Self-Calibration for RF Hardware

Publish Date: Aug 29, 2016 | 2 Ratings | 5.00 out of 5 | Print

Overview

The term “calibration” can mean a number of different things based on the system and context in which it is used. In the case of test and measurement equipment, calibration is the process of verifying a device against a measurement standard under controlled and specific conditions. While calibration is a high level process, the amount of verification and adjustment that is performed can differ based on the design of the device. The main focus of this article will be on self-calibration, however a high-level overview of calibration terms will be discussed to better describe the purpose and importance of self-calibration and how it differs from calibration performed by a metrology lab. In this article, external calibration refers to the calibration of a device by a metrology lab while self-calibration refers to the calibration of a module using its own on-board reference standard that has previously been calibrated to a traceable standard via an external calibration.

Table of Contents

  1. Verification and Adjustment
  2. Accuracy Drift, Temperature Variation, Calibration Interval, and Self-Calibration
  3. External Calibration vs. Self-Calibration
  4. How to Perform Self-Calibration
  5. Conclusion
  6. Related Links

1. Verification and Adjustment

In the context of calibration, verification is the process of measuring a device to check that it is operating within its published specifications. Typically, when a metrology lab receives a device for calibration, it is first verified and the measurements are recorded. These are referred to as the “as-found” measurements. Based on these results, if the device requires adjustment, an adjustment will be performed followed by another verification. These post adjustment data points are referred to as the “as-left” measurements.

Adjustment is the process by which the module’s calibration constants are updated in device memory for more accurate performance. Often, the term calibration comes with an assumption that adjustment will be performed, but this is not always the case. In such cases, the “as-found” data will be the same as the “as-left” data.

Figure 1. As-Is and As-Found calibration data flow chart

 

Back to Top

2. Accuracy Drift, Temperature Variation, Calibration Interval, and Self-Calibration

The accuracy of a device will naturally degrade over time with normal use due to the unavoidable drift of its board-level components. Devices that require calibration usually have a recommended calibration interval. The calibration interval is a recommendation from the manufacturer based on the module’s components and how their accuracy is expected to drift over time. Some applications may require a shorter calibration interval due to tighter measurement tolerance requirements. The need for a shorter calibration interval is evaluated by the customer.

Along with time, temperature variation plays another large role in the accuracy of a device.  Therefore, external calibration and adjustment are performed at a specific temperature that is recorded and stored on the module’s onboard memory. We need a way to adjust for the differences between the external calibration temperature and the actual operating temperature. On top of these benefits, depending on the module, self-calibration can improve behaviors such as image rejection that are not covered with external calibration.

Self-calibration is a much shorter process than an external calibration. As well, it can be performed at any time. For this reason, self-calibration is a great troubleshooting step for NI RF Instruments. When measurements seem incorrect or unexpected, a self-calibration can be performed with little effort. While self-calibration does not explicitly point to a component or problem area, performing self-calibration ensures that the device is as accurate as possible. This can be useful for troubleshooting issues such as unexpected spurious performance or measurement inaccuracies.

 

Back to Top

3. External Calibration vs. Self-Calibration

As mentioned above, external calibration is the process by which a module’s performance is verified against known references. Calibration constants as well as the date, time, and temperature of the latest external calibration are stored in the module's on-board memory. In addition to better performance, some metrology labs offer external calibration with traceability and accreditation if the application or industry requires compliance to a calibration standard. For more information about NI’s calibration services, or to get the calibration certificate for NI modules, see the links below. To check the recommended external calibration interval for a module, see the module’s specification document.

NI Calibration Services
NI Instrument Certificate Generator

In contrast to external calibration, self-calibration is performed more often and uses an accurate onboard reference to compensate for board level temperature variations. During each self-calibration correction values are stored in the onboard memory for self-calibration, however the external calibration data remains unchanged.

 

Back to Top

4. How to Perform Self-Calibration

Self-Calibration of NI Vector Signal Analyzers and Vector Signal Generators

The easiest way to perform a self-calibration on NI Vector Signal Analyzers and Generators is to use the device’s Soft Front Panel (RFSA or RFSG Soft Front Panels) as seen in Figure 2 below. The Self Calibration button can be found in both soft front panels by navigating to Device/System >> Calibration. In the case of the PXIe-5644R/5645R/5646R, the Self Calibration button in the Soft Front Panels performs a partial self-calibration. Partial calibration is discussed in more detail later in this article.

Figure 2. Self-calibrating through the NI-RFSA Soft Front Panel

Additionally, for Vector Signal Analyzers and Vector Signal Generators (excluding the PXIe-5644R/5645R/5646R) self-calibration can be performed through the NI-RFSA and NI-RFSG LabVIEW API’s respectively using the niRFSA Self Cal VI or niRFSG Self Cal VI. This can also be performed in C using the niRFSA_SelfCalibrate and niRFSG_SelfCalibrate functions.

Note: Most of the LabVIEW calibration functions are also available in C. There is a link to the C Function Reference for calibration functions in the Related Links section below.

Figure 3. niRFSA Self Cal VI is used to perform Self-calibration

For multi-module analyzers and generators such as the PXIe-5668R or PXIe-5673E, module specific self-calibration can be performed in NI Measurement and Automation Explorer (NI MAX). Module specific self-calibrations can also be done through the driver; however, a full self-calibration of the system may still be required. The niRFSA Is Self Cal Valid VI can be used to determine if there is valid self-calibration data on the device. It also returns an array of valid steps that can be directly provided to the steps to omit input of the niRFSA Self Cal VI. Omitting steps is discussed in the next section.

Figure 4. niRFSA Is Self Cal Valid VI is used to determine if self-calibration is needed

Omitting Self-Calibration Steps on Vector Signal Analyzers

Omitting steps allows the user to decrease the time of self-calibration by targeting specific sub-systems of the device. Each Vector Signal Analyzer device is different in the steps it performs during self-calibration. For each module, a list of specific steps that can be omitted can be found in the device specific section of the NI-RFSA Help documentation. For instance, the PXIe-5668R self-calibrates the following parameters:

  • Preselector alignment
  • IF flatness
  • Gain reference
  • LO self-calibration

The niRFSA Self Cal VI provides an array input for omitting steps. As mentioned above, not all step options apply to every analyzer. If any of the steps specified to omit are not supported by the analyzer, the driver will return an error.

Figure 5. Omitting Steps on the RFSA Self-Calibration Process

It should be noted that omitting steps can invalidate self-calibration data which in turn causes associated specifications to not be guaranteed. Conversely, omitting valid steps does not invalidate any specifications.

Self-Calibration of NI Vector Signal Transceivers

The Vector Signal Transceivers (VST’s) can be self-calibrated using the NI-RFSA Soft Front Panel and NI-RFSG Soft Front Panels. Additionally, the niVST Self Calibrate VI in the VST Instrument Design Libraries (IDL) API can be used.

In addition to the options described, there is a specific utility included in the NI-RFSA, NI-RFSG, and the VST IDL driver installations that can be used to calibrate the NI PXIe-5644R/5645R/5646R devices. The utility can be found in Windows by navigating to Start>>All Programs>>National Instruments>>Vector Signal Transceivers>>VST Self Calibrate.

Partial Self-Calibrate Range

Additionally, NI Vector Signal Transceivers feature the ability to perform a partial self-calibration. Partial self-calibration allows the user to perform self-calibration on only the necessary frequency and reference level ranges required for the application. It is also possible to choose which types of corrections to apply by omitting steps that are not required for the application. This allows the user to tailor the self-calibration to their specific application.  Data from a partial self-calibration is not stored on the VST’s FLASH memory as it is for a normal self-calibration. Instead, it is stored in the kernel device memory and consequently this data is lost on reboot.

Partial self-calibration takes a fraction of the time of a full self-calibration. This minimizes the time and the temperature drift between the partial self-calibration and measurement which in turn improves measurement accuracy.
There are two ways to perform a partial self-calibration. Since the VST has both an analyzer and a generator, niRFSA RF In Self-Calibrate Range VI and niRFSG RF Out Self-Calibrate Range VI can be used to perform partial self-calibration on each respectively. Alternatively, niVST RF In Self Calibrate Range VI and niVST RF Out Self Calibrate Range VI can be used when working from the open FPGA sample projects.

Note: You can also omit steps as part of a partial calibration as described previously.

 

Back to Top

5. Conclusion

Key Concepts:

  • Self-calibration compensates for inaccuracies caused by component drift time and temperature variations and is required to meet some analyzer and generator specifications. It also provides image rejection and other corrections that external calibration does not depending on the module.
    • Can be performed in the appropriate device Soft Front Panel or through the supported driver API. In the case of the PXIe-5644R/5645R/5646R, the self-calibration utility can also be used.
  • Self-calibrations are performed in steps that vary depending on the device.
    • On some NI Vector Signal Analyzers, steps can be omitted to speed up self-calibration if needed.
    • The niRFSA Is Self Cal Valid VI returns an array of steps that contain valid self-calibration data which can be passed to the steps to omit input of the niRFSA Self Cal VI.
  • In the case of the PXIe-5644R/5645R/5646R, a partial self-calibration can be performed over a smaller frequency and reference level range. This allows the user to select a smaller range of frequencies and reference levels specific to the current application to reduce self-calibration time.
  • Self-calibration is a useful troubleshooting resource for NI RF Instruments and should be the first step when diagnosing unexpected measurements.

As discussed in this article, self-calibration is a necessary and useful tool for performing accurate measurements with NI Vector Signal Analyzers, Generators, and Transceivers. For more information, refer to the Related Links section below and contact NI Support at support@ni.com for technical assistance.

 

Back to Top

6. Related Links

Back to Top

Bookmark & Share


Ratings

Rate this document

Answered Your Question?
Yes No

Submit