Difference between revisions of "TODO"
m |
m (→MAIN ISSUE OF THE DAQ) |
||
| (15 intermediate revisions by 4 users not shown) | |||
| Line 1: | Line 1: | ||
| − | + | See also: | |
| − | + | [[Data decode TODO|https://wiki.fe.infn.it/besiii/index.php/Data_decode_TODO]] | |
| − | + | [[GUFI TODO|https://wiki.fe.infn.it/besiii/index.php/GUFI_TODO]] | |
| − | + | == MAIN ISSUE OF THE DAQ == | |
| − | * | + | * [[Trigger match problems]] |
| + | * [[Noise peaking time]] | ||
| + | * [[Peaks and holes in time distribution]] | ||
| + | * [[Rate set and rate measured]] | ||
| + | * L1 strip X have a threshold of 4-12 fC --> [[:File:Soglie L1X.png]] | ||
| + | * CGEM-IT thresholds (no L1x) have a thresholds above 3 fC (about 20%) --> [[:File:Soglie noL1X.png]] | ||
| − | + | == Old noisy channel == | |
| − | + | The charge measured with random trigger is above 10fC, up to 50 fC. | |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
This is due to few channels with very high noise: | This is due to few channels with very high noise: | ||
* FEB_label = 21, chip 2, channel 3 | * FEB_label = 21, chip 2, channel 3 | ||
| − | * FEB_label = 23, chip | + | * FEB_label = 23, chip 1, channel 26 |
* FEB_label = 36, chip 2, channel 3 | * FEB_label = 36, chip 2, channel 3 | ||
* FEB_label = 43, chip 2, channel 45 | * FEB_label = 43, chip 2, channel 45 | ||
| + | |||
| + | The following channel are not used in GRAAL: (chip,channel) --> (43,3), (46,26), (73,3), (87,45) and they are different from the one shown above | ||
3/4 of these channels (ch. 3 and 45) are located on the edge of the anode-FEB connector (see [[L2_FEB_pinout]]). | 3/4 of these channels (ch. 3 and 45) are located on the edge of the anode-FEB connector (see [[L2_FEB_pinout]]). | ||
| − | + | == Solved == | |
| − | |||
| − | |||
| − | |||
| + | * Charge decrease as the threshold increases -> due to some error in setting the threshold. Now if the thr_sigma*n_sigma < 1, threshold is increased by 1. | ||
| − | + | * The reconstruction is too slow with event.C -> lSolved by changing the arrays to vectors. | |
| − | == | + | * FEB from 0 to 15 are noisy at channel 62 -> removed the channel and use as new TP channel |
| + | * Peaking background on FEB=6 && channel = 62 --> removed the channel | ||
| + | * Only 33% of the hits has a cluster on both sides of the L2, while in principle it should be 100% --> An external pulse has been installed and short data taking (1 hour) are proposed WITHOUT restart of the ROC | ||
| + | * The tracking efficiency is between 50 and 70%. It has to be understood the source of the data loss --> fixed. at 2020.01.20 is about 92 % | ||
| + | * The communication of 12 chip with the ROC is not properly working --> signal cables has to be changed in order to remove the 8b/10b errors as much as possible | ||
| + | * Inconsistency between RUN 134 and RUN 139 has to be understood --> test has to be proposed --> due to gas problem | ||
| + | * FEB_label 2 and 21 dont have calibration. Right now we are using random calibration for EFINE and TFINE --> fixed | ||
| + | * Incongruity in the number of triggers and checks -> check the trigger system -> checked in RUN | ||
==== No signal from FEB 2 and 30 ==== | ==== No signal from FEB 2 and 30 ==== | ||
| − | This is due to missing calibration files (raw data should be present) | + | This is due to missing calibration files (raw data should be present): |
* FEB 30 (calibration files available) has been restored | * FEB 30 (calibration files available) has been restored | ||
* FEB 2 (calibration files not available) will be provided soon with a temporary mock-up calibration (based on TP scan performed with GUFI) | * FEB 2 (calibration files not available) will be provided soon with a temporary mock-up calibration (based on TP scan performed with GUFI) | ||
Latest revision as of 01:56, 25 January 2020
See also:
https://wiki.fe.infn.it/besiii/index.php/Data_decode_TODO
https://wiki.fe.infn.it/besiii/index.php/GUFI_TODO
MAIN ISSUE OF THE DAQ[edit | edit source]
- Trigger match problems
- Noise peaking time
- Peaks and holes in time distribution
- Rate set and rate measured
- L1 strip X have a threshold of 4-12 fC --> File:Soglie L1X.png
- CGEM-IT thresholds (no L1x) have a thresholds above 3 fC (about 20%) --> File:Soglie noL1X.png
Old noisy channel[edit | edit source]
The charge measured with random trigger is above 10fC, up to 50 fC. This is due to few channels with very high noise:
- FEB_label = 21, chip 2, channel 3
- FEB_label = 23, chip 1, channel 26
- FEB_label = 36, chip 2, channel 3
- FEB_label = 43, chip 2, channel 45
The following channel are not used in GRAAL: (chip,channel) --> (43,3), (46,26), (73,3), (87,45) and they are different from the one shown above
3/4 of these channels (ch. 3 and 45) are located on the edge of the anode-FEB connector (see L2_FEB_pinout).
Solved[edit | edit source]
- Charge decrease as the threshold increases -> due to some error in setting the threshold. Now if the thr_sigma*n_sigma < 1, threshold is increased by 1.
- The reconstruction is too slow with event.C -> lSolved by changing the arrays to vectors.
- FEB from 0 to 15 are noisy at channel 62 -> removed the channel and use as new TP channel
- Peaking background on FEB=6 && channel = 62 --> removed the channel
- Only 33% of the hits has a cluster on both sides of the L2, while in principle it should be 100% --> An external pulse has been installed and short data taking (1 hour) are proposed WITHOUT restart of the ROC
- The tracking efficiency is between 50 and 70%. It has to be understood the source of the data loss --> fixed. at 2020.01.20 is about 92 %
- The communication of 12 chip with the ROC is not properly working --> signal cables has to be changed in order to remove the 8b/10b errors as much as possible
- Inconsistency between RUN 134 and RUN 139 has to be understood --> test has to be proposed --> due to gas problem
- FEB_label 2 and 21 dont have calibration. Right now we are using random calibration for EFINE and TFINE --> fixed
- Incongruity in the number of triggers and checks -> check the trigger system -> checked in RUN
No signal from FEB 2 and 30[edit | edit source]
This is due to missing calibration files (raw data should be present):
- FEB 30 (calibration files available) has been restored
- FEB 2 (calibration files not available) will be provided soon with a temporary mock-up calibration (based on TP scan performed with GUFI)
