| ID |
Date |
Author |
Subject |
|
52
|
Thu Mar 11 11:45:52 2010 |
Stefan Ritt | Readout of DRS Data |
| Hao Huan wrote: |
|
Hi Stefan,
thanks to your help I can now successfully keep the Domino wave running at a stable frequency and maintain the channel cascading information in the Write Shift Register. (Since you told me WSR always reads and writes at the same time, I think I need to rewrite the information back every time after reading out from WSR to decide from which channel my data come, don't I?)
|
Yes you do. But if you have WSRLOOP=1 in the config register, this is done automatically. So the SR output is visible at the pin and will be fed back into the input.
| Hao Huan wrote: |
|
However I'm still having difficulty in reading out from the DRS cells. I use the ROI readout mode and assume as long as I give a pulse on RSRLOAD the data will come out one by one.
|
That's not correct. Have a look at Figure 14 of the datasheet. Do you see a single RSRLOAD pulse or many? There is only one RSRLOAD pulse to initialize the readout shift register, then the cells are clocked by SRCLK pulses.
| Hao Huan wrote: |
|
Also I read in the datasheet that WSROUT will give RSR output when DWRITE is low. Sometimes I see some random bits from this output and sometimes I see all zero's. What is the reasonable output I should see from WSROUT, say, when I'm running in the transparent mode with DWRITE low?
|
A single RSRLOAD pulse loads the RSR with a "1" at the domino stop position and "0" in all other places. A pulse on SRCLK shifts this "1" down the RSR. When it arrives at cell #1023, it will be visible for one clock cycle at WSROUT. The "double" functionality of WSROUT has the following background: Assume you use channel cascading 2x2048. Now the domino wave stopps in cell 1020 of the first channel for example. You have to read cells 1020,1021,1022,1024 of the first channel, then you continue with 0,1,2 on the second channel. But how do you know that you have to switch channels after the first four clock cycles? The SROUT output encodes the stop position (in this case 1020), but it needs 10 clock cycles before the information is available, so you don't have it after four cycles. That's where WSROUT comes into play: Since it outputs RSR bit by bit, it will show three "0", then a "1", when you are at cell 1023. Then you know that you have to switch channels immediately. That's why I output RSR via WSROUT if DWRITE is low.
|
|
676
|
Thu Mar 22 14:36:01 2018 |
Phan Van Chuan | Read the CalibrateWaveform | Helo
I'm building an application for reading waveforms from the DRS4 board to PC. However, I am having problems reading calibration data from EEPROM on DRS4 board. The calibration data is read through the function reference:
void DRSBoard :: ReadCalibration (void)
...
ReadEEPROM (1, buf, 1024 * 32);
for (i = 0; i <8; i ++)
for (j = 0; j <1024; j ++) {
fCellOffset [i] [j] = buf [(i * 1024 + j) * 2];
fCellGain [i] [j] = buf [(i * 1024 + j) * 2 + 1] / 65535.0*0.4+0.7;
}
ReadEEPROM (2, buf, 1024 * 32);
for (i = 0; i <8; i ++)
for (j = 0; j <1024; j ++)
fCellOffset2 [i] [j] = buf [(i * 1024 + j) * 2];
...
The Calibrate Waveform is performed by:
int DRSBoard::CalibrateWaveform(unsigned int chipIndex, unsigned char channel, unsigned short *adcWaveform, short *waveform, bool responseCalib, int triggerCell, bool adjustToClock, float threshold, bool offsetCalib)
.....
for (j = 0; j < n_bins; j++) {
value = adcWaveform[j] - fCellOffset[channel+chipIndex*9][(j*skip + triggerCell) % kNumberOfBins];
value = value / fCellGain[channel+chipIndex*9][(j*skip + triggerCell) % kNumberOfBins];
if (offsetCalib && channel != 8)
value = value - fCellOffset2[channel+chipIndex*9][j*skip] + 32768;
...
. Because the calibration data reads incorrectly, the Calibrate Waveform does not do it.
Can read calibration data from EEPROM by any command via Oscilloscope application or DRS Command Line Interface application?
Thank you for your help!!!! |
|
677
|
Fri Mar 23 09:39:55 2018 |
Stefan Ritt | Read the CalibrateWaveform | You don't have to read and calibrate the waveforms in your user code, but can rely on the DRS.cpp library to do that. Just look at the drs_exam.cpp program coming with the distribution. It uses the function b->GetWave() to retrieve the calibrated waveform. If you like, you can look into that function to learn how to apply the calibration, but I can tell you that it's a bit complicated. Since each event starts at an arbitrary stop cell in the DRS4, you have to "rotate" the calibration array. Then you do actually four calibrations in a row (cell, readout, gain and range).
Stefan
| Phan Van Chuan wrote: |
|
Helo
I'm building an application for reading waveforms from the DRS4 board to PC. However, I am having problems reading calibration data from EEPROM on DRS4 board. The calibration data is read through the function reference:
void DRSBoard :: ReadCalibration (void)
...
ReadEEPROM (1, buf, 1024 * 32);
for (i = 0; i <8; i ++)
for (j = 0; j <1024; j ++) {
fCellOffset [i] [j] = buf [(i * 1024 + j) * 2];
fCellGain [i] [j] = buf [(i * 1024 + j) * 2 + 1] / 65535.0*0.4+0.7;
}
ReadEEPROM (2, buf, 1024 * 32);
for (i = 0; i <8; i ++)
for (j = 0; j <1024; j ++)
fCellOffset2 [i] [j] = buf [(i * 1024 + j) * 2];
...
The Calibrate Waveform is performed by:
int DRSBoard::CalibrateWaveform(unsigned int chipIndex, unsigned char channel, unsigned short *adcWaveform, short *waveform, bool responseCalib, int triggerCell, bool adjustToClock, float threshold, bool offsetCalib)
.....
for (j = 0; j < n_bins; j++) {
value = adcWaveform[j] - fCellOffset[channel+chipIndex*9][(j*skip + triggerCell) % kNumberOfBins];
value = value / fCellGain[channel+chipIndex*9][(j*skip + triggerCell) % kNumberOfBins];
if (offsetCalib && channel != 8)
value = value - fCellOffset2[channel+chipIndex*9][j*skip] + 32768;
...
. Because the calibration data reads incorrectly, the Calibrate Waveform does not do it.
Can read calibration data from EEPROM by any command via Oscilloscope application or DRS Command Line Interface application?
Thank you for your help!!!!
|
|
|
393
|
Mon Nov 17 16:36:18 2014 |
Mickey Chiu | Raspberry Pi drsosc does not exit properly | When running drsosc on a raspberry pi, it seems the exit doesn't seem to work at all. This is true for the "exit" button on the window, or the file menu exit, or the "x" on the window. I end up having to kill drsosc manually from the command line. This wouldn't be such a bad thing except that it doesn't seem to store any settings when killed in this way. I'm wondering if anyone else sees the same thing, or if there is a fix out there, before I go and delve into why. |
|
394
|
Tue Nov 25 14:06:34 2014 |
Stefan Ritt | Raspberry Pi drsosc does not exit properly |
| Mickey Chiu wrote: |
|
When running drsosc on a raspberry pi, it seems the exit doesn't seem to work at all. This is true for the "exit" button on the window, or the file menu exit, or the "x" on the window. I end up having to kill drsosc manually from the command line. This wouldn't be such a bad thing except that it doesn't seem to store any settings when killed in this way. I'm wondering if anyone else sees the same thing, or if there is a fix out there, before I go and delve into why.
|
Unfortunately I don't have a pi here right now, so I cannot reproduce your problem. I checked on a linux system and it worked fine with wxWidgets 3.0.1 and GTK2 2.20. The wxWidget library sends an wxID_EXIT event to DOFrame::OnExit, which then closes the window. The destructor of DOFrame then calls SaveConfig() to save the current settings. Maybe you can debug this.
/Stefan |
|
631
|
Fri Oct 13 03:39:01 2017 |
Jonathan Wapman | Raspberry Pi Connection Failure | I am currently attempting to use a raspberry pi to connect to the DRS 4 board. I whenever I try to use the DRS Command Line TOol, Revision 21435 to connect to the drs board, I get the error
"musb_open: libusb_open() error -3"
"USB successfully scanned, but no boards found"
"No DRS Boards Found".
I successfully compiled the libusb driver before compiling the drs software 5.0.6, and installed all other listed packages in the install instructions. |
|
632
|
Mon Oct 16 15:35:22 2017 |
Stefan Ritt | Raspberry Pi Connection Failure | Have you tried as root? Maybe you miss some permissions.
Stefan
| Jonathan Wapman wrote: |
|
I am currently attempting to use a raspberry pi to connect to the DRS 4 board. I whenever I try to use the DRS Command Line TOol, Revision 21435 to connect to the drs board, I get the error
"musb_open: libusb_open() error -3"
"USB successfully scanned, but no boards found"
"No DRS Boards Found".
I successfully compiled the libusb driver before compiling the drs software 5.0.6, and installed all other listed packages in the install instructions.
|
|
|
76
|
Wed May 5 22:30:50 2010 |
Ignacio Diéguez Estremera | Random noise spec in datasheet | Hi,
According to DRS4's datasheet, the random noise is 0.35mVrms. Is this the input equivalent noise voltage? It is computed over the 0-950MHz frequency band?
Regards. |
|
77
|
Thu May 6 08:15:39 2010 |
Stefan Ritt | Random noise spec in datasheet |
| Ignacio Diéguez Estremera wrote: |
|
Hi,
According to DRS4's datasheet, the random noise is 0.35mVrms. Is this the input equivalent noise voltage? It is computed over the 0-950MHz frequency band?
Regards.
|
You cannot compare the DRS4 noise directly with an amplifier for example. The noise mainly comes from variations of the charge injection into the storage cells, and some noise during the readout process, which happens in a completely different frequency domain than the sampling.
So what I did is to keep the inputs open, measure a 1024-bin waveform, and compute the RMS of this waveform. So I believe that this is kind of equivalent noise voltage from 1-950 MHz. It does not start from zero since very low frequency noise (like 50 Hz) just causes a baseline shift and does not influence the RMS, but this is not so important since in most applications people do an event-by-event baseline subtraction to get rid of low frequency noise in their apparatus. The 0.35 mV RMS also depend on the electronics around the chip. On our USB evaluation board the noise it typically smaller (0.31 mV RMS), while in some VME board we measure 0.42 mV RMS. If you do the perfect analog design around the chip, you can maybe push this maybe even lower. |
|
262
|
Tue Jun 18 14:19:39 2013 |
Stefan Ritt | ROOT program to decode binary data from DRSOsc | Please find attached a simple ROOT based program (http://root.cern.ch) to decode binary data from the DRSOsc program. It assumes that all four channels were recorded. If this is not the case, the program can be adjusted accordingly.
To use it, simply type (assuming that you have written a data file "test.dat" with DRSOsc):
root [0] .L decode.C+
Info in <TUnixSystem::ACLiC>: creating shared library /tmp/./decode_C.so
root [1] decode("test");
Info in <TCanvas::MakeDefCanvas>: created default TCanvas with name c1
1927 events processed
"test.root" written
root [2]
If you have turned on the clock on channel4 of the DRS4 evaluation board, it will produce a plot like this:
/Stefan |
| Attachment 1: decode.C
|
#include <string.h>
#include <stdio.h>
#include "TFile.h"
#include "TTree.h"
#include "TString.h"
#include <iostream>
struct Header_t {
char event_header[4];
unsigned int serial_number;
unsigned short year;
unsigned short month;
unsigned short day;
unsigned short hour;
unsigned short minute;
unsigned short second;
unsigned short millisecond;
unsigned short reserved1;
float time[1024];
};
struct Waveform_t {
char chn1_header[4];
unsigned short chn1[1024];
char chn2_header[4];
unsigned short chn2[1024];
char chn3_header[4];
unsigned short chn3[1024];
char chn4_header[4];
unsigned short chn4[1024];
};
void decode(char *filename) {
Header_t header;
Waveform_t waveform;
Double_t t[1024], chn1[1024], chn2[1024], chn3[1024], chn4[1024];
Int_t n;
// open the binary waveform file
FILE *f = fopen(Form("%s.dat", filename), "r");
//open the root file
TFile *outfile = new TFile(Form("%s.root", filename), "RECREATE");
// define the rec tree
TTree *rec = new TTree("rec","rec");
rec->Branch("t", &t ,"t[1024]/D");
rec->Branch("chn1", &chn1 ,"chn1[1024]/D");
rec->Branch("chn2", &chn2 ,"chn2[1024]/D");
rec->Branch("chn3", &chn3 ,"chn3[1024]/D");
rec->Branch("chn4", &chn4 ,"chn4[1024]/D");
// loop over all events in data file
for (n=0 ; fread(&header, sizeof(header), 1, f) > 0; n++) {
// decode time
for (Int_t i=0; i<1024; i++)
t[i] = (Double_t) header.time[i];
fread(&waveform, sizeof(waveform), 1, f);
// decode amplitudes in mV
for (Int_t i=0; i<1024; i++) {
chn1[i] = (Double_t) ((waveform.chn1[i]) / 65535. - 0.5) * 1000;
chn2[i] = (Double_t) ((waveform.chn2[i]) / 65535. - 0.5) * 1000;
chn3[i] = (Double_t) ((waveform.chn3[i]) / 65535. - 0.5) * 1000;
chn4[i] = (Double_t) ((waveform.chn4[i]) / 65535. - 0.5) * 1000;
}
rec->Fill();
}
// draw channel #4
rec->Draw("chn4:t");
// print number of events
cout<<n<<" events processed"<<endl;
cout<<"\""<<Form("%s.root", filename)<<"\" written"<<endl;
// save and close root file
rec->Write();
outfile->Close();
}
|
|
361
|
Wed Jul 30 17:05:06 2014 |
Stefan Ritt | ROOT program to decode binary data from DRSOsc |
| Stefan Ritt wrote: |
|
Please find attached a simple ROOT based program (http://root.cern.ch) to decode binary data from the DRSOsc program. It assumes that all four channels were recorded. If this is not the case, the program can be adjusted accordingly.
To use it, simply type (assuming that you have written a data file "test.dat" with DRSOsc):
root [0] .L decode.C+
Info in <TUnixSystem::ACLiC>: creating shared library /tmp/./decode_C.so
root [1] decode("test");
Info in <TCanvas::MakeDefCanvas>: created default TCanvas with name c1
1927 events processed
"test.root" written
root [2]
If you have turned on the clock on channel4 of the DRS4 evaluation board, it will produce a plot like this:
/Stefan
|
I updated this ROOT program for the new format used with the V5 boards. It's now called "read_binary.C". Usage stays the same. There is also a standalone C program "read_binary.cpp". Both are attached. |
| Attachment 1: read_binary.C
|
/*
Name: read_binary.C
Created by: Stefan Ritt <stefan.ritt@psi.ch>
Date: July 30th, 2014
Purpose: Example program under ROOT to read a binary data file written
by the DRSOsc program. Decode time and voltages from waveforms
and display them as a graph. Put values into a ROOT Tree for
further analysis.
To run it, do:
- Crate a file test.dat via the "Save" button in DRSOsc
- start ROOT
root [0] .L read_binary.C+
root [1] decode("test.dat");
*/
#include <string.h>
#include <stdio.h>
#include "TFile.h"
#include "TTree.h"
#include "TString.h"
#include "TGraph.h"
#include "TCanvas.h"
#include "Getline.h"
typedef struct {
char time_header[4];
char bn[2];
unsigned short board_serial_number;
} THEADER;
typedef struct {
char event_header[4];
unsigned int event_serial_number;
unsigned short year;
unsigned short month;
unsigned short day;
unsigned short hour;
unsigned short minute;
unsigned short second;
unsigned short millisecond;
unsigned short reserved1;
char bs[2];
unsigned short board_serial_number;
char tc[2];
unsigned short trigger_cell;
} EHEADER;
/*-----------------------------------------------------------------------------*/
void decode(char *filename) {
THEADER th;
EHEADER eh;
char hdr[4];
unsigned short voltage[1024];
double waveform[4][1024], time[4][1024];
float bin_width[4][1024];
char rootfile[256];
int i, j, ch, n, chn_index;
double t1, t2, dt;
// open the binary waveform file
FILE *f = fopen(Form("%s", filename), "r");
if (f == NULL) {
printf("Cannot find file \'%s\'\n", filename);
return;
}
//open the root file
strcpy(rootfile, filename);
if (strchr(rootfile, '.'))
*strchr(rootfile, '.') = 0;
strcat(rootfile, ".root");
TFile *outfile = new TFile(rootfile, "RECREATE");
// define the rec tree
TTree *rec = new TTree("rec","rec");
rec->Branch("t1", time[0] ,"t1[1024]/D");
rec->Branch("t2", time[1] ,"t2[1024]/D");
rec->Branch("t3", time[2] ,"t3[1024]/D");
rec->Branch("t4", time[3] ,"t4[1024]/D");
rec->Branch("w1", waveform[0] ,"w1[1024]/D");
rec->Branch("w2", waveform[1] ,"w2[1024]/D");
rec->Branch("w3", waveform[2] ,"w3[1024]/D");
rec->Branch("w4", waveform[3] ,"w4[1024]/D");
// create canvas
TCanvas *c1 = new TCanvas();
// create graph
TGraph *g = new TGraph(1024, (double *)time[0], (double *)waveform[0]);
// read time header
fread(&th, sizeof(th), 1, f);
printf("Found data for board #%d\n", th.board_serial_number);
// read time bin widths
memset(bin_width, sizeof(bin_width), 0);
for (ch=0 ; ch<5 ; ch++) {
fread(hdr, sizeof(hdr), 1, f);
if (hdr[0] != 'C') {
// event header found
fseek(f, -4, SEEK_CUR);
break;
}
i = hdr[3] - '0' - 1;
printf("Found timing calibration for channel #%d\n", i+1);
fread(&bin_width[i][0], sizeof(float), 1024, f);
}
// loop over all events in data file
for (n=0 ; n<5 ; n++) {
// read event header
i = fread(&eh, sizeof(eh), 1, f);
if (i < 1)
break;
printf("Found event #%d\n", eh.event_serial_number);
// reach channel data
for (ch=0 ; ch<5 ; ch++) {
i = fread(hdr, sizeof(hdr), 1, f);
if (i < 1)
break;
if (hdr[0] != 'C') {
// event header found
fseek(f, -4, SEEK_CUR);
break;
}
chn_index = hdr[3] - '0' - 1;
fread(voltage, sizeof(short), 1024, f);
for (i=0 ; i<1024 ; i++) {
// convert data to volts
waveform[chn_index][i] = (voltage[i] / 65536. - 0.5);
// calculate time for this cell
for (j=0,time[chn_index][i]=0 ; j<i ; j++)
time[chn_index][i] += bin_width[chn_index][(j+eh.trigger_cell) % 1024];
}
}
// align cell #0 of all channels
t1 = time[0][(1024-eh.trigger_cell) % 1024];
for (ch=1 ; ch<4 ; ch++) {
t2 = time[ch][(1024-eh.trigger_cell) % 1024];
dt = t1 - t2;
for (i=0 ; i<1024 ; i++)
time[ch][i] += dt;
}
// fill root tree
rec->Fill();
// fill graph
for (i=0 ; i<1024 ; i++)
g->SetPoint(i, time[0][i], waveform[0][i]);
// draw graph and wait for user click
g->Draw("ACP");
c1->Update();
gPad->WaitPrimitive();
}
// print number of events
printf("%d events processed, \"%s\" written.\n", n, rootfile);
// save and close root file
rec->Write();
outfile->Close();
}
|
| Attachment 2: read_binary.cpp
|
/*
Name: read_binary.cpp
Created by: Stefan Ritt <stefan.ritt@psi.ch>
Date: July 30th, 2014
Purpose: Example file to read binary data saved by DRSOsc.
Compile and run it with:
gcc -o read_binary read_binary.cpp
./read_binary <filename>
This program assumes that a pulse from a signal generator is split
and fed into channels #1 and #2. It then calculates the time difference
between these two pulses to show the performance of the DRS board
for time measurements.
$Id: read_binary.cpp 21438 2014-07-30 15:00:17Z ritt $
*/
#include <stdio.h>
#include <fcntl.h>
#include <unistd.h>
#include <string.h>
#include <math.h>
typedef struct {
char time_header[4];
char bn[2];
unsigned short board_serial_number;
} THEADER;
typedef struct {
char event_header[4];
unsigned int event_serial_number;
unsigned short year;
unsigned short month;
unsigned short day;
unsigned short hour;
unsigned short minute;
unsigned short second;
unsigned short millisecond;
unsigned short reserved1;
char bs[2];
unsigned short board_serial_number;
char tc[2];
unsigned short trigger_cell;
} EHEADER;
/*-----------------------------------------------------------------------------*/
int main(int argc, const char * argv[])
{
THEADER th;
EHEADER eh;
char hdr[4];
unsigned short voltage[1024];
double waveform[4][1024], time[4][1024];
float bin_width[4][1024];
char rootfile[256];
int i, j, ch, n, chn_index;
double t1, t2, dt;
char filename[256];
int ndt;
double threshold, sumdt, sumdt2;
if (argc > 1)
strcpy(filename, argv[1]);
else {
printf("Usage: read_binary <filename>\n");
return 0;
}
// open the binary waveform file
FILE *f = fopen(filename, "r");
if (f == NULL) {
printf("Cannot find file \'%s\'\n", filename);
return 0;
}
// read time header
fread(&th, sizeof(th), 1, f);
printf("Found data for board #%d\n", th.board_serial_number);
// read time bin widths
memset(bin_width, sizeof(bin_width), 0);
for (ch=0 ; ch<5 ; ch++) {
fread(hdr, sizeof(hdr), 1, f);
if (hdr[0] != 'C') {
// event header found
fseek(f, -4, SEEK_CUR);
break;
}
i = hdr[3] - '0' - 1;
printf("Found timing calibration for channel #%d\n", i+1);
fread(&bin_width[i][0], sizeof(float), 1024, f);
}
// initialize statistics
ndt = 0;
sumdt = sumdt2 = 0;
// loop over all events in the data file
for (n= 0 ; ; n++) {
// read event header
i = fread(&eh, sizeof(eh), 1, f);
if (i < 1)
break;
printf("Found event #%d\n", eh.event_serial_number);
// reach channel data
for (ch=0 ; ch<5 ; ch++) {
i = fread(hdr, sizeof(hdr), 1, f);
if (i < 1)
break;
if (hdr[0] != 'C') {
// event header found
fseek(f, -4, SEEK_CUR);
break;
}
chn_index = hdr[3] - '0' - 1;
fread(voltage, sizeof(short), 1024, f);
for (i=0 ; i<1024 ; i++) {
// convert data to volts
waveform[chn_index][i] = (voltage[i] / 65536. - 0.5);
// calculate time for this cell
for (j=0,time[chn_index][i]=0 ; j<i ; j++)
time[chn_index][i] += bin_width[chn_index][(j+eh.trigger_cell) % 1024];
}
}
// align cell #0 of all channels
t1 = time[0][(1024-eh.trigger_cell) % 1024];
for (ch=1 ; ch<4 ; ch++) {
t2 = time[ch][(1024-eh.trigger_cell) % 1024];
dt = t1 - t2;
for (i=0 ; i<1024 ; i++)
time[ch][i] += dt;
}
t1 = t2 = 0;
threshold = 0.3;
// find peak in channel 1 above threshold
for (i=0 ; i<1022 ; i++)
if (waveform[0][i] < threshold && waveform[0][i+1] >= threshold) {
t1 = (threshold-waveform[0][i])/(waveform[0][i+1]-waveform[0][i])*(time[0][i+1]-time[0][i])+time[0][i];
break;
}
// find peak in channel 2 above threshold
for (i=0 ; i<1022 ; i++)
if (waveform[1][i] < threshold && waveform[1][i+1] >= threshold) {
t2 = (threshold-waveform[1][i])/(waveform[1][i+1]-waveform[1][i])*(time[1][i+1]-time[1][i])+time[1][i];
break;
}
// calculate distance of peaks with statistics
if (t1 > 0 && t2 > 0) {
ndt++;
dt = t2 - t1;
sumdt += dt;
sumdt2 += dt*dt;
}
}
// print statistics
printf("dT = %1.3lfns +- %1.1lfps\n", sumdt/ndt, 1000*sqrt(1.0/(ndt-1)*(sumdt2-1.0/ndt*sumdt*sumdt)));
return 1;
}
|
|
747
|
Fri Mar 8 19:35:11 2019 |
Abaz Kryemadhi | ROOT Macro for newest software | The older root macro did not work for me for data acquired with the newest software.
so for the newest software and multiple boards, I modified the read_binary.cpp into read_binary.C for those who like to use the root macro, see the attachment.
|
| Attachment 1: read_binary.C
|
/*
Name: read_binary.C
Created by: Stefan Ritt <stefan.ritt@psi.ch>
Date: July 30th, 2014
Modified By: Abaz Kryemadhi
Date: March 7th, 2019
Purpose: Example program under ROOT to read a binary data file written
by the DRSOsc program. Decode time and voltages from waveforms
and display them as a graph. Put values into a ROOT Tree for
further analysis.
To run it, do:
- Crate a file test.dat via the "Save" button in DRSOsc
- start ROOT (type root)
root [0] .L read_binary.C+
root [1] decode("test.dat");
*/
#include <fcntl.h>
#include <unistd.h>
#include <math.h>
#include <string.h>
#include <stdio.h>
#include "TFile.h"
#include "TTree.h"
#include "TString.h"
#include "TGraph.h"
#include "TCanvas.h"
#include "Getline.h"
#include "TAxis.h"
typedef struct {
char tag[3];
char version;
} FHEADER;
typedef struct {
char time_header[4];
} THEADER;
typedef struct {
char bn[2];
unsigned short board_serial_number;
} BHEADER;
typedef struct {
char event_header[4];
unsigned int event_serial_number;
unsigned short year;
unsigned short month;
unsigned short day;
unsigned short hour;
unsigned short minute;
unsigned short second;
unsigned short millisecond;
unsigned short range;
} EHEADER;
typedef struct {
char tc[2];
unsigned short trigger_cell;
} TCHEADER;
typedef struct {
char c[1];
char cn[3];
} CHEADER;
/*-----------------------------------------------------------------------------*/
//int main(int argc, const char * argv[])
void decode(char *filename) {
FHEADER fh;
THEADER th;
BHEADER bh;
EHEADER eh;
TCHEADER tch;
CHEADER ch;
unsigned int scaler;
unsigned short voltage[1024];
double waveform[16][4][1024], time[16][4][1024];
float bin_width[16][4][1024];
int i, j, b, chn, n, chn_index, n_boards;
double t1, t2, dt;
//char filename[256];
char rootfile[256];
int ndt;
double threshold, sumdt, sumdt2;
double sum, baseline, max,amplitude1,amplitude2, amplitude3,amplitude4;
// open the binary waveform file
FILE *f = fopen(filename, "rb");
if (f == NULL) {
printf("Cannot find file \'%s\'\n", filename);
return;
}
//open the root file
strcpy(rootfile, filename);
if (strchr(rootfile, '.'))
*strchr(rootfile, '.') = 0;
strcat(rootfile, ".root");
TFile *outfile = new TFile(rootfile, "RECREATE");
// define the rec tree
TTree *rec = new TTree("rec","rec");
rec->Branch("t1", time[0][0] ,"t1[1024]/D");
rec->Branch("t2", time[0][1] ,"t2[1024]/D");
rec->Branch("t3", time[0][2] ,"t3[1024]/D");
rec->Branch("t4", time[0][3] ,"t4[1024]/D");
rec->Branch("w1", waveform[0][0] ,"w1[1024]/D");
rec->Branch("w2", waveform[0][1] ,"w2[1024]/D");
rec->Branch("w3", waveform[0][2] ,"w3[1024]/D");
rec->Branch("w4", waveform[0][3] ,"w4[1024]/D");
rec->Branch("amplitude1", &litude1,"amplitude1/D");
rec->Branch("amplitude2", &litude2,"amplitude2/D");
rec->Branch("amplitude3", &litude3,"amplitude3/D");
rec->Branch("amplitude4", &litude4,"amplitude4/D");
// create canvas
TCanvas *c1 = new TCanvas();
// create graph
TGraph *g = new TGraph(1024, (double *)time[0][0], (double *)waveform[0][0]);
// read file header
fread(&fh, sizeof(fh), 1, f);
if (fh.tag[0] != 'D' || fh.tag[1] != 'R' || fh.tag[2] != 'S') {
printf("Found invalid file header in file \'%s\', aborting.\n", filename);
return;
}
if (fh.version != '2') {
printf("Found invalid file version \'%c\' in file \'%s\', should be \'2\', aborting.\n", fh.version, filename);
return;
}
// read time header
fread(&th, sizeof(th), 1, f);
if (memcmp(th.time_header, "TIME", 4) != 0) {
printf("Invalid time header in file \'%s\', aborting.\n", filename);
return;
}
for (b = 0 ; ; b++) {
// read board header
fread(&bh, sizeof(bh), 1, f);
if (memcmp(bh.bn, "B#", 2) != 0) {
// probably event header found
fseek(f, -4, SEEK_CUR);
break;
}
printf("Found data for board #%d\n", bh.board_serial_number);
// read time bin widths
memset(bin_width[b], sizeof(bin_width[0]), 0);
for (chn=0 ; chn<5 ; chn++) {
fread(&ch, sizeof(ch), 1, f);
if (ch.c[0] != 'C') {
// event header found
fseek(f, -4, SEEK_CUR);
break;
}
i = ch.cn[2] - '0' - 1;
printf("Found timing calibration for channel #%d\n", i+1);
fread(&bin_width[b][i][0], sizeof(float), 1024, f);
// fix for 2048 bin mode: double channel
if (bin_width[b][i][1023] > 10 || bin_width[b][i][1023] < 0.01) {
for (j=0 ; j<512 ; j++)
bin_width[b][i][j+512] = bin_width[b][i][j];
}
}
}
n_boards = b;
// initialize statistics
ndt = 0;
sumdt = sumdt2 = 0;
// loop over all events in the data file
for (n=0 ; ; n++) {
// read event header
i = (int)fread(&eh, sizeof(eh), 1, f);
if (i < 1)
break;
printf("Found event #%d %d %d\n", eh.event_serial_number, eh.second, eh.millisecond);
// loop over all boards in data file
for (b=0 ; b<n_boards ; b++) {
// read board header
fread(&bh, sizeof(bh), 1, f);
if (memcmp(bh.bn, "B#", 2) != 0) {
printf("Invalid board header in file \'%s\', aborting.\n", filename);
return;
}
// read trigger cell
fread(&tch, sizeof(tch), 1, f);
if (memcmp(tch.tc, "T#", 2) != 0) {
printf("Invalid trigger cell header in file \'%s\', aborting.\n", filename);
return;
}
if (n_boards > 1)
printf("Found data for board #%d\n", bh.board_serial_number);
// reach channel data
for (chn=0 ; chn<4 ; chn++) {
// read channel header
fread(&ch, sizeof(ch), 1, f);
if (ch.c[0] != 'C') {
// event header found
fseek(f, -4, SEEK_CUR);
break;
}
chn_index = ch.cn[2] - '0' - 1;
fread(&scaler, sizeof(int), 1, f);
fread(voltage, sizeof(short), 1024, f);
for (i=0 ; i<1024 ; i++) {
// convert data to volts
waveform[b][chn_index][i] = (voltage[i] / 65536. + eh.range/1000.0 - 0.5);
// calculate time for this cell
for (j=0,time[b][chn_index][i]=0 ; j<i ; j++)
time[b][chn_index][i] += bin_width[b][chn_index][(j+tch.trigger_cell) % 1024];
}
}
// align cell #0 of all channels
t1 = time[b][0][(1024-tch.trigger_cell) % 1024];
for (chn=1 ; chn<4 ; chn++) {
t2 = time[b][chn][(1024-tch.trigger_cell) % 1024];
dt = t1 - t2;
for (i=0 ; i<1024 ; i++)
time[b][chn][i] += dt;
}
t1 = t2 = 0;
threshold = 0.3;
// find peak in channel 1 above threshold
for (i=0 ; i<1022 ; i++)
if (waveform[b][0][i] < threshold && waveform[b][0][i+1] >= threshold) {
t1 = (threshold-waveform[b][0][i])/(waveform[b][0][i+1]-waveform[b][0][i])*(time[b][0][i+1]-time[b][0][i])+time[b][0][i];
break;
}
// find peak in channel 2 above threshold
for (i=0 ; i<1022 ; i++)
if (waveform[b][1][i] < threshold && waveform[b][1][i+1] >= threshold) {
t2 = (threshold-waveform[b][1][i])/(waveform[b][1][i+1]-waveform[b][1][i])*(time[b][1][i+1]-time[b][1][i])+time[b][1][i];
break;
}
// calculate distance of peaks with statistics
if (t1 > 0 && t2 > 0) {
ndt++;
dt = t2 - t1;
sumdt += dt;
sumdt2 += dt*dt;
}
//Find baseline for channel 3 to get amplitude for ch3
sum=0.0;
for (i=0 ; i<10; i++) {
sum+=waveform[0][2][i];
}
baseline=sum/10;
//Find amplitude for channel 3 (this is example channel )
max=-10000.0;
for (i=0 ; i<1022; i++) {
if (waveform[b][2][i]>max) {
max=waveform[b][2][i];
}
}
amplitude3=max;
// fill root tree
rec->Fill();
//Uncomment the following to see couple waveforms of voltage vs time
/*
// fill graph
for (i=0 ; i<1024 ; i++)
g->SetPoint(i, time[b][2][i], waveform[b][2][i]);
// draw graph and wait for user click
... 20 more lines ...
|
|
727
|
Tue Jan 29 14:43:44 2019 |
Abaz Kryemadhi | ROOT Macro for data acquired with the newest software | Hello,
Is there a root macro for decoding binary data acquired with the newest software for single board or multi-boards daisy chained?
Cheers,
Abaz |
|
730
|
Wed Jan 30 17:08:58 2019 |
Stefan Ritt | ROOT Macro for data acquired with the newest software | This one elog:361 should still work.
Stefan
| Abaz Kryemadhi wrote: |
|
Hello,
Is there a root macro for decoding binary data acquired with the newest software for single board or multi-boards daisy chained?
Cheers,
Abaz
|
|
|
663
|
Fri Mar 2 18:08:55 2018 |
Steven Block | ROI | Hello,
I have a question about how ROI works. From what I have read, it will only save data that ocurs some time [ta] dictated by the user after an event is triggered as well as a small time [tb] before the event. The technical manual seems to indicated that the deadtime assciated with operating in ROI mode can be reduced by the following factor:
 .
Where N is the number of points in the time window (ex. 2048 or 1024). Is it ok to describe this as:
Where N' is the number of samples in the ROI and N is the same as before.
For example, if I were running at 5Gsps (200ps between samples), only recording 1024 samples per event and I had an signal that lasted 2ns, that means the signal would last 10 samples. If I set the ROI to only save 20 samples around this signal, would my Deadtime go to:
 ? (The second portion of this equation comes from a response I recieved earlier, but I just want to make sure I understand this concept properly)
I recognize that the caveat is that this would work only if the signal was detected during acquistion, which leads to my next question. If no signals were detected in the 1024*200ps time frame in ROI mode, would the DRS4 go dead for 32us (using the factor = 1 from above equation), or would it dump the earliest events in the buffer for the more recent ones until it detects a signal?
Finally, I assume this functionality can only be utilized with custom electornics with the DRS4, not the evaulation/demo board, please let me know if this is the case.
Best,
Steven |
|
664
|
Fri Mar 2 20:17:17 2018 |
Stefan Ritt | ROI | N'/N is correct. The 2 us "from the response you got from me" come from the fact that after readout, you have to start the DRS4 again. During this time, the power supply usually becomes slightly unstable, and it takes on the evaluation board about 2us to stabilize it again. Tha't why I add the 2 us. If you don't care about slight offset effect, or if you make a better power supply, you dead time would be 10*30ns = 300ns for 10 samples. Starting the DRS again will take one or two clock cycles from the FPGA, which might add another 30 ns or so, depending on how you program the FPGA. So the best you can achieve for 10 samples is maybe 330 ns, if you have a really good power supply (large capacitors).
You can achieve this functionality with the evaluation board, but you would have to make a special firmware for it.
Stefan
| Steven Block wrote: |
|
Hello,
I have a question about how ROI works. From what I have read, it will only save data that ocurs some time [ta] dictated by the user after an event is triggered as well as a small time [tb] before the event. The technical manual seems to indicated that the deadtime assciated with operating in ROI mode can be reduced by the following factor:
.
Where N is the number of points in the time window (ex. 2048 or 1024). Is it ok to describe this as:
Where N' is the number of samples in the ROI and N is the same as before.
For example, if I were running at 5Gsps (200ps between samples), only recording 1024 samples per event and I had an signal that lasted 2ns, that means the signal would last 10 samples. If I set the ROI to only save 20 samples around this signal, would my Deadtime go to:
? (The second portion of this equation comes from a response I recieved earlier, but I just want to make sure I understand this concept properly)
I recognize that the caveat is that this would work only if the signal was detected during acquistion, which leads to my next question. If no signals were detected in the 1024*200ps time frame in ROI mode, would the DRS4 go dead for 32us (using the factor = 1 from above equation), or would it dump the earliest events in the buffer for the more recent ones until it detects a signal?
Finally, I assume this functionality can only be utilized with custom electornics with the DRS4, not the evaulation/demo board, please let me know if this is the case.
Best,
Steven
|
|
|
665
|
Fri Mar 2 21:05:48 2018 |
Steven Block | ROI | Great! That is very helpful.
One more question. If no signals were detected in the 1024*200ps time frame in ROI mode, would the DRS4 go dead for 32us (or 30us depending on the supply) for, or would it dump the earliest events in the buffer for the more recent ones until it detects a signal to readout? Or rather, does filling the buffer force a readout or can it dynamically shift out old data until it detects a signal to readout.
Steven
| Stefan Ritt wrote: |
|
N'/N is correct. The 2 us "from the response you got from me" come from the fact that after readout, you have to start the DRS4 again. During this time, the power supply usually becomes slightly unstable, and it takes on the evaluation board about 2us to stabilize it again. Tha't why I add the 2 us. If you don't care about slight offset effect, or if you make a better power supply, you dead time would be 10*30ns = 300ns for 10 samples. Starting the DRS again will take one or two clock cycles from the FPGA, which might add another 30 ns or so, depending on how you program the FPGA. So the best you can achieve for 10 samples is maybe 330 ns, if you have a really good power supply (large capacitors).
You can achieve this functionality with the evaluation board, but you would have to make a special firmware for it.
Stefan
| Steven Block wrote: |
|
Hello,
I have a question about how ROI works. From what I have read, it will only save data that ocurs some time [ta] dictated by the user after an event is triggered as well as a small time [tb] before the event. The technical manual seems to indicated that the deadtime assciated with operating in ROI mode can be reduced by the following factor:
.
Where N is the number of points in the time window (ex. 2048 or 1024). Is it ok to describe this as:
Where N' is the number of samples in the ROI and N is the same as before.
For example, if I were running at 5Gsps (200ps between samples), only recording 1024 samples per event and I had an signal that lasted 2ns, that means the signal would last 10 samples. If I set the ROI to only save 20 samples around this signal, would my Deadtime go to:
? (The second portion of this equation comes from a response I recieved earlier, but I just want to make sure I understand this concept properly)
I recognize that the caveat is that this would work only if the signal was detected during acquistion, which leads to my next question. If no signals were detected in the 1024*200ps time frame in ROI mode, would the DRS4 go dead for 32us (using the factor = 1 from above equation), or would it dump the earliest events in the buffer for the more recent ones until it detects a signal?
Finally, I assume this functionality can only be utilized with custom electornics with the DRS4, not the evaulation/demo board, please let me know if this is the case.
Best,
Steven
|
|
|
|
675
|
Mon Mar 19 16:22:42 2018 |
Stefan Ritt | ROI | The DRS4 has an internal storage of 1024 capacitors. They work as a ring buffer, so at 5GSPS you can store 200ns wide signals. After 200ns, the first samples are overwritten by new samples, so you always have the last 200ns of samples stored. Once you trigger the DRS4, this buffer is frozen, and the readout of this buffer causes the dead time. No trigger, no dead time. Hope this answers your question.
Stefan
| Steven Block wrote: |
|
Great! That is very helpful.
One more question. If no signals were detected in the 1024*200ps time frame in ROI mode, would the DRS4 go dead for 32us (or 30us depending on the supply) for, or would it dump the earliest events in the buffer for the more recent ones until it detects a signal to readout? Or rather, does filling the buffer force a readout or can it dynamically shift out old data until it detects a signal to readout.
Steven
| Stefan Ritt wrote: |
|
N'/N is correct. The 2 us "from the response you got from me" come from the fact that after readout, you have to start the DRS4 again. During this time, the power supply usually becomes slightly unstable, and it takes on the evaluation board about 2us to stabilize it again. Tha't why I add the 2 us. If you don't care about slight offset effect, or if you make a better power supply, you dead time would be 10*30ns = 300ns for 10 samples. Starting the DRS again will take one or two clock cycles from the FPGA, which might add another 30 ns or so, depending on how you program the FPGA. So the best you can achieve for 10 samples is maybe 330 ns, if you have a really good power supply (large capacitors).
You can achieve this functionality with the evaluation board, but you would have to make a special firmware for it.
Stefan
| Steven Block wrote: |
|
Hello,
I have a question about how ROI works. From what I have read, it will only save data that ocurs some time [ta] dictated by the user after an event is triggered as well as a small time [tb] before the event. The technical manual seems to indicated that the deadtime assciated with operating in ROI mode can be reduced by the following factor:
.
Where N is the number of points in the time window (ex. 2048 or 1024). Is it ok to describe this as:
Where N' is the number of samples in the ROI and N is the same as before.
For example, if I were running at 5Gsps (200ps between samples), only recording 1024 samples per event and I had an signal that lasted 2ns, that means the signal would last 10 samples. If I set the ROI to only save 20 samples around this signal, would my Deadtime go to:
? (The second portion of this equation comes from a response I recieved earlier, but I just want to make sure I understand this concept properly)
I recognize that the caveat is that this would work only if the signal was detected during acquistion, which leads to my next question. If no signals were detected in the 1024*200ps time frame in ROI mode, would the DRS4 go dead for 32us (using the factor = 1 from above equation), or would it dump the earliest events in the buffer for the more recent ones until it detects a signal?
Finally, I assume this functionality can only be utilized with custom electornics with the DRS4, not the evaulation/demo board, please let me know if this is the case.
Best,
Steven
|
|
|
|
|
59
|
Tue Mar 30 22:57:34 2010 |
Hao Huan | ROFS Configuration | Hi Stefan,
according to the DRS4 datasheet, if we want an input range centered around U0, the ROFS should be 1.55V-U0. However when I read the codes of the evaluation board application, ROFS seems to be 1.6V-1.25*U0 where the coefficient 1.25 is said to come from sampling cell charge injection correction. Is it the right equation to use? What exactly does that charge injection correction mean?
Thanks a lot.
|
|
68
|
Thu Apr 15 13:48:40 2010 |
Stefan Ritt | ROFS Configuration |
| Hao Huan wrote: |
|
Hi Stefan,
according to the DRS4 datasheet, if we want an input range centered around U0, the ROFS should be 1.55V-U0. However when I read the codes of the evaluation board application, ROFS seems to be 1.6V-1.25*U0 where the coefficient 1.25 is said to come from sampling cell charge injection correction. Is it the right equation to use? What exactly does that charge injection correction mean?
Thanks a lot.
|
1.55V-U0 is the theoretical values, but there are certain "dirt" effects like chip-to-chip variation and charge injection. The difference between various chips is easily 20-30mV, so there is not a single "correct" value. The formula 1.6V-1.25*U0 I developed for a special evaluation board, where it kind of worked better than the theoretical value, but I never made systematic studies. One should average over several chips and use some solid average there. Best is if you try both formulas and check what give you the better linearity. |
|