Maximizing Accuracy in Automated Test Systems

Welcome to the Designing Next Generation Test Systems Developers Guide. This guide is collection of white papers designed to help you develop test systems that lower your cost, increase your test throughput, and can scale with future requirements. This white paper provides strategies for maximizing system accuracy. To download the complete developers guide (120 pages), visit ni.com/automatedtest.

Introduction

When designing automated test systems, maximizing accuracy is usually a key concern. Determining how to maximize accuracy can be difficult. Most test engineers turn to the data sheets for the instruments they are evaluating with the hope that these documents provide all of the answers. However, other factors are just as important in maximizing accuracy in your automated test systems.

This paper provides you with five strategies that you can follow to maximize the accuracy of your automated test systems. The five strategies are the following:

Strategy 1 - Understand instrument specifications

When evaluating the accuracy of an instrument, the data sheet is a valuable resource. However, it is important to understand that different instrument vendors oftentimes specify measurement accuracy using either different terminology or similar terminology with different meanings. Thus, it is important to have a clear understanding of all the parameters involved in defining the characteristics of an instrument. Often the terms resolution, precision, and accuracy are used interchangeably, but they actually indicate very different entities. Although common sense indicates that a 6½-digit digital multimeter (DMM) must be accurate to the 6½-digit level, this may not be the case. The number of digits can simply relate to the number of figures that the meter can display and not to the minimum distinguishable change in the input. You need to verify that the instrument sensitivity and effective resolution are enough to guarantee that the instrument will give you the measurement resolution you need.

For example, a 6½-digit DMM can represent a given range with 1,999,999 counts or units. But if the instrument has a noise value of 20 counts peak-to-peak, then the minimum distinguishable change must be at least 0.52 x 20 counts because resolution, which is the smallest amount of input signal change that an instrument can detect reliably, is equal to counts or volts of Gaussian noise multiplied by 0.52. Thus, the effective number of digits (ENOD) for this particular 6½-digit DMM is:

As you can see, the number of digits listed in the data sheet for a DMM is an important piece of information, but it should not be considered the ultimate or only parameter to take into account. By knowing the measurement accuracy and resolution requirements for your automated test systems, you can compute the total error budget of the instruments you are considering and verify that they satisfy your needs. Moreover, do not hesitate to ask vendors to clarify the meaning of the specifications in data sheets because not knowing the true performance of your instruments could lead to costly errors.

To better understand the specifications of the instruments that you are evaluating, read the Understanding Instrument Specifications -- How to Make Sense out of the Jargon application note.

Strategy 2 - Consider calibration requirements

Regardless of the accuracy of the instruments you select for your automated test systems, it is important to realize that the accuracies of the electronic components used in all instruments drift over time. The effects of time in service as well as environmental conditions add to this drift. As time progresses, changes in component values cause greater uncertainty in your measurements. To resolve this issue, your instruments must be calibrated at regular intervals.

External calibration is the comparison of instrument performance to a standard of known accuracy. The result of an external calibration may be documentation showing the deviation of a measurement from the known standard, but most often it also includes adjusting the measurement capability of the instrument to ensure that its measurement accuracy is within vendor-provided limits. Many vendors provide graduated accuracy tables (see Figure 1), which offer a clear uncertainty profile from the last external calibration of an instrument.

Figure 1. Graduated accuracy tables provide you with a clear uncertainty profile from the last external calibration of an instrument.

To have an instrument externally calibrated, you can send it back to the vendor, or you can send it to a calibration or metrology laboratory. Additionally, you may have external calibration capabilities at your facility. Regardless of how you will externally calibrate your instruments, it is important to realize that external calibration intervals for a particular type of instrument are not always the same for different vendors. One vendor may have an external calibration interval of one year for a function generator, while another vendor's external calibration interval for a function generator with equivalent or better accuracy specifications may be two years. Choosing the second instrument reduces the cost of maintaining your automated test system. When selecting instruments, be careful to consider the external calibration intervals (see Figure 2).

[+] Enlarge Image
Figure 2. Not all instruments of the same type have equivalent external calibration intervals. When selecting instruments, consider the external calibration intervals.

In addition to external calibration, instruments for some vendors include self-calibration functionality. Instruments that offer self-calibration include hardware resources such as precision voltage references so you can quickly calibrate the instrument without removing it from the test system or connecting it to external calibration hardware. Self-calibration is not a replacement for external calibration, but it does provide a method of improving instrument measurement accuracy between external calibration intervals.

Maintaining properly calibrated instruments reduced measurement errors, improves consistency between measurements, and provides you assurance that you are making accurate measurements. Visit the Understanding Calibration tutorial to find additional calibration resources.

Strategy 3 - Be aware of the operating environment

Not all instruments have the same environmental specifications. The storage and operating temperature and relative humidity specifications can vary between vendors. Your automated test systems may be in an office-type environment where temperature and humidity are tightly controller, but they may be in a factory or other industrial setting. At the very least, it is important to be aware of the environmental specifications for your instruments and understand how they can affect measurement accuracy.

As an example, traditionally, DMMs are external calibrated at a particular temperature, and this calibration is characterized and specified over a limited temperature range, usually ±5 ºC (or even ±1 ºC in some cases). Thus, whenever the DMM is used outside of this temperature range, its accuracy specifications must be derated by a temperature coefficient, usually on the order of 10% of the accuracy specification per ºC. At 10 ºC outside of the specified range, you may have twice the specified measurement error, which can be a serious concern when absolute accuracy is important.

Keeping the environmental temperature of a precision instrument within ±5 ºC can be challenging in a production environment, or in an automated test system composed of multiple instruments. Instruments in a system are subject to temperature rise caused by inherent compromises in air circulation and other factors. If the changes in temperature exceed these limits, and tight specifications are required, then recalibration is also required at the new temperature. Take, for example, the 10 VDC range on traditional DMMs. A DMM may have an accuracy of:

In this specification, if you apply 5 V to the input, the error is:

This is the traditional method of specifying accuracy. If the ambient temperature is outside of the 18 to 28 ºC range, the user needs to derate the accuracy using the temperature coefficient (tempco). With the traditional method, the only way to achieve the specified accuracy outside of the 18 to 28 ºC range is to fully recalibrate the system at the desired temperature. Of course, this process is often impractical and expensive. In the same example, if the DMM ambient temperature is 50 ºC, perhaps due to stacking of many instruments in a rack with limited airflow, and the tempco is specified as:

tempco = (5 ppm of reading +1 ppm of range)/ºC

Then the additional error is:

This error at 50 ºC ambient temperature is nearly 5X greater than the specified 1-year accuracy. To eliminate errors caused by the operating environments of your automated test systems, instruments from certain vendors include features such as self-calibration (as discussed previously). This feature results in highly accurate, ultrastable instruments at any operating temperature, even well outside of the traditional 18 to 28 ºC range. To revisit the previous DMM example, the additional error introduced by temperature coefficient using a National Instruments PXI-4070 DMM with self-calibration would be fully covered in its 90-day and 2-year specifications and would be:

tempco with self-cal: < (0.3 ppm of reading + 0.3 ppm of range)/ºC
(already accounted for in the specification)

Figure 3. To eliminate errors caused by the operating environment of your automated test systems, instruments are available with self-calibration.

More information regarding the effect of your automated test systems' operating environments on accuracy is available in the Understanding the FlexDMM Architecture white paper.

Strategy 4 - Use proper fixturing

Connecting your automated test system to the device under test (DUT) may be as simple as cabling from the instruments to a breakout box or screw terminals and connecting to the DUT for systems with less than 50 test points or only a few instruments. For larger systems with hundreds of test points, multiple instruments, reconfigurable system requirements, and/or frequent connects/disconnects, an approach such as a mass interconnect system is usually required.

In either situation, it is important that your fixturing be composed of cabling designed to maximize measurement accuracy. Low-quality cabling can have a significant negative impact on the accuracy of your automated test systems. Cabling that is designed to maximize measurement accuracy exhibits characteristics such as low leakage and low thermal emf (see Figure 4).

[+] Enlarge Image
Figure 4. Low-leakage, low-thermal-emf cables help to maximize accuracy in your automated test systems.

As discussed previously, a mass interconnect system is a mechanical fixture designed to facilitate the connection of a large number of signals either coming from or going to a DUT. This usually entails some mechanical enclosure through which all signals are routed from instruments (typically in a rack) to the DUT, making it easy to quickly change out DUTs (see Figure 5). A mass interconnect system also protects the cable connections on the front of the instruments from repeated connect/disconnect cycles. Instrument cable connections that have experienced excessive connect/disconnect cycles are subject to wear and damage, which can degrade measurement accuracy.

[+] Enlarge Image
Figure 5. A mass interconnect system is a mechanical fixture designed to facilitate the connection of a large number of signals either coming from or going to a DUT (image courtesy of Virginia Panel Corporation).

Strategy 5 - Take advantage of synchronization

Another aspect of accuracy in automated test systems is phase accuracy - the degree to which the timing of signals being acquired and generated are precisely correlated. Synchronization, specifically hardware synchronization, minimizes the skew between instruments, which maximizes correlation. For example, if your automated test system includes two oscilloscopes that simultaneously acquire data from the DUT, unless the oscilloscopes use synchronized start triggers and sample clocks, it is almost impossible to compare the phases of the acquired signals.

Another example of using hardware synchronization to maximize phase accuracy in your automated test system is phase-locked looping the sample clocks for an arbitrary waveform generator (ARB) and oscilloscope to the same stable reference clock in a stimulus-response test. If precise hardware synchronization is not employed in a stimulus-response test, a fractional number of cycles of the analog wave being generated by the ARB will be acquired by the oscilloscope. When the acquired sine wave is analyzed using an FFT, spectral leakage is present as "skirts" in the spectrum, as is illustrated by the white trace on the graph in Figure 6. The use of phase-locked looping synchronization eliminates a fractional number of cycles from being acquired by the oscilloscope. This, in turn, eliminates the spectral leakage, as is illustrated by the red trace in Figure 6.

Figure 6. Synchronization improves phase accuracy in your automated test systems. For example, this graph illustrates hardware synchronization eliminating spectral leakage in a stimulus-response test.

Test platforms provide significantly different levels of hardware synchronization. Some offer limited functionality, and others, such as PXI, offer very sophisticated hardware synchronization resources. PXI has a high-performance timing and synchronization bus built into the backplane, which eliminates the need to use external cabling between instruments. Using the integrated PXI timing and synchronization bus, the instruments in your automated test systems can be synchronized to the sub-nanosecond level. More information regarding the synchronization of PXI instruments is available in the National Instruments T-Clock Technology for Timing and Synchronization of Modular Instruments tutorial.

Conclusion - Maximizing Accuracy in Automated Test Systems

Most likely, maximizing accuracy is important to you when designing automated test systems. The five steps discussed in this white paper - understand instrument specifications, consider calibration requirements, be aware of the operating environment, use proper fixturing, and take advantage of synchronization - provide you with a roadmap for maximizing accuracy in your automated test systems.

Relevant NI Products and Whitepapers

National Instruments, a leader in automated test, is committed to providing the hardware and software products engineers need to create these next generation test systems.

Software:

• NI TestStand Test Management Framework
• LabVIEW Graphical Programming Environment
• Signal Express Interactive Measurement Software
  

Hardware:

• Modular Instruments (Oscilloscopes, Multimeters, RF, Switching, and more)
• Multi-function Data Acquisition
• PXI System Components (Chassis and Controllers)
• Instrument Control (GPIB, USB, and LAN)
  

White Papers:
NI offers a Designing Next Generation Test Systems Developers Guide. This guide is collection of whitepapers designed to help you develop test systems that lower your cost, increase your test throughput, and can scale with future requirements. To download the complete developers guide (120 pages) , visit ni.com/automatedtest.

2 ratings | 5.00 out of 5

Read in English Spanish | Print | PDF

Reader Comments | Submit a comment »