Change page style: 

Observing Overheads

Current estimates are that the overheads associated with each new science target (for target acquisition, telescope, WFS and instrument re-configuration etc) depend upon the guide configuration chosen for the observation.

Setup times for various observation types are:

Mode Setup times
NIFS+PWFS2 (non-AO) 11 mins
NIFS+Altair-NGS 11 mins
NIFS+Altair-LGS 25 mins
NIFS OIWFS
add 5 mins
NIFS Coronagraphy
add 4 mins

Long observations may be split over several nights to better accommodate them in the queue. Please allow for one acquisition for every 1.5 - 2 hours of observing when calculating the time required for your project.

Acquiring with the coronograph requires some additional iterations during acquisition to ensure an accurate centering behind the occulting disk. Likewise, acquiring the OIWFS star incurs an additional overhead of approximately 5 minutes. These overheads are in addition to the baseline set-up time for a given guide configuration. For example, the assumed overhead for an NGS-AO coronographic observation including an OIWFS star (recommended) is 11+4+5 = 20 minutes.

Dithering overheads: 30s per dither step

For the small 3" x 3" FOV of NIFS, it will be necessary to dither the telescope off the science source to acquire a sky spectrum for nearly all observations (see examples in the OT Library).

For automated sequences of exposures (dither patterns) the estimated on-source efficiency is ~40%; (i.e. 60%; of the elapsed time is used for detector readout, telescope offsetting, WFS re-acquisition etc). This estimate is quite realistic for a typical target->sky->target->sky... observations.

Very short exposures will have lower efficiency because of the fixed overhead per image (up to 30 seconds for dithers). Longer exposures will have higher efficiency.

Readout overheads

Observations in the low noise readout modes incur higher overheads. Standard "bright object" readout requires 5 seconds per image, medium read mode is 21 seconds per image, and the "faint object" readmode, which provides the lowest 6e- readnoise, requires 85 seconds to read out each frame. Also, readout overheads accumulate for every coadd, so a "faint object" observation that has 8 coadds will have 8 x 85 = 680 seconds of readout overhead! This is not recommended. For optimal efficiency, it is best to chose a readmode based on the brightness of your target and the length of the individual exposures.

Readmode read noise readout time (per coadd)
Bright source 18 e- 5.3 sec
Medium source 9 e- 21.2 sec
Faint source
6 e- 84.9 sec

Grating change overhead: 65s

A new OT for 2006B CfP should have all the above NIFS overheads properly handled, then "Timeline" in the "Basic Sequence Component" should be close to the actual exection time (i.e., wall clock time).

  • The current OT assumes a fixed overhead (0.17 min = 10.2 sec) for all telescope & instrument setups. Actual times for some of these setups are quite off from the fixed overhead of 10.2 sec. Two main discrepancies are offsets (or dithering) and grating changes. For each dither position, overhead is ~30 sec (instead of 10.2 sec) and a grating change takes typically ~65 sec.
  • The "Planned Observing Time" value from the current OT includes only actual exposures plus the initial target setup time. It does not include any telescope and/or instrument overheads after the initial target setup. A true observing time can be obtained by adding up all items which appear in the "Timeline". To see this "Timeline", click sequence -> TimeLine.