Deorbit guidance, navigation and flight control software operates through the transition DAP to provide maneuvering of the spacecraft to the OMS deorbit ignition attitude, OMS thrusting commands, OMS engine gimbaling for thrust vector control and RCS thrusting commands, in conjunction with use of the DAP similar to that for orbit insertion.

In returning home, the orbiter must be sufficiently decelerated by an OMS retrograde burn that when it enters the atmosphere, it maintains control and glides to the landing site. For the nominal end of mission, a retrofiring of approximately 2.5 minutes is performed at the appropriate point in the vehicle's trajectory. For this maneuver, the orbiter is positioned in a tail-first thrusting attitude. Deorbit thrusting is nominally accomplished with the two OMS engines and must establish the proper entry velocity and range conditions. It is possible to downmode to one OMS engine (with RCS roll control) or, in the event that both OMS engines malfunction, to plus X aft RCS jets.

Approximately four hours before deorbit, the environmental control and life support system radiator bypass/flash evaporator system is checked out, since the flash evaporator is used to cool the Freon-21 coolant loops when the ECLSS radiators are deactivated and the payload bay doors are closed. The high-load evaporator cools the coolant loops until the ECLSS ammonia boilers are activated by the GPCs at an altitude of some 140,000 feet. The orbiter IMUs are aligned, the star trackers are deactivated and the star tracker doors are closed.

About one hour before deorbit, the crew members take their seats. The spacecraft is then manually maneuvered using the RCS jets to the deorbit attitude (retrograde). About 30 minutes before deorbit, the OMS is prepared for deorbit thrusting. This consists of OMS thrust vector control gimbal checks, OMS data checks, orbiter vent door closure and single auxiliary power unit start. At the completion of the single OMS deorbit burn, the crew manually maneuvers the spacecraft to the required entry attitude (nose first) using the RCS jets. The propellants remaining in the forward RCS are dumped through the forward RCS engines, if required, and the two remaining APUs are started and remain operating through entry and landing rollout. Thermal conditioning of the spacecraft's hydraulic fluid system is also begun, if required.

The deorbit phase of the mission includes the deorbit burn preparations, including the loading of burn targets and maneuvering to burn attitude; the execution and monitoring of the burn; reconfiguration after the burn; and a coast mode until the atmosphere (and dynamic pressure buildup) is reached at approximately 400,000 feet. This is called the entry interface.

The deorbit and entry flight software is called OPS 3. Major mode 301 is a deorbit coast mode in which deorbit targets can be loaded, although the burn cannot be executed in this mode. This mode is necessary to execution of the burn. After the burn, a software transition is made to another coast mode, major mode 303, which is used to prepare for penetration into the atmosphere.

During the deorbit phase, navigation again propagates the orbiter state vector based upon a drag model or upon inertial measurement unit data if sensed vehicle accelerations are above a specified threshold. During OPS 3, navigation maintains and propagates three orbiter state vectors, each based on a different IMU. From these three state vectors, a single orbiter state vector is calculated using a mid value selection process and is passed on for use by guidance, flight control, dedicated display and CRT display software. Three separate state vectors are propagated to protect the onboard software from problems resulting from two IMU data failures. In such a case, once the bad IMU is detected and deselected, the state vector associated with the remaining good IMU will not have been polluted. This three-state vector system is used only during OPS 3 since this phase is most critical with respect to navigation errors and their effects on vehicle control and an accurate landing.

Another feature available during this phase is the software's computation of a statistical estimate of the error in the state vector propagation, which is used later in flight when external sensor data are available. Also, in this phase, it is possible for the crew or the Mission Control Center to input a delta state vector to correct navigation.

Guidance during deorbit is similar to that used in the orbit insertion phase. The PEG 4 scheme is used to target the deorbit burn and guide the vehicle during the burn, although the required conditions are different. The deorbit burn targets are for the proper conditions for entry interface, including altitude, position with respect to the Earth and thus the landing site, and satisfaction of certain velocity/flight path angle constraints. Together these ensure that the vehicle can glide to the landing site within thermal limits. Deorbit burn targets are specified before flight for a nominal mission, but it is possible for the ground to uplink changes or for the flight crew to recompute them using an onboard hand-held calculator program. It is also possible to specify that OMS fuel be wasted during the burn (burned out of plane) to establish an acceptable orbiter center of gravity for entry.

The crew is responsible for loading these targets on the deorbit maneuver execute display. Guidance then computes the necessary vehicle attitude to be established before the burn and displays it to the crew. As in OPS 1, it is possible to load an external delta-velocity (PEG 7) target, but this option is not normally used.

Flight control during the deorbit phase is similar to that used during orbit insertion-i.e., the transition DAP is once again in effect.

The flight crew interfaces with the guidance, navigation and flight control software during the deorbit phase via CRT display inputs, RHC/THC maneuvers and ADI monitoring. The major CRT display used is the deorbit maneuver execute. In major modes 301 and 303, the display is deorbit maneuver coast; whereas in major mode 302, it is deorbit maneuver execute. This display is identical in format to the OMS-1 maneuver execute and orbit maneuver execute displays, although there are some differences in its capabilities between OPS 1, 2 and 3. It is used to set up and target the OMS burn, to specify fuel to be wasted during the burn, to display the required burn attitude, to initiate an automatic maneuver to that attitude and to monitor the progress of the burn.

Another CRT display available during the deorbit phase is the horizontal situation. During deorbit preparation, the crew may verify that the display is ready for use during entry (correct runway selection, altimeter setting, etc.), but its other capabilities are not utilized until after entry interface.

The flight crew's task during this phase includes entering the correct burn targets in the deorbit maneuver execute display and maneuvering to burn attitude, either automatically or manually using the RHC. The burn itself is typically executed in auto, and the flight crew's task is to monitor the burn's progress in terms of velocities gained and OMS performance.

In cases of OMS failures (engine, propellant tank, data path), the flight crew must be prepared to reconfigure the system to ensure that the burn can safely continue to completion, that sufficient RCS propellant remains for entry and that the orbiter center of gravity stays within limits.