From jgt@slac.stanford.edu Sat Aug 12 14:47:06 2006 Date: Fri, 16 Jun 2006 09:59:00 -0700 From: John Gregg Thayer To: D.A. Smith Subject: Re: req't 5.2.11 -- 10 uS event time accuracy (fwd) Howdy, So here, in a nutshell, is what I do for my "offline" analysis. (run 77005274 is perfectly representative.) First, I extract the packets from the EPUs and M7(APIDs 0x3bc,0x3bd,0x3be, 0x3fc) and using FSW debug tools (qsep_print) dump the headers of the CCSDS packets. From this, we check that the TIMETONE was being processed by the LAT. (You can see this from the timestamps in the log, just as well, but Rich Baun wanted to see the timestamps for the data packets that were used in the next step.) I then run a root script on the digi file for the run, and create a plot of the time between the PPS and the Event as recorded in the GEM contribution. (I've attached the plot from run 77005274 to this email.) The dump and the plot are both included in the Shift log for each run, and can be accessed by clicking on the "Dir" link. The names of the files are deltaPpsTime_XXXX.pdf and qsep_print_XXXX.txt. In the development of this amazingly interesting script, we checked that the absolute value could be changed in a predictable way by introducing a fixed delay in the external trigger, and indeed this plot would change accordingly. Similarly, the value of 7-9 is reasonable for when there is no additional delay. (If there were no delay of the PPS in the VSC, then one would expect this value to be equal to the window width, which is 12 in the standard LAT configurations.) Please let me know if there's anything else I can do for you. (Or if you want to see my pathetic root script.) Gregg D.A. Smith wrote: > > Hello Gregg, > > In Pat's message from 2 months ago (see way below), he wrote "test > results will be analyized and documented in LAT-TD-xxxxx." Yesterday I > asked him if xxxxx exists and who's writing it. His answer? "You are." > !!! Funny guy. > > All kidding aside, I'm quite willing to do it. But all I know about the > LAT 13x tests is some doc I read and your brief words a while back. If > I'm gonna "analyse and document" then I need data files and more > specific tips & pointers. > > So I just went to Anders et al runs data base and searched for LAT-13x, > and found lots of stuff (an example, just below). > > Which leads me to think that run 77005274 would be a good one. > > That run triggered on PPS... It's easy for me to spit out the dates of > the 601 events in there... it's clearly written "no GPS lock"... Once I > got the dates, what do I do with 'em? > > Let me put that differently: you conceived & executed that test (with a > little help from your friends) to demonstrate that there's <10 uS jitter > between PPS trigger time and LAT timestamp. I know how to get the latter > but not the former. > > Conclusion: I volunteer to do this chore but I need help getting > started. Help? > > all the best, David. > > > GLAST Shift Logbook > Day Shift for Wed, May 24, 2006, 08:00:00 > > Continue with LAT CPT, switching to Configuration 2. > +++ Begin run for the record +++ > Run LAT002 Power up in Configuration 2, Turbo, per LAT-PS-07836-16 para 5.1 > [skip to ...] > Run LAT232 GRB Handling per LAT-PS-07836-16 para 5.19 > Run LAT302 ACD CPT per para 5.23 > Run LAT402 CAL CPT per para 5.25 > Run LAT502 TKR CPT per para 5.27 > Run LAT132 LAT-Spacecraft Interface Test per para 5.11 > --- end run for the record --- > > [19:37] PH: start LAT132 - LAT/spacecraft Interface Test, > lat-aids-02006, step 40, Seq 251. > [19:46] PY: ERROR (I don't know if these are expected...) > following is example of errors in the script engine shell: > > ERROR:fsw:SIU: 19:42:55.719807 uLTX0 LTX_resultList MSG GAPBEGIN Message > blackout starts (trying to report 0x7c170e97) > ERROR:fsw:SIU: 19:42:56.225362 uMSG MSG_outputScan MSG GAPEND Message > blackout ends (20 messages dropped) > ERROR:fsw:SIU: 19:42:56.249173 uLTX0 LTX_resultList LCS LCBTXERR LCB > transmit reports error 0x00000005 > ERROR:fsw:SIU: 19:42:56.249203 uLTX0 SendSlaveDone LSM BADSEND Fail on > sending message to EPU nid=3 LcbStatus=0x7c170e97 > ERROR:fsw:SIU: 19:42:56.302178 uSTX LTX_resultList MSG GAPBEGIN Message > blackout starts (trying to report 0x7c170e97) > ERROR:fsw:SIU: 19:42:56.624668 uMSG MSG_outputScan MSG GAPEND Message > blackout ends (12 messages dropped) > ERROR:fsw:SIU: 19:42:56.670312 uLTX0 LTX_resultList LCS LCBTXERR LCB > transmit reports error 0x00000005 > ERROR:fsw:SIU: 19:42:56.670342 uLTX0 SendSlaveDone LSM BADSEND Fail on > sending message to EPU nid=4 LcbStatus=0x7c170e97 > ERROR:fsw:SIU: 19:42:56.722789 uLTX0 LTX_resultList MSG GAPBEGIN Message > blackout starts (trying to report 0x7c170e97) > ERROR:fsw:SIU: 19:42:57.225352 uMSG MSG_outputScan MSG GAPEND Message > blackout ends (13 messages dropped) > [19:49] PY: finish LAT132 - LAT/spacecraft Interface Test, > lat-aids-02006, step 40, Seq 251. LatSc1553Interface.py. > run ID 077005273, passed > > [19:52] DR: start LAT132 - LAT/spacecraft Interface Test, > lat-aids-02006, step 40, Seq 251, 1PPS external trigger > > > > ---------- Forwarded message ---------- > Date: Mon, 17 Apr 2006 08:10:39 -0700 > From: "Hascall, Patrick A" > To: D.A. Smith > Cc: "Grove, Eric" > Subject: RE: req't 5.2.11 -- 10 uS event time accuracy (fwd) > > Dave, > > Here's what we have in the verification plan > > Title: Accuracy of Event Time Measurements > Requirement: The time accuracy of event time measurements shall be < > 10 microseconds relative to spacecraft time. > Verification Summary: The accuracy of the 1PPS signal is an SC > requirement. Verification of LPS41 is to be performed in parallel with > the verification of LPS749, LPS827, LPS829, and LPS1298 documented by > LVP217, LVP243, LVP244, and LVP446 respectively. In test cases LAT131 - > LAT Config 1 LAT/Spacecraft Interface Test, LAT132 - LAT Config 2 > LAT/Spacecraft Interface Test, LAT133 - LAT Config 3 LAT/Spacecraft > Interface Test, and LAT134 - LAT Config 4 LAT/Spacecraft Interface Test, > (see the LAT Performance and Operations Test Plan, LAT-MD-02730, > paragraphs 6.7.2.10.1, 6.7.2.10.2, 6.7.2.10.3, and 6.7.2.10.4), > configure the LAT to trigger off of the 1PPS signal and to filter out > muons. Note the start time for the first GPS Time Message sent by the > VSC. Collect muons for 10 minutes minimum and correlate the timestamp > provided by FSW in the Science Data with the time contained in the GPS > Time Messages. Verify an event is received once per second with each > event timestamp being wit > hin 10 uS > of the corresponding GPS time for that one second time period. Test > results will be analyized and documented in LAT-TD-xxxxx. > > It turns out that Eric is the "requirement owner" in this matrix. > > The test that implements it is LAT13x. The high level definition is in > section 6.7.2.10 in the Performance and Operations Test Plan (attached). > It is being implemented by Gregg Thayer. We are having the second table > top tomorrow. That means that he's generated a script and we will walk > through it comparing what he has done to the planned requirements > verification. Specific time for the table top has not yet been set. > > Pat > > -----Original Message----- > From: D.A. Smith [mailto:smith@cenbg.in2p3.fr] > Sent: Sunday, April 16, 2006 1:01 AM > To: Hascall, Patrick A > Subject: Re: req't 5.2.11 -- 10 uS event time accuracy (fwd) > > > Hello Pat, > > This Tuesday the Pulsar science group will meet and one this we'll be > discussing is the publication describing the LAT that is being written > now. In particular, contributions of the pulsar group to that paper. > > In that context, I asked Eric the question below, and he re-directed me > to you. Could you please point me to the document that describes what > test or meas't will be done? Is there a physicist chore that you might > need done pertaining to that test, that we should volunteer for? > > all the best, David. > > > =+=+=+=+=+=+=+ > David A. Smith, GLAST / IN2P3 / CNRS > Centre d'Etudes Nucleaires de Bordeaux-Gradignan B.P. 120, 33175 France > vox: +33 5 57 12 08 91. fax: +33 5 57 12 08 01 > > ---------- Forwarded message ---------- > Date: Sat, 15 Apr 2006 19:22:46 -0400 (EDT) > From: J. Eric Grove > To: D.A. Smith > Cc: eric.grove@nrl.navy.mil, eduardo@slac.stanford.edu, > D. Dumora > Subject: Re: req't 5.2.11 -- 10 uS event time accuracy > > Dave, > > Yes, this is under discussion. I suggest you guys connect with Pat > (hascallp@slac.stanford.edu). There is indeed a test plan. > > Eric > > >> >> Eric and Eduardo, >> >> Has req't 5.2.11 of LAT-SS-00010 for 10 uS event time accuracy (goal >> is >> 2uS) already been demonstrated? It's under control? How who when? >> >> The LAT instrument paper draft is circulating, and I was wondering if >> in the pulsar section we could (or want to) have a plot (or just a >> number) showing that we have indeed demonstrated that glast will have >> pulsar timing capability. >> >> all the best, David. >> >> =+=+=+=+=+=+=+ >> David A. Smith, GLAST / IN2P3 / CNRS >> Centre d'Etudes Nucleaires de Bordeaux-Gradignan B.P. 120, 33175 >> France >> vox: +33 5 57 12 08 91. fax: +33 5 57 12 08 01 >> > > > -- > J. Eric Grove > Code 7650 > Naval Research Lab > Washington, DC 20375 > > Phone: 202 767 3112 > Fax: 202 767 6473 > Email: eric.grove@nrl.navy.mil [ Part 2, Application/PDF 5.3KB. ] [ Unable to print this part. ]