This table is a calendar of the operational manoeuvers and SW changes
made in the receivers. The comments are based on the descriptions
given by Tom Meehan. The date is followed by an ON or an OFF depending on
what is the status of atmospheric
occultations for that day.
SAC-C:
| |
Software-related event | Hardware-related event |
| Thu, 10 May 2001 - ON | The SW update to the A side is complete and appears to be working properly.
The receiver ran for 12 hours last night without a reset while producing real-time
Nav and POD data.
Tonight, we'll enable the occultation mode. With any luck, we'll have several megabytes of occultations to analyze by Monday morning. Initial parameters for occultations will be the same as CHAMP except 50 Hz data will be enabled on L1-CA only. |
|
| Wed, 16 May 2001 - ON |
The SAC-C BlackJack seems to be producing occultations consistently. However, the new SW is only on the A-side electronics. The receiver reset rate is about the same as before (2 hours) but the phase break problem (due to the oscillator fluctuation) is extremely frequent. This decreases the volume of useable occultations for now. | |
| Wed, 16 May 2001 - OFF | Some SW was uploaded and the sample rate of the POD data was changed on SAC-C. The SW update was part of a bug fix testing of the tracking SW (L2 50 Hz and better L2 acquisition). The sample rate change was from 10 secs to 1 sec on the POD data so one could better evaluate the oscillator problem on the A-side electronics. | |
| Fri, 6 Jul 2001 - ON | The SAC-C receiver has been doing occultations since Friday morning. | |
| Mon, 9 Jul 2001 - ON | This morning, we sent a modified tracking library to improve L2 tracking
and extend the flywheeling time from 7 seconds to 12 seconds (as set
on CHAMP). The max POD sats was reduced from 9 to 6 at this time to
see if the reset performance improves. Currently there are about 7 resets/day
with occultations enabled and 9 sats for POD.
L2 phase at 50 Hz was enabled so we could check an upcoming fix to the CHAMP tracking library. |
|
| Tue, 10 Jul 2001 - OFF & ON | SW upload and commands planned for yesterday's contacts were
not sent until this morning. So we shouldn't see any changes to the tracking
data until tonight's data download. Also, the SHM should be back on following tonight's contact @ 02:32..PRN 19 is currently in "L2-C/A" mode. This is evidently causing some disturbance to the receiver since we seem to have a "reset festival" when PRN 19 is in the mix. This test is supposed to be completed in a couple of days. |
|
| Mon, 15 Jul 2002 - ON | A SW upload to add the RTG capability started over the weekend. As of this morning enough has been uploaded to disturb the regular occultation scheduling. We expect occultations to be "messed up" for the next 2 days. | |
| Fri, 19 Jul 2001 - OFF | We tried to update the SAC-C receiver with the same changes as CHAMP, but ran into trouble after the Occ-Manager library was uploaded. The trouble was with retrieving files from the SAC-C server onThursday. This prevented from verifying proper uploading of the second library. By the time the problem was fixed, the weekend (UT) had arrived so commanding was finished until Monday. Unfortunately, with only the new Occ-Man library, no clock sats are being tracked. Soooo, occcultation data is trash for these last 3 days. | |
| Tue, 31 Jul 2001 - ON | SAC-C began producing occultations (with a clk-diff sat) this morning around 03:00 UT. We haven't examined any of the data yet except to verify the new SW is running and the clk-diff data is there. | |
| Thu, 02 Aug 2001 - ON | We've looked at about 48 hours of SAC-C data following the SW update.
There is some status and a question. My analysis should be considered anecdotal:
1) The scheduler seems to work better than before. 15 occultations failed out of 279 attempts in 36 hours. About half were due to missing clock sat data. Either it failed to acquire the clock sat or lost it shortly after acquiring it. About 25% were bad because the clock sat chosen was too low and went into occultation about 30 seconds after starting. The other 25% are miscellaneous failures. On an estimate the total attempts would have been higher by about 20 if there were no resets. In general we waste less mass storage on-board because the occ-clock sat pairs are better synchronized (we used to collect about 5 seconds of occ data before clock data started). 2) L2 phase/amplitude jumps at the start of tracking are fixed. 3) There is an error in the data formatting for clock-sats:
We may be losing about 10% of the total due to resets and scheduling problems (not counting how many more we might get if we could do multiple occs). It may take about 2-weeks work to get about 5% back by tweaking the scheduler/tracking code more. |
|
| Wed, 08 Aug 2001 - OFF | At 02:56 UT A command was sent to expand the occultation parameters so that a wider aft viewing angle and allow a greater separation angle between the clk-diff and occ sat. The purpose was to get the number of occs back up over 200+ as before.This set the SAC-C occultation antenna OFF! Until the next command contact at about 02:30 UT tomorrow, there will be lots of attempts for occs in the schedule but no actual occ data. | |
| Thu, 9 Aug 2001 - OFF | Occ data is flowing but it seems forced into a 10 second data rate! The problem with the SHM commanding was found but now I need to upload another MCP to properly clear this bug. I've duplicated the problem in the lab but have no simple solutions except to update the MCP which will take a couple of days. | |
| Mon, 13 Aug 2001 - ON | New SW was uploaded to SAC-C over the weekend which fixes the ability
to control Magnetometer (SHM) data. SHM data is currently set "OFF". Occultations
are enabled as of yesterday at 14:00 UT. Occultation parameters have been "widened" to 55 degrees azimuth with the acceptable clk-diff sat angle increased to 70 degrees.L1&L2 50 Hz data is enabled still on SAC-C but SAC-C is still running the SW which does not produce L2 50HZ data on the clk-diff link. This mode will be set OFF later this week. SAC-C has 7 POD sats enabled. |
|
| Sat, 18 Aug 2001 - OFF & ON | New tracking SW was uploaded to SAC-C on Friday. This version
has the CHAMP SW improvements from last week but does not have the L2 tracking
mods that caused so much trouble for occultation retrievals. L2 tracking is
back to being highly coupled with L1 as before. I've looked through about
20 occultations, and see no signs of the early loss of lock seen with the
last CHAMP update.
In the two 50 Hz examples I examined, L1-L2 50 Hz data appears to have greatly diminished 1Hz phase "spikes" in the so-called forward difference d(L1-L2)/dt. Hopefully this update will be a good compromise for both problems. If all goes as planned, by Monday: |
|
| Thur, Oct 04 2001 - OFF & ON | A new library uploaded to SAC-C. No operational parameters were changed but additional resets will be found in the data corresponding to the upload times. |
|
| Frid, 05 Oct 2001 - OFF | On Friday morning, the last few packets of a new library were uploaded to SAC-C. Unfortunately, the command to re-start normal tracking scheduling did not seem to make it to the receiver. | |
| Tues, 09 Oct 2001 - ON | The SAC-C receiver has not been tracking all weekend. Things were successfully restarted @ 14:30 and occultations enabled.Data look normal. |
|
| Mon, 29 Oct 2001 - ON | The tracking SW was modified today at 13:30 UT. Among the expected changes is an occultation rate increase to 100 Hz from 50 Hz. |
|
| Tues, 30 Oct 2001 - ON | SAC-C 100 Hz data has started (10/30 at 14:45). | |
| Tues, 06 Nov 2001 - OFF | SAC-C has temporarily stopped tracking due to a commanding problem at ~22:00. I hope to get it back up this afternoon at 22:00. | |
| Fri, 16 Nov 2001 - ON | Two libraries were uploaded to SAC-C since yesterday. These are
supposed to improve the success ratio of occ+calibration. Both libraries are
needed in order to have the proper effect. The second library is not running
as of today at 16:00 UT so it may be few hours until things are running properly.
Since this configuration has the ability to schedule reflection data. It has been trying to take these "snapshots" from the nadir antenna since the scheduler was loaded. But, since the tracking library is not the correct version it continually fails. This puts a lot of "IonoStart" messages in the data file. Further testing of the s/w will continue. |
|
| Tues, 20, Nov 2001 - ON | The receiver was configured to take some reflections snapshots with the nadir antenna. Those data were tagged with antenna 10 or '0xa'. This mode may have affected the occultation data output due to a much larger column of 100 Hz reflections data than expected. | |
| Wed 21 Nov 2001 - ON | The command was sent to turn reflections mode off and to continue
with normal 100 Hz occultations. A fix was added to better avoid choosing clock sats that could go into occultation themselves. This should be a < 5% effect. Let me know if you get time to look. |
|
| Wed, 19 Dec 2001-ON | Drag-Makeup Maneuver next Wednesday 2001/12/19 at 12:48:00 | |
| Wed, 16 Jan 2002-ON | Propulsion maneuver on January 16 th at about 14:47 UT (Back up Jan 18 th 14:34 UT). |
|
| Wed, 30 Jan 2002-OFF | We're doing an upload this week of a bunch of OS changes. The upload started as a something with little impact on the tracking but has spiraled into a major outage. Assume no data from SAC-C until next week at this point. | |
| Wed, 13 Feb 2002-OFF | Drag-Makeup Maneuver Wednesday 2002/02/13 16:30. | |
| Sun, 03 Mar 2002 -ON | Occultations were enabled at around 3:30 UT on Sunday. So far things look OK. Both ionospheric and atmospheric modes are enabled. A small percentage of "events" are seen with the scheduler still, but at first glance the production looks much healthier. |
|
| Wed, 27 Mar 2002-ON | Lowered cutoff elevation from -20 to -30 degrees. Coincidentally, reset rate increases. Iono parameters were set to the earlier values, but the receiver is still resetting frequently. |
|
| Fri, 29 Mar 2002-OFF | Occultations off. | |
| Sun, 31 Mar 2002-ON | Mass memory download filesize has dropped again starting MM20020331_024130. Normal size is now 20955330. (I write this on 2002-04-05.) | |
| Mon, 1 Apr 2002 - ON | Reset rate is back to normal. | |
| Tue, 2 Apr 2002 - ON | Large sections of the data have unphysical large negative P2 pseudorange. | |
| Wed, 3 Apr 2002 - ON | Occultations restarted, but TKM reports format is incorrect. | |
| Thu, 4 Apr 2002 - ON | After re-starting the occultations but with 8 sats max for POD the receiver seems to have settled back to normal behavior. The occultation data itself is not formatted correctly so a command was prepared this morning to fix this.When occultations were re-started the configuration was changed to raise the starting altitude to about 120 km and to turn amplitudes off on L2. This command seems to be corrupted somehow . | |
| Tue, 23 Apr 2002 - ON | The command was sent yesterday to switch from the "B"-side electronics to the "A"-side. The data from that side looks fine. The task now is to update the SW from the May 2001 version now on the A-side. After the SW is up to data I'll have a better idea how much better (hopefully) the tracking will be compared to the recently deteriorated B-side data.This SW update will take all week. | |
| Wed, 24 Apr 2002 - OFF | Drag-Makeup Maneuver Wednesday 2002/04/24 13:16:00 UT. | |
| Mon, 29 Apr 2002 - ON | SAC-C SW update is slowly progressing. The plan is to switch over to the A-side HW only during the commanding and then back to the B-side for the next 12 or so hours. This way, as long as the B-side is functioning OK we can continue to get most of our science data while this SW update goes on. | |
| Thu, 23 May 2002 - ON | The data after 14:06 today looks better. On first inspection the data dropouts are gone. It could be the steps taken by the SAC-C folks fixed things. Iono tracking was enabled after 14:06 (an earlier attempt at commanding this failed) | |
| Tue, 28 May 2002 - ON | The receiver command response was very erratic so the configuration file was erased. This file contains the latest command configurations saved in non-volatile memory. When erased, the default configuration was loaded which starts the occs around 80 km and only takes iono data down to -20 degrees elevation. |
|
| Thu, 30 May 2002 - ON | Done a quick check of the occ data. It appears OK with higher starting atmo-occs and lower ending iono-occs. The 50Hz is enabled on L1&L2 phase and L1 amplitudes for both Occ and ref sats. This is close enough to the limit of our memory capacity that I did not want to risk turning on 50 Hz amplitudes for L2. | |
| Mon, 29 Jul 2002 - ON | Check of status: Iono occs are currently set to cut off at 200 km. Atmospheric occs are set to start at 135 km. | |
| Mon, 14 Oct 2002 - OFF | Beginning this week I plan to modify the tracking SW on SAC-C to begin open-loop type data output. This mod will be a multi-step process since some of these mods need to be testedin-orbit. | |
| Sun, 03 Nov 2002 - ON | SAC-C receiver returned to a "normal" tracking mode (~
01:00UT , 11/03/2002). I am experimenting with a mod to the |
|
| Wed, 15 Nov 2002 - ON | Open-loop format code.For now, SAC-C should appear to be back to normal with RTG data OFF as well. | |
| Sat, 17 Nov 2002 - ON | SAC-C goes into Emergency Mode | |
| Tue, 17 Dec 2002 - ON | SW upload to SAC-C Tonight at 01:30 UT | |
| Thur, 10 Apr 2003 - ON | A new Master Control Program (MCP) was uploaded on 4-10. However,
I could not check its performance until this morning when downloaded mass-memory
files were available from the SAC-C server. This new MCP is designed to handle rising occultations (more SW uploads required though) and turns clock steering OFF during an occultation. This is designed to "relax" highly non-linear oscillator frequency changes which make using an atmospheric model less accurate during open-loop mode. |
|
| Sat, 12 Apr 2003 - OFF | Occultations have been OFF since about 16:00 UT on Sat, 4-12, because the new command for enabling occultations was not properly transmitted. | |
| Mon, 14 Apr 2003 - ON | Occultations seem to be back ON as of about 03:15 UT today. We haven't looked at data quality stuff yet just verified they're being scheduled. | |
| Sat, 10 May 2003 - ON | On Saturday, an update to the TrackNPack library was uploaded to
SAC-C. The new library was not loaded by the receiver until Sunday. Two
things were supposed to be fixed. One, was a bug which caused occasional calibration
data to be missing. The other was supposed to reduce the noisy highrate L2 data in either the calibration or occultation link. I haven't checked for the first bug fix but I can see the second is not better; its worse! I'll need to figure out why my ground testing scenario (where things work fine) does not represent the SAC-C environment and try for another code update tonight. |
|
| Sat, 17 May 2003 - ON |
An updated version of TrackNPack started running.
It was uploaded on Wednesday but its starting to take too long to
get resets. I'll have to start forcing them! The non-Open Loop data should be fine. Open-loop data seems to be intermittently output. I'd like to verify that the 50 Hz L2 phase noise is now down to reasonable levels. The P1 SNR dropped (relative to CA). SNR drop is a scale factor from open-loop that crept into 1 sec SNR scaling. To see same SNR as before, multiply P1&P2 SNR by 240/150. Sorry for the trouble. |
|
| Sat, 21 June 2003 - ON |
First test of rising occs on forward antenna. That works. We have collected 1 sec tracking data from the forward antenna on SAC-C. SNRs look good on all three signals (CA,P1,P2). However, rising occ data collection confuses the scheduling of the clock reference satellite used for setting occultations. In the data starting Saturday at 02:20 UT, several of the setting occultations are without a clk reference satellite. I'll correct this. |
|
| Mon, 30 JUN 2003 - ON |
On Saturday, two libraries were uploaded to SAC-C.
The changes involved a time-based OL mode whereby each 24 hours between
4-8 UT, 100 Hz OL occultation data would be taken instead of the
standard 50 Hz FW data. Also, between 5-9 UT, reflections data would
be taken, excluding atmospheric occultation data. The remaining hours were
to be devoted to FW occultation data. Better commands for controlling rising
occultation data were also added. The first library loaded successfully, but the second was missing 2 packets so we ended up with a partial load which led to some erratic performance over the weekend. Rising occs were continuously scheduled, no OL data was recorded.The additional SHM data causes an overrun of the mass memory by about 5-6 hours per 24. Needless to say, the SHM data will be commanded OFF this morning and a reload of the missing library is planned for this morning. Hopefully by tomorrow the data will be back to normal. I'm planning more uploads throughout the week to sort out OL scheduling and to further modify the reflections data collection. Normal FW occultation data should continue to flow. |
|
| Thu, 07 JUL 2003 - ON |
Further tests of OL and rising occultation
SW are foreseen for the next weeks in preparation for COSMIC. This will increase
the probablity of coverage gaps. |