STScI Logo

Constructing Visits

GHRS Instrument Handbook

In putting together your Phase I proposal here are some things to remember.


You will need to allow time for an acquisition in virtually all cases. An ONBOARD ACQ with the GHRS should be used unless it will clearly not work. An FOS-assisted acquisition is a satisfactory alternative, but early- and interactive acquisitions are to be avoided.

If the STEP-TIME you need to get 100 or more counts in the peak pixel exceeds about 2 seconds then you have a faint-target acquisition, and you should use the longer time on the Phase I work sheets. This is because for faint targets the LOCATE phase of the acquisition takes a lot longer, just as the spiral search part at the beginning does. The LOCATE phase accurately centers your target in the LSA.

For very bright stars you may have to acquire with MIRROR-A2 or MIRROR-A1. In either case you must allow for more time for the acquisition because internal calibrations of the mirror locations take longer (because they reflect less light).


Ask for a wavecal if you believe the default wavelengths are not adequate for your needs and allow time accordingly. A typical wavecal requires about one minute of exposure time, or about 5 minutes of spacecraft time.

Side Switching

When you use G140L, G140M, Echelle-A, MIRROR-A1 or MIRROR-N1, you are using Side 1 of the GHRS. When you use G160M, G200M, G270M, Echelle-B, MIRROR-A2 or MIRROR-N2, you are using Side 2. You may switch back and forth between these two detectors, but doing so requires a wait of about 40 minutes each time to allow for the detector to reach stability. Thus it is generally to your advantage to construct visits so that they only use one Side or the other, or at least so switching is minimized. It may often be possible to have the side switch occur during target occultation if the switch coincides with the boundary between whole orbits of observation.

Memory Usage

A long series of observations with the GHRS on a single target with many different grating settings can result in a command load for the telescope that exceeds its memory. This is most frequently encountered with echelle observations at many wavelengths. See Section on page 52.