Beruflich Dokumente
Kultur Dokumente
Disclaimer
The material contained in this document is provided “as is” and is subject to being changed,
without notice, in future editions. Further, to the maximum extent permitted by applicable law,
Vallen Systeme disclaims all warranties, either expressed or implied with regard to this
specification and any information contained herein, including but not limited to the implied
warranties of merchantability and fitness for a particular purpose. Vallen Systeme shall not be
liable for errors or for incidental or consequential damages in connection with the furnishing,
use, or performance of this document or any information contained herein.
Vallen Systeme shall not be liable for any direct, indirect, consequential or incidental damage
arising out of the use or inability to use of the AMSY-6 and the equipment delivered with it.
Vallen Systeme reserves the right to charge for any efforts taken to remedy any problems for
which we are not responsible.
Revision Record
Date Changes
04-2017 Updated info, added global coordinate info, changes related to USB3.1
This document:
• Describes the operation of the AMSY-6 system
• Provides hints on how to perform AE-measurements with AMSY-6
• Outlines the most important tools and procedures required for system maintenance and test
This document is one out of four documents describing the AMSY-6 system. Additional
information can be found in:
• AMSY-6 System Specification
• AMSY-6 System Description
• Vallen AE-Suite Software Manual
Contents
1 Introduction ....................................................................................................................... 8
2 Installing Vallen AE-Suite software................................................................................. 9
2.1 Installation requirements................................................................................................. 9
2.2 User Account Control settings for Windows 7 and Windows VISTA .............................. 9
2.3 Obtaining the software .................................................................................................. 10
2.4 Pre installation .............................................................................................................. 10
2.5 Software installation ...................................................................................................... 11
2.6 Installation Troubleshooting .......................................................................................... 15
2.6.1 Installation process does not continue after internet activation ........................... 15
2.6.2 AMSY-6 hardware detection failed ...................................................................... 15
2.6.3 Hardware initialization reports a firmware mismatch ........................................... 15
2.7 Updating software ......................................................................................................... 15
2.8 Upgrading software ....................................................................................................... 16
3 Setting up measurement hardware ............................................................................... 17
3.1 Position of mains switch ............................................................................................... 17
3.2 Chassis setup ............................................................................................................... 17
3.3 Multi chassis setup ....................................................................................................... 17
3.4 Elements of the chassis ................................................................................................ 18
3.4.1 Control panel elements ........................................................................................ 19
3.4.2 ASIP-2 front panel elements ................................................................................ 23
3.4.3 Back panel elements ............................................................................................ 25
4 Extending hardware........................................................................................................ 28
4.1 Adding ASIP-2 .............................................................................................................. 28
4.2 Adding external parameter inputs ................................................................................. 28
5 Mounting AE-Sensors .................................................................................................... 29
5.1 Basic information about AE-sensors ............................................................................ 29
Literature:
Indicates additional literature references for further reading
Special Feature:
Indicates a unique feature of Vallen hard- or software
UAC (user account control) has to be disabled for installation purposes and running the AE-
suite software. In order to do so, follow these steps.
Windows VISTA
1. Launch MSCONFIG: StartRun MSCONFIG
2. Click on the Tools tab. Scroll down till you find "Disable UAC/User account control". Click
on that line.
4. A CMD window will open. When the command has been executed, you can close the
window.
5. Close MSCONFIG. You need to reboot the computer for changes to apply.
You can re-enable UAC by selecting the "Enable UAC" line and then clicking on the Launch
button. However, enabled UAC will prevent execution of any AE-Suite software programs.
2. Click on the Tools tab. Scroll down till you find “UAC/User account control settings". Click
on that line.
4. A pop up menu opens where the UAC setting can be changed by use of a slide control.
Disable UAC by setting it to the lowest value and confirm it with “OK”
5. Close MSCONFIG. You need to reboot the computer for changes to apply.
You can re-enable UAC by choosing a higher security setting. However, enabled UAC will
prevent execution of any AE-Suite software program.
2. Remove older versions of the AE-Suite software. We strongly recommend a backup of files
that you would like to keep (e.g. library files and vac files). Data files (primary, transient and
transient feature data files) will not be deleted during uninstall process.
Internet download
If the AE-Suite software is obtained by downloading it from the internet:
1. Store and unzip the AESuiteYYYY.MMDD.zip (e.g. to your Windows desktop). Open the
folder to which AE-Suite was unzipped and read the ReadMe.rtf first, please.
Alternatively, choose “KeyCode Request Form” and fill out the required fields.
Send the document by email or fax. We will respond with a KeyCode which you should
paste into the lower field.
After successful installation of the driver you can carry out the hardware detection by
clicking “Next >”.
The newest software version can be obtained from the Vallen website. Please refer to chapter
2.3 for obtaining the software.
SP1L, SP2L
These two LEDs are reserved for future usage.
Master (green)
The Master LED indicates the master chassis in a multiple chassis setup. The LED is off in all
slave chassis during hardware detection.
Full (red)
This LED is always off unless PC resources are running low or a process such as an updating
or scanning anti-virus software locks down CPU. A blinking Full LED indicates that the chassis
output buffer runs about 80% full (in this case the chassis enters long duration mode). This
situation is indicated by the “E” hit flag. Once the buffer is emptied the Full LED will cease
blinking.
Full LED is constantly on, if the output buffer gets an overflow (fatal, non-recoverable error).
This LED is controlled for each chassis individually.
Please “Pause” Acquisition to reset a “Full” LED that was permanently on.
USB (green)
When Acquisition is running, this LED is switched on whenever data is transferred over USB. If
USB LED is off no USB data transfer happened for over 1s.
No Pol (red)
This LED is always off. It will be on in similar cases as documented for Full LED. No Poll LED is
switched on, if the chassis was not polled for data by the acquisition program for at least 5
seconds. Power down AMSY-6 and switch it on again to reset No Poll LED.
It is required to check and repair PC configuration, if this LED gets on.
This LED is controlled for each chassis individually.
Please “Pause” Acquisition or disable recording by use of AE enable/disable switch to reset “No
Poll” LED if it was on.
If No Poll LED cannot be reset power down AMSY-6.
Long (yellow)
The LED is usually off. If on, the Long LED indicates that at least one channel in the chassis
entered long duration mode.
An AE-channel enters long duration mode, if a channel’s buffer runs half full. With AMSY-6 this
can only happen if hit rate exceeds 100 000 hits/s per chassis. When in the Long Duration
Mode, the Duration Discrimination Time of the AE-channel is (temporarily) set to 98ms. This
causes the hits currently processed in the channel to last until termination by the timeout feature
(approx. 100ms after the start of the hit). This effectively reduces the maximum feature data set
rate to 10 hits per second.
A low hit rate ensures that buffers can be emptied preventing loss of data.
During the Long Duration Mode, the “D-Flag” is set in the hit data sets, identifying data
generated during Long Duration Mode. (Find in help text under “Hit Flags” more details about
flags in hit data sets). Such data does not retain individual hit information; therefore it is
unsuitable for calculation of location, for counting hits, and for processing distributions and
correlations of the AE feature set (amplitude, risetime, duration, energy, counts). On the other
hand, cumulative information about number of cascaded hits (CHIT, see section 0), counts
(CCNT, see section 0) and energy (CENY, see section 0) is completely maintained during Long
Mode. The Long LED goes off when the buffers in all channels have been emptied.
This LED is controlled for each chassis individually.
Pulsing (green)
The Pulsing LED indicates that a chassis is generating pulses. This LED is only active in the
master chassis of a multiple chassis setup and indicates also pulsing slave chassis. A “C” hit-
flag indicates a pulsing channel while a “c” hit-flag indicates a receiving channel.
In a slave chassis this LED is always off.
Run (green)
In a master chassis this LED is on, if the acquisition program is in recording mode.
If recording is disabled by hardware (AE Switch disabled or /DISABLE signal active) when
recording mode is started, this LED remains off, until recording is enabled by hardware. Then it
stays on until acquisition is suspended.
While this LED is off, the time counter is not running, the parametric counter PCTA is frozen,
and the parametric counter PCTD is cleared. No data sets are being generated when the Run
LED is off.
AE-Disable Switch
This switch is located to the left of the “AE disable” LED. Pushed to the right (towards the LED),
the switch disables data acquisition (AE and TR).
It can be used to disable acquisition of data when, for example, work on the test object has to
be carried out or when it is clear that by an external incident false data would be generated.
Data acquisition is disabled as long as “AE disable” LED is blinking. Blinking shall draw
operator’s attention not to forget the disable state of measurement hardware.
TR-Disable Switch
This switch is located left of the “TR disable” LED. Pushed to the right, it disables TR-data
(waveform) acquisition only, but AE-data will still be recorded.
This can be used to reduce the amount of data stored to disk: e.g. enable TR recording only
from time to time may provide useful information but reduces the data volume considerably.
TR Disabled LED
The TR Disabled LED indicates that transient recording is disabled. This LED is on if:
• transient recording is disabled globally in the acquisition setup
• TR disable switch is in disable position
• AE-disable LED is on
It is blinking when the acquisition program is in recording mode and recording is disabled by AE
or TR switch (blinking has priority, except TR data acquisition is not enabled).
Threshold LED
This green LED is on for at least 30 ms, when the respective channel detects a threshold
crossing. The LED flashes yellow, if the AE signal - before digital filtering - exceeds 95% of the
input range. This indicates that the preamplifier output is nearby or above the saturation point.
Selected LED
This green LED indicates when the channel is selected for audibility (see Audio Selection
Switch). The LED flashes yellow, when this channel is in pulsing mode. In pulsing mode a pulse
of programmable amplitude is passed through the ASIP-2 to the corresponding BNC connector.
It can be passed through the preamplifier up to the sensor’s piezoelectric element in order to
excite an acoustic event for a sensor coupling test.
Open LED
This LED indicates if the DC current consumption at the BNC socket is less than 8-12mA. This
indicates that no preamplifier is connected to the BNC-connector.
The Open LED makes it easy to find out to which channel a certain preamplifier is connected:
By removing the cable from the preamplifier, the corresponding Open LED turns on.
DC OUT LED
This LED is green, when the channel is set (by software) to 50 Ohm input impedance with an 8-
28V DC supply for a preamplifier. This is called the DC-50 Ohm mode.
rd
This LED is yellow, when the channel is set to 3 input mode with an 8-28V DC supply e.g. for a
rd
preamplifier. On default, 3 input mode means 200 Ohm. Note: The 200 Ohm resistor can be
replaced by another resistor or by a constant current diode in order to use a preamplifier with
ICP interface.
The DC OUT LED is OFF, when the channel is set to AC-mode. Then the connector does not
deliver DC and the input impedance is 100kOhm. Paralleling up to 20 channels in AC mode to
one in DC@50 Ohm mode does not cause a significant amplitude reduction.
Alarm connector
The Alarm connector provides access to 4 signals:
• Warning
• Alarm
• SP 0 and SP1 (two spare signals) for future developments
The respective pins are driven low by AMSY-6 (usually by use of the Alarm Manager) if the
Warning, Alarm, SP0 or SP1 LED is on. Hence the Alarm connector provides an interface for
external applications to read out the status of the LEDs. If the pins are driven low externally, the
according LED on the Control Panel will be switched on.
Externals Connector
The Externals Connector is a D-Sub connector which provides access to following functions
• External record control: enables or disables data acquisition. Driving pin #8 low disables
acquisition of AE- and TR-data.
• Control of the digital parametric counter PCTD: by use of pins 11 to 14 the PCTD can be
configured. Pin #11 is input for PCTD-Clock. Every rising and falling edge of digital signal
increments/decrements counter. Up-counting is enabled if pin 12 is set high, down-counting
if pin 12 is set low. PCTD is enabled if pin 13 is high, otherwise it is disabled. Pin 14
governs PCTD-reset: high or open resets counter, low enables storage of current counter
value. PCTD can only be reset if PCTD is enabled (pin13 is high).
• I²C interface (prepared for future functionality): I²C Bus can be extended through the
Externals connector, so application or user specific external extension modules could be
controlled by this bus. Since the I2C bus is controlled by the NIOS processor the
development of such modules can only be performed by Vallen Systeme. Currently it is
only in use for internal verification.
• User defined DAC Output (prepared for future functionality)
USB port
The USB port is used for connection to a PC supporting USB 3.1 Gen 1. The 5V line on the
USB cable is not used in AMSY-6, so no USB-power is required.
A special USB connector type was chosen that avoids unwanted cable separation by extra
strong retention force.
Audio output
The audio jack is a 3,5mm TRS type connector with both audio channels driven from the same
mono signal. It can be used to connect either one passive 4 or 8 ohm speaker or two 8 ohm
speakers in parallel.
It is not allowed to connect a TS type connector (mono), a headphone or an active speaker to
this connector.
If an external audio device is attached, internal speaker is disabled. Internally 8 ohm speakers
are used, so parallel operation of one speaker at the rear and one speaker at the front is
possible (e.g. in MB19-V1).
The signal on the audio jack is chassis specific and identical with the signal for the internal
speaker.
Address selector
The rotary switch defines the address of the chassis, which can be any number between 1 and
9. In a multiple chassis setup (see section 3.3) each chassis must have a unique address. The
chassis with the lowest address is automatically the master chassis of the setup. A chassis with
address 0 or multiple coupled chassis with the same address are not allowed.
The Acquisition software automatically detects reports and rejects invalid address selection.
Pulse out
Each chassis has its own pulser module, generating pulses under control of software. Within
each chassis, this pulse goes to all ASIP-2 and to this BNC connector.
The Pulse out connector should only be used for equipment testing purposes.
The pulse at this connector can have up to 450Vpp, with approximately 2 µs rise time.
Connecting any other instrument (e.g. preamplifier without pulse-through function or an Ethernet
network) to this connector might cause permanent damage to such instruments.
Pulse-Through
A chassis is able to generate an electrical pulse. Such a pulse can be used to excite the piezo
element in the sensor which will induce a surface displacement leading to excitation of a surface
wave. For an electrical pulse to reach the AE-sensor it has to by-pass the ASIP-2 and
preamplifier. Pulse-through describes the capability of a preamplifier to connect-through an
electrical pulse to an AE-sensor.
More information can be found in the separate document “Acoustic Emission Preamplifiers”
Since the generation of AE events are asynchronous (stochastic), the AE events do not always
readily separate from one another, especially at high AE event rates. The system requires
directives to detect (the threshold = voltage level to trigger a hit; see section 7.8.2) and separate
(duration discrimination time and the rearm time; see section 7.8.2) discrete AE bursts.
Two types of measurements are done which do not rely on the separation of individual hits:
• Status data e.g. root mean square status (RMSS). RMS is acquired for each channel for
time intervals of specified length. It averages just noise excluding hits.
• Parametric data: parametric data is acquired at certain time intervals (interval setting:
please refer to the online help file) independent of hits. Additionally parametric data sets are
stored in between the interval setting if hits occur.
The subsequent and final step is starting the data recording (see section 7.9)
Figure 6: hardware detection tab in Edit Mode after very first connection of a chassis to a PC
For a proper initiation and in order to continue with the acquisition setup, an input device per
channel has to be defined. Please see or continue to section 7.4.4 for more information.
Chan.
A graphical symbol in the “Chan.” column identifies the channel’s status. In the cell right to the
symbol a logical channel number can be specified.
HW
Values in cells of this column indicate the hardware address of a channel in format xx.yy.z.
xx indicates number of chassis as set by address selector switch at rear side of a chassis.
yy indicates the address of an ASIP-2 board.
z indicates one of the two channels of an ASIP-2 board. z can be “a” or “b”.
Input device
For more information see section 7.5.
Pulser func.
The pulser function can be set to “None”, “Pulse-through” or “AST” (automatic sensor test). A
“None” setting prohibits that a pulse is connected through to the preamplifier/sensor of the
channel. Such a setting is useful in case preamplifiers are used that do not support pulsing
function.
If an input device is selected that does not support “Pulse-through” or “AST” no changes can be
made.
Figure 11: configuration files can be imported via “More…” button and selecting “Import
configuration…”
• Frontend Filter: The front end filter can be activated and defined here.
• Pulser: It contains the settings for the pulser for the sensor coupling test
• Special: This setting dialog is used for changing between true energy and signal strength
and others.
7.8.2 AE-Channels
AE-channel setup is most important acquisition parameter setup. It defines start and end criteria
of hits, the filter settings, transient record trigger settings and gain setting of each channel
individually. Double clicking on a channel will open the channel’s edit menu.
Main settings for most tests are:
• Threshold
• Rearm Time (RAT)
• Duration Discrimination Time (DDT)
• Application specific filters
These settings are usually adjusted to achieve the following goals:
• Optimize sensitivity range while limiting false-hits from noise (governed by threshold)
• Limit overlap of hits (governed by DDT)
• Prevent reflections from effecting AE parameters (governed by RAT)
• Increase signal to noise ratio (governed by application specific filters)
Initial settings for threshold, rearm time and duration discrimination time may be:
Material Threshold Rearm Time Duration Discrimination Time
[dBAE] [ms] [ms]
Metals 30 - 40 1,0 – 4,0 0,40 – 2,00
FRP 40 - 55 0,3 – 2,0 0,15 – 1,00
The settings, especially the threshold setting, can and should be determined experimentally.
The start of a hit is easily determined by the first threshold crossing. In order to detect a first
threshold crossing it must be made sure that a previous hit has ended. A hit has ended when
Duration Discrimination time (DDT) given in µs expired without any threshold crossing. After
Figure 13: AE-channels tab with summary of settings. Double click on a channel will open its
edit menu
Factory defaults such as Metallic Pressure Vessel, Composite Pressure Vessel, Tank Floor
Standard or Fatigue Test contain default settings for these applications. Nevertheless
acquisition parameters must be fine-tuned for a specific test object.
Threshold
Determining the correct threshold setting is often defined in the test instruction. If a test
instruction is not required for an AE-measurement, a detection threshold can be determined by
following procedure:
A good practice determining an acquisition threshold is by acquiring data with a low threshold
for 5 to 10 minutes. A low threshold setting would be 22dBAE to 24dBAE. Environmental
conditions during this “noise” test have to be similar compared to the actual test (e.g. pump is
running, etc.). The peak noise amplitude within this period has to be determined. Adding 6dB to
this peak noise amplitude will result in a good threshold setting.
Rearm time
Rearm time is governed by the lifetime of a burst emitted by an AE-source. Rearm time should
be in the order of burst lifetime. Lifetime determines when maximum amplitude of a burst falls
below detection threshold. Hence lifetime is a function of the strength of an AE-source and
attenuation of the material.
Calculating lifetime needs knowledge of
• attenuation per unit distance (α),
• expected maximum strength of a source (peak amplitude, Apk),
• detection threshold (Adet) and
• speed of sound (c) in the material.
A pk − Adet
t life =
αc
Usually an artificial source, such as the Hsu Nielsen source can be used in determining burst
lifetime. Rearm time should be chosen in range of burst lifetime.
TR-Acquisition Parameters
For more information, please see chapter 7.9.
Input Settings
This section summarizes input device settings. It is important to check “Calc. gain” setting,
which must correspond to gain of the input device plus additional gain of Range setting.
Otherwise amplitudes will be over- or underestimated.
Parametric Timing
Interval specifies the time interval between stored parametric data sets during time periods in
which no hits are detected. The data sets generated in this time interval (0.2ms..600s) are
called time driven parametric data. Parametric data sets are stored more often while hits are
detected. (See “clock”). These data sets are called hit driven parametric data.
Clock defines the parametric sampling interval (0,2 to 10ms). Clock values less than 1ms should
be avoided for large channel applications. Otherwise the bus gets overloaded with parametric
data.
PCTA-Trigger (PA0)
Level and Hysteresis settings are used for cycle counting. PCTA is a counter that an increase
by one every time the voltage at the parametric input PA0 has passed through the hysteresis
band from Level + Hysteresis to Level.
Record Control
Recording is disabled if the voltage at the parametric input specified by "PAx input" exceeds the
upper limit. Also, recording is disabled if the voltage at the parametric input specified by "PAx
input" is below the lower limit.
An AMSY distinguishes between AE-feature data and transient data. First and foremost
Continuous Mode recording will acquire AE-feature data. Additionally transient data can be
• A single channel for indefinite time (i.e. until manually stopped by user),
• All available channels in repeating succession for indefinite time,
• All available channels in succession for one time.
“Label” allows inserting user text with a time stamp into acquired data stream. A label may
indicate certain interesting occasions during a test or can be used to structure the test
sequence. During analysis labels can be used as starting- or halt points.
8.2.1 Assembling individual hits into an event data set (FHCDT condition)
The main task of the Event Builder is identifying a first hit of a hit sequence caused by an
acoustic emission event. Subsequent hits of the sequence will contribute to an event data set.
When acoustic emission events are well separated in time then their hit sequences are also well
separated in time. This separation in time, when no hits occur, is used to determine a first hit
caused by an acoustic emission event.
Identifying a first hit is achieved by a data processing condition based on the First Hit Channel
Discrimination Time (FHCDT) parameter. This condition determines the arrival time difference to
the latest hit occurred before. Evaluation of this condition yields two results:
• Arrival time difference is larger than FHCDT: the currently processed hit is a first hit caused
by an acoustic emission event. An active assembling process is closed if a first hit is
detected. A new event assembling process is started.
• Arrival time difference is less than FHCDT: If an event assembling process is active the
currently processed hit is considered to be a sub-hit of an active event data set.
An example of two hit sequences is shown in figure 20. The graph shows time axes (x-axis) of 4
different AE-channels labeled channel 1 to channel 4. Deltoid shaped objects indicate hits that
occur in these channels. The deltoid shape shall represent the characteristic of hits having a
short rise time and longer ring down. The left corner of the deltoid shaped object indicates the
arrival time, i.e. the first threshold crossing. The first hit sequence, hits marked 1 to 4, is
separated by a period of ∆t54 from a second hit sequence, consisting of hits marked 5 to 7.
FHCDT has to be smaller than ∆t54 in order that the Event Builder is able to identify hit marked 5
as a first hit. On the other hand FHCDT has to be larger than the largest time difference
between two consecutive hits of an event data set in order that hits 1 to 4 or similarly hits 5 to 7
are grouped into an event data set.
The time window of an event data set must be at least of the length it takes an elastic wave to
propagate the maximum distance, dmax. When c is the speed of sound of the elastic wave, the
result is:
𝑑𝑚𝑚𝑚
DT1X-Max = 1.5 ×
𝑐
Please note that for a safety margin a factor 1.5 has been accounted for.
Channel group
There may be one or more channel groups inside a location processor. With multiple channel
groups more complex structures can be covered. An AE-channel can belong to more than one
channel group. A location algorithm has to be assigned to each channel group. However each
channel group can have a different location algorithm.
Location errors
Location accuracy can be adversely influenced by
• A different wave mode than the assumed one determines the arrival time.
• A wave takes a different propagation path than assumed by the algorithm
• Burst of two or more sources overlap at the sensor
• Sources emit bursts in such a quick succession, that there is not enough time for the burst
in the structure to decay, therefore they do not represent a “new” event.
Diagrams
Several different types of diagrams are available: 2D-, 3D-graphs, TR-diagrams. TR-diagrams
show waveforms (transient data) in time and/or frequency domain. Many properties of those
graphs can be adjusted by the operator: the results (attributes) to be shown (e.g. Amplitude vs.
time, or Duration vs. Energy...), left + right axis legend, caption (top) and comment text (bottom).
Legends can be edited or made invisible and/or not printable. Diagrams can show distributions,
correlations (point plots), histories, etc., depending on the selected kind of results. Each
diagram can show several planes. Data represented on a plane can be filtered, the color and
shape of symbols can be selected and more.
Listing
A listing presents data in a table format. The attributes which are listed and their ordering can be
user selected.
Still Image
Still images can be inserted and displayed on different pages of the VisualAE setup. This kind of
visual can be used to place a company logo on an analysis page.
Directory Hits vs
5) For instance, if a +/- 10 Volt signal corresponds to a -10 to 10 kN load with positive
voltage corresponding to tension, the following text should be entered:
Name Load
Offset 0 mV
Properties Menu
12) Call up the diagram's property dialog
Right click on the diagram to get the pop-up menu, choose the properties item
13) Set up a right axis
14) In the attribute menu of the diagram dialog, check the box by the right vertical axis
attribute
15) By the right attribute arrow, change the right
vertical axis attribute to a parametric Select
the right arrow to get the attribute menu: a
dialog will appear with a note describing
which steps must be used to get a right
axis. These steps are repeated and fulfilled
by following the next steps.
Under the Parametric Results, choose the
PA0U item (Note: if no parametric scaling
was implemented in step 1, choose the PA0
item and this result will appear in millivolts).
16) Add a right axis distribution plane to the
diagram
Go to the distribution menu (select the
//distribution\\ tab at the top of the dialog)
Click on the right vertical axis (PA0U vs.
Time[s]) inside the center of the box.
Click on the [[Add]] button immediately
below the dialog's center region
17) In the Plane Settings Dialog under //plane style\\ tab: change the plane type to line history
and change the plane's color to a color other than green.
18) Change the plane's legend
Change the menu by selecting the //plane legend\\ tab at the top of the plane settings
dialog
Replace the default macro (=[filter]) with a description of the parametric as in step
Creating parametric conversion either appropriate text (Load) or better yet the following
macro can be used: =[YR-ATTRUnitLong]
Select [[OK]] at the bottom of the plane settings dialog.
Actual Example
In advanced materials, more than one path to failure can occur and it can be very difficult and
time consuming to analyze and predict how these materials behave. Fiber reinforced composite
materials are the most prevalent example of materials that readily exhibit multiplicity of failure
modes. The diagram which accompanies this example is from a composite material taken to
failure.
When loaded to failure, many composite lay-ups will produce matrix cracking first. Matrix
cracking is the cause of the knees in stress-strain curves. As matrix cracking represents
permanent damage, it is necessary to recommend that composite materials be used only below
this knee point. The change in deflection of the stress-strain curve can be hard to detect and it
may be hard determine when it occurred. Even when the transition is easy to determine, it is
often hard or impossible to determine the extent of the matrix cracking: how soon matrix
cracking occurs before the knee point and how long matrix cracking continues after the knee
point, if at all. One basic method for determining the knee point from the stress-strain curve is to
draw a tangent to the load curve at the initial part of the curve and another at the high end of the
curve and to see where the tangents intersect.
The data shown comes from a lay-up with 0, 45 and 90 degree plies. In load portion of the
diagram, several problems can be seen: there is a "toe" at the beginning of the load curve, the
change in slope is quite small and the stress-strain curve gives almost no indication how long
the transition occurs. A "toe" in a stress-strain is a common feature; it occurs at low load levels
and is caused by incomplete load transfer. This problem corrects itself at higher load levels.
Tangents drawn to the lower part and higher part of the curve are highly variable; you may find it
easy to justify any point between 150 seconds (1100 N) and 250 seconds (2200 N) as the
transition point of this curve.
The AE is readily detectable and two phases can be seen in the diagram. In the first, the AE
increases to a steady rate which holds about constant between 150 and 250 seconds. After 250
seconds the AE increases dramatically until the first load drop. In initial area and around the
region where the slope changes, matrix cracking can be verified under the microscope. In the
second phase which concludes with the failure of several plies, the ply failure can be visibly
observed by the unaided eye. In fact the AE confirms that matrix cracking occurs at a range of
loads, predominantly between 150 and 250 seconds. The matrix cracking is not completely
restricted to this period while the development of ply failure occurring at 350 seconds is
accelerating from at least 250 seconds into the test.
Directory Amp vs
Properties Menu
6) Call up the diagram's property dialog
Right click on the diagram to get the pop-up menu
Choose the properties item
7) Set up a right axis menu of the diagram dialog, check the box by the right vertical axis
attribute
8) By the right attribute arrow, change the right vertical axis attribute to an RMSS (under
other results RMS background noise from status set) or RMS-background (under Hit
results called RMS background noise). Note: RMS-status and RMS background are
stored differently: RMS background is stored with each hit, while RMS status are stored
Actual Example
Welding, like other processes, creates distinct regions of high RMS. In particular, when welding
the RMS level rises to a fairly constant value during welding and falls back when welding is
complete. Weld cracking, on the other hand, generates readily detectable AE hits that is easily
discerned from the RMS as dots in this representation. Cracking may occur during welding,
immediately after welding or long periods after welding depending on the mechanism which
generates it.
In this case an intentionally bad weld is made so that it will crack. The actual welding is
performed in about one minute long, which can be seen in the increase in RMS from 0-60
seconds. During welding, various sources, including phase transformation, initial contact
between the welding device and the material and surface oxidation occur. After welding, as
stress relief and diffusion occurs between base and weld materials, cracking can be seen
starting immediately with welding and continuing for at least one half hour of welding. For some
weld types cracking may be detected in welds days after the weld is made. The time delay with
the weld is also an indication of the cause of the AE. Early cracking can indicate stress release
due to cool down, for instance, while cracking long after welding indicates slower processes,
such as Hydrogen induced cracking from Hydrogen diffusion.
Directory Location
For this example, values from 4 sensors of a larger interlocking array are used (see diagram
above)
10) Verify
The [[Verify]] button will automatically check for location settings that are questionable.
Any combinations of settings that are found questionable will be reported with a short
explanation. This can be helpful in determining if some of the location settings should be
modified.
11) Add a comment to the processor.
An advantage of giving a processor or diagram a comment is that in the structure tree the
processor/diagram will have a more descriptive identity.
For this example, a comment "Planar Location Processor for Channels 7, 8, 11 and 12"
was used.
Properties Menu
17) Call up the diagram's property dialog
Right click on the diagram to get the pop-up menu
Choose the properties item
18) Adjust the plane symbol and legend for the left axis.
Select the left axis plane by double clicking on the distribution icon under the left vertical
axis
Select a light symbol (e.g. light green), or if there are a great deal of locatable events
choose Correlation (dots)
Select [[OK]] at the bottom of the plane settings dialog.
19) Select [[OK]] at the bottom of the diagram properties dialog.
Actual Example
Noise is of constant importance in any AE test. It is usually best if you can identify and filter
noise
• at its source (e.g. mechanically dampening it),
• by its frequency (e.g. applying appropriate frequency filters)
• by its amplitude range (e.g. increasing thresholds or using a floating threshold) or
• by its spatial extent (e.g. using guard sensors)
There is sometimes no alternative to sorting and filtering data after it has been recorded.
The diagram in the example, counts versus duration correlation, represents a check on the
resonance dependence of the collected hits. In this diagram, a great deal of the data shows a
linear correlation where counts/duration line falls on the resonance frequency of the sensor.
There is a second group of data, however, which lies much below the resonance correlation
line. This is low amplitude signals which have extended duration because the reoccurring
source continues to excite the threshold. In this case, there is rubbing noise on the structure.
Several correlation graphs (counts-duration, counts-amplitude, duration-amplitude) are being
used as auxiliary graphs for the purpose of identifying noise in structural testing. They help
Directory Hits vs
Properties Menu
6) Call up the diagram's property dialog
Right click on the diagram to get the pop-up menu
Choose the properties item
7) Adjust Attributes
By the left vertical axis, change the drop down menu from linear scale to logarithmic
scale.
Actual Example
When there is a high hit activity rate during a test, the size of the data file can be excessive.
Also, the frequency hits overlap one another is likely to be much greater than with slower activity
rates. Especially in composite materials large amounts of AE activity can be expected. The
number of hits and the peak rate in hits/second will both be high. How uniformly the total
monitored activity changes with threshold can be determined from a cumulative amplitude plot.
In this example, over 20'000 hits were monitored from a composite loaded until failure with a
monitoring threshold of 50 dB. In the example diagram, it can be seen that if a 55 dB threshold
had been chosen, only 10'000 hits would be expected and at 60 dB only 5'000 hits. On the other
hand, if one extrapolates this graph, one might expect 50'000 hits if the threshold were lowered
to 45 dB and more than 100'000 hits at 40 dB. Since this curve appears uniform, one might
want to use a higher threshold (such as 50 or 55 dB) if one is only interested in statistically
characterizing this material. The higher threshold will limit the chance that hits will be
Directory Root
Actual Example
Location requires the grouping of multiple hits into events and calculating the position of the
events. In the example diagram, AE-source location on a gas storage sphere is shown in 3
dimensions. This example shows only a small extract of a full scale test. Only such events are
selected which occur in a limited region of the vessel. The concentration of events is
emphasized by the clusters.
Simple Correlation
This AE-diagram can easily be modified to
represent any correlation. Other common
correlations include
• Amplitude Vs. Duration
• Amplitude Vs. Energy
• Amplitude Vs. Risetime
Channel Activity
This AE-diagram can easily be modified to
represent any channel activity, for instance
energy or counts.
A right axis attribute can be added to
contrast to AE parameters against channel.
RMS Status
An AE-diagram showing RMS-status can
be used to determine the general
background level for each AE channel
before and during testing.
For RMS-status, it only makes sense to
view individual channels. Other time
distributions with separate channel planes
can be obtained by modifying the y-axis
attribute.
Stress-Strain
Any parametrics that are sampled at the
parametric input can be plotted against
each other in this type of AE-diagram.
Depending on the parametric input, graphs
such as load-displacement, strain-
temperature can be generated, for
instance. Care must be taken when the
parametric range is not monotonically
increasing, a filter can be implemented to
limit the time range if necessary.
Cascade Comparison
This AE-diagram can show how well hits
were separated. Nearly identical cascaded
hits and hits indicate good separation. Two
alternative versions of this graph may also
be helpful:
• Counts, Cascaded-Counts vs. time
• Energy, Cascaded Energy vs. time
Poor separation may be improved by
adjusting rearm time, duration
discrimination time and threshold; however
there are other effects to consider.
Linear Location
This AE-diagram has been put under a
location processor.
These examples found in the library are only the beginning, and a very small fraction of the AE-
diagrams possible in VisualAE. Simple modifications can be used by the user to obtain endless
variations on these AE-diagrams:
In VisualAE, any axis attribute (amplitude, energy) can be changed to another attribute from
within the AE-diagrams "property" dialog.
VisualAE allows easy addition and subtraction of "planes" from any existing diagram that
expands or removes the change and representation of the data.
A specific selection of data (filtered) can be chosen for each plane separately.
Any modifications to graphs can be stored in VisualAE's library of previously defined AE-
diagrams. These graphs can be recalled and modified. Each Library entry has specific
designation as to its origin (user-given title) and its author (these AE-diagrams have the author
specified as Vallen-Systeme).
Background-burst overlay
The background noise level along with the
detection of AE signals are depicted in this
VisualAE AE-diagram. Both left and right
axis scales are fixed to correspond (1 µV=0
dB).
Typically, when the background
(continuous) noise level is 20 dB below the
threshold, the continuous RMS will not be
contributing to the collected hits, while
when the RMS is within 15 dB or less of the
threshold, continuous background will be
contributing to the threshold crossings.
Note when the RMS level is very low
(below 5 µV) the measurement is in the
digital steps and is less exact.
Speed of sound
Another source of information from an
attenuation study (see example 2.4) is
propagation delay. By plotting an AE-
diagram with arrival time differences along
with a distribution (acting as a trend),
VisualAE can make the velocity from such
a graph easy to visualize.
Highlighting results
Activity during increased loading is an
important criterion for structural evaluation,
however it can often be difficult to easily
represent.
This VisualAE AE-diagram uses planes to
highlight the regions of increased AE
activity that corresponds to subsequent
load increase areas.
Perspectives of AE 3D diagrams
The image below shows perspectives of the same VisualAE's 3D AE-diagrams, with graduated
levels.
Not only can the graphs be rotated and stretched easily, the graphs can be shown with and
without perspective viewing, without axis and with various representations (blocks are used at
right, correlation, nets, pyramids, etc. can also be used.
Figure 21: results of location algorithm on cylindrical part of hull. Clustering summarizes active
areas on hull.
Diagrams show results for each channel group in figure 21 and figure 22, individually. Since end
caps are modeled by spherical location processor, results are available in 3D.
Figure 24: location results of lower end cap. Figure 25: location results of upper end cap.
Results are available in 2D. Results are available in 2D.
2R
x=� R cos(LATI) sin(LONG)
R − R cos(LATI)
2R
y=� R sin(LATI) sin(LONG)
R − R cos(LATI)
R, LATI and LONG correspond to radius of end cap, latitude and longitude.
In this example a fourth location processor, LC4, is inserted. This location processor consists of
three channel groups, also. One of them for cylindrical part of hull and one for each end cap.
End cap location algorithm is a planar one and sensor positions are calculated according to
Lambert’s projection rule.
Location processor Lc4: Hull and End
Caps (planar) – LAMBERT Projection
approximates spherical end cap by a
plane.
Filter processors Fl20 (“Hull CHGPR”),
Fl24 (“Upper Half Dome CHGRP”) and
Fl22 (“Lower Half Dome CHGRP”) are
channel group filters that let pass only
data from one channel group which is
linked to a certain part of the hull. These
filter guarantee that in subsequent visuals
only data from one channel group is
displayed (or in subsequent processors
only data from one channel group is
available).
Resulting location diagrams of end caps
are shown below.
Table 1: comparison of location results on lower end cap. Results of spherical location
processor are shown in group “Spherical”. Results of planar location algorithm with
orthographically projected sensor positions are shown in group “orthographic
projection”. Results of planar location algorithm with Lambert’s projected sensor
positions are shown in group “Lambert’s projection”. LUCY result in red font indicates
that it exceeded an arbitrarily set threshold and location result is therefore considered
unreliable.
Location results of DSET 615557 and 619077 of spherical location processor are not shown in
location plot (see figure 22) because their LUCY exceeds an arbitrarily set threshold. In contrast
planar location algorithm with Lambert’s projection of sensor positions locates all but one event
(DSET 592316). However, this located event is questionable since closer analysis of sequence
of hit channels indicates that two events were overlapping, one in vicinity of channel 50 (AE
Table 2: comparison of location results on upper end cap. Results of spherical location
processor are shown in group “Spherical”. Results of planar location algorithm with
orthographically projected sensor positions are shown in group “orthographic
projection”. Results of planar location algorithm with Lambert’s projected sensor
positions are shown in group “Lambert’s projection”. Red font in LUCY column
indicates results that are not shown in diagram because of an arbitrarily set LUCY
filter. Red font in LONG / LATI columns indicate positions that are mapped to lower
part of sphere.
Location results of hits with low amplitude are in general afflicted with larger error or uncertainty.
Because of a LUCY filter which was set to an arbitrary value not all located hits are shown on
the upper end cap (see figure 23). Planar location processor with Lambert’s projection of sensor
positions locates all locatable events but one (last one in table 2). However location result of
DSET 646554 is questionable since only minimum number of sensors contributed to location
result and peak amplitude was close to threshold of 40dBAE.
Location results of planar algorithm fit better to spherical location results when Lambert’s
projection is used. Lambert’s azimuthal equal-area projection distorts distances close to equator
but not to same extent as orthographic projection does. Hence Lambert’s projection is more
suitable for approximating a half sphere.
In all cases, located AE-sources should be verified after the test. This is especially true for
planar approximations of end caps.
Figure 29: location diagram showing results of both end caps and cylindrical part of hull in one
diagram.
HITS: Hits
This returns the value 1 for each hit data set, to be used in cumulative- or rate plots.
Visuals placed behind a location processor usually use only first hits of events. In case another
hit parameter is used as axis attribute the user can select if only first hits of events or all sub hits
of an event shall be used for plotting the diagram. Separate counters per channel are used, if
one axis is assigned to CHAN.
THR: Threshold
The first crossing of the threshold level is the start of a hit. The end of the hit is reached when
the input voltage does not cross the threshold for a specified period of time (called the duration
discrimination time). The threshold is a dB-value of same range and resolution as the peak
amplitude.
In case of a floating threshold, the threshold can vary during a test. Because of this, the
threshold is not only a parameter but also a result.
X, Y, Z:
Local coordinates; local coordinates are used by the location algorithm. Each sensor group has
its own local coordinate system.
For circular structures, the valid range for X is (-0.5...+0.5) * circumference.
In case of cylindrical structures Y-axis runs parallel to the axis of revolution.
For spherical results, X corresponds to the projection of the radius vector onto the x-axis.
For spherical results, Y corresponds to the projection of the radius vector onto the y-axis.
For spherical results, Z corresponds to the projection of the radius vector onto the z-axis.
GX, GY, GZ
Global coordinates; shall be used in 3D diagrams. Global coordinates take into consideration
translation and rotation of sub-parts (e.g. heads and barrel) of an object such as a pressure
vessel.
LONG: Longitude
This means the degree longitude of a sphere. LONG is available for spherical location, only.
GpHI
Historic Index; describing the relation of the most recent hits to all previous hits
GpRs
Grading Result (in development)
GpMHI
Maximum historic index
GpCt
Number of hits used for calculation.
PA
Peak Amplitude of a transient that has been recorded to a transient date page
RT
Rise time of a transient that has been recorded to a transient data page. Rise time is the time
from first threshold crossing to time of peak amplitude.
Dur
Duration of a transient that has been recorded to a transient data page
CTP
Counts to peak, meaning number of positive threshold crossings from first threshold crossing to
peak amplitude based on a transient that has been recorded to a transient data page.
FI
Initiation frequency calculated as counts to peak divided by risetime.
FR
Reverberation frequency calculated as ring down counts divided by ring down time, whereby
ring down counts is simply CNTS-CTP and ring down time is Dur-RT.
Tx_FFT.CoG
h
Frequency of center of gravity (similar to FCOG) of x time segment
Tx_FFT.FoM
th
Frequency of maximum FFT amplitude (similar to FMXA) of x time segment
Tx_Avg
RMS/AVG FFT amplitudes of time segment x
2. Choose the desired start and end criteria in the ‘project settings’
3. In the listing ‘Properties’, tab ‘Attributes’ select the data (columns) you to display and export
5. Make a right mouse click on the listing, select ‘Copy’ and then which type of export you
want (ASCII file to clipboard or HDD, bitmap, directly to printer…) and follow the
instructions.
2. Choose the desired start and end criteria in the ‘project settings’.
4. make a right mouse click on the diagram, select ‘Copy’ and then which type of export you
want ASCII file to clipboard or HDD, bitmap) and follow the instructions.
If an external data set is to be correlated with the AE data set, a synchronization point is very
useful. A singular event (such as any simulated AE) that also registers on the other
measurement device can serve this purpose. The time of this data set can be used as a
reference on the AE-data.
15.2 Printing
Creating hardcopies directly from VisualAE is simple and convenient. Any Windows-supported
printer can be used. A dialog window offers a printer-preview and many possibilities for
adjusting the appearance of the printed page (landscape or portrait, size, position on page,...).
The above mentioned details do not completely describe all features of VisualAE. The target of
this description is to enable you to find your way through VisualAE, learn its use and get
confident in a minimum of time. VisualAE is supported by extensive, context sensitive help texts.
You can also quickly find information of interest over the help text's list of contents and index.
16.1 Evaluators
Evaluators are the front-end of the Alarm system. An Evaluator defines a monitoring condition
consisting of two independent OR linked conditions. One of these conditions is responsible for
triggering warnings, the other one for triggering alarms. Depending on the result of these two
conditions an Evaluator has three evaluation states:
• NULL: warning – and alarm condition evaluate false
• Warning: warning conditions evaluates true
• Alarm: alarm condition evaluates true. Alarm evaluation state will always override Warning
evaluation state.
Whenever the monitoring condition is evaluated true, the Evaluator triggers the Signal which is
connected to it.
Pre-defined Evaluators are called Watchdogs. Currently, nine Watchdogs are implemented.
Limits of a Watchdog’s monitoring conditions can be changed by the user any time. Watchdogs
and their monitoring condition can be accessed from within the Alarm Manager.
User defined Evaluators are called Alarm Processors (see section 8.10). Alarm Processors are
only available in VisualAE. An Alarm Processor is fully configurable by a user.
The Alarm Processor has two tabs for setup called Alarm and Comment. Use the Alarm tab for
setting up monitoring condition. Comment tab can be used to give Alarm Processor a
meaningful name (Title field) for referencing it in the Data Processing Structure. Furthermore
Comment field can be used for specific information which will be written to log file if the Alarm
Processor triggers its Signal.
16.3 Signals
A Signal is the interface between Evaluators and Actions. At the input side a Signal is
connected to an Evaluator. At the output side a number of Actions can be connected to a
Figure 30: VisualTR window showing one page with three TR-diagrams each having a client
window
VisualTR offers TR-Diagrams, TR-Pages and TRWindows. The TR-Diagram shows one
waveform in time and/or frequency domain. Any data set of the selected TR-file can be shown in
any TR-Diagram. Each diagram can independently scale, scroll, zoom, split into a time and a
frequency graph, be copied (ASCII or graphical bitmap format) to any other Windows application
via clipboard, and more.
Several diagrams can be grouped into a TR-Page. TR-Pages are groups of TR-diagrams, ideal
for arranging signals for contrast, presentation and/or printout. For each page a header and a
bottom text can be edited and printed. Several pages can be set-up. Tabs are provided for the
easy selection of a page by a single mouse click.
Several pages can be grouped into a TR-Window. TR-Windows contain one or several pages
showing waveforms of the same file. To compare waveforms from different files, more than one
TR-Window can be simultaneously active in VisualTR.
This hierarchy lets the user conveniently compare waveforms of, e.g.:
TR-Filter, TR-Unifier
TR-Filter allows one to apply a digital (software) filter to already recorded data and store the
result. Thereby TR-data recorded with different hardware filters can be unified.
TR-Unifier is a tool to change sample rate and page length (number of samples per waveform)
after recording. This is very useful if data acquired with different settings shall be analyzed with
VisualClass.
Classifier
Classifier provides an interface for employing a VisualClass classifier to transient data.
One of the impressive features of VisualClass is its capability to visualize the separation of
classes in so-called feature-feature projection plots.
The power of VisualClass is based on two points: Its visual transparency and its extensive
automated processing. For each step of the classifier development process VisualClass offers
tools which automatically optimize the results.
When starting a classification, pre-selected prototype data for each class is normally not
available. Sometimes it is not even known how many classes might be separable in one
transient data file. Finding qualified prototype signals out of the data manually is in most cases a
very difficult and time consuming task. VisualClass supports you with its unsupervised learning
strategies:
• Feed a convenient part of your transient data file into VisualClass and specify the feature
extraction parameters for the signals. VisualClass will automatically analyze this data and
search for similarities among the signals. If a set of similar signals is found, then these
signals are collected in a cluster. Frequently, more than one solution can be found and
VisualClass shows all cluster combinations that could be detected in the data.
• All cluster combinations are stored. The user can easily pick out one cluster combination
consisting of several clusters and handle these clusters like pre-defined prototype data.
Now all mentioned methods can be used on these classes to train a classifier.
• When the classifier is built, it can be run on any other transient data file. VisualAE™ can
filter signals out of unknown data which match the classes the closest. With the provided
utility programs these signals can easily be added to the prototype data of the
corresponding classes.
• On the other hand, data can be filtered to identify the waveforms that are furthest from all
classes. These signals can be selected to create a new class. This makes the prototype
data step by step more and more representative for all kind of signals to be expected in
your special Acoustic Emission application.
The first problem may occur more often when the event lifetime in a structure is long compared
with the time needed to travel across the structure. The second and third problem is related. A
saturation limit can be calculated when event lifetime corresponds to the reciprocal of event
rate. Measurable overlapping becomes significant, when these two parameters are within only
several orders of magnitudes from one another.
Adet = Anoise + X
A value of 6dB for X is useful.
A pulsing channel always produces a response to its pulse. It is usually less than responses of
receiving channels because the pulse is fed over an attenuator (36dB) into the AE-channel. This
produces a time stamp of the emitted pulse.
A pulsing table is generated before and after the test in order to prove a constant quality of the
coupling. In case of very long tests a pulsing table can be generated in regular intervals.
In the end the corresponding pulsing tables are compared, i.e. the cell values are compared
pair-wise. If a deviation occurs (differences larger than 5-10dB) the coupling of the individual
channel(s) with large deviation have to be checked again by Hsu-Nielsen sources (see section
20.1.2). If the Hsu-Nielsen source test reveals a degradation of sensitivity a change in long term
coupling quality has to be assumed and if possible accounted for in the analysis.
Automated way
The automated way requires the definition of a location processor, in which the positions of the
sensors are defined. Pulsing tables can use the sensor position information of the location
processor to calculate the speed of sound automatically between each sensor pair.
Large differences in speed of sound in a plate can be caused by influence of wave modes.
Sensors in near vicinity of pulsing channel are more likely to be hit by a fast symmetrical wave
than sensors farther away.
21.1.1 Requirements
SysVeri requires a fully operational AE system model AMSY-5 or AMSY-6 including computer to
control the AMSY (see corresponding AMSY-x system specification for required operating
system and Vallen AE-Suite software release), software Microsoft Word 2000 or later, and
Vallen AE-Suite software release 2005.0401 or later, with software Key ‘SysVeri’ enabled.
For chassis verification, the following is also required:
• A programmable DC-calibrator with RS232 cable (CbD in Fig. 2), to stimulate the
parametric inputs. SysVeri-software supports the control of model “M2000” over RS232.
Software also supports the use of any manually controlled DC-Calibrator, e.g MIVC222-
HPII. Required accuracy: 0,015% at 10V, output range: 0 to +10V.
• BNC cable, BNC-T-connectors and adapter for the calibrator output (usually 4mm banana
sockets) to connect the DC-output voltage of the DC-calibrator to all parametric inputs in
parallel.
• An already verified Reference Board to measure the calibration pulse:
a) ASIP-2/A or ASIP-2/S board (with software releases higher than R2008.0101), or
b) ASIPP board with dummy high-pass (HP4-dum) and dummy low-pass (TP4-dum) filter
module.
For AE-channel verification the following is also required:
• A programmable Arbitrary Function Generator, to stimulate programmed waveforms of AE
bursts. SysVeri-software supports control of model “Agilent 33220A” over Ethernet or USB,
alternatively model “Agilent 33120A” over RS232 with appropriate cable to PC.