31jan03 - updated on 11feb03 Phase II testing summary: ------------------------- Participants: Frederic Gueth (IRAM), Jerome Pety (IRAM), Leonardo Testi (Arcetri, SSR), Chris Wilson (McMaster, SSR), Alain Coulais (Obs Paris), Andrew Baker (Max Planck), Joe Schwarz (ESO), Debra Shepherd (NRAO), Steve Myers (NRAO) Tutorial participation: ----------------------- AIPS++ tutorial by Debra and Steve: Frederic, Jerome, Leonardo, Chris, Andrew, Joe, Alain CLIC tutorial by Jerome and Frederic: Debra, Steve, Joe IRAM provided notes, fluxes, and ALMA-TI FITS files for 2 more projects: -------------------------------------------------------------- Project h121 - line observations of an evolved shell Project i028 - CO mosaic of an outflow Phase II testing during the week of 25jan03: -------------------------------------------- Chris: 07feb97 GG Tau - 1mm and 3mm continuum processed together Project h121 - 01may98: 3mm 12CO1-0 line & continuum. reduced & imaged sucessfully. - 05may98: reduced sucessfully but not imaged yet. (ran out of time) Jerome: 07feb97 GG Tau - 1mm and 3mm continuum processed together Project h121 - 09may98: reduced & imaged 12co1-0 line successfully. Leonardo: 07feb97 GG Tau - 1mm continuum reduced and imaged after 3mm continuum reduction during tutorial. Frederic: Project i028 - Outflow mosaic fields were calibrated successfully but could not image because PdBI fits transofrmation did not transfer field positions correctly. Andrew: 07feb97 GG Tau - 1mm and 3mm continuum processed together. Also began installing AIPS++ on personal linux machine at Max Planck and converted 3 PdBI observations to ALMA-TI FITS format for future processing. Problems identified fixed? ---------------------------------------------------- ------ - No 3mm USB choice in AIPS++ iramcalibrater.g yes - verified - When spectral set up changed for a source, yes - verified FLUX did not work. - Flags were taken into account when calculating almost - needs flux during AIPS++ FLUX but the average weight verification of the data used to determine source flux and antenna efficiency did not. flux is OK now, efficienceis need to be fixed - AIPS++ iramcal.rf: A typo in the name field caused no functionto fit a bandpass solution to all fields - Can't abort out of AIPS++ iramcal functions due no to lack of id(). - AIPS++ FLUX 'fixed' parameter doesn't appear to no work in GUI mode (but does work when running from script) - AIPS++ needs a stop/abort function when deriving no flux solutions in interactive mode - the data set should not be changed if you abort. - AIPS++ phcor gave an error when running from glish yes - needs because of an index out of range issue. verification (the function still worked) - Plots should not show the flagged data in AIPS++ no rf, phase, flux, amp. It can be confusing to the user if flagged data shows up. - 05may98 data set: AIPS++ phcor has an error: ? 0 or negative parameter in the data. But the function ran OK, Not sure if this is an AIPS++ bug or PdBI data error. - PdBI did not transfer mosaic positions to fits. yes - needs (fixed by Robert Lucas) verification - Antenna positions are wrong in the AIPS++ MS no (uv data was correct) - Error bars on the rf plots of channel data appear no to be inconsistent with the dispersion between channels. - IRAM/PdBI data reduction cookbook requires -- modifications. Comments from Joe Schwarz & Chris Wilson were given to Debra on 31jan03. Debra will provide a new version by 7feb03. AIPS++ & CLIC tutorials identified several enhancements required for processing millimeter interferometric data: ---------------------------------------------------------------- Enhancement requests: - AIPS++ needs to plot total power, Tsys, pointing, & water column versus time like the CLIC first look plots to be able to identify likely bad data and flag it. This is needed for any millimeter processing. AIPS++ said they will incorporate this new functionality. - AIPS++ should plot ALMA_FLAG_CORR_ROW results during phcor to see qualitatively how many scans were switched with the NO_PHAS_CORR data. A glish script was drafted outside of the iramcalibrater.g script that will be incorporated in the iramcal.phcor function. This plot should also be available in the pgplot of normalized antenna efficiency that is created during the flux determination. - During the flux calibration, AIPS++ should be able to limit the time range over which a flux solution will be found. This can be done in a very limited way now by using msplot to temporarily flag data during the flux cal and then unflagging the data when finished. - AIPS++ should be able to limit the time range on the GSPLINE fits to solve for phase or amplitude versus time. - AIPS++ should be able to choose which reference antenna to use during calibration. - AIPS++ should suppress the "no data found" warning during the msplot.locate function unless there is no data found in the entire msplot locate box. Comment from Joe Schwarz on GUIs: --------------------------------- - GUI panels like the 'msplot data selection' 1) wastes a lot of screen space (too much blank area) 2) Doesn't help to guide the user's eye (parameters are listed but not in subgroups on the same screen) Because of 1) and 2), screens are confusing and difficult to document/explain to the user.