Difference between revisions of "Minutes-7-17-2014"

From GlueXWiki
Jump to: navigation, search
(→‎Agenda)
m (Text replacement - "/halldweb1.jlab.org/" to "/halldweb.jlab.org/")
 
(10 intermediate revisions by one other user not shown)
Line 9: Line 9:
 
# Other
 
# Other
  
−
[https://halldweb1.jlab.org/wiki/index.php/Connect_to_Bluejeans_Meetings Instructions for Bluejeans meeting connection]
+
[https://halldweb.jlab.org/wiki/index.php/Connect_to_Bluejeans_Meetings Instructions for Bluejeans meeting connection]
  
 
FDC meeting ID: 290664653
 
FDC meeting ID: 290664653
−
 
−
<!--
 
  
 
= Minutes =
 
= Minutes =
  
−
Participants: Curtis, Naomi (CMU), Luke, Eugene, Dave, Chris, Nick, Simon, Mike, Beni, Fernando, and Lubomir (JLab).
+
Participants: Curtis, Naomi (CMU), Sean (NU), Luke, Dave, Chris, Mike, Simon, Beni, Fernando, and Lubomir (JLab).
  
−
== Gas system ==
+
== CDC update ==
  
−
- Last Thursday after the FDC meeting Dave and Lubomir connected the power of the gas controls in the gas room to a UPS. When the power was restored the 4-20mA setting of the two masters were different (40mA as usual), but also the settings of one slave channel did change, that never happened before. Dave will see if he can read the settings and either restore them correctly, or issue an alarm. This has to be discussed also with Hovanes.
+
Mike made a report, linked above:
  
−
- On Saturday the switch on the CO2 bottles didn't work when one set of bottles got empty; it was supposed to switch to the full set of bottles. For one-two hours there was no CO2 flow into the FDC mixing tank, and 0.2l/m (instead of 0.9) was going into the CDC mixing tank. The bottom half of the CDC and 25% of the FDC channels tripped. Dave can implement an alarm, but everybody agrees some kind of a messaging system is needed, as well. Scott contacted the gas company to replace the switch. This switch was working before, and we have the same switch for the Argon bottles.
+
- first slide shows the positions of the channels with noise as analyzed from the fADC data.
  
−
== DAQ ==
+
- next several slides demonstrate the "missing bit" problem. In one case all the channels in one module have that. Naomi: such problem was fixed by re-floating the circuit board. 
  
−
- Beni: The attempt to read synchronously the BCAL and the CDC failed. It's not clear why, the ROCs where receiving the data but then they got lost somewhere. Will try it again tomorrow.  
+
- last slide shows the mean and the RMS of the pedestals. Questions about the meaning of the overflow/underflow in these histograms: Mike will work on that.
  
−
- Beni: there's a problems with the relation between the pedestal positions and the DAC values in the fADC. The relation is different for the CDC and the FDC. For the CDC the pedestals are at ~500 and all the DAC values are set to 0x9000 (hex), and for the FDC the pedestals are at ~1000 but the DAC values are set all at 0x8000 (see raw data plots linked to item 4). Fernando explained this by the difference in the pre-amp gains (factor of 4).
+
- Mike: the strain relief of the signal cables is not good enough: sometimes may lose contact. Fernando will find a way to improve it.
  
−
== CDC update ==
+
- According to Mike the pedestal subtraction in the Dave's code is done correctly, so this is not an explanation for the low CDC efficiency as discussed yesterday in the calibration meeting.
−
 
+
−
- Mike studied the noise in the CDC using raw fADC data. He applies a negative threshold (below the pedestal) and plots all the channels that pass the threshold as function of the sample number: there's clear correlations between the channels. The positions of these noisy channels is shown in the plot attached: mostly at the 6-9 o'clock sector when looking downstream.
+
−
 
+
 
== FDC update ==
 
== FDC update ==
  
−
- Lubomir looked at the raw fADC data: plots linked above represent the distributions of the ADC values (from all the samples) as function of the channel. The plots are very useful since one can identify different kinds of problems: noisy channels when the area below the pedestal is populated, missing bit (like in the CDC plot ch.63), detector problem when the signals are small (like chan.71 in the FDC plot), connection issue when there's only a thin pedestal. Another problem is like ch.69 FDC plot: there's no pedestal visible, all values are overflown, however, if you look with a scope with differential probe we do see normal signals. If you connect another "good" cable to the same fADC connector everything is OK. What is then the difference between the fADC and the differential probe? Is this problem related to the pedestal problem Beni discussed above?
+
- Lubomir presented a table (xls file linked) with all the problematic cathode channels as identified from the latest DAQ runs. Luke checked these channels with a scope which information is included in the Q-column with color. Information about Anatoly's measurements with scope in March/April (right after insertion) and the attempts to fix these channels is also included.
  
−
- Summary of these observation are shown in a spreadsheet linked above (only half of the crates are included so far). The number of bad detector channels is about the same as we had right after the installation as studied with a scope.
+
- For 69 of the channels (pink or red color) we suspect the problem is at the detector side (mostly conductive tape, but also connector or pre-amp issues). In the packages (1 to 4) the number of such channels is 4,1,41,23. Ten of these channels were OK in April, but had problems before, except two of them.  
  
−
- Luke showed his studies (linked) of the noise in the FDC using a scope. He was looking for similar correlated noise that is seen in the CDC. This kind of noise is well visible only on these long strip that have short strips next to them (due to having ground from both sides of the frame). Luke's summary: most of the noise in the first and second package and mostly in the outer cells (one and six), frequencies 200kHz, 80kHz and 40kHz, amplitudes 10-15mV. Interestingly, if the cable is not well plugged into the adapter card, one can see the same noise on all the channels and the noise increases when the card is close to the VME crates (see picture).
+
- We have 11 channels (green) where we see signals with the scope but not with fADC. However, the fADC channels are OK with another cable; Fernando suspect cable/connector problem. In the fADC data you see overflow or a fixed value at ~2000 and Beni made a list of such channels before.
  
 +
- Finally, there are 9 channels (yellow) with no signal, but also almost no noise. Some of these were identified before to have problems with the 3M connectors, but the rest may have other issues.
 +
 +
== Electronics ==
 +
 +
- Fernando looked for EMI noise in the hall at the platform. He didn't find anything in the kHz range. The only source of very low noise was the mass flow meters for the BCAL Nitrogen. Dave proposed to check this by turning them off for a while. However, Fernando was not able to get very close to the CDC due to the target work there.
 +
 +
- Starting middle of August - September Fernando's group will start checking/fixing/re-routing cables that were not installed properly. Expect short interruptions at that time.
 +
 
 
== Engineering ==
 
== Engineering ==
  
−
- Bill (not at the meeting): the microscope holder for the cathode scan is ready. Need to connect the motor the stage and to wire the motor controller and the stoppers (Dave). Luke will work on the scanning project.
+
- Luke, with help from other people, started working on the scanning project (picture attached): the microscope has been attached to the motion stage, the motor can now be controlled from the computer. The shaft that connects the motor is missing but Tom ordered a new one; the encoder still needs to be connected.
 +
 
 +
== Other ==
  
−
-->
+
- Beni:it's time to start using alcohol in the gas system. Will ask Scott to fill the alcohol bubblers.

Latest revision as of 05:07, 1 April 2015

July 17, 2014 FDC+CDC meeting

Agenda

  1. DAQ (Beni)
  2. CDC status File:MStaibCDCUpdate17July2014.pdf (Mike, Beni)
  3. FDC status (Lubomir, Luke)
  4. Electronics (Chris, Nick)
  5. Engineering Cathode Scan system (Bill)
  6. Other

Instructions for Bluejeans meeting connection

FDC meeting ID: 290664653

Minutes

Participants: Curtis, Naomi (CMU), Sean (NU), Luke, Dave, Chris, Mike, Simon, Beni, Fernando, and Lubomir (JLab).

CDC update

Mike made a report, linked above:

- first slide shows the positions of the channels with noise as analyzed from the fADC data.

- next several slides demonstrate the "missing bit" problem. In one case all the channels in one module have that. Naomi: such problem was fixed by re-floating the circuit board.

- last slide shows the mean and the RMS of the pedestals. Questions about the meaning of the overflow/underflow in these histograms: Mike will work on that.

- Mike: the strain relief of the signal cables is not good enough: sometimes may lose contact. Fernando will find a way to improve it.

- According to Mike the pedestal subtraction in the Dave's code is done correctly, so this is not an explanation for the low CDC efficiency as discussed yesterday in the calibration meeting.

FDC update

- Lubomir presented a table (xls file linked) with all the problematic cathode channels as identified from the latest DAQ runs. Luke checked these channels with a scope which information is included in the Q-column with color. Information about Anatoly's measurements with scope in March/April (right after insertion) and the attempts to fix these channels is also included.

- For 69 of the channels (pink or red color) we suspect the problem is at the detector side (mostly conductive tape, but also connector or pre-amp issues). In the packages (1 to 4) the number of such channels is 4,1,41,23. Ten of these channels were OK in April, but had problems before, except two of them.

- We have 11 channels (green) where we see signals with the scope but not with fADC. However, the fADC channels are OK with another cable; Fernando suspect cable/connector problem. In the fADC data you see overflow or a fixed value at ~2000 and Beni made a list of such channels before.

- Finally, there are 9 channels (yellow) with no signal, but also almost no noise. Some of these were identified before to have problems with the 3M connectors, but the rest may have other issues.

Electronics

- Fernando looked for EMI noise in the hall at the platform. He didn't find anything in the kHz range. The only source of very low noise was the mass flow meters for the BCAL Nitrogen. Dave proposed to check this by turning them off for a while. However, Fernando was not able to get very close to the CDC due to the target work there.

- Starting middle of August - September Fernando's group will start checking/fixing/re-routing cables that were not installed properly. Expect short interruptions at that time.

Engineering

- Luke, with help from other people, started working on the scanning project (picture attached): the microscope has been attached to the motion stage, the motor can now be controlled from the computer. The shaft that connects the motor is missing but Tom ordered a new one; the encoder still needs to be connected.

Other

- Beni:it's time to start using alcohol in the gas system. Will ask Scott to fill the alcohol bubblers.

Retrieved from "https://halldweb1.jlab.org/wiki/index.php?title=Minutes-7-17-2014&oldid=65652"