Last Update: 20080331

The all-sky performance of pointing with HARP has always been poor - see for instance the report of 20080104 - with rms scatters in elevation, in particular, regularly exceeding 3". But how much of this is due to the K-mirror misalignments and how much to the fundamental HARP pointing model ? It is assumed that the the HARP model will be similar to the RxA model, bar some collimation terms, and that any K-mirror misalignments add to the error budget - but in what proportion ?

To isolate these two issues we record here the result of performing pointings, which use K-mirror positions that align the sky in an AZ-EL orientation on HARP, followed by science observations, specifically jiggle-maps, which generate an RA-DEC orientation of the sky at the array. Point-like, bright line-sources are used. The AZ-EL fivepoint routinely yields the pointing offsets in (az,el) from the model, while the (RA-DEC) jigglemap of the same source may be analyzed for the centroid - and the two compared.

Ideally, the jiggle-map is sandwiched between two fivepoints so that the quality of the fivepoint may be assessed. Data have been acquired as follows:

                                             IMG_ANG    (uaz,uel)     (uaz,uel)
     UT    #  target    az el    daz   del   obs calc     before    after-before
                                                          after
20080116  30  crl618   395 69   -1.3  -2.2   -65  -63   -0.2 +4.8 
                                                        -0.6 +4.9    -0.4  +0.1  
20080117  55 irc+10216 253 70   +1.2  +1.3          3   +0.4 +1.9 
                                                   93   -0.3 +2.1    -0.7  +0.2
20080117  78 irc+10216 277 20   -0.3  +2.1        -53   -4.5 +1.6 
                                                   37   -4.2 +2.0    +0.3  +0.4

                                                       before only
20080121  10  crl618    41 67   -0.7  -5.0         17   -0.0 +7.0
                                                  -73
20080121  24* AP Lyn   369 48   -1.6  -0.2    62? -60   -3.3 -1.4
                                                  
20080121  36  AP Lyn   345 46   -0.6  +0.9       -104   -0.3 -6.6
                                                  -14
20080121  48  AP Lyn   331 33   -5.8  -2.1   -76  -77   +0.8 -3.5
                                                   13
20080121  71  GL1922   140 33   -3.9  -0.1    75   75   -1.3 +4.6

     UT    #  target    az el    daz   del   obs calc     before    after-before
                                                          after
20080127  41 IRC+10216 108 71   -0.5  +3.5    48   48   +1.4 +7.8 
                                                          -    -       -     -
20080127  62  RXBoo     72 41   +1.0  -1.6   -50  -53   +0.3 +9.8  
                                                        +0.1 +6.9    -0.2  -2.9
20080127  76  GL5254   226 27   -2.7  -0.9    70   69   -5.6 +6.0  
                                                  -21   -7.4 +6.2    -1.8  +0.2
20080212  48 IRC+10216 270 42   -0.6  +0.9   -33  -33   +2.3 -2.4
                                                   57   -0.5 -2.2    -2.8  +0.2
20080212  59   RxBoo   314 81   +0.2  +1.6    40   40   +1.1 +2.5
                                                  -50   +0.2 +4.3    -0.9  +1.8
20080213  20    RAur    14 54   -2.1  -8.8   -59  -59   -0.2 +7.0
                                                        +0.0 +2.8    +0.2  -4.2
20080213  38  OH231.8  199 54   -2.1  +2.0    35   36   -0.9 +0.7
                                                        -1.1 +3.8    -0.2  +3.1
20080213  54 IRC+10216 268 48   -1.3  +0.5    63   63   +0.7 +2.6
                                                  -27   +0.3 +2.1    -0.4  -0.5
20080214  38 IRC+10216 145 82   +1.0  -0.4   +26   26   +1.2 +0.8
                                                        +1.9 -1.1    +0.7  -1.9
20080214  44  OH231.8  227 41   -2.2  -0.2    -4   -4   -0.4 +8.3
                                                   86   -0.1 +3.8    +0.3  -4.5
20080214  66 IRC+10214 273 32   +3.9  -0.6   -43  -42   -3.1 +1.9
                                                   48   +0.0 +0.9    +3.1  -1.0
20080214  70   X Her   392 55   -0.2  -3.7   -83  -82   -0.3 -0.7
                                                        -0.4 -1.5    -0.1  -0.8
20080214  73 IRC+20362  85 49   +2.0  +1.7   +39  -51   -2.2 -3.3
                                                          -    -       -     -
20080215  43 IRC+10216 153 82   +0.7  +2.5   +18  108   +3.6 -2.8
                                                        +3.5 -2.0    -0.1  +0.8
20080215  77 IRC+20362  87 56   +2.6  -1.8   +47  -43   -1.8 -3.2
                                                        -2.0 -2.7    -0.2  +0.5
20080216  39  CRL618   301 30   -1.4  +0.2   -56  -56   +0.2 -0.4
                                                   34   -0.9 -0.1    -1.1  +0.3
20080218   8  CRL618   306 57   +1.9  -1.5   -54  -53   +0.0 +1.9
                                                   37   +1.2 +2.7    +1.2  +0.8
20080225  39    CIT6    34 77   -0.9  -1.9   -51  -51   +0.2 -3.1
                                                        +0.3 -2.3    +0.1  +0.8  
20080225  42 IRC+10216 203 83   -0.9  +4.3   +60   61   -0.5 -0.4
                                                        -2.2 +3.4    -1.7  +3.8
20080226  18 IRC+10216 144 82   +0.2  +1.8   +27   27   +1.5 -2.0
                                                        +1.4 -2.1    -0.1  -0.1
20080228  41 IRC+10216 262 59   -0.8  +0.8   +76   76   -2.8 +3.2
                                                  -14   -2.6 +3.4    +0.2  +0.2
20080301  38  CRL618   303 50   +0.0  -0.2   -52  -52   +0.7 +0.0
                                                   38   +1.6 +1.0    +0.9  +1.0
20080303  54   CIT6    367 79   -0.0  -1.5   -18  -19   +0.4 -1.3
                                                 -109   +0.4 -0.6    +0.0  +0.7
20080311  13  CRL618   312 63   -3.1  +5.1   -57  -57   +1.5 -1.1
                                                        +1.8 -0.2    +0.3  +0.9 
20080311  16   CIT6    426 40   +1.9  +0.4   -51  -51   -0.5 -0.8
                                                        -0.0 -1.7    +0.5  -0.9
20080312  35 IRC+10216 230 80   +0.0  +2.4   +31  +32   +3.7 -3.3
                                                        +2.7 -3.1    -1.0  +0.2
20080312  65  GL1922   173 45   -2.1  -0.1   +52  +52   +1.0 +5.2
                                                        +0.1 +4.4    -0.9  -0.8
20080315  62  GL1922   165 44   +0.6  +2.3   +60  +60   +0.8 +3.0
                                                        +1.1 +1.6    +0.3  -1.4
20080330* 50 IRC+10216 256 67   -3.7 +10.7   +88  +88   -2.0 -3.2
                                                   -2   -1.2 -6.8    +0.8  -3.6
20080330* 75  GL1922   142 34   -1.5  +0.6   ---  +74   -0.4 +1.4
                                                        -0.4 +1.4     ???   ???
20080331  35  CRL618   301 28   +2.9  -2.7   -57  -57   +2.1 -4.6
                                                  +33   +2.6 -4.0    +0.5  +0.6 
20080331  64 IRC+10216 274 31   +3.6  +0.3   +46  +46   -2.0 +2.8
                                                  -44   +1.3 +1.3    +3.3  -1.5
                               ------------   
               N=38  Mean       -0.3  -0.0
                     s.d.        2.0   2.5

Notes

• the values daz,del are the centroids of the jigglemaps w.r.t. the 'before' pointing, while the uaz,uel are the offsets of the pointings from the pointing model
• The K-mirror gui km_pointing provides the "ImageAngle" (IMG_ANG) during the observation. This is usually recorded, but is also calculated using a) the known circumstances at the mid-point of the observations and b) the first two rules for calculating the term 'RPA'. This is nonetheless a somewhat unpredictable component of the K-mirror orientation, which takes the value 0 or 90 or -90 or +-180 degrees. My code is often unable to distinguish between two possible values on the basis of maximum tracking time prior to the mirror hitting one of its limits, and is a priori unable to apply the third rule, and so often both remaining options are given, one above the other. Apart from the entry for 20080121_00024 (see below), the table suggests nothing amiss in the orientation of the array during these observations. (Uncertainties of a couple of degrees may be expected in the observed position: it needs to be made at mid-observation, and may be changing quickly at certain positions on the sky).
• a value of -60 is adopted for the IMG_ANG for 20080121_00024
• The data of UT20080330 are suspect, due to a possible SMU power failure - see JCMT fault [20080330.002]) - and the second of the two measures on that day were preceded and followed by pointings that are recorded with identical offsets (to 4dpl.). Neither is included in the analysis.

The offsets (of the RA-DEC centroids) are plotted as a function of IMG_ANG and as a function of (E-2K) below:

One egregious outlier (that with the most negative elevation residual) is removed prior to solving for the K-mirror coefficients. The residuals are plotted in the AZ/EL plane below left, and (below right) after correction by the best-fitting K-mirror model:

The data are joined in order of the K-mirror angle. Data such as these may eventually show the systematics that we also seek via the HARP pointing 'arcs'. After eliminating the outlier, the remaining (33) data, for which the K-mirror angle has been explicitly recorded, produce the following solution for the K-mirror coefficients:

   N     xc     yc     x0     y0     xd     yd   |  Az-rms-El
  20   -1.6   -0.8   -1.9   -1.7   -2.1    1.7   |  2.0   1.2   (20080222 *)
  22   -1.7   -0.6   -1.8   -1.9   -2.1    1.5   |  2.0   1.3   (20080225)
  24   -1.6   -0.6   -1.7   -1.9   -2.0    1.3   |  1.9   1.2   (20080228)
  26   -1.6   -0.5   -1.7   -1.7   -1.9    1.3   |  1.8   1.2   (20080301)
  28   -1.5    0.1   -1.2   -1.1   -1.2    1.2   |  1.9   1.7   (20080311)
  30   -1.5    0.0   -1.3   -1.3   -1.4    1.1   |  1.9   1.7   (20080312)

  33   -1.1    0.0   -0.5   -1.6   -0.8    0.3   |  2.0   1.7   (20080331)
  32   -0.7   -0.2    0.4   -1.0   -0.3    0.3   |  1.8   1.8   ..........
  31   -1.0   -0.5   -0.1   -1.2   -0.5    0.9   |  1.8   1.3   ..........

* - An analysis of HARP allsky pointing data from ~mid-February using the 20080222 results is given here.

The resulting performance, (2.0",1.7") rms, is only a slight improvement on the raw performance of (2.1",2.0"), although there is clear condensation in the centre of the distribution. The removal of two further (successively most distant) outliers may be justified, leading to a solution (!) with essentially nominal performance characteristics:

However, applying the solution to the 99 regular HARP (AZ/EL) fivepoint data collected during the last week of HARP operations (UT20080325-29) effected no improvement in rms pointing performance : (2.6",3.7") to (2.5",3.9").