This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
operations:pointing [2012/05/03 04:20] Jim Lovell |
operations:pointing [2017/05/30 05:41] Jamie McCallum |
||
---|---|---|---|
Line 1: | Line 1: | ||
====== Antenna Pointing Solutions ====== | ====== Antenna Pointing Solutions ====== | ||
+ | |||
+ | |||
===== Reduction of observations and producing a model ===== | ===== Reduction of observations and producing a model ===== | ||
+ | |||
+ | There are two methods of producing a pointing model for the 12m antennas. There's the 'old' way using octave scripts and the power sensor or the 'new' way using the built-in PCFS pointing routines. The 'new' way is recommended because the Field System no longer uses the pointing model in the antenna controller by default. However it's important to keep the pointing models synchronised between the PCFS and antenna controller because some applications (like spacecraft tracking) still require the model in the controller. | ||
+ | |||
+ | The model parameters are implemented slightly differently in the PCFS and the Controller. Here are | ||
+ | some notes on the differences between the internal PCFS model and the model implemented within the controller: {{:operations:model_params_12m_controller_and_pcfs.pdf|Model Param notes}}. Note the sign change in one of the parameters. | ||
+ | |||
+ | |||
+ | ==== Obtaining a pointing model with the PCFS ==== | ||
+ | It's a good idea to allow about 24h of observing to obtain a new pointing solution. Then some additional time is required (a few hours) to confirm that the new solution is working. | ||
+ | |||
+ | - Start up the field system software as usual and make sure the DBBC is set up in a standard R1/R4 configuration. | ||
+ | - It's probably worth using eRemoteControl for this too and setting up the log monitor so you will be alerted if there's a problem. | ||
+ | - The pointing routine will append (or make) a log file called ''/usr2/log/sx.log'' which is later used to make a pointing solution. If this file already exists, consider renaming it so there's a fresh one for the new solution. | ||
+ | - Then, in the field system: <code>proc=point | ||
+ | initp | ||
+ | acquire</code> | ||
+ | - The Field System will then automatically choose a pointing source and determine an offset. It will continue to do this indefinitely for a range of sources. Let it run for 24h. | ||
+ | |||
+ | === During the pointing run === | ||
+ | |||
+ | Lots of messages are produced from the fivpt (pointing) routine. It's worth watching the output (initially anyway) to check it's making sense: | ||
+ | * [[operations:documentation:fivept_output|What does the fivpt output mean?]] | ||
+ | * [[operations:documentation:pointing_monitoring|Monitoring and plotting pointing data]] during a pointing run | ||
+ | |||
+ | === Ending a pointing run === | ||
+ | |||
+ | When you think you have enough data, stop the pointing run in the field system by killing the pointing routines, zeroing the pointing offsets and send the antenna to stow: | ||
+ | <code>kill | ||
+ | azeloff=0d,0d | ||
+ | source=stow | ||
+ | </code> | ||
+ | |||
+ | === Finding a pointing solution === | ||
+ | |||
+ | Open a terminal on the Field System (make sure X11 is being forwarded) then start up the pointing solution GUI: | ||
+ | <code>pdplt &</code> | ||
+ | Load the log file with ''File -> New''. To cycle through the plots use the ''Graph'' menu or ''n'' and ''p''. By default the fit residuals are shown but you can also plot the raw data under the ''Data'' menu. Obviously bad data can be flagged by clicking on points. To remove outliers automatically, try the ''Edit -> x-sigma'' option to remove points more than a specified number of sigma out. After flagging points, you can use the ''Edit -> Reprocess'' option to re-fit the data. If necessary, do some more flagging and reprocess. | ||
+ | |||
+ | The ''Edit -> Modify Parameters'' option lets you control how and which parameters are used. The default (params 1 through 8 and number 23 set to 1, the rest 0) seems to work well. Note that parameters 15, 16, 17 and 18 (frequency 2 tilt) have been used at the GSFC 12m to improve the solution. However, it's worth producing a solution using the default parameters too because this can be copied to the controller (which doesn't support the other parameters). | ||
+ | |||
+ | The ''Statistics'' menu shows you how good the fit is. | ||
+ | |||
+ | When you are satisfied with the pointing solution: | ||
+ | <code> File -> Save </code> | ||
+ | This will save two new files in ''/usr2/log''. The ''xtr'' file contains the data which can be re-loaded and reduced if necessary. The ''err'' file contains the new model: look for ''new_model'' in the file. | ||
+ | |||
+ | === Updating the pointing model === | ||
+ | To update the pointing model (which will go into ''/usr2/control/mdlpo.ctl'') use the ''update'' command. If the new model is called err2013.043, type | ||
+ | <code>update 2013.043</code> | ||
+ | Then check the new file to see if it's been updated. | ||
+ | |||
+ | To copy the model into the controller, use HMI | ||
+ | ==== Obtaining a pointing model with the Power Sensor ==== | ||
+ | |||
An octave script (fitter.m) is available for reducing scans made using "ramps" in the HMI controller and the "monochanstart.sh" scripts in ''~oper/Agilentu2k/''. The monochanstart.sh script reads from the power sensor and writes it to a file - usually ~oper/Agilentu2k/sensordata.txt. To use the script, start octave and run ''fitter'' from the command line. You will be prompted for a file name and the mjd day and seconds of the scan. The output is the amplitude (in dB), the FWHM and central location (in seconds). To best use this, scans made using a ramp where the nominal position is reached at a defined epoch are recommended as this allows for simple estimation of the pointing offset (fitter central location - ramp epoch)*(ramp rate in degrees/sec). The script is made up of three files - fitter.m, fittererr.m and julday.m and requires the debian packages octave, octave-io and octave-optim to be installed. | An octave script (fitter.m) is available for reducing scans made using "ramps" in the HMI controller and the "monochanstart.sh" scripts in ''~oper/Agilentu2k/''. The monochanstart.sh script reads from the power sensor and writes it to a file - usually ~oper/Agilentu2k/sensordata.txt. To use the script, start octave and run ''fitter'' from the command line. You will be prompted for a file name and the mjd day and seconds of the scan. The output is the amplitude (in dB), the FWHM and central location (in seconds). To best use this, scans made using a ramp where the nominal position is reached at a defined epoch are recommended as this allows for simple estimation of the pointing offset (fitter central location - ramp epoch)*(ramp rate in degrees/sec). The script is made up of three files - fitter.m, fittererr.m and julday.m and requires the debian packages octave, octave-io and octave-optim to be installed. | ||
===== Hobart 12m model ===== | ===== Hobart 12m model ===== | ||
+ | |||
+ | ==== Version 2016.135 ==== | ||
+ | |||
+ | Taken following the installation of the new cable wrap. | ||
+ | |||
+ | Newly derived coefficients: | ||
+ | <code> | ||
+ | 00008 2017 135 00 53 37 0 | ||
+ | 90.0000 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 | ||
+ | -0.7106876969 0.0000000000 0.0228546374 0.0497939363 0.0030318052 | ||
+ | -0.0002992165 0.1175717711 0.0146358637 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0019028779 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | </code> | ||
+ | |||
+ | The HMI controller was updated 30/05/17 | ||
+ | |||
+ | |||
+ | ==== Version 2016.235 ==== | ||
+ | |||
+ | Taken following re-installation of S/X feed and movement of subreflector to original position. | ||
+ | |||
+ | Newly derived coefficients: | ||
+ | <code> | ||
+ | 00007 2016 235 23 14 09 0 | ||
+ | 90.0000 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 | ||
+ | -0.4333212972 0.0000000000 0.0158737078 0.0405583084 0.0002418841 | ||
+ | 0.0006486370 0.1193821207 0.0167687163 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0006459429 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | </code> | ||
+ | |||
+ | ^ FS Coeff ^ HMI Coef ^ New FS model ^ | ||
+ | | P1 | G2| -0.4333 | | ||
+ | | P3 | G3| 0.0159 | | ||
+ | | P4 | -G1| 0.0406 | | ||
+ | | P5 | G5| 0.0002 | | ||
+ | | P6 | -G4| 0.0006 | | ||
+ | | P7 | G7| 0.1194 | | ||
+ | | P8 | G8| 0.0168 | | ||
+ | | P23 | G9| 0.0006 | | ||
+ | |||
+ | ^ HMI Coef ^ Value ^ | ||
+ | | G1 | -0.0406 | | ||
+ | | G2 | -0.4333 | | ||
+ | | G3 | 0.0159 | | ||
+ | | G4 | -0.0006 | | ||
+ | | G5 | 0.0002 | | ||
+ | | G6 | not used | | ||
+ | | G7 | 0.1194 | | ||
+ | | G8 | 0.0168 | | ||
+ | | G9 | 0.0006 | | ||
+ | |||
+ | Field system and HMI have been updated | ||
+ | |||
+ | ==== Temporary HMI model ==== | ||
+ | This is a temporary model applied in HMI during broadband tests with the Callisto system. | ||
+ | |||
+ | Original model: | ||
+ | <code> | ||
+ | P1 -0.0335 | ||
+ | P2 -0.4380 | ||
+ | P3 0.0066 | ||
+ | P4 -0.0026 | ||
+ | P5 -0.0023 | ||
+ | P6 0 | ||
+ | P7 0.1538 | ||
+ | P8 0.0214 | ||
+ | P9 0.0026 | ||
+ | </code> | ||
+ | |||
+ | Additional offsets are | ||
+ | <code> | ||
+ | Az = +2.7 arcmin = 0.045 deg | ||
+ | El = -4.0 arcmin = -0.0667 deg | ||
+ | </code> | ||
+ | |||
+ | The DC terms in the pointing model are P2 (Az) and P7 (El), so new model is: | ||
+ | <code> | ||
+ | P1 -0.0335 | ||
+ | P2 -0.393 | ||
+ | P3 0.0066 | ||
+ | P4 -0.0026 | ||
+ | P5 -0.0023 | ||
+ | P6 0 | ||
+ | P7 0.08713 | ||
+ | P8 0.0214 | ||
+ | P9 0.0026 | ||
+ | </code> | ||
+ | |||
+ | |||
+ | |||
+ | ==== Version 2016.151 ==== | ||
+ | 2 June 2016 | ||
+ | |||
+ | Observations over 24he. Little change to rms before and after | ||
+ | |||
+ | <code> | ||
+ | Data: | ||
+ | xEl mean -0.0013 | ||
+ | xEl RMS 0.007 | ||
+ | El mean -0.015 | ||
+ | El RMS 0.008 | ||
+ | |||
+ | After solution: | ||
+ | xEl RMS 0.007 | ||
+ | El RMS 0.009 | ||
+ | </code> | ||
+ | |||
+ | Solutions applied err2016.151 | ||
+ | |||
+ | {{:operations:opsmac1.png?200|}} | ||
+ | |||
+ | |||
+ | |||
+ | ==== Version 4.2 ==== | ||
+ | |||
+ | Model updated from observations taken over a few days around 2014/205. Slight dependance on Elevation offset with Azimuth removed. Otherwise looks good. | ||
+ | |||
+ | ==== Version 4.1 ==== | ||
+ | |||
+ | Slight offset in elevation. Model updated from observations on 2013/323 | ||
+ | |||
+ | ==== Version 4.0 ==== | ||
+ | |||
+ | Taken using PCFS on 2013 day 043/044. Coefficients used during pointing: | ||
+ | |||
+ | <code> | ||
+ | * 0 0 0 0 0 0 0 | ||
+ | * 90.0 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 | ||
+ | * -0.4263 0.0 0.0048 0.0363 -0.0004 | ||
+ | * -0.0026 0.1906 0.0250 0.0 0.0 | ||
+ | * 0.0 0.0 0.0 0.0 0.0 | ||
+ | * 0.0 0.0 0.0 0.0 0.0 | ||
+ | * 0.0 0.0 0.0 0.0 0.0 | ||
+ | * 0.0 0.0 -0.0014 | ||
+ | </code> | ||
+ | |||
+ | Newly derived coefficients: | ||
+ | <code> | ||
+ | 00002 2013 044 23 30 55 0 | ||
+ | 90.0000 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 | ||
+ | -0.4380196333 0.0000000000 0.0065650782 0.0334785432 -0.0023106572 | ||
+ | 0.0025938717 0.1538095325 0.0214055236 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0025560106 0.0000000000 0.0000000000 | ||
+ | 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 | ||
+ | </code> | ||
+ | |||
+ | In HMI these are translated to: | ||
+ | |||
+ | ^ Coeff ^ Old model ^ New model ^ | ||
+ | | P1 | -0.0363 | -0.0335 | | ||
+ | | P2 | -0.4263 | -0.4380 | | ||
+ | | P3 | 0.0048 | 0.0066 | | ||
+ | | P4 | -0.0026 | -0.0026 | | ||
+ | | P5 | -0.0004 | -0.0023 | | ||
+ | | P6 | not used || | ||
+ | | P7 | 0.1906 | 0.1538 | | ||
+ | | P8 | 0.0250 | 0.0214 | | ||
+ | | P9 | -0.0014 | 0.00255 | | ||
+ | |||
+ | Field system and HMI have been updated | ||
==== Version 3.4 ==== | ==== Version 3.4 ==== | ||
30 June 2011\\ | 30 June 2011\\ | ||
Line 17: | Line 241: | ||
P8 = 0.0000\\ | P8 = 0.0000\\ | ||
P9 = 0.0029\\ | P9 = 0.0029\\ | ||
- | |||
==== Version 3.3 ==== | ==== Version 3.3 ==== | ||
17 Feb 2011\\ | 17 Feb 2011\\ | ||
Line 150: | Line 373: | ||
===== Katherine 12m model ===== | ===== Katherine 12m model ===== | ||
+ | ==== Version 2016.151 ==== | ||
+ | |||
+ | Observations over 24h following recent maser maintenance. Little change to rms before and after | ||
+ | |||
+ | <code> | ||
+ | Data: | ||
+ | xEl mean 0.0027 | ||
+ | xEl RMS 0.007 | ||
+ | El mean -0.0118 | ||
+ | El RMS 0.0118 | ||
+ | |||
+ | After solution: | ||
+ | xEl RMS 0.0126 | ||
+ | El RMS 0.013 | ||
+ | </code> | ||
+ | |||
+ | Solutions applied err2016.151 | ||
+ | |||
+ | {{:operations:ke_pointing_2016_151.png?200|}} | ||
+ | |||
==== Version 1.0 ==== | ==== Version 1.0 ==== | ||
Line 197: | Line 440: | ||
===== Yarragadee 12m model ===== | ===== Yarragadee 12m model ===== | ||
+ | |||
+ | ==== Version 2016.151 ==== | ||
+ | |||
+ | Observations over 24h following recent maser maintenance. Little change to rms before and after | ||
+ | |||
+ | <code> | ||
+ | Data: | ||
+ | xEl mean -0.0006 | ||
+ | xEl RMS 0.011 | ||
+ | El mean 0.015 | ||
+ | El RMS 0.008 | ||
+ | |||
+ | After solution: | ||
+ | xEl RMS 0.012 | ||
+ | El RMS 0.010 | ||
+ | </code> | ||
+ | |||
+ | Solutions applied err2016.151 | ||
+ | |||
+ | {{:operations:yg_pointing_2016_151.png?200|}} | ||
+ | |||
+ | This solution was put into HMI on 22 June 2016, ~23:45 UT (day 174). Parameters are: | ||
+ | |||
+ | <code> | ||
+ | G1 -0.0105 | ||
+ | G2 -0.0070 | ||
+ | G3 0.0180 | ||
+ | G4 0.0094 | ||
+ | G5 -0.0024 | ||
+ | G7 0.2059 | ||
+ | G8 0.0158 | ||
+ | G9 0.0007 | ||
+ | </code> | ||
+ | |||
+ | Note that these were initially put in to HMI on June 20 but the sign for G4 was wrong. Not sure if that had a significant impact or not. | ||
+ | |||
+ | Also updated HMI (on day 175, 00 UT) with revised station coordinates. Only sig change was to elevation which was 50m too low. | ||
+ | |||
+ | <code> | ||
+ | Lat, Long, Elevation: | ||
+ | Before: -29.0465, 115.3467, 200.0 | ||
+ | After: -29.0469, 115.3465, 250.48 | ||
+ | </code> | ||
+ | |||
+ | |||
+ | |||
==== Version 1.5 ==== | ==== Version 1.5 ==== | ||
Line 373: | Line 662: | ||
Az 20.21 = -1141 (i.e. -0.1141 deg) | Az 20.21 = -1141 (i.e. -0.1141 deg) | ||
</code> | </code> | ||
- |