Keysight Technologies Power Meter Measurement Speed Optimization Solution

Abstract

In RF and microwave power testing, especially in manufacturing environments, test speed is often a key factor affecting overall efficiency. Keysight Technologies has proposed a series of practical methods and technical recommendations to optimize measurement speed for its power meter and power sensor systems. This article systematically sorts out nine optimization practices, aiming to help users significantly shorten test time and improve production efficiency while ensuring measurement accuracy.

1. Selection of Measurement Query Methods

The power meter supports three main query commands: MEASure?, READ? and  FETCH?. Understanding their differences is critical for speed optimization:

MEAS?: A composite command that automatically completes initialization, configuration and reading. It is easy to operate but slower, and will overwrite some instrument settings (such as trigger mode, number of averages).

READ?: Also a composite command, but it allows users to adjust settings such as the number of averages before execution. It is more flexible and faster than  MEAS? 

FETCH?: It only reads completed measurement results from the output buffer without performing initialization, so it has the fastest speed and is suitable for high-power signals. However, it may return inaccurate data if averaging is not completed during low-power measurement.

Recommendation:
Use  FETCH? in scenarios that require high power and high speed; use  READ? in applications that need to balance speed and accuracy; unless you pursue simple operation, try to avoid using  MEAS?.

2. Reasonable Configuration of Measurement Averaging

The averaging function is used to suppress noise and improve measurement accuracy, but it will increase measurement time. The power meter supports manual setting of averaging count (1–1024) or enabling automatic averaging mode.

For high-power signals (e.g. +10 dBm):the noise is low, the averaging count can be set to 1, and using  FETCH?  allows fast acquisition of accurate results.

Low-power signals (such as –40 dBm):the noise is relatively large, so a higher number of averages (e.g., 128) is required. If you use  FETCH? and the waiting time is insufficient, it may lead to inaccurate data.

Recommendation:
Dynamically select the averaging strategy based on signal power. For high-power signals, averaging can be turned off or reduced to improve speed; for low-power signals, ensure that averaging is completed, and use  READ? or MEAS? when necessary.

3. Flexible Application of Trigger Modes

The power meter supports three trigger modes:

Free Run (Free Run):Continuous measurement, suitable for continuous wave or repetitive signals.

Single Trigger (Single Shot):After one measurement is executed on each trigger, it returns to the idle state.

Continuous Trigger (Continuous Trigger):It depends on internal or external trigger signals and is suitable for time-gated measurement of pulsed, burst and other signals.

Speed Comparison (take N1912A + N1921A as an example):

Free Run mode FETCH? About 40 ms per point;

Continuous Trigger with FETCH? can be speeded up to 4.4 ms/point.

Recommendation:
Select the appropriate trigger mode based on the signal type. Use Free Run for continuous signals, use Continuous Trigger for burst signals, and combine with FETCH? to maximize the speed.

4. Power Sensor Measurement Speed Modes

The power sensor provides three measurement speed modes:

Normal Mode (Normal):Full functionality is supported, the speed is slow (approx. 50 ms/point).

Double Mode (Double):Speed is doubled (approx. 25 ms/point), and all functions remain supported.

Fast Mode (Fast):The fastest speed (up to 1000+ points/second), but functions such as averaging and limit comparison are disabled. It is suitable for scenarios that require extremely high speed and relatively loose accuracy requirements.

Recommendation:
Prioritize Fast mode when permitted; use Double mode if complete functions are required.

5. Buffer Mode Improves Batch Measurement Efficiency

When a large amount of data needs to be collected (e.g., 1000 readings), the traditional method requires multiple queries and has low efficiency. After enabling buffer mode, the instrument can cache multiple readings internally first and then upload them all at once, which significantly reduces communication overhead.

Setting command:TRIG:COUNT 50 (supports up to 50, in Fast mode).

Speed comparison:No Buffer is about 5.5 ms/point, Buffer 50 can be increased to 1.35 ms/point, and the speed is increased by about 4 times.

Recommendation:
Be sure to enable buffer mode in batch measurement scenarios and set the buffer size reasonably.

6. Unit Selection: Watt is better than dBm

The power meter performs internal calculations in linear units (watts). If you select dBm for the output, logarithmic conversion is required, which increases processing time.

Speed Comparison:In the Fast + Buffer 50 mode, using watt units can increase the speed by approximately 10% compared to dBm.

Applicable Scenarios:This improvement is applicable to thousands of high-volume measurements; the impact on a small number of measurement points is not significant.

Recommendation:
If allowed by the test system, set the output unit to watts, and then convert it to dBm in the software later.

7. Output Format: Binary is better than ASCII

The power meter supports two output formats:

ASCII: Approximately 17 bytes per reading.

Real (Binary): Only 9 bytes per reading.

The binary format can significantly reduce the bus data volume and improve transmission efficiency. In the Fast + Buffer 50 mode, combined with watt units, it enables high-speed measurement of  1 ms/point .

Recommendation:
Force use of  FORMAT REAL in high-speed acquisition systems.

VIII. Synchronous Control with *OPC?

When performing time-consuming operations such as sensor zeroing and calibration, query timeout or data error may occur without synchronous waiting. The *OPC? command returns a flag after all operations are completed, achieving precise synchronization between the instrument and the computer, and avoiding the waste caused by manually setting fixed waiting time.

Recommendation:
In automated test procedures, enable the  *OPC? synchronization mechanism for all operations that may cause delays.

IX. External Trigger Improves System Calibration Speed

In frequency sweep or power sweep calibration, the traditional method requires step-by-step setting of the signal source and waiting for power meter measurement, which is time-consuming. Through the external trigger function, hardware synchronization between the signal source and the power meter can be achieved, greatly improving scanning efficiency.

One-way external trigger:the signal source triggers the power meter, and a reasonable dwell time needs to be set.

Two-way external trigger:achieves handshake synchronization, no dwell time is required, and the speed is faster.

Actual Measurement Comparison (596 frequency point sweep):

Conventional Method:20.7 s

Unidirectional External Trigger:16 s

Bidirectional External Trigger:15.37 s

Recommendation:
It is recommended to prioritize the bidirectional external trigger architecture in calibration systems, and cooperate with the buffer and FETCH? to achieve high-speed sweeping.

Conclusion

Optimizing power measurement speed is a systematic project that requires comprehensive adjustments in multiple dimensions including query method, averaging strategy, trigger mode, sensor speed, buffer, unit format, synchronization control and external trigger. Through its flexible instruction set and hardware design, Keysight provides users with a wealth of tuning methods. In practical applications, users should first clarify the characteristics of the test signal and accuracy requirements, then make targeted configuration in combination with the above suggestions, so as to maximize test efficiency without sacrificing measurement quality.

Appendix: Related Keysight Literature

• 《Choosing the Right Power Meter and Sensor》– 5968-7150E

• 《4 Steps for Making Better Power Measurements》– 5965-8167E

• 《Maximizing Measurement Speed Using P-Series Power Meters》– 5989-7678EN

• This article is compiled based on Keysight application note 《Practices to Optimize Power Meter/Sensor Measurement Speed and Shorten Test Times》 (Document No. 5990-8471EN, December 2017), and covers its nine optimization practices, which applies to power meter and sensor systems including 8480, N8480, E Series, P Series, U2000 Series, etc.

Contact Us>>