15A.2 Measurement Configuration Considerations

Back

This section describes how the analog input (AI), analog output (AO), timing, and triggering

configuration affect your measurements.

15A.2.1 Input Signal Considerations

One must consider the following configuration before acquiring dynamic signals and performing shock

and vibration measurements.

15A.2.1.1 Input Pseudodifferential and Differential Configuration

DSA devices such as the National Instruments PXI-4461 supports two terminal configurations for AI,

differential and pseudodifferential. The term pseudodifferential refers to the fact that there is a 50 W

resistance between the outer BNC shell and chassis ground. One can configure the NI PXI-4461 input

channels on a per-channel basis. Therefore, you can have one channel configured for differential mode

and the other channel configured for pseudodifferential mode. Configure the channels based on how the

signal source or DUT is referenced. Refer to Table 15A.1 to determine how to configure the channel based

on the source reference.

If the signal source is floating, use the pseudodifferential channel configuration. A floating signal

source does not connect to the building ground system. Instead, the signal source has an isolated groundreference

point. Some examples of floating signal sources are outputs of transformers without grounded

center taps, battery-powered devices, nongrounded accelerometers, and most instrumentation

TABLE 15A.1 Input Channel Configuration

Source Reference Channel Configuration

Floating, ground referenced Pseudodifferential

Ground referenced Differential

Virtual Instrumentation for Data Acquisition, Analysis, and Presentation 15-75

© 2005 by Taylor & Francis Group, LLC

microphones. An instrument or device that has an isolated output is considered to be a floating signal

source. It is important to provide a ground reference for a floating signal. If no ground-reference point is

provided — for example, in selecting differential mode with a floating microphone — the microphone

outputs can drift outside the NI PXI-4461 common-mode range.

If the signal source is ground-referenced, use either the differential or pseudodifferential channel

configurations. A ground-referenced signal source connects in some way to the building system ground.

Therefore, it is already connected to a ground-reference point with respect to the NI PXI-4461, assuming

the PXI or CompactPCI chassis and controller are plugged into the same power system. Nonisolated

outputs of instruments and devices that plug into the building power system fall into this category.

Provide only one ground-reference point for each channel by properly selecting differential or

pseudodifferential configuration. If you provide two ground-reference points — for example, if you select

pseudodifferential mode with a grounded accelerometer — the difference in ground potential results in

currents in the ground system that can cause measurement errors. The 50 W resistor on the signal ground

is usually sufficient to reduce this current to negligible levels, but results can vary depending on the

system setup.

The NI PXI-4461 is automatically configured for differential mode when powered on or when power is

removed from the device. This configuration protects the 50 W resistor on the signal ground.

15A.2.1.2 Gain

DSA devices such as the NI PXI-4461 often offer variable gain settings for each AI channel. Each gain

setting corresponds to a particular AI range, and each range is centered on 0 V. The gain settings are

specified in decibels (dB), where the 0 dB reference is the default input range of ^ 10 V.

Positive gain values amplify the signal before the analog-to-digital converter (ADC) digitizes it. This

signal amplification reduces the range of the measurement. However, amplifying the signal before

digitization allows better resolution by strengthening weak signal components before they reach the ADC.

Conversely, negative gains attenuate the signal before they reach the ADC. This attenuation increases the

effective measurement range though it sacrifices some resolution for weak signal components.

In general, select the voltage range that provides the greatest dynamic range and the least distortion.

For example, consider an accelerometer with a 100 mV/g sensitivity rating with an absolute maximum

output voltage of 5 Vpk. In this case, the ^ 10 Vpk is appropriate, corresponding to 0 dB gain. However,

the ^ 3.16 Vpk setting maximizes the dynamic range if one knows the stimulus is limited, for example, to

20 g or 2 Vpk.

15A.2.1.3 Input Coupling

One can configure each AI channel for either alternating current (AC) or direct current (DC) coupling.

If you select DC coupling, any DC offset present in the source signal is passed to the ADC.

The DC-coupling configuration is usually best if the signal source has only small amounts of offset

voltage or if the DC content of the acquired signal is important.

If the source has a significant amount of unwanted offset, select AC coupling to take full advantage of

the input dynamic range.

15A.2.1.4 Integrated Electronic Piezoelectric Excitation

If you attach an Integrated Electronic Piezoelectric Excitation (IEPE) accelerometer or microphone to an

AI channel that requires excitation from your DSA device, you must enable the IEPE excitation circuitry

for that channel to generate the required current.

One can independently configure IEPE signal conditioning on a per-channel basis. It is common to set

the excitation from 0 to 20 mA with 20 mA resolution.

A DC voltage offset is generated equal to the product of the excitation current and sensor impedance

when IEPE signal conditioning is enabled. To remove the unwanted offset, enable AC coupling. Using DC

coupling with IEPE excitation enabled is appropriate only if the offset does not exceed the voltage range

of the channel.

15-76 Vibration and Shock Handbook

© 2005 by Taylor & Francis Group, LLC

15A.2.1.5 Nyquist Frequency and Bandwidth

Further discussion of DSA measurement configuration requires a brief introduction of two concepts:

* Nyquist frequency

* Nyquist bandwidth

Any sampling system, such as an ADC, is limited in the bandwidth of the signals it can represent.

Specifically, a sampling rate of fs can only represent signals with a maximum frequency of fs=2: This

maximum frequency is known as the Nyquist frequency. The bandwidth from 0 Hz to the Nyquist

frequency is the Nyquist bandwidth.

15A.2.1.6 Analog-to-Digital Conversion

ADC is discussed in Chapter 16. DSA devices commonly use a conversion method known as delta – sigma

modulation. If the data rate is 51.2 kS/sec, each ADC actually samples its input signal at 6.5536 MS/sec,

128 times the data rate, and produces one-bit samples that are applied to the digital filter. This filter then

expands the data to 24 bits, rejects signal components greater than the Nyquist frequency of 25.6 kHz,

and digitally resamples the data at 51.2 kS/sec.

The one-bit, 6.5536 MS/sec data stream from the ADC contains all of the information necessary to

produce 24-bit samples at 51.2 kS/sec. The delta – sigma ADC achieves this conversion from high speed to

high resolution by adding a large amount of random noise to the signal so that the resulting quantization

noise, although large, is restricted to frequencies above the Nyquist frequency, 25.6 kHz in this case. This

noise is not correlated with the input signal and is almost completely rejected by the digital filter.

The resulting output of the filter is a band-limited signal with a large dynamic range. One of the

advantages of a delta – sigma ADC is that it uses a one-bit digital-to-analog converter (DAC) as an

internal reference. As a result, the delta – sigma ADC is free from the kind of differential nonlinearity

(DNL) and associated noise that is inherent in most high-resolution ADCs.

15A.2.1.7 Antialias Filters

A digitizer may sample signals containing frequency components above the Nyquist limit. The process by

which the digitizer modulates out-of-band components, returning them to the Nyquist bandwidth, is

known as aliasing. The greatest danger of aliasing is that there is no straightforward way to know whether

it has happened by looking at the ADC output. If an input signal contains several frequency components

or harmonics, some of these components maybe represented correctly while others are aliased.

Low-pass filtering to eliminate components above the Nyquist frequency, either before or during the

digitization process, can guarantee that the digitized data set is free of aliased components. The NI PXI-

4461 employs both digital and analog low-pass filters to achieve this protection.

In addition to the ADC built-in digital filtering, DSA devices may also feature a fixed-frequency analog

filter. The analog filter removes high-frequency components in the analog signal path before they reach

the ADC. This filtering addresses the possibility of high-frequency aliasing from the narrow-bands that

are not covered by the digital filter.

15A.2.1.8 Input Filter Delay

The input filter delay is the time required for digital data to propagate through the ADC digital filter, For

example, a signal experiences a delay equal to 6.3 msec at 10 kS/sec. This delay is an important factor for

stimulus – response measurements, control applications, or any application where loop time is critical. In

this case, it is often advantageous to maximize the sample rate and minimize the time required for 63

sample clock cycles to elapse.

The input filter delay also makes an external digital trigger appear to occur 63 sample clocks later than

expected. Alternatively, the acquired buffer appears to begin 63 samples earlier than expected. This delay

occurs because external digital triggering is a predigitization event.

Virtual Instrumentation for Data Acquisition, Analysis, and Presentation 15-77

© 2005 by Taylor & Francis Group, LLC

15A.2.1.9 Overload Detection

It is desirable to ensure that the DSA device includes overload detection in both the analog domain

(predigitization) and digital domain (postdigitization). An analog overrange can occur independently

from a digital overrange, and vice versa. For example, an IEPE accelerometer might have a resonant

frequency that, when stimulated, can produce an overrange in the analog signal. However, because the

delta – sigma technology of the ADC uses very sharp antialiasing filters, the overrange is not passed into

the digitized signal. Conversely, a sharp transient on the analog side might not overrange, but the step

response of the delta – sigma antialiasing filters might result in clipping in the digital data.

Modern DSA devices allow you to programmatically poll the digital and analog overload detection

circuitry on a per-channel basis to monitor for an overload condition. If an overload is detected, consider

any data acquired at that time corrupt.

15A.2.2 Output Signal Considerations

This section describes the theory of operation of the output components of DSA devices such as the NI

PXI-4461.

15A.2.2.1 Output Pseudodifferential and Differential Configuration

The output channel terminal configuration options are very similar to those for the input channels. The

NI PXI-4461 output channels are configurable on a per-channel basis. As with the input channels, you

should configure the output channel based on how the DUT is referenced. Refer to Table 15A.2 to

determine how to configure the output channel based on the DUT reference.

If the DUT inputs are floating, use the pseudodifferential channel configuration. The term

pseudodifferential refers to the fact that there is a 50 W resistance between the outer BNC shell and chassis

ground. A floating DUT does not connect in any way to the building ground system. Instead, the DUT

has an isolated ground-reference point. Transformer inputs without center ground taps, battery-powered

devices, or any instruments that have an isolated input are all examples of floating DUTs. One should

provide a ground-reference for a floating DUT input. If no ground-reference point is provided — for

example, in selecting differential mode with a floating shaker table input amplifier — the outputs can

float outside the common-mode range of the amplifier input.

If the DUT input is ground referenced, use the differential channel configuration. A single-ended DUT

connects in some way to the building system ground. Therefore, it is already connected to a groundreference

point with respect to the NI PXI-4461, assuming the PXI or CompactPCI chassis and controller

are plugged into the same power system. Nonisolated inputs of instruments that plug into the building

power system fall into this category.

You should provide only one ground-reference point for each channel by properly selecting the

differential or pseudodifferential configuration. If you provide two ground-reference points — for

example, by selecting the pseudodifferential output mode for a single-ended amplifier as the DUT — the

difference in ground potential results in currents in the ground system that can cause errors in the output

signal. The 50 W resistor on the signal ground is usually sufficient to reduce this current to negligible

levels, but results can vary depending on the system setup.

The NI PXI-4461 is automatically configured for the differential mode when powered on or when

power is removed from the device. Using the differential mode by default protects the 50 W resistor on

the signal ground.

TABLE 15A.2 Output Channel Configuration

DUT Reference Output Channel Configuration

Floating Pseudodifferential

Ground referenced Differential

15-78 Vibration and Shock Handbook

© 2005 by Taylor & Francis Group, LLC

15A.2.2.2 Attenuation

Modern DSA devices offer variable gain settings for AO. Most gain settings correspond to a particular AO

range, always centered at 0 V. These gain settings can be specified in dB, where the 0 dB reference

corresponds to the default output range.

Table 15A.3 summarizes the three output gain options available on the NI PXI-4461.

In general, select the gain that provides the greatest dynamic range and the least distortion. The ^1 Vpk

setting maximizes the dynamic range if you know the stimulus is limited to, for example, 0.5 Vpk. You can

minimize system distortion by providing sufficient headroom between the stimulus setting (0.5 Vpk) and

the range setting (^1 Vpk). In some cases in which distortion performance is critical, you can reduce the

overall dynamic range to improve the distortion characteristics by selecting the ^ 10 Vpk setting.

15A.2.2.3 Digital-to-Analog Conversion

Digital-to-analog conversion (DAC) is discussed in Chapter 16. The delta – sigma DACs on the NI PXI-

4461 function in a way analogous to delta – sigma ADCs. The digital data first passes through a digital

interpolation filter, then the resampling filter of the DAC, and finally goes to the delta – sigma modulator.

In the ADC, the delta – sigma modulator is an analog circuit that converts high-resolution analog

signals to high-rate, 1-bit digital data, whereas in the DAC the delta – sigma modulator is a digital circuit

that converts high-resolution digital data to high-rate, 1-bit digital data. As in the ADC, the modulator

frequency shapes the quantization noise so that almost all of its energy is above the Nyquist frequency.

The digital 1-bit data is then sent directly to a 1-bit DAC. This DAC can have only one of two analog

values, and therefore is inherently perfectly linear.

15A.2.2.4 Anti-imaging and Interpolation Filters

A sampled signal repeats itself throughout the frequency spectrum. These repetitions begin above one

half the sample rate, fs; and, theoretically, continue up through the spectrum to infinity. Images remain

in the sample data because the data actually represent only the frequency components below one half fs

(the baseband).

15A.2.2.5 Output Filter Delay

Output filter delay, or the time required for digital data to propagate through the DAC and interpolation

digital filters, varies depending on the sample rate. This delay is an important factor for stimulus –

response measurements, control applications, and every application where loop time is critical.

Навигация