / unknown	Measure it and label it

4. Check multiple frequencies

   Place markers at several points across the band and note the readings. If the value varies by more than the tolerance above, the attenuator may have issues.

Measuring Frequency Flatness

Flatness is the variation in attenuation across the rated frequency range. A flat attenuator maintains its rated value from low frequencies through the upper end of its band.

1. Widen the sweep if needed

   Make sure the sweep covers the full rated frequency range.

2. Auto-scale the S21 trace

   Adjust the scale so you can see small variations. Set the reference to the nominal attenuation value and use 1 dB/div or 0.5 dB/div.

3. Use delta markers to measure flatness

   Place the reference marker at the center of the band. Then move a second marker to find the highest point and lowest point of the S21 trace. The difference between these extremes is the flatness.

   Example: if the highest reading is -9.7 dB and the lowest is -10.4 dB, flatness is 0.7 dB.

Flatness Expectations

Attenuator Quality	Flatness (DC - 1 GHz)	Flatness (DC - 2 GHz)
Precision	±0.2 dB	±0.4 dB
Good quality	±0.5 dB	±0.8 dB
Budget / homebrew	±1.0 dB	±2.0 dB or worse

Flatness usually degrades above 1 GHz because parasitic reactances in the resistors become significant. An attenuator that looks fine at 500 MHz may roll off substantially at 1.5 GHz.

Measuring Input/Output Match (S11)

A good attenuator presents close to 50 ohms at both ports. Measuring return loss tells you how well it matches.

1. Terminate the far end

   Connect a 50 ohm SMA load to Port 2 end of the attenuator, then connect the attenuator input to Port 1.

2. Read S11

   The S11 trace shows return loss in dB. More negative values indicate a better match.

3. Apply the rule of thumb

   Return loss should be at least twice the attenuation value:

   Expected return loss >= 2 x attenuation (dB)

   A 6 dB pad should show at least -12 dB return loss. A 10 dB pad should show at least -20 dB. A 20 dB pad should show -40 dB or better.

4. Check across the band

   Return loss typically degrades at higher frequencies. Note where it starts to fall off.

Why Attenuators Improve Match

An attenuator absorbs reflected energy on both passes — once going toward the mismatch and once coming back. This is the “return loss improvement” property:

Improvement ≈ 2 × attenuation value (dB)

A 6 dB pad placed in front of a load with 6 dB return loss (3:1 SWR) improves the system return loss to approximately 18 dB (1.3:1 SWR). This is why pads are commonly used to tame difficult loads.

Reversibility Check

A properly built resistive attenuator is a reciprocal device — it should measure identically in both directions.

1. Measure S21 in the forward direction

   Note the attenuation value and shape of the trace.

2. Flip the attenuator

   Reverse the input and output connections.

3. Measure S21 again

   Compare the two traces. They should overlay within measurement uncertainty (typically ±0.1 dB).

4. Investigate asymmetry

   If the readings differ by more than 0.3 dB, the attenuator may have:

   • A damaged connector on one end
   • A cold solder joint
   • Internal construction that is not symmetric (some designs are intentionally directional, but most fixed pads are not)

Practical Examples

Verifying a Homebrew Pi-Network Pad

Testing a hand-built attenuator on a PCB or inside an SMA barrel:

1. Measure S21 from 50 kHz to 1 GHz

   Note the nominal attenuation. Compare to the design value based on the resistors you used.

2. Check flatness

   Homebrew pads often use through-hole or chip resistors with parasitic inductance. Look for attenuation that increases above 500 MHz — this is normal and depends on resistor type and layout.

3. Measure S11

   If return loss is worse than expected, check resistor values with a multimeter. A 5% tolerance resistor may shift the match noticeably.

4. Compare to the design table

   Use the reference table below to verify that your resistor values are correct for the target attenuation.

Testing an Unknown Surplus Attenuator

Common problems with cheap or surplus attenuators from online marketplaces:

• Wrong value: The label says 10 dB but it measures 8.3 dB. This is surprisingly common with budget attenuators.
• Poor flatness above 500 MHz: Attenuation rolls off or develops ripple at higher frequencies due to poor internal construction.
• Bad match at one port: One connector may be damaged or internally the pad is poorly built.
• Frequency-dependent value: The attenuator reads close to spec at low frequencies but drifts at the upper end of its range.

If you find an attenuator that measures well, label it with the actual measured value and the frequency range you verified. “9.8 dB, flat to 1 GHz” is more useful than the “10 dB” printed on the housing.

Quick Reference: Pi-Network Resistor Values

Standard 50 ohm Pi-network attenuator resistor values for verification. If you built it yourself, these are the values to check with a multimeter.

Attenuation	Series R (ohm)	Shunt R (ohm)
1 dB	5.8	870
2 dB	11.5	436
3 dB	17.6	292
6 dB	37.4	150
10 dB	71.2	96.2
20 dB	248	61.1
30 dB	790	53.3

Caution

Attenuators have power ratings. The NanoVNA-H outputs approximately -10 dBm (0.1 mW), so there is no risk to any attenuator during testing. However, never leave a high-power transmitter connected to the NanoVNA-H — the input circuitry can be damaged by signals above +10 dBm.

Troubleshooting

S21 reading does not match the rated value

• Verify calibration is still valid (check the THRU reference)
• Check adapters for damage or looseness
• Confirm connector types are mated correctly (SMA vs. RP-SMA is a common mix-up)

Return loss is worse than 2x the attenuation value

• Ensure the far-end termination is a good 50 ohm load
• The attenuator may have a damaged connector
• At higher frequencies, parasitic effects in the resistors degrade the match

Flatness degrades sharply above a certain frequency

• This is normal behavior — all resistive attenuators have a usable frequency ceiling
• Chip resistors and compact construction extend that ceiling
• Wirewound or large through-hole resistors roll off earlier

Next Steps

• Characterizing a Filter - Similar S21/S11 measurement techniques applied to filters
• Your First S21 Measurement - Basics of transmission measurement
• Full Calibration - Ensure your calibration is accurate before critical measurements