Line ops last night were cancelled due to a “no-go” by thetelescope assembly subsystem. A problem had been found that could not enableobservations tonight. It was a call the science team did not want to hear, butit was the right call. This exercised the reason why there is a “readinessreview” before going out to execute a complex activity. A plan was put in placefor the 1st shift when they get in at 7am (0700h) today (Wed) to address theproblem and report back during the day. If all goes well, a crew-briefing willbe scheduled again at 2130h tonight and we can resume lineops at 2300h.
If we were observing using a ground-based telescope, wewatch the weather. A seasoned ground-based observer watches the humidity. Youcan often get obsessed looking at trends in pressure, temperature, etc. It’simportant as you may need to replan your allotted observation time if you losea night (or nights) to the weather-gods.When I assisted with a balloon launch last summer at Ft. Sumner, NM, we’dgather daily to address the winds. Winds were most stable at dawn so we’d haveour “crew briefing” at 3 or 4am with readiness to roll out at 5am with the hopeto launch in the next hour or so (it would take nearly an hour to do the roll-outof the balloon and the He fill). Yes, sometimes the call would be made at 3amfor a “no-go” or even as late as right before the fill. And then you roll backthe balloon to the hangar. Last Sept, we launched on the 3rd attempt. All rocket launches also watch the weather and have various sub-system “go/no-go” checks.
SOFIA ops are not so different from those other examples.
So, we replan again. We have three remaining nights left inthe schedule, two this week and one contingency night next week, which nowseems to be required. Also, we’ve started looking at the flights scheduled fornext week, to see what tests planned in flight would supersede the line opstests to allow to compress our “line ops” schedule. Now, this is a calculatedrisk since the purpose of line ops is to test the system end-to-end beforeflight. So essentially you want to run the key components you plan to test inflight on the ground first.
What are line ops anyway? It’s not as “dramatic” as theactual flight, but it serves very important purposes to follow our observationplan end-to-end, address timing issues, and most importantly, communication betweenpeople and communication between people & machines. The plane is towed outon the runway to a viewing position safe from any active runway traffic, andpreferably in a location far from buildings or lights to obstruct viewing angles.We operate on plane-provided power. We command the telescope door to open,configure the telescope, check it out, power the science instrument, and startrunning through a series of discrete tests, some of which are to be run exactlyon the flights, and other diagnostic tests that are needed that would otherwisetake up the valuable flight time.
One of the tests we want to do is test the “nod” function ofthe telescope and how the data sets we collect affect our observing strategyoptimization (ahem, improve signal to noise). In mid-IR astronomy, the skybackground is “brighter” than our targets. In fact, we often cannot see ourtargets in the original raw data until we do a “background subtraction.” So weuse the telescope’s secondary mirror to “chop” a source back & forth (as itwould appear on our detector) at a fast rate. And then we would command thetelescope to “nod” to a different part of the sky. And repeat the process of“chopping” and “nodding” over a pre-planned orientation, both “throw distance”and “angle.”
You can read more about Chopping at Nodding at Why Chopping & Nodding is needed for SOFIA FORCAST Observations
An example taken from PDF on Signal to Noise Improvement by Chop/Nods sums it up nicely.
So we’ll be exercising things like this during the line ops,exploring the same technique for different roll angles because when it comes toyour science target which can be anywhere in the sky, we’d like to understandthe system performance and, if any, limitations.
We have other tests planned like assessing the detector biasperformance, looking at flexure of our alignment, particular for our grism modewhere we have narrow slits, optimizing a new flat field technique, and runningthrough the science scripts to checking for timing and fix any commandingerrors.
So fingers crossed, we will get on sky tonight, on the tarmacat Palmdale, CA. The skies have been clear the last two nights, so we theweather gods have been kind. We now need the electrical-power-subsystem gods tobe kind.