== MAMBO2/ABBA2 Checklist ==

This document describe the checks and test that '''must''' be done before and during the Bolometer pool, as well as the frequency needed for these tests. (Maintained by GQ.)

The tests are divided into different groups, as they supposed to be carried by different groups of the IRAM staff (operators, AoDs, or Rxs)

<<TableOfContents(5)>>

=== AOD CHECKLIST ===

These items are explained below, in detail. This list is intended as a checklist. Do you have all these points under control?
 

 * Any observer from a B-rated project has 4 hours of observation granted. But just this. Any increase of time has to be or
agreed by the pool coordinator, or based in that no other projects can be observed at that moment.

 * Teach the observers on the general aspects of the observation '''even''' the expert MAMBO observers ('''weekly'''):
    * CookbookPako. 
    * When to switch to heterodyne ?
    * What is a spiky scan ? 
    * What is a jump in signal ?
    * Not to open a second monitor at flexible@mrt-lx3:monitor/
    * How to reduce a pointing and a focus

  Do you know the answer to these questions? If not, ask the pool coordinator well in advance ( to avoid the problems to be not noticed)

 * Start the PoolGuide. (not now)
 * Observations on absorber (AbsorberObs, '''at least once per week!''')
   * Do when the weather is too bad even for heterodyne observations.
   * Do if the scans are too spiky
 * Do a [[http://www.iram.es/IRAMES/mainWiki/FlatField|Beam Map]]

 * Check the cable connections of ABBA (if ABBA does not react)

 * Monitors: Monitor, MonCoo, MonInst, MonitorLoop, MonitorList

 * Basic Mopsic Reduction: ooCSF, mapCSF,...

 * Database reduction: possible errors and how to solve them.

 * How to introduce the scans by hand.


=== KNOWN PROBLEMS ===

 * Right-edge problem

=== AOD CHECKLIST explained ===

 * When to switch to heterodyne ?

 If the weather is too bad to observe with MAMBO we have to change to 3mm backup observations. 

 * What is a spiky scan ? 

 There is a clear difference between a spike and skynoise.

 * What is a jump in signal ?

 The signal must be "flat". It can be noisy, but it should be flat around a certain value (the value of the flux). In the monitor the subscans are delimited by a red line at the start of the subscan and a green line at the end. If there are jumps between a green and a red line (i.e. between subscans) this is not important.

 * How to reduce a pointing and a focus

In the pool account, enter in the reduction directory. There run mopsic. To reduce the last pointing run @psn. For the last focus @fsn.


 * Monitors: Monitor, MonCoo, MonInst, MonitorLoop, MonitorList

The monitor and the MonCoo are those regularly running in the control room. The Monitor shows the result of the present scan. The MonCoo shows the tracking of the source. In case Absorber measurements are done in stow, the monInst monitor must be opened. The instructions to open this monitor can be found bellow. This monitor is intended to see jumps or oscillations in the signal. At the top-left side of the monitor, it is shown the total power channels. The signal of these channels can show a slope, but not jumps.
The MonitorLoop, MonitorList are monitors that reproduce what was seen in the observations.

   * Open the Monitor of instabilities

    In case bolometer observations are done in stow position (i.e. absorber), the mopsic Monitor will not be able to reduce the data, since it understands that there are severe tracking errors. In this case the Monitor of Instabilities must be opened instead of the mopsic Monitor.

    The first step is to stop the loop of the Monitor by pressing ctrl+c. Then we exit from mopsic by typing <exit>. We should change from the folder called monitor/ where we are, to that called monInst/. Once there, we type mopsic @ monInst, and the monitor of instabilities will start to run.



 * Basic Mopsic Reduction: ooCSF, mapCSF,...

Read the mopsic Cookbook. The pool coordinator can provide some examples for practice.

 * Database reduction: possible errors and how to solve them.

The main problem is a failure in the reduction. The first step is to check the log files, located at flexible:reduce/, e.g. mopsicoocsf.txt. For instance:


{{{
  MOPSIC
 26.10.2010, Robert W. Zylka

W-SIC,  Session is not interactive, EDIT and MEMORY off
opLM: The LOG & MES files are MOPSIC.LOG1 & MOPSIC.MES1
      User Gildas using procedure ooCSF.mopsic
I-ASTRO_INIT, Ephemeris opened
init: INPUT LIST initialized
loadList: Loading the inList onOff.LIST
             2 observations in the inList
selInList:    2 OBJECTS with TYPE(s) o, selected
qSortOlist: List sorted according to Julian Date
prList: List of INPUT observations, page 1
    1 J0816+48   1199.1 20101115 55515.2396  9.1373 -1.00 o abba-20101115s21-imb
    2 J0816+48   1199.1 20101115 55515.2545  9.4964 -1.00 o abba-20101115s22-imb
  #  object     lambda   date      MJD       LST    elev obsp   file
setIDFN: iram30m-abba-20101115s21-imb.fits
setTel: Telescope set to IRAM 30m
getKindRC: RC   121 is not data
reaimbfitspht1: doZig T tIncS  0.520000 phTimeS  0.260000 smplRateHz  3.84615
reaIMBFitsSu: subScan 11 DATE-OBS incompatible with slowTraceRate=1: 31.001
calcEfBMsep: Effective beam sep.   32.0000 arc sec
trErStat2: trErAz: min   -0.193 max    0.163 <>   0.00036 <rms>    0.0529 asec
           trErEl: min   -0.492 max    0.510 <>   0.00070 <rms>    0.1492 asec
chkTrEr2: All not blanked coordinates have Az tracking error <     3.00 arc sec
chkTrEr2: All not blanked coordinates have El tracking error <     3.00 arc sec
maskSig: 4 records blanked (from in total 2181)
reaIMBFits: Source J0816+48

!!!!!!!!!!!!!!!!!!!!!!!!!!!!
 No calibration available
!!!!!!!!!!!!!!!!!!!!!!!!!!!!

getCmd: ERROR executing READ
I-ERROR,  Occured in /sw/gildas/gildas-exe-apr10/pro/oo_rea.mopsic at Line   20
I-ERROR,  Called by Loop  1 (#  1.000    )                                at line    2
I-ERROR,  Called by /sw/gildas/gildas-exe-apr10/pro/oo_run.mopsic at Line   84
I-ERROR,  Called by /sw/gildas/gildas-exe-apr10/pro/ooCSF.mopsic at Line   52
F-PAUSE,  Session is not interactive
F-SYSTEM,  Exit code           44 from call to SYSEXI
}}}

This log shows that there is no calibration (i.e. rcp data) available, therefore the calibration does not work.

However the most typical problem is that the file containing the opacities needed for the calibration is corrupted by the gopool. This file gets corrupted when any register is written twice. The file is flexible:reduce/dafs/current.TAUS . This problem is easy to  solve. Ask the pool coordinator in person how to solve it.


 * How to introduce the scans by hand.

 In the pool account, go to observationData/imbfits/ and type >update YYYYMMDDs#SCAN NUMBER#
 

=== BOLOMETER TEST CHECKLIST ===

This section details the test to be done with the bolometer, previous to the pool observations. '''Both general tests and particular session tests must be done.'''
Also the Alignment test in case MAMBO has been reinstalled.

GENERAL TESTS:

 * Perform Skydips (discrete and slewed). '''Slewed skydips is still not working correctly'''

 * Slew by ~270 degrees before doing a pointing. 

 * Do [[http://www.iram.es/IRAMES/mainWiki/AbsorberObs|absorber]] observations.

 * If weather is good and stable enough do a [[http://www.iram.es/IRAMES/mainWiki/FlatField|Beam Map]]

 * Test the normal observing modes.

 * Test the PoolGuide tool and the online database reduction.

ALIGNMENT TEST LIST:

 * If MAMBO2 has been reinstalled first a optical alignment must be done.

 * Do a pointing model. During night covering all the sky in az. and el. The coverage can be checked with the regular mopsic macros used to plot the pointing corrections.

 * The pointing corrections should similar than those before. 

 * The pointing corrections should be the same for different bolometers. Use the macros roff20, roff53, roff8, roff3 which are located in different regions of MAMBO.

 * The spillover in sources below some Jy should show two halves of a circle. However this pattern has not been seen for years. Nowadays the know-but-bad patter is ok, since we have not been able to get anything better (so far).


TEST PRIOR TO POOL S09:

 * Check if there is signal jumps. If so, change the ABBA2 version and check again.

 * Beam stability tests. To check for the reasons on the elongation show in the deep field maps.

 * Do on-off observations at 2 deg. from the sun to check its effects.

=== ABBA2 TEST CHECKLIST ===

This tests are required mainly if the ABBA2 version has changed.

 * Verify that configuration changes works e.g. setting the gain. (e.g. pointing on a planet with different gains.)

 * Change observing mode (i.e total power and wobbler switching

 * Offline stability test. [[http://mrt-lx1.iram.es/mainWiki/NcsOfflineTests|ABBA offline tests]]
     Note that this tests only reproduce the 1-phase observations (i.e. total power).

'''Other ABBA issues'''

 * How to open the ABBA2 monitor: 
{{{ vncviewer -shared mrt-abba:2 }}}
 

<!> If ABBA2 needs to be rebooted, the ABBA2 software is initialized automatically. If two ABBA2 processes are running it will not work at all.

 * Change to ABBA2 previous software version: SwitchAbbaSoftVersion 


'''ABBA2 Labview versions'''


|| Version || Pool || Known problems ||
|| 20090710a || PoolW0809 || ABBA2 Chashes twice per day||
|| 20090812a || PoolS09 || Sync. problems for skydips (25% lost), 1 crash per day ||

=== Some notes on Operator duties ===
 * Check the bolometer temperature
 * Check that the paddle motor works.
 * Fill the bolometer: https://gra-lx1.iram.es/mainWiki/BolometerCooldownInstructions

<!> Remember to remove the cable from the box when He3 is recycled manually

=== Some notes on receiver engineer duties ===
 
 * Check the resistance
 * Check battery voltage 
 * Check the Bias voltage 
 * Check the blank/sync Signal at ABBA2 with an oscilloscope. ('''During tests, and/or if strong spikes appear''')
 * Check the analog signals after the preamplifier while covering the entrance window with a metal plate. ('''During tests, and/or if strong spikes appear''')
----
=== USEFUL DOCUMENTATION ===

 * G.Siringo, 15.4.2008, "ABBA2 short description" [[attachment:abba2_short_description.pdf]]
 * Cable connections between MAMBO2 preamplifiers and the ABBA2 backend: [[attachment:abba2-cables2.png]] (30-May-2011, G.Quintana-Lacaci and confirmed by J.Penalver)

----

=== PROBLEMS AND SOLUTIONS ===

This is a quick guide through normal problems that appeared in the past and the possible reasons and solutions.

 ||'''PROBLEM'''||'''REASON'''||'''SOLUTION'''||
 ||Bololog is not updated|| goPool has not been sent|| Ask the operator to do so||
 ||Mopsic Monitor is not working || There is other monitor running and the DAFS files have been corrupted || Stop the second monitor and recreate the DAFS ||
 ||ABBA does not start getting data||Wobbler is not wobbling|| Ask the operator||
 ||ABBA process (labview) do not start || Other Labview might be running || check the processes and kill all labviews ||
 ||ABBA does not start getting data||Wobbler works fine but the signal did not reach ABBA|| Check the cable connections with ABBA||
 ||The skydip results are abnormally high||The paddle is put in from of the beam||Ask the operator to try it by hand or contact the Rx people||