PDS_VERSION_ID = PDS3 RECORD_TYPE = STREAM OBJECT = TEXT PUBLICATION_DATE = 2000-11-02 NOTE = "MGS RST Instrument Health Report HEA0307A.TXT" END_OBJECT = TEXT END Miscellaneous ============= Vincent Hung reports that SA's are working on our intermittent query and CD writer problems. Engineering Data ================ Engineering Data Statistical Summary from CSV2STAT: First SCET: 2000-10-31T00:06:58.771 Last SCET: 2000-10-31T23:59:58.759 CHANNL CHANNEL TITLE DN-LO DN-HI EU-LO EU-HI EU-AVG/STDEV RECDS ------ ------------- ----- ----- --------- --------- ---------/-------- ----- F-0190 HGA_AZ_ANG 6424 12534 1.57E+00 3.06E+00 2.86E+00/9.04E-03 2554 F-0195 HGA_EL_ANG -11233 11089 -2.74E+00 2.71E+00 -8.74E-01/3.25E-02 2554 L-0034 MOT1_RANGING 0 0 5105 L-0036 MOT1_TLM_MOD 0 0 5109 L-0037 MOT1_USO_ENA 0 0 5105 L-0044 MOT2_RANGING 1 1 5105 L-0046 MOT2_TLM_MOD 0 1 5109 L-0047 MOT2_USO_ENA 1 1 5105 L-0053 RF_SW_OSC 2 2 5109 L-0060 TWTA1_FILMNT 0 0 5105 L-0061 TWTA1_HV 0 0 5108 L-0070 TWTA2_FILMNT 0 1 5108 L-0071 TWTA2_HV 0 1 5105 L-0080 USO_PWR 0 0 5105 L-0090 KaBLE_STAT 0 0 5105 L-0091 KaBLE_ENABLE 1 1 5108 L-0111 MOT1_RCV_AGC 255 255 -1.53E+02 -1.53E+02 -1.53E+02/7.66E-07 5105 L-0121 MOT2_RCV_AGC 159 255 -1.53E+02 -1.26E+02 -1.50E+02/1.16E-01 5105 L-0131 TWTA1_HLX_I 0 1 -3.84E-02 -1.85E-02 -3.82E-02/2.36E-05 5105 L-0132 TWTA1_ANOD_V 0 1 -1.00E+03 -1.00E+03 -1.00E+03/1.45E-02 636 L-0141 TWTA2_HLX_I 17 42 2.86E-01 7.85E-01 4.63E-01/2.60E-03 5105 L-0142 TWTA2_ANOD_V 61 229 -8.66E+02 -5.30E+02 -6.30E+02/5.94E+00 636 L-0200 USO_REG_V 203 205 4.06E+00 4.10E+00 4.08E+00/2.67E-04 635 L-0201 USO_OVEN_V 128 130 2.56E+00 2.60E+00 2.58E+00/3.34E-04 635 T-0302 EPC1_T 188 197 -4.65E+00 -7.38E-01 -2.98E+00/3.28E-02 637 T-0303 EPC2_T 182 192 -2.43E+00 1.71E+00 -8.65E-01/4.58E-02 637 T-0310 RF_ISOLTR1_T 229 233 -6.05E+00 -1.15E+00 -3.34E+00/8.95E-03 5109 T-0311 RF_ISOLTR2_T 225 233 -6.05E+00 3.13E+00 -1.54E+00/2.61E-02 5109 T-0312 TWT1_T 228 233 -6.05E+00 -2.48E-02 -2.80E+00/3.75E-02 637 T-0313 TWT2_T 219 233 -6.05E+00 8.56E+00 -9.22E-01/1.76E-01 637 T-0315 USO_T 145 148 1.94E+01 2.03E+01 1.98E+01/8.92E-03 637 T-0316 KA_AMP_T 215 219 8.56E+00 1.16E+01 9.74E+00/2.40E-02 642 L-0200, L-0201, and T-0315 are nominal. Open Loop Data ============== Data Collection --------------- We have transferred the following open loop data to Stanford: YYYY/DDD DSS DSP SAMP BW DSP BIT REC RECS RECS MAX START STOP RATE MODE RESN LEN MSSG SNR -------- --- -------- -------- ---- ---- ---- ---- ---- ----- ---- ----- 2000/306 45 19:14:00 19:21:00 5000 2000 2 12 1666 2101 0 44.3 2000/307 65 05:03:00 05:10:00 5000 2000 2 12 1666 2101 0 43.4 2000/307 65 07:00:00 07:07:00 5000 2000 2 12 1666 2101 0 45.6 2000/307 65 08:58:00 09:05:00 5000 2000 2 12 1666 2101 0 41.7 2000/307 65 10:56:00 11:03:00 5000 2000 2 12 1666 2101 0 46.5 mk_rpt_line: 0 channels with preplook.log files 2000/307 15 12:53:00 13:00:00 5000 2000 2 12 1666 2101 0 47.0 2000/307 15 14:51:00 14:58:00 5000 2000 2 12 1666 2101 0 47.6 Anomalies --------- Data from occultation 7394i (DSS 65, start time 2000/306-05:03:00) have carrier level fluctuations of +/-0.5 dB. Data from occultation 7396i (DSS 65, start time 2000/306-08:58:00) have a noise background that decreases linearly by about -1.5 dB over the course of the 7 minute recording. Closed Loop Data ================ Data Collection --------------- No new data. Analyses -------- Boris Semenov has calculated the differences between spacecraft position vectors reported in the JPL and Goddard orbit reconstructions. He showed a plot of the results at the SDVT meeting on 4 October. The plot is reporduced here as Figure HEA0307A.PS1. Frank Lemoine returns to GSFC tomorrow. Prior to leaving Toulouse, he forwarded the (slightly edited) comments below about MGS one-way tracking data and their utility in orbit determination: Since the last MGS Radio science team meeting, and Dave Hinson's comments on the use of the one-way Doppler data, I've been testing addition of the one-way doppler from MGS to our OD processing. Below is a summary of arcs in July-Sept 2000 where the orbits were calculated using (A) 2/3 way Doppler only, and (B) the 2/3 way Doppler plus one-way Doppler. A frequency bias is adjusted for each pass of one-way data. Also, the one-way data is downweighted by 2X in the OD processing compared to the two-way data. The orbits with the one-way Doppler appear more consistent - particularly in the radial direction. The consistency in the cross and along-track directions is also noticeably improved. The processing of the one-way Doppler is an extremely labor intensive operation. The ODF has gobs and gobs of one-way data. However, easily 40 to 50 percent of the data from the ODF were deleted by the time the final GEODYN run was made - and this after doing multiple runs to edit and clean up the data. In these runs, the one-way data were only selected when there were large gaps (four hours or more) in the 2/3 way tracking. The RMS to the one-way data seems quite impressive - comparable to the two-way RMS of fit in many cases. However, if one examines the residual plots, one can see patterns in the residuals that could be signs of remaining signals due to attitude mismodeling, accidental inclusion of data during occultation, or some other unknown issues. To make sure the one way data were "real", I double checked the times of the usable data on one day for a six hour period (000730) against times in an SOE file which gave the start and stop of occultation. Indeed, there were no "imaginary" data points. In other words there were no points in occultation that shouldn't be in the OD run. Georges Balmino wrote and supplied a subroutine to edit data for an oblate planet + arbitrary radius (atmosphere) for the GRGS processor (GINS). The subroutine is modular so that we can eventually include this directly in GEODYN - hopefully in the near future. It seems clear the one-way data should be included systematically in the MOLA processing - as the data seem helpful - but some work will need to be done to automate the processing. Since the data require so much editing and clean-up before use (not necessarily an easy task). Blindly adding a SELECT card and turning the crank does not work very well. One needs to start with an apriori one-way bias, and a converged orbit determined from the two-way data and pre-edit the e+09 residuals before even getting started. It would be helpful to know why we have so much more one-way data than two way data as it implies the DSN spends a lot of time just listening to MGS - as opposed to time spent tracking in two-way mode. As we know, the time spent in two-way tracking mode is quite circumscribed (cf, see the residual plots for the two-way tracking). I hope what we have here are "real" data, and not data fabricated by a DSN subprocessor for the pleasure and torture of data analysts .... On the geodesy.gsfc.nasa.gov anon ftp site, (cd to dist/one-way) I've placed residual files for the 2-way Doppler and the 1-way doppler for various arcs in august 2000. I would be curious what RST members at Stanford make of some of the residual patterns for the one way data. For example, some of the patterns around 410 to 414 km seem quite curious. Orbit from 2/3 way doppler only -------------------------------- RMS (rad, cross-tr, along-tr) MGS orbit overlaps with mgm1004d for arc set orbit6e.articulate (all units are meters) 000725.000801.9912.000730.000805.RMS .1031 11.7052 4.166 000730.000805.9912.000803.000810.RMS .9623 13.4555 15.847 000803.000810.9912.000808.000815.RMS 1.5175 20.2107 19.500 000808.000815.9912.000813.000820.RMS .4135 18.0222 7.219 000818.000825.9912.000823.000830.RMS 2.8376 8.6947 50.075 000823.000830.9912.000828.000904.RMS 1.6435 47.4773 8.654 000828.000904.9912.000902.000909.RMS 13.9440 8.0459 56.005 Orbits from 2/3 way Doppler + One-way Doppler --------------------------------------------- RMS (rad, cross-tr, along-tr) MGS orbit overlaps with mgm1004d for arc set orbit6f.articulate (all units are meters) 000725.000801.000730.000805.RMS .0869 8.0547 2.8031 000730.000805.000803.000810.RMS .1636 9.2713 4.5967 000803.000810.000808.000815.RMS .8525 14.8993 7.1191 000808.000815.000813.000820.RMS .9549 24.8063 9.6735 000818.000825.000823.000830.RMS .3133 3.6431 2.1243 000823.000830.000828.000904.RMS .0551 17.3204 6.4721 000828.000904.000902.000909.RMS .9303 6.7224 5.4292 ===================================================================== RMS of fit for arcs with 2/3 way AND 1 way Doppler --------------------------------------------------------------------- In OD: sigma of two-way data = 0.1 Hz; sigma of one-way data = 0.2 Hz --------------------------------------------------------------------- 0.284 RU approx = 1.0 m; range is round trip. Multiply Hz by approx 35.46 to convert to mm/s (round trip). arc start stop 1wyDoppler 2wyDoppler Range RMSPOS Nobs RMS Bobs RMS Nobs (m) Hz Hz RU 000725 18:10 000801 15:10 12138 0.0093 5727 0.0127 0 9.3 000730 15:50 000805 15:20 8604 0.0067 4352 0.0093 0 15.4 000803 12:00 000810 16:50 10364 0.0060 4740 0.0085 0 15.8 000808 01:50 000815 16:30 11023 0.0067 3991 0.0098 0 10.0 000813 13:20 000820 10:30 5485 0.0055 3461 0.0064 0 17.7 000818 18:50 000825 13:50 8517 0.0051 3874 0.0060 0 41.6 000823 16:20 000830 04:50 7173 0.0052 5890 0.0059 0 14.6 000828 14:40 000904 16:30 5507 0.0051 5679 0.0052 0 16.7 RMS of fit for arcs with 2/3 way Doppler only arc start stop 1wyDoppler 2wyDoppler Range RMSPOS Nobs RMS Bobs RMS Nobs (m) Hz Hz RU 000725 18:10 000801 15:10 5727 0.0121 0 195.7 000730 15:50 000805 15:20 4353 0.0090 0 21.9 000803 12:00 000810 16:50 4740 0.0080 0 23.8 000808 01:50 000815 16:30 3991 0.0075 0 149.5 000813 13:20 000820 10:30 3461 0.0064 0 26.1 000818 18:50 000825 13:50 3874 0.0056 0 41.8 000823 16:20 000830 04:50 5890 0.0058 0 21.0 000828 14:40 000904 16:30 5679 0.0050 0 32.9 RMS POS= Initial epoch position uncertainty (dependent on data weighting, data quality and distribution, and orbital geometry over the arc.) ============================ The processing for the arcs above is the same; IE all adjust state, cd/day, Cr/arc, accels during memntum dumps, and frequency bias/pass for the one-way Doppler. All reductions are with mgm1004d (GMM2B) (80x80) (the spherical harmonic coefficients are on CD MORS_1008 in SHA/GGM2BC80.SHA). Ancillary Data ============== Data Collection --------------- We have obtained the following files from the FIS: Original File Name Name on SOPC -------------------------------- ------------ AMD: amdgen.00-306.out 0304305A.AMD SAK: mgs_solar_array_2000-305.ck 0305305A.SAK TCK: mgs_spice_c_kernel_2000-305.ck 0305305A.TCK SOE: mm087a.01.tsoe 0317321A.SOE We have obtained the following files from NAIF: Original File Name Name on SOPC -------------------------------- ------------ AGK: mgs_hga_ck_2000-10-23.bc 0297297A.AGK MOLA POD Team Orbit Solutions ============================= Data Collection --------------- We have obtained the following files from GEODESY: Original File Name Name on SOPC -------------------------------- ------------ GDN: gdn_ql17_009221750_009291700.tsp 0266273A.GDN GDN: gdn_ql17_009271650_010041610.tsp 0271278A.GDN GDN: gdn_ql18_007251910_008011510.tsp 0207214B.GDN GDN: gdn_ql18_007301850_008051520.tsp 0212218B.GDN GDN: gdn_ql18_008031300_008101650.tsp 0216223B.GDN GDN: gdn_ql18_008080250_008151630.tsp 0221228B.GDN GDN: gdn_ql18_008131420_008201030.tsp 0226233B.GDN GDN: gdn_ql18_008181950_008251350.tsp 0231238B.GDN GDN: gdn_ql18_008231720_008300450.tsp 0236243B.GDN GDN: gdn_ql18_008281540_009041630.tsp 0241248B.GDN Figure Captions =============== Figure HEA0307A.PS1: Difference between components of spacecraft position vectors in IPN and GDN files from reconstructed orbits during February, March, and April 2000. Figure was created by Boris Semenov of NAIF to illustrate approximate level of 'noise' between the two solutions. It does not presume to argue which is the better solution -- only to show where they differ and by about how much. 'Downtrack' differences tend to be largest, but never more than +/-1 km. Figures ======= Plot(s) illustrate some of the points made in the text. Each plot is a PostScript file, created and printable on our SOPC. For report text file HEAydddC.TXT Figure N will be labeled HEAydddC.PSN and sent as the N+1st e-mail message of a set. We assume that you can strip off the e-mail headers and either display or print copies of the figures on a machine in your area. People who have indicated that they cannot handle these files by e-mail are not included in the e-mail distribution of figures.