Info

Compute noise-equivalent temp, difference, NEAT

NEAT = Ng/AN

NEAT = 0.3 K

Temperature limit the instrument can resolve

9.5.6 Assess Life-cycle Cost and Operability of the Payload and Mission

In addition to trades between minimal and desired performance, spacecraft designs are heavily driven by cost Several approaches have been proposed and implemented to treat cost as an independent variable. For our purposes, it is sufficient to note that trading cost and performance means it is no longer sufficient to state the mission requirements clearly and realistically. Rather, the mission requirements become involved in the iterative process of design (see Fleeter [1996]). There are several excellent descriptions of how to include cost as a system parameter rather than a given; see for example Shishko and Jorgensen [1996],

Working through the trade-offs associated with cost, performance, and requirements in this early stage of payload definition keeps payload designers focused on the best sensor characteristics to maximize mission performance and minimize cost. Designers sometimes have a tendency to want to perform a purely analytical evaluation of the costs and benefits of various design options. Unfortunately, the relative benefits of different design features are difficult if not impossible to quantify in ail unambiguous and universally accepted manner. Analysis can be very useful for providing a common footing and level playing field for die different design attributes. Ultimately, however, judgments about satisfying mission objectives within cost and schedule constraints rely on human insight, adding to the difficulty and importance of this portion of the payload definition process.

Once we determine the final payload type and basic payload performance requirements, then payload final design can commence. The final payload design could be as simple as an evaluation of existing payloads that are available, or it could involve detailed design, fabrication, and testing of an entirely new instrument The final step in the payload definition and sizing process is the decision to procure or fabricate the spacecraft payload.

Integrating a payload into a spacecraft design introduces several practical considerations for the other payload subsystems. These derived requirement can have a significant impact on the rest of the spacecraft Table 9-16 contains an overview of some of the accommodation aspects of a payload as it impacts the other spacecraft subsystems. Resolving the impact of these requirements means we must assess the performance, cost, and technical risk of each subsystem to accommodate the payload.

9.6 Examples

We present two examples of remote sensing payload designs—one very preliminary and one very mature—to give an indication of the beginning and ending points of the design process. Sec. 9.6.1 provides an initial assessment of a payload to fulfill the FireSat mission. Sec. 9.6.2 describes features of the MODerate-Resolution Imaging Spectroradiometer (MODIS), one of the primary sensors on board the Earth Observing System EOS-AMI spacecraft, which has a fire detection capability.

9.6.1 The FireSat Payload

To illustrate the preliminary design process for payloads, we will estimate the basic parameters for the FireSat payload developed throughout Chaps. 1-8 and earlier in Chap. 9. We cannot expect to carry out a detailed design without substantial input from an IR payload designer. Still, we would at least like to know whether the FireSat payload is the size of a shoebox or the size of a truck.

TABLE 9-16. Impact of Remote Sensing Pay loads on the Spacecraft Design. The table summarizes requirements in other elements of the spacecraft design that must be present to support a remote sensing payload.

Impact Area

Typical Requirements to Support Payload

Additional Considerations

Structure

Mount the optical instruments Isostatically to the spacecraft bus

Do not apply excessive forces or torques to the payload instrument

Make the mounting structure or base plate for optical components stiff enough to prevent any misalignment when subjected to the forces and vibrations of launch

Carefully analyze aging of material (e.g., stress release in metal parts), humidity release, transition to micro-gravity, and acceleration forces

Typical stability requirements at critical locations within the optical instrument housing are in the (im and mdeg range

Thermal

Make large opto-mechanical assemblies temperature stabilized or isothermal

Operate refractive optical systems typically within a specific temperature range to achieve required performance (frequently they employ semi-active temperature control)

Make reflective systems entirely from the same material which leads to a compensation of thermal effects (typically done for cryogenic optical systems)

Large reflective systems (which use Zerodur, Aluminum, or Beryllium or newly developed materials such as SiC or CsiC as materials for the mirrors) and mounting structures (which use composite materials) are temperature sensitive and may require semi-active temperature control of structure and/or mirrors

Temperature gradients in optical components can severely degrade performance

External Alignment

Align the optical axis of the instrument and/or the line of sight of the pointing device with an external reference on the spacecraft External alignments may need to be on the order of 1 arc sec

Use reference cubes to achieve alignment

External alignment requires a calibrated optical bench.

Pointing

For monocular optical instruments, make the pointing requirements on the order of 0.1 to 0.01 of the swath width, typically

For stereoscopic instruments, automated digital terrain mapping requires pointing knowledge of 1/5 of a pixel

Mount attitude determination sensors (e.g., star sensors) to the Instrument (not the bus) to minimize the effects of thermoelasticity

Do pointing by maneuvering the spacecraft or by pointing devices (such as pointing mirrors for the front of the Instrument or gimbals for the entire instrument)

Assembly Integration and Verification

Optical Instruments require clean rooms and clean laminar air flow benches for ail integration and verification activities

Clean room requirements typically range from 100 to 100,000 ppm

Cleaning optical surfaces is generally not possible

During exposure to the environment, use cleanliness samples to verify the level of contamination

System

Accommodation

Sensor must have an unobstructed field-of-view

Sensor must have a guard cone to prevent performance degradation due to stray light

Avoid pointing toward the Sun

Orient radiators and passive coolers for infrared systems to prevent interference with optical devices

Calibration devices impose geometric constraints with respect to the optics of the system and the orbit

The FireSat altitude trade led to a preliminary altitude, h = 700 km. From this, we can determine the angular radius of the Earth, p.

A key parameter in the system design is the minimnm elevation angle, £, at which the system can work. We do not have an estimate of that yet, but we do know that IR payloads do not work well at small elevation angles. Therefore, we will tentatively assume a minimum elevation angle of 20 deg, recognizing that this may be a very critical trade at a later stage. With this assumption, we can compute the nadir angle range, % the maximum ground-track angle or swath width, A, and the maximum range to the target, D, from the formulas in Sec. 5.2:

sin 7/ = cos £ sin p 7/ = 57.9 deg From Eq. (5-25a) (9-25)

D = Re (sin X / sin rf) = 1,580 km From Eq. (5-27) (9-27)

These equations imply that the sensor on board the spacecraft will have to swing back and forth through an angle of ±57.9 deg to cover the swath. The swath width on the ground will be 2 x 12.1 = 24.2 deg wide in Earth-central angle, with a maximum distance to the far edge of the swath of 1,580 km. Had we been able to work all the way to the true horizon (e = 0), the maximum Earth central angle would be 90 - p = 273 deg, and the swath width would be 55 deg. Increasing the minimnm elevation angle to 20 deg has very dramatically reduced the size of the available swath.

We next find the orbit period, P, and longitude shift per orbit, AL, (Sec. 7.2):

P = 1.659 x 10-4 x (6,378 + h)3« = 98.8 min From Eq. (7-7) (9-28) AL = 1.65 x (360/24) = 24.8 deg From Eq. (5-17) (9-29)

Therefore, at the equator, successive node crossings are 24.8 deg apart Notice that this is slightly larger but very close to the 24.2 deg swath width which we computed above. This is an important characteristic for FireSat. It would be extremely valuable to have the swaths overlap so that every FireSat spacecraft can cover all locations on the Earth twice per day. Therefore, in designing the payload, we should work hard to maintain either the altitude or the minimum elevation angle to provide some swath overlap. Doing so could dramatically reduce the number of spacecraft required and therefore the cost of the system.

As Fig. 9-24 shows, the swath width does not need to be quite as large as the spacing between nodes along the equator. Even at the equator, it is enough to have a swath width equal to S, the perpendicular separation between the ground tracks. In Chap. 7, we selected an inclination for FireSat of 55 deg to cover up to 65 deg latitude. Consequently, we can use the spherical triangle ABC shown in the figure to compute S as follows:

S = sin-1 (sin 24.8 deg sin 55 deg) = 20.1 deg (9-30)

The perpendicular separation between the orbits at the equator is 20.1 deg. Because the swath width is 24.2 deg, we now have some overlap margin even at the equator and substantial margin at higher latitudes, which are the primary areas of interest We could, therefore, increase the minimum elevation angle to 25 deg. This would be a reasonable option. At present, we instead choose to hold e at 20 deg and to provide some margin on altitude and elevation angle for later payload trades.

Fig. 9-24. Computation of FireSat Ground-Track Parameters.

We next compute the required resolution and data rates for FireSat. From Table 1-5, we initially estimated the needed ground resolution as 30 m. Because this is meant to be a very low cost system, we will assume that the required resolution, dn is at nadir so that from an altitude of 700 km we have:

Had we made this requirement at the maximum slant range of 1,580 km, the required resolution would have been 0.001 deg.

Using this resolution, we can follow the procedure outlined in Table 9-15 to compute the data rate for FireSat as 85 Mbps. This data rate from the FireSat sensor is very high. However, we will be able to reduce it in many ways. We could process the data on board or, more simply, turn off the payload over the oceans or other areas where fire detection is of marginal utility. For now, we will leave the value as computed so that we remain aware of the data rate out the sensor, recognizing that this will be need to be reduced later in the system design.

We next compute mapping and pointing budgets for FireSat. We do not have a firm mapping requirement, but we do have some broad sense of what is needed. We begin, therefore, with a rough estimate of performance parameters and create the mapping error as a function of the elevation angle shown in Fig. 9-25A. In this figure, we have used a 0.1-deg nadir angle and azimuth errors corresponding to a relatively inexpensive pointing system based on an Earth sensor. We know we can go to a more expensive system if necessary. In looking at Fig. 9-25A, we see that the mapping error at our chosen minimum elevation angle of 20 deg is between 6 and 8 km. While we are not

Fig. 9-24. Computation of FireSat Ground-Track Parameters.

certain what our mapping requirement is, we are reasonably sure that it is smaller than 6 km. We need to locate fires more accurately than this. Note also that the accuracy has been set almost entirely by our crude attitude number of 0.1 deg.

The next most critical parameter is the 1 -km error in target altitude. This means that we assume we can determine the altitude of the fire above the Earth to 1 km—a reasonable accuracy with an oblate Earth model. But significantly improving this accuracy would require carrying a map of the altitudes of all of the regions of interest That could be very difficult, particularly in mountainous areas, and would cost a lot more money. Therefore, it is of little value to drive the error in nadir angle down below approximately 0.05 deg because it would no longer be the dominant error source. Fig. 9-25B shows the curves that we would achieve with the error in nadir angle reduced to O.OS deg and all of the other error sources remaining the same. Now the contribution of the errors in nadir angle and target altitude are comparable, so we will use this budget to establish a preliminary mapping requirement of 5.5 km at a 20-deg elevation angle, and 3 .5 km at a 30-deg elevation angle. This may still be considerably more erode than we would like, so we may need to revisit this issue.

A. 0.1 deg Nadir Angle Error B. 0.05 deg Nadir Angle Error

Fig. 9-25. FlreSat Mapping Budget Reducing the nadir angle error below 0.05 deg will have relatively little Impact on the overall mapping error because of the 1 km target altitude error. Compare with Fig. 5-22.

A. 0.1 deg Nadir Angle Error B. 0.05 deg Nadir Angle Error

Fig. 9-25. FlreSat Mapping Budget Reducing the nadir angle error below 0.05 deg will have relatively little Impact on the overall mapping error because of the 1 km target altitude error. Compare with Fig. 5-22.

Equipped with the analysis of the mission geometry, we turn our attention to the process described in Table 9-15, these computations allow us to evaluate the optical, signal processing, and radiometric performance of the instrument The third column in that table summarizes the results of the computations for a whiskbroom sensor design for the FiieSat mission.

The example FireSat design addresses only initial feasibility of the instrument Several challenges remain with this design and addition iterations need to be made in the context of mission requirements and constraints. The computed data rate of 85 Mbps will present a design challenge, as will the multiple pixel scanner needed to scan 256 pixels simultaneously. This will require all 256 pixels to be read out in parallel, and the signal processing will need to be designed accordingly. These features present a particularly demanding element of the initial design.

0 0

Post a comment