Difference between revisions of "Minutes-2-27-2014"

From GlueXWiki
Jump to: navigation, search
m (Text replacement - "/halldweb1.jlab.org/" to "/halldweb.jlab.org/")
 
(7 intermediate revisions by one other user not shown)
Line 4: Line 4:
 
# Installation status [http://www.jlab.org/Hall-D/detector/fdc/installation/FDCinst02272014_1.JPG]
 
# Installation status [http://www.jlab.org/Hall-D/detector/fdc/installation/FDCinst02272014_1.JPG]
 
# Engineering (Bill, Dave)
 
# Engineering (Bill, Dave)
−
## Cathode scan
+
## [http://www.jlab.org/Hall-D/detector/fdc/calibration/FDC_Cathode_strip_measurement.xlsx Cathode scan]
 
## LV fast shutdown
 
## LV fast shutdown
 
# Electronics  [http://www.jlab.org/Hall-D/detector/fdc/installation/Tests_Feb_2014.xlsx problematic channels] (Fernando, Chris, Nick)
 
# Electronics  [http://www.jlab.org/Hall-D/detector/fdc/installation/Tests_Feb_2014.xlsx problematic channels] (Fernando, Chris, Nick)
−
# Mini-DAQ status [https://halldweb1.jlab.org/elog-halld/FDC FDC E-log] (Beni)
+
# Mini-DAQ status [https://halldweb.jlab.org/elog-halld/FDC FDC E-log] (Beni)
 
# Other
 
# Other
−
 
−
<!--
 
  
 
= Minutes =
 
= Minutes =
  
−
Participants: Eugene, Fernando, Bill, Dave, Nick, Chris, Simon, Beni, and Lubomir.
+
Participants: Eugene, Bill, Dave, Nick, Chris, Simon, Beni, and Lubomir.
  
 
== Installation ==
 
== Installation ==
  
−
- On Thursday last week we inserted the FDC into the magnet (see pictures attached). It was taken out of the magnet on Monday. The test was successful and we learned how to better handle the cables during insertion, what we discussed on the meeting. As Chris suggested, during the extraction, we used a strap to keep the cables up when moving the FDC. Bill suggested, for the final insertion, to use bearings (at least three) attached to the rails to be able to move the cables together with the rest, FDC and the mesh. Another discussion: how to attach the cables to the face of the magnet. On Monday we already exercised this at one quadrant using holding rings attached to the magnet. The main problem is that one has to spread the cables evenly around the periphery to make space for the CDC and also to be able to work on the BCAL modules. Beni proposed to extend the mesh further at the upstream side to cover completely the CDC, however the question was if there will be space to work around the BCAL modules. Bill proposed to build a ring (attached to the platform) with the same diameter as the mesh to hold the cables and, at the same time, be able to work on the other detectors. Fernando proposed to build a bigger movable ring attached to the face of the magnet, to support the cables.  
+
- A cable holder frame has been installed at the top on face of the magnet (picture attached). It's too far from the mesh and the cables will sag (Nick), but better than before. We may need such frames also on the sides. New cable tray will be installed on the right side of the magnet to avoid crossing with the BCAL water pipes. Tom is working on the cable cradle that will slide together with the FDC on bearings (Bill). Bill is manufacturing the FDC-to-mesh connections made of metal and plastic to insulate the two. We also have to cut the mesh loop. Need to have also caps for the Fluorinert tubes to be used  when the cooling system is disconnected from the FDC.  
  
−
- Before the insertion, the cooling system failed again, the reasons are unknown so far, but there was no LV at this time. During the insertion the Fluorinert houses were disconnected and taped with duck tape at the chiller side (due to the lack of proper caps), as a result we lost >1 gal of Fluorinert: need proper caps for the final insertion!
+
- Yesterday Lubomir replaced the mass flow controller (6000 sccpm range) of the temporary gas system with a re-calibrated one of 500 sccpm. It turned out that to have the same flow on the rotameter we had to set the flow at 200 sccpm or less, to be compared to setting of the previous flow controller of 400sccpm. It turned out also that gas bottle pressure goes down slower than before, but we need more days to figure this out. If this is the case then the old flow controller was leaky, but at the same time it is strange that we are flushing now successfully all the four packages with just 200sccpm total; normally we needed such a flow for one package only.
−
 
+
−
- Tom informed us that the power in the Hall and the CC will be OFF for 3-4 days starting February 20 (next Thursday). For the FDC we need the gas running all the time and will have to replace the mass flow controller (which requires power) with a rotameter. To avoid potential high pressure in the chamber with such a system Beni suggested installing a bubbler at the gas input.
+
  
 
== Engineering ==   
 
== Engineering ==   
  
−
- Bill will travel to Hexagon Metrology in Charlotte, NC next Wednesday/Thursday to scan the cathode.
+
- Bill returned from Hexagon Metrology, Charlotte, NC, and showed us pictures of the MCC machine and a lot of data: the xlsx file is attached above containing all the raw data. The cathode was scanned optically and automatically with a small step (5mm?) along the strips recording the edges of the strips. Sometimes the machine re-scanned the same point, sometime was obstructed by a defect, or by the tape on the back of the cathode that connects the foils. In any case, this is not a problem since the number of points is huge. Eugene proposed for a better visualization to make a scattered plot of all the (x,y) points. This will help to decide how to analyze the data.  
−
 
+
−
- We need to modify the mesh to make sure there are no closed loops and also to insulate it from the FDC packages (now it's connected through two metal plates). The mesh will have a separate ground. Bill will work on that.
+
−
 
+
−
- Dave started working on the hardware shut off of the LV crate from the chiller. He needs to identify how exactly to interpret the signals about the chiller status and then will figure out with Fernando what the PCB should be doing. The board can be mounted on the same rack (back side) where the LV crate is. As for the MPOD controller settings Dave will talk to Hovanes.
+
  
 +
- Dave figured out how to get the status signals from the chiller: it turned out the schematics in the chiller documentation were not up to date. Dave had already everything connected, waiting only for a power supply, so maybe tomorrow we will be able to test the fast shut down. First Hovanes will try it on a different crate. Dave received the flow sensor for the chilled water but not yet the PLS module to be used for the flow monitoring. 
  
 
== Electronics ==
 
== Electronics ==
  
−
- Fernando did noise/signal measurements on the FDC: good news, the noise at/around FDC is low: 8mV peak-to-peak. Will report at the collaboration meeting. Eugene: expect many more new devices to be turned on in the Hall. 
+
- Nick has a spare F1TDCV3 module so we can install it right away.
  
−
- Lubomir prepared a table (linked above) with the problematic channels found so far with a scope or, for the wires, with the Beni's mini-DAQ.  The problems are grouped in several types:
+
- Nick and Chris fixed one very bad flat cable connector with mixed wires at the detector side. There are still several 3M connectors to be fixed, but we want to make sure with the mini-DAQ that the problem is there. According to Chris may take many hours to fix such a connector.  
−
 
+
−
#conductive tape issues (or related to that): 13 channels found so far with a scope and the shaper board (doesn't see the inner 8 channels), some were fixed already, half of the problems are next to the cooling manifolds where it's difficult to bend the tubes.
+
−
#two groups of four pre-amps connected to the same LV cable showing similar problem: one of the channels saturates and the signals on all the channels of the other cards have a long (500ns) tail. Fernando will look at it, but suggested first replacing the cable, then the pre-amps.
+
−
#~10 cards with cable issues, meaning if you replace the cable it works. Nick and Chris will come tomorrow and fix the bad connectors at the detector side.
+
−
#6 wire cards with mostly one channel not working as found by the Beni's tests. Most of these cards are at the same position at the four packages (C1,S1-24) and these channels were working before the FDC insertion. Chris: we had to use force when inserting the FDC and may have pulled on the whole bundle that connects the cables at the same positions at the packages. All these channels have been fixed by pushing the cards back in the pre-amp connector.
+
−
#4 cards all in the first package and always at the same place w.r.t to the cell: S49-72 with 5 to 10 first channels showing 1-30kHz noise at 8V threshold. The noise goes down to 10-500Hz at 10V. The most interesting is that the noise is there even if turn off the HV and that there's no noise when looking with an analog pre-amp. Fernando will look into this.
+
−
#2 cards with very small (in amplitude) LVDS signals. Fernando: check cables, replace pre-amps.
+
−
 
+
−
- All the negative (field) HV channels have been checked by Anatoly measuring the capacitance at the electronics end of the HV cable. It ranges from 1.9 to 2.5 uF. If the cable is disconnected from the detector side the capacitance is 1.2 uF. We conclude the negative HV connections are OK.
+
  
 
== Mini-DAQ ==
 
== Mini-DAQ ==
  
−
- Beni presented his latest results from the mini-DAQ (E-log entries 138-141). Beni has now all the crates (four) with TDCs read out so we can test all the wires at the same time. The plots show the problems discussed above.
+
- Beni: according to Cody the new and working firmware with the trigger window set by programmable parameters should be ready by the end of the week. Then we need from Bryan Moffit the software to set up the parameters. Eugene: this work is on the critical path now and everybody should know it has high priority. In case the firmware is not ready by the beginning of the next week then we will ask Cody to have these parameters hard coded. If this doesn't work we may consider going back to the Gerard's firmware that is slow but operational, however we don't want to go that back (Eugene).
−
 
+
−
- Beni proposed to swap the top and bottom cables as connected to each of the TDCs, so that "it makes sense", and he did it. Now there's correspondence between the detector and the TDC channel numbering. If everybody agrees we will have to re-label the connectors and change the translation table.
+
−
 
+
−
-->
+

Latest revision as of 06:58, 1 April 2015

February 27, 2014 FDC meeting

Agenda

  1. Installation status [1]
  2. Engineering (Bill, Dave)
    1. Cathode scan
    2. LV fast shutdown
  3. Electronics problematic channels (Fernando, Chris, Nick)
  4. Mini-DAQ status FDC E-log (Beni)
  5. Other

Minutes

Participants: Eugene, Bill, Dave, Nick, Chris, Simon, Beni, and Lubomir.

Installation

- A cable holder frame has been installed at the top on face of the magnet (picture attached). It's too far from the mesh and the cables will sag (Nick), but better than before. We may need such frames also on the sides. New cable tray will be installed on the right side of the magnet to avoid crossing with the BCAL water pipes. Tom is working on the cable cradle that will slide together with the FDC on bearings (Bill). Bill is manufacturing the FDC-to-mesh connections made of metal and plastic to insulate the two. We also have to cut the mesh loop. Need to have also caps for the Fluorinert tubes to be used when the cooling system is disconnected from the FDC.

- Yesterday Lubomir replaced the mass flow controller (6000 sccpm range) of the temporary gas system with a re-calibrated one of 500 sccpm. It turned out that to have the same flow on the rotameter we had to set the flow at 200 sccpm or less, to be compared to setting of the previous flow controller of 400sccpm. It turned out also that gas bottle pressure goes down slower than before, but we need more days to figure this out. If this is the case then the old flow controller was leaky, but at the same time it is strange that we are flushing now successfully all the four packages with just 200sccpm total; normally we needed such a flow for one package only.

Engineering

- Bill returned from Hexagon Metrology, Charlotte, NC, and showed us pictures of the MCC machine and a lot of data: the xlsx file is attached above containing all the raw data. The cathode was scanned optically and automatically with a small step (5mm?) along the strips recording the edges of the strips. Sometimes the machine re-scanned the same point, sometime was obstructed by a defect, or by the tape on the back of the cathode that connects the foils. In any case, this is not a problem since the number of points is huge. Eugene proposed for a better visualization to make a scattered plot of all the (x,y) points. This will help to decide how to analyze the data.

- Dave figured out how to get the status signals from the chiller: it turned out the schematics in the chiller documentation were not up to date. Dave had already everything connected, waiting only for a power supply, so maybe tomorrow we will be able to test the fast shut down. First Hovanes will try it on a different crate. Dave received the flow sensor for the chilled water but not yet the PLS module to be used for the flow monitoring.

Electronics

- Nick has a spare F1TDCV3 module so we can install it right away.

- Nick and Chris fixed one very bad flat cable connector with mixed wires at the detector side. There are still several 3M connectors to be fixed, but we want to make sure with the mini-DAQ that the problem is there. According to Chris may take many hours to fix such a connector.

Mini-DAQ

- Beni: according to Cody the new and working firmware with the trigger window set by programmable parameters should be ready by the end of the week. Then we need from Bryan Moffit the software to set up the parameters. Eugene: this work is on the critical path now and everybody should know it has high priority. In case the firmware is not ready by the beginning of the next week then we will ask Cody to have these parameters hard coded. If this doesn't work we may consider going back to the Gerard's firmware that is slow but operational, however we don't want to go that back (Eugene).

Retrieved from "https://halldweb1.jlab.org/wiki/index.php?title=Minutes-2-27-2014&oldid=65841"