Babysitting

Document Sample
Babysitting Powered By Docstoc
					                                           21 July 2004




           Babysitting
              AD
           from MCR
                    Contents

1. Introduction.

2. Switching between experiments.

3. Routine monitoring of the AD

4. What to do if there is no beam reaching the
experiment – some hints to help in identifying
problems.

5. Contacts.
1. Introduction.
As from September 2002, the AD will be run by a team of 5 AD
supervisors (ADS). The ADS on duty will be responsible for the
AD operation for one week at a time, Monday 07:00 to Friday
23:00. A second person from the ADS team is designated
“backup” whose role is to assure the continued supervision in case
the regular ADS has been working during the previous night.

The babysitting starts when the AD supervisor leaves for the day.
The responsibility of the MCR staff from that moment and until
the following morning (weekdays) is the following:

  - Monitor the functioning of the AD
  - Switch between the 4 experiments as instructed by the ADS
    at 17:30 each day. Presently, switching takes place at 23:00
    and 07:00.
  - Deal with security and access (as now)
  - Perform some basic troubleshooting in case of problems.
2. Switching between experiments.
Following the instructions of the ADS and after confirming with
the “new” and “old” experiments, switching is done with a special
application program found at: “control” – “ADE switching”.
If the program is already open at the console, please exit and start a
new instance. Wait until the present switching contitions have been
detected (green box as below), then click on the new experiment.

At present, following experiments are running :
ATRAP1
ATRAP2
ATHENA
DEM (ACE : Antiproton Cell Experiment)
ASACUSA (Note: a second dialog box will prompt for with RFQD
or without RFQD. The ADS will give instructions about which
setup is presently in use)




                    AD experiment switching dialog box
When switching between DEM(ACE) and other experiments, note
that DEM(ACE) ejection takes place at 300 MeV/c and the others
at 100 MeV/c. To change extraction momentum, open “cycle”,
“cycle editor”,then select “mode”, “injec-ejec mode” set ejection
Flat top to FT3C and C-ejection to -838 ms for 300 MeV/c ejection
as below. Then select “File”, “Write Cycle”.

                  300 Mev/c (FT3C) C-ejection -838 ms




                   100 Mev/c (FT4B) C-ejection -30 ms
When the Cycle editor has finished the “Write Cycle”, Open
“Cycle”, “RF cycle editor” and select “Macros”, then click either
“Standard operation” for 100MeV/c or “Enable 300 MeV/c
ejection”.




Once the switching is done (this will take a few minutes), check
with the experiment if they are happy with the beam.

If not, go to chapter 4………
3. Routine monitoring of the AD
 - To monitor the beam as it arrives at the experiments, use the
   following URL:s:

   - ATRAP : http://pcepad205/ppac.html

   - ATHENA: http://pcathena10.cern.ch/
               “monitor” – “online display”

   - ASACUSA: http://pspc7750.cern.ch/screens.jpg
              (RFQ screens only)

 - Use the Schottky DSP intensity measurement (“measur” –
   “Schottky”) to monitor beam survival during the cycle (see
   also chapter 4).

 - Use the FFT Schottky measurement (VIP : DR.SCHOTTKY-
   TA and also displayed on the AD vistar) to monitor beam
   behaviour on the cooling flattops (see also chapter 4).

 - Use the AD vistar to monitor proper cycling.

 - Use nAos global AD “pbar injection+ejection” (see also
   chapter 4).
4. What to do if there is no beam reaching
the experiment.

 - Is the AD cycling?

 - Is beam injected into the AD?

 - Is the beam lost during the cycle?

 - Is the beam ejected?

 - Is the ejection line properly setup? (check
   this first if problems occur after an
   experiment switchover)

 - If the problem cannot be identified and fixed
   – call the AD piquet !

 - Note: all following screen copies represent a
   situation when the AD is working correctly.
Is the AD cycling?
  - Check if the AD is cycling on the AD vistar.
  - The correct situation for beam request and cycle running can
    be seen below (cycle with beam, permanent request) :




                    “Cycle” – “cycle control”

Also:

Is the AD beam and external conditions in PS ok ?
Is beam injected into the AD?
  - check the 4 bunches, PS and AD kickers with nAos global :
  “pbar injection+ejection”




  - FTA.TFA9012 is the first trafo in the injection line.
  - DR.UHZA shows particles circulating during the 6 first turns
    in the AD ring using one of the ring pickups.

- The injected beam should be visible on the schottky FFT display,
VIP – “DR.SCHOTTKY-TA” and is also visible on the AD vistar:

Beam just after injection and before start of stochastic cooling at
3.5 GeV/c:
Is beam injected into the AD? (continued)
   - Check inj. Line transmission on TFA9012 and 9053
   - Check if beam hits the production target on MTV9064
   - Check MTV:s before target MTV9003, 9031 and 9043
   - Check injection line elements – “working sets” - “Injection”
Note: Several pulsed elements including the injection septum
(DR.SMI5306) are stopped during part of the AD cycle and will
therefore sometimes display erratic AQN.
Trafos after target (TFA6006 and 5302) indicate arbitrary values
Is beam injected into the AD? (continued)

- Check the inj. kickers via special AP: “Control” – “KFA55
  control”
  - 10 MHz cavities are used for bunch rotation at injection.
    Without these, very little beam is decelerated.
  “working sets” – “longitudinal” – “RF_HL_C10”




Is the beam lost during the cycle?
  - The DSP based longitudinal schottky analysis system is a very useful
    tool to monitor beam survival during the AD cycle. Below is an
    example of bunched beam intensity measurements during one cycle:
    “measur” – “Schottky” (bunched beam only during ramps).
Presently, unbunched measurements are added to the graph which now looks
like this (reliable measurements are highlit, compare with above) :




Note variable scale on time axis, here we are looking at 2 cycles.
Is the beam lost during the cycle? (continued)

Using the above program, numerical values at several points
during the cycle can also be displayed (“options” – “number of
particles”). DE.TFA7049 is situated in the beginning of the
ejection line and is common for all experiments.




Another useful tool to monitor beam survival during the cycle as
well as the effect of stochastic and electron beam cooling is the
Schottky FFT display (VIP – “DR.SCHOTTKY-TA”). This
display is now also integrated in the AD vistar:

Beam just after injection and before start of stochastic cooling at
3.5 GeV/c:




Is the beam lost during the cycle? (continued)
Same beam after stochastic cooling at 3.5 GeV/c and before start
of deceleration:
Here we have decelerated to 2GeV/c and are ready to start the
stochastic cooling:




Still at 2GeV/c, but now after the cooling:




Is the beam lost during the cycle? (continued)
This picture is taken at 300 MeV/c and before the start of the
electron cooling:
Same beam after electron cooling at 300 MeV/c:




Now we are at 100MeV/c, before electron cooling:




Is the beam lost during the cycle? (continued)
Here the beam is cooled at 100 MeV/c and is ready for ejection.
(note that the intensity seems to have increased, this is however not
the case, it is merely some sort of instabilities indicating a very
cool beam.):




Is the beam lost during the cycle? (continued)
  - Check AD ring power supplies (“working sets” – “main_ring”)
- Note: To switch ON/OFF the ring main and the DR.BHZTRxx+xx
    supplies (the first 23 supplies in the ws; from DR.BHZ to
    DR.BHZTR51+52),
ONLY DR.BHZ, DR.QUAD and DR.QUAD-MAIN2 should be
controlled. (the remainder will be switched automatically)
Supplies from DR.DHZ0204 to DR.QSK4308 are controlled normally.
The AD has to be cycling and you have to wait until the end of cycle
before the ring supplies come ON.
- ccv:s are not valid (GFAS control), aqn:s are ok – throughout the cycle.




Is the beam lost during the cycle? (continued)
  - Check the deceleration RF-cavity

“working sets” – “longitudinal” – “RF_HL_C02”




Is the beam ejected?
  - check the ejection with nAos global :
“pbar injection+ejection”
  - DA.PULONG shows the bunch at the last turn before
     extraction, also useful for bunchlength estimation.
  - DE.TFA7049 shows the bunch in the beginning of the
     ejection line (common part for all experiments)
  - DR.KFE35+50 : ejection kicker




  - Check the common part of the ejection line “working sets” –
    “ejection_100MeV”.




  Is the beam ejected? (continued)
  - Special AP for the ejection kickers “control” – “KFA35
    control”




Is the ejection line properly setup?
 - Check the ejection line power supplies :
      o “working sets” – “asacusa_de1”
      o “working sets” – “athena_de2”
      o “working sets” – “atrap_de3”
      o “working sets” – “ace_dem”
 - Hardcopies of the reference settings for the 4 beamlines can
   be found in 4 green folders marked “ATHENA , “ATRAP”
   and “ASACUSA” and “ACE”.
 - MWPC:s are used for beamline checking/steering. Remote
   control is available from the AD console manager. Watch
   out, there is a 25-30 s delay for IN/OUT movement and also
   for the data acquisition.
 - Reference MWPC beam profiles can be found on the AD
   webpage. http://psdoc.web.cern.ch/PSdoc/acc/ad/index.html
      o “setting up de1 asacusa”
      o “setting up de2 athena”
      o “setting up de3/de4 atrap”
 - Layout of 3 beamlines : (note vertical deflection for ATRAP
   lines)




5. Contacts.
- AD piquet      163822
- ASACUSA        79812
- ATRAP          79813
- ATHENA         79814
- ACR            76688/76689



AD piquet:
P.Belochitskii
L.Bojtar
B.Dupuy
P.Fernier
P.Readman

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:112
posted:9/16/2012
language:Unknown
pages:23