Teledyne SP Devices ADQ14-FWPD User manual

ADQ14-FWPD
User Guide
Author(s): Teledyne SP Devices
Document ID: 18-2074
Classification: Public
Revision: PA1
Print date: 2018-04-06

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
Contents
1 Introduction 3
1.1 Definitions&Abbreviations ................................... 3
2 Features 3
2.1 Specification ........................................... 4
2.2 Overview ............................................. 4
3 Getting Started 7
3.1 SDKInstallation ......................................... 7
3.1.1 Installing the SDK (Windows) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.1.2 InstallingtheSDK(Linux) ................................ 7
3.1.3 Unsupported Software Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 CollectingData.......................................... 8
4 Detailed System Description 8
4.1 Digital Baseline Stabilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.2 MovingAverage ......................................... 8
4.3 PulseSpecification........................................ 9
4.3.1 EdgeWindows...................................... 10
4.3.2 BaselineTracking .................................... 12
4.4 TriggerFeatures ......................................... 12
4.4.1 TimestampReset .................................... 12
4.4.2 TriggerBlocking ..................................... 13
4.4.3 SynchronizationSignals................................. 14
4.5 Coincidence............................................ 14
4.6 DetectionWindow ........................................ 15
4.7 DataMultiplexer ......................................... 16
4.8 PulseCharacterization...................................... 16
4.8.1 PulseWidth........................................ 17
4.8.2 PulsePeakValue .................................... 17
4.8.3 Pulse Peak Value Timestamp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.8.4 Overflow ......................................... 17
4.9 Histograms ............................................ 17
4.9.1 Overflow ......................................... 18
4.10Padding.............................................. 19
4.11UserSpaceRecord ....................................... 19
4.11.1FixedLengthRecord................................... 19
4.11.2VariableLengthRecord ................................. 19
4.12DataCollectionModes...................................... 20
4.12.1RawPulseData ..................................... 20
4.12.2EveryNthRecord .................................... 20
4.12.3PulseMetadata...................................... 21
4.12.4 Raw Pulse Data with Padding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 1 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
5 User Application Example 22
5.1 CExample ............................................ 24
5.1.1 PythonScripts ...................................... 24
5.2 DiskStreamingExample..................................... 24
5.3 PulseCharacterizationGUI ................................... 25
6 Troubleshooting 25
6.1 ManagingLicenseFiles ..................................... 25
6.1.1 ReadingtheDNA .................................... 25
6.1.2 Updating the Digitizer License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 2 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
1 Introduction
This document is the user guide for the ADQ14 digitizer running the pulse detection firmware (-FWPD)
option. Equipped with the pulse detection firmware, the ADQ14 digitizer provides the user with sophisti-
cated tools to identify, collect and analyze regions of interest in the data stream in real-time. This relaxes
the requirements on the host computer in terms of sustained bandwidth of the device-to-host interface,
storage space and offline data parsing.
1.1 Definitions & Abbreviations
Table 1lists the definitions and abbreviations used in this document and provides an explanation for each
entry.
Table 1: Definitions and abbreviations used in this document.
Term Explanation
ADC Analog-to-digital converter
API Application programming interface
DBS Digital baseline stabilization
FWPD Pulse detection firmware for ADQ14
GSPS 109samples per second
GUI Graphical user interface
Horizontal parameters Used to describe parameters and settings which are specified in
samples (SI unit seconds). For example, the length of a record
of data.
LEW Leading edge window
MSPS 106samples per second
TEW Trailing edge window
Vertical parameters Used to describe parameters and settings which are specified in
ADC codes (SI unit Volt). For example, the level trigger thresh-
old.
2 Features
This section presents the specification of ADQ14-FWPD along with brief descriptions of some of its core
features. Detailed information may be found in Section 4. Additionally, information pertaining to the
ADQ14 platform in general and not this firmware specifically may be found in the ADQ14 manual [1].
However, if the documents give conflicting information, this document takes precedence for devices
running this particular firmware option.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 3 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
2.1 Specification
The specification for the pulse detection firmware is presented in Table 2. For the general specification
of the ADQ14 digitizer, please refer to the ADQ14 data sheet [2].
Table 2: Specification for ADQ14-FWPD.
Item Min Max Unit
General
Pulse width 1 - Sample
Pulse separation 1 - Sample
Record length (500 MSPS, 1 GSPS) 4 232 - 1 Sample
Record length (2 GSPS) 8 232 - 1 Sample
Pulse characterization
Pulse width 11216 - 1 Sample
Pulse peak timestamp 2-232 - 1 Sample
Average pulse rate (1 GSPS) - 250 106Pulses / s
Average pulse rate (2 GSPS) - 500 106Pulses / s
Average pulse rate for histograms - 33 106Pulses / s
Histogram bin value - 1 048 575 -
1See Section 4.8.1
2See Section 4.8.3
2.2 Overview
Fig. 1presents a block diagram outlining the main features of the data path. The diagram is drawn for a
four-channel device but applies to one- and two-channel devices as well.
•Input range
ADQ14 digitizers with the –VG option offer the possibility to tune the input range of the digitizer.
•DC offset
The analog signal may be subjected to a DC offset. If only pulses with a certain polarity is expected,
the user may wish to offset the analog signal to place the baseline close to the maximum or minimum
ADC code. This allows pulses with amplitudes that would normally saturate the ADC to be safely
collected, thus increasing the dynamic range of the digitizer.
•ADC
The analog signal is sampled at 500 MSPS (–A devices), at 1 GSPS (–C devices) or at 2 GSPS
(–X devices).
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 4 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
Figure 1: A block diagram presenting the main features of the ADQ14-FWPD data path.
•Digital baseline stabilizer
The digital baseline stabilizer (DBS) provides a highly accurate and stable baseline, specifically
intended for pulse applications. Additional information on the DBS is provided in Section 4.1.
•Moving average
The moving average function tracks the baseline of the signal using a configurable window looking
at the last 100 samples. Tracking the baseline allows the user to input the pulse specification rela-
tive to the baseline rather than in absolute ADC codes. The moving average function is discussed
further in Section 4.2.
•Acquisition engine
The acquisition engine is tasked with creating records of data. Each channel is configured inde-
pendently and requires a trigger condition and a specification of the horizontal parameters.
–Trigger
The trigger condition may be selected from one of the three global trigger sources: external
trigger, internal trigger or software trigger. The latter is mainly used for debugging purposes
and has limited practical application. In this case, the selected source is shared among all
channels using the global trigger source. Another option is to use the level trigger, in which
case each channel operates independently since the trigger condition is detected by analyzing
the local data stream. The ways to trigger a record are discussed throughout Section 4,
particularly in Sections 4.4,4.6 and 4.11.
–Coincidence
The coincidence function allows the user to impose additional conditions to the trigger mecha-
nism. Each channel has a configurable logic OR expression which needs to evaluate to true in
order for a record to be generated from a trigger event. An example of a coincidence condition
is provided below.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 5 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
Example
“A trigger on channel A should be accepted and create a record if channel B has detected
a trigger in the last 100 ns.”
In addition to the trigger condition and coincidence masking, the definition of a region of interest
may be refined further by using a detection window (Section 4.6). This window defines a period in
time during which pulses are accepted for analysis. Any pulses outside of the window are rejected
and are prevented from propagating to the user.
•Data multiplexer
The data multiplexer allows the user to select which data stream to input to the pulse characteriza-
tion process. In this way, the data collection modes (discussed in Section 4.12) make it possible to
verify the system by observing the raw pulse data together with its corresponding metadata. The
data multiplexer is discussed in Section 4.7.
•Pulse characterization
The pulse characterization function offers real-time analysis of the pulse data where three key
metrics are extracted: the peak value, the width (time-over-threshold) and the time of the peak value
relative to the trigger timestamp. The pulse metadata format is described in detail in Section 4.12.3.
Additionally, the metadata collection mode (discussed in Section 4.12) allows the user to reduce
the output data rate further by choosing to only output the pulse metadata.
•Histogram
The pulse peak value and width are forwarded to two independent histograms which are tasked
with keeping track of the frequency of occurrence of events of a certain type. Details on the his-
togramming feature may be found in Section 4.9.
•Padding grid
The padding grid defines a window in time which acts as a guide when the digitizer inserts padding
data. This padding data is required in applications which require a certain minimum activity regard-
less of the event frequency while still maintaining a high throughput capacity of the device-to-host
interface.
Inserting padding data into the data stream introduces overhead and thus lowers the effective
bandwidth of the device-to- host interface (the bandwidth used to transfer useful data). This is
an inevitable consequence since the concept of padding targets the trade-off between throughput
and latency. The goal is to provide the user with the tools needed to find a balanced compro-
mise between the application’s requirements. The padding mechanism is described in detail in
Section 4.10.
•Output engine
The output engine offers flexible control over the flow of data. There are five different data collection
modes ranging from only pulse metadata to raw pulse data with padding. The output engine is
described in detail in Section 4.12.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 6 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
3 Getting Started
This section provides information on how to interface with ADQ14 and the pulse detection firmware.
Table 3describes the peripherals and their usage.
Table 3: Peripheral connections on ADQ14.
Connector Description
TRIG External trigger input/output
SYNC Synchronization input/output
A Channel A input (one, two and four-channel devices)
B Channel B input (two and four-channel devices)
C Channel C input (four-channel devices)
D Channel D input (four-channel devices)
CLK External reference clock input
3.1 SDK Installation
The Software Development Kit (SDK) contains the ADQAPI, drivers, examples and documentation re-
quired tor successfully interface with the digitizer. The installation procedure for Microsoft Windows and
Linux is described in the following sections.
3.1.1 Installing the SDK (Windows)
For Microsoft Windows the SDK is installed by running
ADQ-setup_rXXXXX.exe
where XXXXX is the version number. After the installation the example code is located in
<Path to installation directory>/C_examples/
and the documentation in
<Path to installation directory>/Documentation/
3.1.2 Installing the SDK (Linux)
The SDK is supported for a number of Linux distributions and versions. The required files are all included
in
ADQ_SDK_linux_rXXXXX.tar.gz
where XXXXX is the version number. The archive contains the SDK installation files, example code and
documentation. The README file, located in the root directory of the archive, describes the installation
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 7 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
procedure in detail for the different distributions.
3.1.3 Unsupported Software Features
Listed below are the features supported by ADQ14 which are not supported by ADQ14-FWPD,
•ADCaptureLab
•LabView API
3.2 Collecting Data
The FWPD C example is provided together with the SDK. For Windows it is located in
<Path to installation directory>/C_examples/ADQAPI_FWPD_example
after installing the SDK, and for Linux in the
examples/ADQAPI_FWPD_example
directory of the SDK archive. This example illustrates the capabilities of ADQ7-FWPD, and can be used
as-is. However, the example is primarily intended to be used as a guide for creating applications tailored
to a specific use case.
4 Detailed System Description
This section provides additional details on important system features.
4.1 Digital Baseline Stabilization
For optimal performance in pulse applications, it is advantageous if the baseline of the signal is locked
to a specific value. This is the purpose of the Digital Baseline Stabilizer (DBS). The signal processing
block is placed directly after the ADC in the data path.
The DBS uses a blind identification process operating in the background. When activated by the user,
DBS will constantly monitor and follow variations in the baseline and correct for time-varying behavior
such as temperature variations. The baseline is adjusted to the target value with 22-bit precision.
4.2 Moving Average
The moving average function computes an average of samples earlier in time using a memory with
capacity to store the previous 100 samples. From this data set, the user can select a range of samples
which should be used to compute the moving average value attributed to the current sample. The range
is specified as a window which has a length and an offset from the current sample. Both the length and
the offset may be given in the range from 0 to 100 samples. However, the sum of the two parameters
may not exceed 100 samples.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 8 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
The refresh rate of the moving average value is tied to the device sample rate. For –A, –C and –X
devices the value is updated every 2, 4 and 8 samples, i.e. every 4 ns.
The purpose of the moving average function is to allow the user to input the pulse specification
(discussed in Section 4.3) relative to the baseline rather than in absolute ADC codes.
4.3 Pulse Specification
In order for the system to sift through the constant stream of data separating regions of interest from
uninteresting regions, the user has to tune the system to look for pulses fulfilling a certain criteria—the
pulse specification.
A pulse is defined by a trigger level and three hysteresis (offset) values: the reset hysteresis, the
trigger arm hysteresis and the reset arm hysteresis. These four values are specified by the user together
with a polarity setting, indicating if the target pulse is
•positive, i.e. the pulse grows toward higher ADC codes from the baseline or
•negative, i.e. the pulse grows toward lower ADC codes from the baseline.
The arithmetic involved in computing the absolute detection levels of the pulse differs depending on the
polarity. Equations (1) and (2) present the relations for positive and negative pulses, respectively.
Positive pulses
Reset level =Trigger level -Reset hysteresis
Trigger arm level =Trigger level -Trigger arm hysteresis
Reset arm level =Reset level +Reset arm hysteresis
(1)
Negative pulses
Reset level =Trigger level +Reset hysteresis
Trigger arm level =Trigger level +Trigger arm hysteresis
Reset arm level =Reset level -Reset arm hysteresis
(2)
Fig. 2places the levels discussed above into the context of a positive pulse. The figure consists of five
subfigures:
•Fig. 2a—the reset level and events
The reset level is computed by subtracting the reset hysteresis from the trigger level. When the
rising edge of the pulse crosses the trigger level a trigger event is generated. Conversely, when
the falling edge of the pulse crosses the reset level a reset event is generated. An event always
coincides with a sample point and is chosen as the first sample on or beyond the threshold.
These two events are the heart of the pulse detection firmware as they frame the region of
interest, aiding subsequent functions such as pulse characterization.
•Fig. 2b—the trigger arm level
The trigger arm level is computed by subtracting the trigger arm hysteresis from the trigger level.
In order for the rising edge to generate a trigger event, the system has to be armed. For a positive
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 9 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
pulse, this means that the signal has to visit ADC codes equal to or less than the trigger arm level.
Following a trigger event, the system is automatically disarmed until the arming criteria is met once
again.
•Fig. 2c—the reset arm level
The arming mechanism for the reset event is identical to that of the trigger event apart from one
detail: they are mirror versions of each other. In order for the falling edge of the pulse to generate a
reset event, the signal has to visit ADC codes equal to or greater than the reset arm level. Following
a reset event, the system is automatically disarmed until the arming criteria is met once again.
•Figs. 2d and 2e—false events
The arming mechanisms are what enables the protection of the data stream against false events.
However, this protection hinges on the hysteresis values input by the user to work correctly. As
shown in Fig. 2d, badly configured arm levels may have the same effect as not having them at all.
The arming levels should be set at least in the same range as the noise to be effective, as
shown in Fig. 2e.
For completeness, the conditions to generate a trigger or reset event are stated in (3) and (4) for positive
and negative pulses, respectively. The data at time instance nis denoted x[n].
Positive pulses
Trigger event: x[n]≥Trigger level
Reset event: x[n]≤Reset level
(3)
Negative pulses
Trigger event: x[n]≤Trigger level
Reset event: x[n]≥Reset level
(4)
4.3.1 Edge Windows
The values discussed in the previous section target the vertical characteristics of a pulse. How the
horizontal characteristics of a pulse are treated by the pulse detection firmware are discussed in more
detail in Sections 4.11 and 4.12. However, this section will introduce the leading edge window (LEW)
and the trailing edge window (TEW). The names hint at their purpose; the former defines a window in
time before the trigger event while the latter defines a window after the reset event, as shown in Fig.3.
These windows extend the region of interest outward from the pulse.
Note
The leading edge window and trailing edge window are two fixed-length regions that surround the
pulse and allow the user to extend the region of interest.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 10 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
(a) The reset level is determined from the trigger level and
the reset hysteresis.
(b) The trigger arm level is determined from the trigger
level and the trigger arm hysteresis. The level is used to
arm the mechanism generating trigger events.
(c) The reset arm level is determined from the reset level
and the reset arm hysteresis. The level is used to arm the
mechanism generating reset events.
(d) Badly configured arm levels may yield false events from
the incorrectly detecting pulses within the signal noise.
(e) Setting the arm levels greater than the noise variation
protects against false events.
Figure 2: Figs. 2a–2c demonstrates the pulse definition in the context of a positive pulse. Figs. 2d and 2e
highlights the purpose of the arm levels.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 11 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
Figure 3: The leading edge window and the trailing edge windows are used to extend the region of
interest relative to the trigger event and reset event, respectively.
4.3.2 Baseline Tracking
The baseline tracking offered by the moving average function (Section 4.2) allows the user to specify
the trigger level relative to the baseline instead of in absolute ADC codes. Since all the other levels are
measured relative to the trigger level, a necessary feature is locking the baseline value when a trigger
event occurs. This ensures that the relative specification remains constant for the duration of the pulse
(possibly longer if variable length mode is active, see Sections 4.12.1 and 4.11.2). The value remains
locked until the end of the trailing edge window, at which point the baseline tracking is resumed.
Note
The baseline is tracked up until the moment of a trigger event, where its current value is locked. The
value remains locked until the trailing edge window has been acquired.
4.4 Trigger Features
This section describes features relating to the external trigger interface: blocking triggers, generating
synchronization signals to external equipment and resetting the internal timestamp of the digitizer. In-
formation regarding how a trigger event is used to acquire data is discussed in other sections, namely
Sections 4.6,4.11 and 4.12.
4.4.1 Timestamp Reset
The timestamp reset feature allows the user to use an external signal to reset the timebase of the digitizer.
This function is used to synchronize the timebase of multiple digitizers or to relate the timebase to some
external event, useful for applications involving some repeated process.
The timestamp reset engine listens for external events on either the external trigger connector (TRIG)
or the synchronization connector (SYNC). The selected source will be referred to as the timestamp reset
source throughout this section. Each trigger source has an associated edge specification to select if the
rising and/or falling edges should generate trigger events. This global edge specification is used by the
timestamp reset engine and may be set by calling SetTriggerEdge().
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 12 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
Controlling the trigger blocking feature involves three functions in the ADQAPI: SetupTimeStamp-
Sync(),ArmTimeStampSync() and DisarmTimeStampSync(). The timestamp reset feature can be con-
figured in two modes:
First event Mode index 0
Once the timestamp reset engine is armed, the first event observed on the timestamp reset source
will reset the timestamp.
Every event Mode index 1
Once the timestamp reset engine is armed, all subsequent events observed on the timestamp
reset source will reset the timestamp.
4.4.2 Trigger Blocking
The trigger blocking function generates a blocking window that may be used together with the timestamp
reset feature to ensure that the digitizer does not create records with unsynchronized timestamps.
The trigger blocking engine listens for external events on either the external trigger connector (TRIG)
or the synchronization connector (SYNC). The selected source will be referred to as the trigger blocking
source throughout this section. The trigger blocking engine uses the global edge specification set by
calling SetTriggerEdge().
Controlling the trigger blocking feature involves three functions in the ADQAPI: SetupTriggerBlock-
ing(),ArmTriggerBlocking() and DisarmTriggerBlocking(). The blocking function can be config-
ured in four modes:
Once Mode index 0
Once the trigger blocking engine is armed, this mode inhibits the creation of records until the first
event on the trigger blocking source has been observed. At this point, the blocking is disengaged
and all subsequent triggers are allowed to propagate.
Window Mode index 1
Once the trigger blocking engine is armed, this mode inhibits the creation of records until an event
on the trigger blocking source has been observed. Following an event, the blocking is disengaged
for a fixed duration (the window length) and triggers are allowed to propagate. Any additional
events on the trigger blocking source that occurs during the window are ignored, i.e. they do not
reset the window.
Gate Mode index 2
Once the trigger blocking engine is armed, this mode inhibits the creation of records until an event
on the trigger blocking source has been observed. Following an event, the blocking is disengaged
for as long as the trigger blocking source signal remains ‘logic high’ or ‘logic low’ (depending on
the source’s edge specification) and is reengaged once an event of the opposite type is detected.
Inverse window Mode index 3
Once the trigger blocking engine is armed, this mode allows triggers to propagate until an event
on the trigger blocking source has been observed. Following an event, the blocking is engaged
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 13 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
for a fixed duration (the window length), preventing triggers from propagating and creating records
during this time.
The trigger blocking feature can be queried by the user for the number of blocking events by calling
GetTriggerBlockingGateCount(). Additionally, the information is included in the record header and
represent the gate counter’s value at the time the record was triggered. Combining the timestamp reset
and trigger blocking features allow the user to keep track of the number of times the timestamp has been
reset. In this way, the data from two batches separated by one or several resets of the timestamp are
able to be related to each other in time.
Note
The timestamp reset and trigger blocking features may be combined to make the gate counter count
the number of timestamp resets.
4.4.3 Synchronization Signals
In applications where the digitizer is used to generate the trigger signal to other equipment, the method
of choice is to use the internal trigger and output this periodic pulse on the external trigger connector
(TRIG). The frame synchronization feature introduces an additional signal synchronized to the internal
trigger which may be output on the synchronization connector (SYNC). This signal acts as a frame for
the free-running trigger signal and has three configuration parameters: the frame length, the frame factor
and the internal trigger edge to use as the reference point for the frame. These parameters are specified
by calling the function SetupFrameSync(). Additionally, the output has to be activated by calling Enable-
FrameSync().
The frame generator listens to the internal trigger signal counting the number of edges matching the
edge type specified by the user. Once this number is equal to the frame factor, the output is pulled high
for a duration equal to the frame’s length. Fig. 4presents an example scenario.
SYNC
(Frame sync output)
TRIG
(Internal trigger output)
Frame length
Frame factor
Frame length
Frame factor
Figure 4: An illustration of the frame synchronization feature. In the depicted scenario, the frame factor
is set to 7.
4.5 Coincidence
The coincidence feature allows the user to further refine the definition of a region of interest. It offers
a mechanism to constrain the conditions of acquiring a pulse by querying the other channels for recent
events.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 14 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
Each channel has an associated mask and a window length. The masking expression is interpreted
as a bit mask specifying the conditions in which a trigger event is accepted for the channel ‘owning’ the
mask.
A bit in the mask has a position and a value. The position indicates to which channel a coincidence
condition should be defined. The value 1implies a connection while 0implies no connection. For ex-
ample, if 0b1100 is used as a mask for channel A, this indicates that a trigger event on channel A must
coincide with recent events on channels C and D in order to be allowed to propagate. The recency is
determined by the window length of the coinciding channels.
Non-causal coincidence conditions are not supported. That is, the user cannot specify that a pulse
should be accepted on channel A only if there is a pulse on channel B in the next 100 ns.
Important
Non-causal coincidence conditions are not supported.
Example
On a four-channel 1 GSPS device (–4C), channel A has the coincidence mask 0b1010 and a window
length of 1000 samples. The mask and window length for channel B is 0b0101 and 600 samples. The
masks for channels C and D are the default values 0b0100 and 0b1000 and the window lengths are
both 800 samples.
This configuration will only detect a pulse on channel A if any of the following conditions are met:
•There has been an event on channel B in the previous 600 ns.
•There has been an event on channel D in the previous 800 ns.
To detect a pulse on channel B, any of the conditions below needs to be met:
•There has been an event on channel A in the previous 1000 ns.
•There has been an event on channel C in the previous 800 ns.
Since the coincidence masks of channels C and D reference themselves, every event will be accepted
on these channels.
The coincidence function is controlled by the API functions PDSetupTriggerCoincidence() and PD-
EnableTriggerCoincidence(). The latter must be called regardless if the user intends to use the feature
or not.
Important
The user is required to call PDEnableTriggerCoincidence() regardless if the feature is used or not.
To disable the feature, the function’s argument should be 0.
4.6 Detection Window
A detection window defines a period in time in which pulses are accepted and forwarded to the pulse
characterization engine. The detection window has a fixed length and an associated trigger event which
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 15 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
may be the output of the global trigger multiplexer or the per-channel level trigger. Fig. 5demonstrates
detection window where two corner cases are highlighted: any pulse not completely inside the detection
window is rejected. From a technical point of view, completely is defined as both the trigger event and
the reset event (Section 4.3) occurring inside the window.
Figure 5: An example of how a detection window is used to mask undesired pulses. Pulses marked with
a green check mark are accepted. Pulses marked with a red x-mark are rejected.
How the detection window is set up depends on the data collection mode. Refer to Section 4.12 for
details.
4.7 Data Multiplexer
The data multiplexer is used to select any of the input data channels and forward the stream of data
to any of the characterization channels. The multiplexer allows the user to output a single data stream
on multiple channels and may be used to, e.g. verify that the pulse metadata corresponds to the raw
pulse data. Each channel of the data multiplexer can be configured to select data from any of the input
channels.
The multiplexer is controlled with the API call PDSetDataMux()
Example
To connect output channel B (channel number 2) to input channel A (channel number 1), issue
PDSetDataMux(1, 2)
The data multiplexer configuration is reset when the clock configuration is changed, i.e. calling SetClock-
Source() will reset the input channel multiplexer configuration.
4.8 Pulse Characterization
The pulse characterization feature is used to compute key pulse metrics in hardware. The metrics are:
•the pulse width (time-over-threshold),
•the pulse peak timestamp and
•the pulse peak value (extreme value).
These metrics form a pulse metadata packet described in Section 4.12.2.
Pulses are characterized during the detection window, as illustrated in Fig. 5where pulses starting
before the window trigger or ending after the window ends are ignored.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 16 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
4.8.1 Pulse Width
The pulse width is computed as the number of samples between the pulse trigger and reset level. This
value will wrap if the pulse width is larger than 216 - 1 samples.
Note
The pulse width metadata value will wrap if the pulse length is larger than 216 - 1 samples.
4.8.2 Pulse Peak Value
The maximum value or the minimum value of the pulse. For positive pulses the peak is defined as the
maximum value, and for negative pulses the peak is defined as the minimum value
The peak is a 16-bit signed value.
4.8.3 Pulse Peak Value Timestamp
The peak value timestamp is the sample index at which the extreme value occurred, relative to the record
header timestamp. That is, the time in samples between the detection window trigger and the extreme
value.
The timestamp is a 32-bit unsigned value and will wrap if the distance between the detection window
trigger and the pulse is larger than 232 - 1 samples.
If the peak value is reached at multiple samples, the timestamp value will report the last one.
Note
The timestamp value will wrap after 232 - 1 samples.
4.8.4 Overflow
The pulse characterization module can on average process 500 million pulses / second at 2 GSPS and
250 million pulses / second at 1 GSPS. The module will buffer up to 2048 pulses at 2 GSPS (1024 pulses
at 1 GSPS) after which the characterization module overflows.
An overflow condition will set a sticky status bit which can be read with the API call PDGetCharacterizationStatus().
The status bit is cleared when calling PDSetupCharacterization().
Note
The pulse characterization module can, on average, process 500 million pulses / second at 2 GSPS
and 250 million pulses / second at 1 GSPS.
4.9 Histograms
The firmware histogram module provides histograms for the width and peak value of the pulse. The
histograms operation is independent of the data collection mode.
The number of histogram bins are listed in Table 4. The histogram bins are represented by a 20-bit
unsigned number and will saturate at the maximum value (1 048 575). The data is mapped to a bin
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 17 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
number by
Bin[n] =(x+δ)·α
1024 ,(5)
where xis the histogram metadata value, δis an offset and αis a scale factor.
In addition to the normal histogram data bins, each histogram has an overflow and underflow bin.
These bins are used if the resulting bin number from (5) is outside of the allowed range. That is if the bin
number is smaller than zero the underflow bin will be incremented by one. Conversely, if the bin number
is larger the total number of bins the overflow bin will be incremented by one.
The histogram underflow and overflow bins are accessed by calling PDGetHistogramStatus().
Example
Let a certain pulse have the peak value -5000 and the peak value histogram be configured with scale
factor α= 1024 and the offset δ= 4000. The resulting bin number would be -1000, from (5). Since
this is less than 0, the underflow bin would be incremented by one.
The API calls used to interface with the histograms are listed below. Please refer to the ADQAPI reference
guide [3] for details on the usage.
•PDReadHistogram()
•PDReadHistogramStatus()
•PDClearHistogram()
•PDSetupHistogram()
4.9.1 Overflow
The average processing bandwidth of the histogram is roughly 33 million pulses / second. For short
bursts of pulses a buffering FIFO is implemented. If the number of pulses exceeds the processing band-
width, a overflow flag will be set. The flag can be accessed by calling ReadHistogramStatus().
The overflow status is reset by clearing the histograms.
Note
The histogram can process 33 million pulses / second on average. The ReadHistogramStatus() API
call can be used to detect an overflow.
Table 4: The number of bins for the different histogram types.
Type Number of bins
Pulse peak value 16384
Pulse width 4096
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 18 of 28

Classification Revision
Public PA1
Document ID Print date
18-2074 2018-04-06
4.10 Padding
The data padding functionality is used to keep a constant data rate from the digitizer. The padding is
active in the data collection modes pulse metadata and raw pulse data with padding.
The padding block will append zeros to the data to ensure that the total amount of data measured
over one padding frame is at least equal to the minimum frame length. Setting the minimum frame length
to zero disables the padding.
Note
The padding functionality is disabled by setting the minimum record length to zero.
The padding is applied to each padding frame after a certain offset. The padding offset is equal to
the detection window length in pulse metadata mode. The padding is configured using the API call
PDSetupCharacterization(). How the padding frame is set up depends on the data collection mode,
see Table 5.
For pulse metadata mode the padding data is appended to the data record, whereas for the raw pulse
data with padding mode a new record is generated with the padding data.
4.11 User Space Record
User space records are the records of data received by the user. A record consists of a data part and a
header part. The header includes information about e.g. the length and the trigger timestamp.
The contents of the user space records will depend on the data collection mode. In all modes but
the metadata mode, the record will consist of the data sampled by the ADC. In the metadata mode, the
record will consist of the pulse metadata packets.
The raw data records can either have a fixed length or a variable length. Fig. 6illustrates the dif-
ferences between the two record types. The length of a metadata record will depend of the number of
pulses detected.
4.11.1 Fixed Length Record
A fixed length record is defined by the trigger event and the record length. The length of a fixed length
record is specified by the user, and is independent of the data collected. A fixed length record can be
triggered by any of the trigger sources.
4.11.2 Variable Length Record
The length of a variable length record will depend on the data. A variable length record is defined as the
data between the trigger event and reset event. If trailing or leading edge windows are used, these will
also be included in the record.
If two pulses have overlapping trailing and leading edge windows, they will be concatenated into one
single record. This is illustrated in Fig. 6. The variable length mode can only be used with the level
trigger.
ADQ14-FWPD – User Guide www.teledyne-spdevices.com Page 19 of 28