ADC Simulation driver

March 20, 2016

Mark Rivers

University of Chicago

 

Table of Contents

Introduction

This is an EPICS areaDetector driver for a simulated ADC driver. This driver generates times-series data for 8 signals as a 2-D array [8, NumTimePoints]. The signals are common waveforms (sine, cosine, square wave, sawtooth, random noise, etc.) The driver was written in part to test the new NDPluginTimeSeries plugin in ADCore R2-5. But it can also serve as a starting point for writing drivers that are used to collect time-series data like ADCs, waveform digitizers, etc.

Each signal in the driver describes a waveform. The equations governing these signals are similar. The equation for the sine wave is the following:

    Signal[t] = Offset + Noise * random() + Amplitude * sin((time * Frequency + Phase/360.) * 2. * PI);
    

This driver inherits from asynNDArrayDriver. It also implements a number parameters that are specific to the ADC simulation detector. The ADCSimDetector class documentation describes this class in detail.

Simulation driver specific parameters

The simulation driver-specific parameters are the following:

Parameters for the driver as a whole
Parameter Definitions in ADCSimDetector.cpp and EPICS Record Definitions in ADCSimDetector.template
Parameter index variable asyn interface Access Description drvInfo string EPICS record name EPICS record type
P_Acquire asynInt32 r/w Start/stop acquisition. Values are:
0: "Done": Stops acquisition
1: "Acquire": Starts acquisition
SIM_ACQUIRE $(P)$(R)Acquire
$(P)$(R)Acquire_RBV
busy
bi
P_TimeStep asynFloat64 r/w Time per data point SIM_TIME_STEP $(P)$(R)TimeStep
$(P)$(R)TimeStep_RBV
ao
ai
NumTimePoints asynInt32 r/w Number of time points to accumulate before calling registered plugins SIM_NUM_TIME_POINTS $(P)$(R)NumTimePoints
$(P)$(R)NumTimePoints_RBV
longout
longin
P_AcquireTime asynFloat64 r/w Total elapsed time before automatically stopping acquisition. If 0 then acquistion continues until manually stopped. Note that this is "simulated time", i.e. the number of time points since acquisition was started * TimeStep. This may differ from actual time because of the processor speeds, granularity of epicsThreadSleep(), etc. SIM_ACQUIRE_TIME $(P)$(R)AcquireTime
$(P)$(R)AcquireTime_RBV
ao
ai
P_ElapsedTime asynFloat64 r/w Total elapsed time since acquisition was started. SIM_ELAPSED_TIME $(P)$(R)ElapsedTime ai
Parameters for each signal N.
Parameter Definitions in NDPluginTimeSeries.h and EPICS Record Definitions in ADCSimDetectorN.template
P_Amplitude asynFloat64 r/w The amplitude of this signal. SIM_AMPLITUDE $(P)$(R)Amplitude ao
P_Offset asynFloat64 r/w The offset of this signal. SIM_OFFSET $(P)$(R)Offset ao
P_Phase asynFloat64 r/w The phase of the signal in degrees. SIM_PHASE $(P)$(R)Phase ao
P_Noise asynFloat64 r/w The noise of this signal. This value is multipled by a random number between -1 and 1 for each time point. SIM_NOISE $(P)$(R)Noise ao
P_Period asynFloat64 r/w The period of this signal. The frequency is calculated from the period. SIM_PERIOD $(P)$(R)Period ao
P_Frequency asynFloat64 r/o The frequency of this signal, calculated from the period. SIM_FREQUENCY $(P)$(R)Frequency ai

Configuration

The ADCimDetector driver is created with the ADCimDetectorConfig command, either from C/C++ or from the EPICS IOC shell.

int ADCSimDetectorConfig(const char *portName,
                      int numTimePoints, int dataType,
                      int maxBuffers, size_t maxMemory,
                      int priority, int stackSize)
  

The simDetector-specific fields in this command are:

For details on the meaning of the other parameters to this function refer to the detailed documentation on the simDetectorConfig function in the ADCSimDetector.cpp documentation and in the documentation for the constructor for the ADCSimDetector class.

There an example IOC boot directory and startup script (iocBoot/iocADCSimDetector/st.cmd) provided with ADExample.

MEDM screens

The following is the MEDM screen simDetector.adl for the simulation detector.

ADCSimDetector.adl

ADCSimDetector.png

The NDPluginTimeSeries is used to collect the time series data from the ADCSimDetector, perform optional averaging. In this case it is being used in circular buffer mode. The NDArrays it produces are being used by the NDPluginFFT plugin to compute Fast Fourier Transforms.

NDPluginTimeSeries.adl

ADCSimDetectorTimeSeriesPlugin.png

This is a plot for the sin(x)*cos(x) output. This screen contains some of the controls for the time series plugin.

NDTimeSeriesPlot.adl

ADCSimDetectorSinCosPlot.png

This is a combined plot for the square wave output. It contains the time-series data, as well as the real, imaginary, and absolute value of the FFT. Note that the square wave FFT contains the fundamental frequency of 5 Hz, and all of the odd harmonics.

NDPluginTimeSeries.adl

ADCSimDetectorSquareWavePlotAll.png