1. Obtain Reliable RF Measurements
RF measurements are often straightforward in theory, but more difficult when put into practice. You can readily obtain core RF measurements, such as power, frequency, and noise, from a wide offering of contemporary RF instruments. However, getting a result and getting a correct result are often worlds apart. By implementing best practices as part of your overall RF measurement regiment, you can be confident in a reliable, accurate, and repeatable outcome.
2. Understand the Terminology
Terms such as accuracy, repeatability, resolution, and uncertainty are often intermingled or misused in an assortment of RF applications, adding to the confusion and diminishing confidence in the measurement. Understanding a few key terms and their proper context is essential when making RF measurements.
For example, digital displays on instruments are easier to use than an analog gauge when trying to discern the proper reading on an analog scale. However, even if the digital display shows three digits to the right of the decimal for a measurement, you can infer nothing about the resolution and accuracy of the instrument and its measurement.
Because a display shows thousandths of a decibel of power or fractions of a hertz in frequency does not imply that the instrument is capable of measuring such minute changes. Often the number of displayed digits well exceeds the capacity of the instrument to measure at such levels. To fully understand the capabilities of an RF instrument, always refer to the specifications documents or data sheets.
Consistent definitions can reduce confusion in your measurements. The following are a few key terms that you will frequently see in use:
- Resolution – The smallest change an instrument can reliably detect
- Repeatability – The ability to make the same measurement numerous times under the same conditions with the same result
- Uncertainty – A quantization of the lack of knowledge of the exact value being measured
- Accuracy – The ability of an instrument to measure the actual/absolute value of a parameter within some stated error specification
Uncertainty is always present, and an assessment of the sources of errors can help determine measurement uncertainty. In addition to those mentioned above, there are related terms that come into play when tracing performance from the specifications document back to the National Institute of Standards and Technology (NIST) or other standards bodies. Traceability is required to ensure all measurement devices are defined with a common, absolute standard. The term “specification” implies warranted performance that is made with test equipment whose calibrations are traceable to NIST. “Typical” often implies that performance is 100 percent tested, but measurement uncertainties are not included. “Nominal” performance is usually supplemental information and is not generally measured on every instrument.
Accuracy is the ability of an instrument to measure the absolute value of a parameter within some stated range of error. In other words, X plus or minus Y. A measurement of 34 is not meaningful without some error limits (and units). Likewise, an error specification of five is not useful. Even an error specification of five percent is not much help. Is that plus or minus five percent, or plus three percent and minus two percent? To be correct, accuracy should be stated, for example, as 34 volts ±1 volt, 34 volts ±1%, or 34 volts +2/-1 volt.
Take the time to understand RF measurement terms and become familiar with their meaning. The more accurately you communicate about the measurement, the better understanding and confidence in the result.
3. Know Your DUT
The behavior of the device under test (DUT) can significantly impact RF measurements. For example, temperature can impinge on stability, and thus, repeatability. Many RF devices and RF instruments do not internally compensate for temperature variations. As a result, they must be at a stable temperature to minimize measurement errors from drift. The immediate surroundings (for example, air conditioning cycling on and off, covers and panels being removed or added, being outdoors or indoors, and location near heat sources) can contribute considerably. Be aware of the proper warm-up times, DUT cooling requirements, and ambient environment to ensure temperature stability.
In active devices, excessive power can lead to heating. For instance, in testing a high-powered amplifier, the DUT itself can be temperature stable, but what about the downstream components? See if there are switches and attenuators being heated by the amplifier output. Look for errant signals, such as harmonics, that are being generated by the amplifier. Power supply lines are susceptible to environmental noise that can be directly superimposed on the output. And, measuring the linear parameters of an amplifier (gain and phase) while the amplifier is in compression is frustrating to discover after the fact. All impact the accuracy of the RF measurement. Understanding the DUT, its operation, and its impact on RF measurement parameters before the device is measured, leads to meaningful results.
4. Identify the Areas of Uncertainty
Matching RF test equipment data sheet specifications to the measurement requirements of the DUT is not sufficient. This is especially evident in RF measurements where higher frequencies and mismatches, among other things, amplify uncertainties. Errors introduced at each step in the measurement path skew the overall result. When an incorrect measurement occurs, you should first suspect that the measurement is in error before suspecting a faulty DUT.
Understand the key operational specifications of the instruments and apparatus in the measurement path, including the DUT. Among other specifications, know the match, the frequency response, noise figure, and power. Also know the tolerances on these parameters. Remember factors such as:
- RF switch repeatability, aging, and power handling
- The directivity of couplers, the phase stability of cables, and the insertion loss and return loss of adapters
- The impedance quality of circuit board traces, device sockets, and transitions on and off the circuit board
- Electromagnetic interference (EMI) radiated and coupled into the measurement
Items not normally considered such as cooling, harmonics, spurs, and other nonlinear behavior can also add to errors in the measurement. Look at the overall test setup and determine the error contribution of each piece to obtain a realistic assessment of the expected measurement uncertainty. Identify the sources of error and their impact on accuracy, repeatability, and uncertainty. This leads to a better outcome and gives you the ability to more efficiently and meaningfully allocate budget and resources.
5. Pay Attention to All Connections and Components
The cost to develop, design, test, and deliver a product to market is a considerable investment. The success or failure of a company can depend on a product’s performance. There is significant outlay for high-performance RF test equipment, because being able to demonstrate that a new product meets or exceeds a key specification grabs market share. Furthermore it represents a competitive advantage, and a major increase in a company’s revenue stream.
However, a high-performance, expensive, well-calibrated test system and an equally remarkable DUT are not enough. Give equal consideration to the quality and repeatability of the connections and components attaching the test system to the DUT. Perhaps one or two tenths of a dB improvement in a key spec is a competitive advantage. A source and load match (VSWR) of 1:1.5 is good by most standards. But that level of match can introduce an uncorrected mismatch uncertainty of (approximately) ±0.35dB. A key spec of 0.2dB is impossible to prove when it is cloaked by that much uncertainty.
Neglecting items such as cables, switches, attenuators, connectors, sockets, adaptors, and accessories can be detrimental to the entire measurement. Start with the accuracy desired, and then choose components that are appropriate for the measurement. A good rule of thumb is to have measurement system performance at least 10 times as good as the DUT parameter that you are testing.
With a high-quality signal path in place, the next step is to employ good measurement practices. Ensure that you properly clean and store cables, connectors, and adaptors. Even the best cables and adaptors wear out. Discard them. They are consumables in the test process. Take steps to minimize the use of adaptors. Make sure torque wrenches and connector gages are being used routinely and that hot-switching is minimized. Remember proper electro-static discharge (ESD) practices. Even a cascade of the highest quality components between the test system and the DUT can introduce error into the measurement.
6. Select the Right Tool for the Job
What parameters need to be measured and to what level of accuracy largely determines the selection of RF equipment required to test a DUT. Buying the best is a safe tactic, but it wastes budget resources that you could apply to other parts of the measurement. If RF power is the only quantity to be measured, a RF power meter may be a more ideal choice than a vector signal analyzer.
Scalar instruments measure only magnitude (amplitude), while vector instruments measure magnitude and phase. Even if phase is not a required measurement, consider that often vector instruments can offer better error correction because phase information is required to quantify unwanted reflections in the system.
Equating price to performance is not the best rule to follow when purchasing RF equipment. A high-quality swept-tuned spectrum analyzer can take a lot of your budget. Although they are excellent instruments for their intended purpose, with accuracies typically in the range of ± 1 dB or worse, they fail to meet the needs for measuring absolute RF power . In the same way, it is difficult to measure a -155 dBm/Hz noise floor on the DUT if the instrument in use has a noise floor of -140 dBm/Hz.
Consider the right tool for the right job. Overpaying for accuracy or measurements that are not required is a waste of money and resources. Further, it limits the funds for other areas, such as cables and switches that may prove more beneficial to the quality of the measurement.
7. Develop a Process
Once you have identified and implemented best practices, instill them into a routine or process that can be clearly understood and communicated throughout the organization. The consequence is better repeatability and consistency of the RF measurement results.
For example, a frequently asked question concerning processes is “How often should I calibrate?” Many RF instruments tend to be sensitive to changes in the environment. If this is the case, calibrating more often may be necessary. High-accuracy requirements also may drive the need for more frequent calibration. Whatever the case, apply due diligence to understanding the requirements for RF calibration and capture it as a disciplined process.
Processes throughout the design, verification, test, and manufacturing phases can affect RF measurement performance. Think about what operational parameters need to be verified and which ones should be tested in manufacturing. Also consider upstream and downstream processes (for example, rework, solder, assembly, and shielding) that impact the accuracy, repeatability, and uncertainty of RF measurements.
Processes are important to capture and implement good RF practices. This facilitates institutional learning and standardization of sound methodology. Consistency in following an established process throughout the product life cycle can have a major impact on RF parameters and their proper measurement.
8. Improve the Quality of RF Measurements
Making RF measurements can be easy; however, it is a bit more challenging to do them well. Using sound practices and capturing those in a process can improve the quality of RF measurements. There are many ways to identify and implement best practices. Improving RF measurements is a continuous process to gain understanding and put that knowledge into practice. The steps outlined in this paper serve to build a foundation to build from to improve RF measurement skills and get the most from your RF test instruments.