DarkLight: Difference between revisions

From DaqWiki
Jump to navigation Jump to search
 
(12 intermediate revisions by the same user not shown)
Line 158: Line 158:
* JMP2 - left (INP)
* JMP2 - left (INP)
* JMP3 - left (TTL input)
* JMP3 - left (TTL input)
* P0 - both right (NIM output)
* P0, P2 - both left (TTL output)
* P1 - both left (TTL output)
* P1, P3 - both right (NIM output)
* SW1 - 0 (VME address)
* SW1 - 0 (VME address)
* SW2 - 0 (VME address)
* SW2 - 0 (VME address)
Line 167: Line 167:


<pre>
<pre>
LEMO_IN  3-4 | not used
LEMO_IN  1-2 | TTL | 10 MHz clock and GEM trigger from DL-TDC
...
LEMO OUT 7-8 | NIM | 40 MHz clock to P.M. DS-DM TTL clock module
LEMO_OUT 5-6 | TTL | 10 MHz clock and GEM trigger to next PPG32
LEMO_OUT 3-4 | NIM | MPD trigger 100 ns long
LEMO_OUT 1-2 | TTL | busy out to DL-TDC
</pre>
</pre>
Note:
* clock and trigger daisy chain is TTL (changed from NIM)


== Firmware sources ==
== Firmware sources ==
Line 218: Line 228:
Gabby Gelinas
Gabby Gelinas
</pre>
</pre>
= TDC calibration pulser =
TDC calibration pulser settings for 50 kHz rate:
* period: 20000
* duty_cycle: 1000
= Procedure for TDC calibrations =
The FPGA based DL-TDC must be calibrated before looking at the data
and these calibrations must be periodically updated:
* record a cosmic run, mid.lz4 file size 500 Mbytes is good
* analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01574.mid.lz4
* examine plots dltdc/file_time/tdcXX_fine_bins_le and _te, they should look like reference plots, for all channels
* examine plots fine_time_le and fine_time_te, if they are not flat and have severe distortions, calibrations must be updated
* make new fine time calibration file: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01574.mid.lz4 -- --dltdc-finetime
* edit dlcfmdb/dltdc_finetime.txt, add an entry for the new run
* re-analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01574.mid.lz4
* reexamine the fine_time_le and _te plots, they should look flat now
* examine plots dltdc16/per-scintillator fine time plots/dltdc_scNN_t_NN_MM_FF. they should look like reference plots with small or no distortions.
* new calibrations are good. commit and push the new fine time calibration json and timetime.txt files in dlcfmdb.
In addition, TDC offsets must be periodically checked and updated as needed:
* record a pulser run:
* pulser at 50 kHz (see settings above)
* sipm vbias off
* file size 500 Mbytes is good
* analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01575.mid.lz4
* examine the tdc calibration plots, most channels should be good, some channels will show distortions, this as expected
* examine plots dltdc/pulser/tdc_pulser_le_vs_tdc_chan, _te_vs_tdc_chan and _width_vs_tdc_chan, they should be flat is in the reference plot
* _width_vs_tdc_chan should show 8 ns for all channels except tdc32 (A) and tdc34 (T), they should be 3 ns wide.
* if values observed are different, the offset calibration must be updated.
* make a new offset calibration file: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01575.mid.lz4 -- --dltdc-offsets
* edit dlcfmdb/dltdc_offsets.txt, add an entry for the new run
* re-analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01575.mid.lz4 -- --dltdc-pulser
* pulser plots should now look as expected
* new calibrations are good. commit and push the new offsets calibration json file and offsets.txt file in dlcfmdb.


= AAA =
= AAA =

Latest revision as of 23:26, 15 September 2025

DarkLight experiment at TRIUMF

Links

Drawings

SVG format - trigger detector diagram

DL-trigger-detector.svg

SVG format - trigger and clock distribution

DL-trigger-clock.svg

SVG format - FPGA trigger logic

DL-trigger-fpga.svg

PDF format:

DAQ

  • ODB /eq/cb02/settings:
  • dl_ctrl for A+B 0xC0FE
  • dl_ctrl for A*B 0xC0F0
  • A is grand-OR of all scintillators on the TDC-A cable (enabled by dl_trg_mask[7..0], 8 bits for the 8 scintillators)
  • B is grand-OR of all scintillators on the TDC-B cable (enabled by dl_trg_mask[15..8], 8 bits for the 8 scintillators)
  • each scintillator signal (A[7..0] and B[7..0]) is L*R hardwired in dl/dl.sv
  • dl_tdc_mask can be used to disable each of the 32 TDC inputs (see map between TDC inputs and scintillator channels)
  • there is no button to disable chronobox counters and rates for individual CB channels

Detector map from Gabby

  • Electron side detector (A-side)
ch16 - tdc04 - B8 - sc08 - B7 - tdc09 - ch08
ch15 - tdc05 - B8 - sc07 - B7 - tdc08 - ch07
ch14 - tdc12 - B6 - sc06 - B5 - tdc03 - ch06
ch13 - tdc13 - B6 - sc05 - B5 - tdc02 - ch05
ch12 - tdc06 - B4 - sc04 - B3 - tdc11 - ch04
ch11 - tdc07 - B4 - sc03 - B3 - tdc10 - ch03
ch10 - tdc14 - B2 - sc02 - B1 - tdc01 - ch02
ch09 - tdc15 - B2 - sc01 - B1 - tdc00 - ch01

TDC cabling map

This TDC map is hardcoded in the analyzer and in the FPGA trigger logic.

Orange and Black detectors TDC cable connections are the same. 

TDC cable A is tdc00..tdc15
TDC cable B is tdc16..tdc31
A signal is tdc32
B signal is tdc33
T = A*B  is tdc34

TDC end  | SiPM | scintillator | SiPM | TDC end
------------------------------------------------
tdc01    |  B1  |  sc01, sc02  |  B2  | tdc14-15
tdc10-11 |  B3  |  sc03, sc04  |  B4  | tdc67
tdc23    |  B5  |  sc05, sc06  |  B6  | tdc12-13
tdc89    |  B7  |  sc07, sc08  |  B8  | tdc45
------------------------------------------------

DLDB port map

DLDB  scintil     channel     SiPM  scint
port  lator
-----------------------------------------
  0 - sc01 sc02 - ch01 ch02 - eB1 - e1 - e2
  1 - sc02 sc01 - ch09 ch10 - eB2 - e2 - e1
  2 - sc03 sc04 - ch03 ch04 - eB3 - e3 - e4
  3 - sc04 sc03 - ch11 ch12 - eB4 - e4 - e3
  4 - sc05 sc06 - ch05 ch06 - eB5 - e5 - e6
  5 - sc06 sc05 - ch13 ch14 - eB6 - e6 - e5
  6 - sc07 sc08 - ch07 ch08 - eB7 - e7 - e8
  7 - sc08 sc07 - ch15 ch16 - eB8 - e8 - e7
  8 - sc09 sc10 - ch17 ch18 - pB1 - p1 - p2
  9 - sc10 sc09 - ch25 ch26 - pB2 - p2 - p1
 10 - sc11 sc12 - ch19 ch20 - pB3 - p3 - p4
 11 - sc12 sc11 - ch27 ch28 - pB4 - p4 - p3
 12 - sc13 sc14 - ch21 ch22 - pB5 - p5 - p6
 13 - sc14 sc13 - ch29 ch30 - pB6 - p6 - p5
 14 - sc15 sc16 - ch23 ch24 - pB7 - p7 - p8
 15 - sc16 sc15 - ch31 ch32 - pB8 - p8 - p7

FPGA and TDC channel map

sc01 - ch01 - tdc00 - GPIO-1-20
sc01 - ch09 - tdc15 - GPIO-1-21
sc02 - ch02 - tdc01 - GPIO-1-22
sc02 - ch10 - tdc14 - GPIO-1-23
sc03 - ch03 - tdc10 - GPIO-1-31
sc03 - ch11 - tdc07 - GPIO-1-34
sc04 - ch04 - tdc11 - GPIO-1-29
sc04 - ch12 - tdc06 - GPIO-1-32
sc05 - ch05 - tdc02 - GPIO-1-24 not GPIO-1-30
sc05 - ch13 - tdc13 - GPIO-1-25
sc06 - ch06 - tdc03 - GPIO-1-26
sc06 - ch14 - tdc12 - GPIO-1-27
sc07 - ch07 - tdc08 - GPIO-1-35
sc07 - ch15 - tdc05 - GPIO-1-30
sc08 - ch08 - tdc09 - GPIO-1-33
sc08 - ch16 - tdc04 - GPIO-1-28

sc09 - ch17 - tdc16 - GPIO-0-15
sc09 - ch25 - tdc31 - GPIO-0-14
sc10 - ch18 - tdc17 - GPIO-0-13
sc10 - ch26 - tdc30 - GPIO-0-12
sc11 - ch19 - tdc26 - GPIO-0-4
sc11 - ch27 - tdc23 - GPIO-0-1
sc12 - sc20 - tdc27 - GPIO-0-6
sc12 - ch28 - tdc22 - GPIO-0-3
sc13 - ch21 - tdc18 - GPIO-0-11
sc13 - ch29 - tdc29 - GPIO-0-10
sc14 - ch22 - tdc19 - GPIO-0-9
sc14 - ch30 - tdc28 - GPIO-0-8
sc15 - ch23 - tdc24 - GPIO-0-0
sc15 - ch31 - tdc21 - GPIO-0-5
sc16 - ch24 - tdc25 - GPIO-0-2
sc16 - ch32 - tdc20 - GPIO_0-7

PPG32-Rev2 settings

Jumper settings

  • JMP1 - left (INP)
  • JMP2 - left (INP)
  • JMP3 - left (TTL input)
  • P0, P2 - both left (TTL output)
  • P1, P3 - both right (NIM output)
  • SW1 - 0 (VME address)
  • SW2 - 0 (VME address)
  • SW3 - C (VME A24 address 0xC0'0000)

Connections

LEMO_IN  3-4 | not used
LEMO_IN  1-2 | TTL | 10 MHz clock and GEM trigger from DL-TDC
...
LEMO OUT 7-8 | NIM | 40 MHz clock to P.M. DS-DM TTL clock module
LEMO_OUT 5-6 | TTL | 10 MHz clock and GEM trigger to next PPG32
LEMO_OUT 3-4 | NIM | MPD trigger 100 ns long
LEMO_OUT 1-2 | TTL | busy out to DL-TDC

Note:

  • clock and trigger daisy chain is TTL (changed from NIM)

Firmware sources

Build firmware

  • ssh trinatdaq ### must be Ubuntu-20 to run quartus 13.1
  • cd /home/olchansk/git/vme-nimio32-dl-clock/VME-NIMIO32/PPG32-Rev1
  • /daq/quartus/13.1.4.182/quartus/bin/quartus
  • open VME-PPG32 project, processing -> start compilation
  • will produce new sof and pof files (jic file must be regenerated manually)
  • load sof file into FPGA using quartus jtag programmer, then load pof file into EPCS16 flasg using srunner, see below

Load firmware

  • cd ~olchansk/git/vme
  • ./srunner_vme_gef.exe -id -16 /dev/null 0x300020 ### identify EPCS chip
  • ./srunner_vme_gef.exe -program -16 /home/olchansk/git/vme-nimio32-dl-clock/VME-NIMIO32/PPG32-Rev1/VME-PPG32.pof 0x300020
  • ./test_VMENIMIO32_gef.exe --addr 0x300000 --reboot

Check clock status

  • ./test_VMENIMIO32_gef.exe --addr 0x300000 --read 9 ### read clock status, bits are as above
  • ./test_VMENIMIO32_gef.exe --addr 0x300000 --write 9 0 ### reset the pll_unlocked_latch

HP LVPS instructions

From: Gabby Gelinas <ggelinas@triumf.ca>
Date: Mon, 23 Jun 2025 20:29:15 +0000

Hi all,

Including the instructions on how to save the start up settings of the low voltage power supplies in case something changes and it needs to be redone again
later.

1. Set the supplies to what you want them to default to
2. On one supply, press the blue button (says “save” above it)
3. Press "0"
4  Press “enter”
5. Repeat steps 2-4 on the other supply
6. Turn both supplies off
7. Turn both supplies on while holding down the “8” button to make sure it worked. You’ll see a message about loading save 0.

Gabby Gelinas

TDC calibration pulser

TDC calibration pulser settings for 50 kHz rate:

  • period: 20000
  • duty_cycle: 1000

Procedure for TDC calibrations

The FPGA based DL-TDC must be calibrated before looking at the data and these calibrations must be periodically updated:

  • record a cosmic run, mid.lz4 file size 500 Mbytes is good
  • analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01574.mid.lz4
  • examine plots dltdc/file_time/tdcXX_fine_bins_le and _te, they should look like reference plots, for all channels
  • examine plots fine_time_le and fine_time_te, if they are not flat and have severe distortions, calibrations must be updated
  • make new fine time calibration file: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01574.mid.lz4 -- --dltdc-finetime
  • edit dlcfmdb/dltdc_finetime.txt, add an entry for the new run
  • re-analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01574.mid.lz4
  • reexamine the fine_time_le and _te plots, they should look flat now
  • examine plots dltdc16/per-scintillator fine time plots/dltdc_scNN_t_NN_MM_FF. they should look like reference plots with small or no distortions.
  • new calibrations are good. commit and push the new fine time calibration json and timetime.txt files in dlcfmdb.

In addition, TDC offsets must be periodically checked and updated as needed:

  • record a pulser run:
  • pulser at 50 kHz (see settings above)
  • sipm vbias off
  • file size 500 Mbytes is good
  • analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01575.mid.lz4
  • examine the tdc calibration plots, most channels should be good, some channels will show distortions, this as expected
  • examine plots dltdc/pulser/tdc_pulser_le_vs_tdc_chan, _te_vs_tdc_chan and _width_vs_tdc_chan, they should be flat is in the reference plot
  • _width_vs_tdc_chan should show 8 ns for all channels except tdc32 (A) and tdc34 (T), they should be 3 ns wide.
  • if values observed are different, the offset calibration must be updated.
  • make a new offset calibration file: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01575.mid.lz4 -- --dltdc-offsets
  • edit dlcfmdb/dltdc_offsets.txt, add an entry for the new run
  • re-analyze: ./dlana.exe --mt --jsroot -R8082 /daq/data17/ehall/data/run01575.mid.lz4 -- --dltdc-pulser
  • pulser plots should now look as expected
  • new calibrations are good. commit and push the new offsets calibration json file and offsets.txt file in dlcfmdb.

AAA

Retrieved from "https://daq00.triumf.ca/DaqWiki/index.php?title=DarkLight&oldid=8584"