Thermal Subsystem PDR

1
Thermal SubsystemPDR

• Josh Stamps
• Nicole Demandante
• Robin Hegedus
• 12/8/2003

2
Mission Statement

• To maintain temperature range of all hardware
  throughout the duration of flight.

3
Temperature Factors

• Knowing all sources of heat, as well as the
  geometry and thermal characteristics of all
  satellite components, we can predict the
  temperature at any point of interest.

• Heat Sources
• - Sun (Direct Solar)
• - Earth (IR)
• - Earth Reflection (Albedo)
• - Electrical Component
• (Batteries, Solar Panels, etc.)

• Thermal Characteristics
• - Specific Heat
• Measure of how much heat energy per unit
  mass that an object gains or losses in a
  temperature change of one degree.
• - Thermal Conductivity
• Gives value to the heat flux that will
  travel through a material as a result of a
  temperature gradient.
• - Absorptivity, Emissivity
• Each is a percentage of how much incident
  radiation will be absorbed and emitted by an
  element. This is largely dependent on surface
  finish.

4
Method of Attack (TAK)

• TAK III (Thermal Analysis Kit) This software
  is what we utilize to obtain temperature
  predictions for any point in the satellites
  orbit.
• Inputs Heat Sources, Thermal Properties, Node
  Geometry, Node Interfaces (Conductors)
• Outputs Temperatures of each node, at any point
  in time
• Limitations TAK is unable to calculate radiation
  heat transfers in the case of an orbiting
  satellite, and we are left to use alternate
  methods of determining heat losses and gains due
  to radiation. For this we use programs developed
  by Bob Poley at Ball Aerospace.

5
Model Definition

• The physical model is replaced by a collection
  of nodes linked together by conductors.
• Each node tells TAK how much energy the real
  object can hold for each degree Celsius. This is
  based on the specific heat of the material as
  well as its physical mass.
• Conductors inform TAK how heat travels from node
  to node. This is based on
• - conductivity in the case of conduction heat
  transfer
• - emmissivity, absorptivity, and the incident
  radiation for the case of radiative conductors.

6
Node Identification

• - -
  Each node is given a five digit
  identifier.
• The first two numbers identify which side of the
  satellite we are referring too.
• The second number set refers to the depth of
  the node on whichever side it is located.
• The last three numbers are for the node which is
  on a certain side, at a certain depth.

7
Node Identification Continued

• Side Indices
• 1 Zenith (Boom) Panel
• 2 Nadir (Camera) Panel
• 3,4,5,7,8,9 Side Panels
• 6 Internal Panels (example Torque Rods)
• 10,11 Aerofins
• 12 FITS Panel
• 13,14,15,16,17,18,19,20,21 Tip Mass

8
Node Identification Continued

• Depth Index
• 0 Screws, Washers, etc.
• 1 Most external components (Solar Panels)
• 2 MLI nodes
• 3 Frame Nodes (ISOGRID)
• 4 Component Box Nodes

9
Node Identification Continued

• Node Index
• Letter Prefix is not part of the node index,
  but merely identifies the geometry of the node
• A - .9855 x 8.255 x .5 (in)
• B - .5 x .5 x .25 (in)
• C - .12 x .25 x 1.9418 (in)
• D - .656 x 11.781 x .25 (in)

NADIR ISOGRID (Side 2, Depth 3)
10
Node Identification Continued

• Node Index
• A - .25 x .3125 x 12 (in)
• B - .25 x .3125 x 7.875 (in)
• C - .25 x .5 x .8115 (in)
• D - .25 x .5 x .5 (in)
• E - .25 x .125 x 1.8832 (in)
• F - .25 x .125 x 1.5 (in)
• G - .25 x .125 x 1.75 (in)
• H - .25 x .125 x 1.6819 (in)

SIDE Panel ISOGRID (Sides 3,4,5,7,8,9,10,11 Depth
3)
11
Node Identification Continued

• Node Index
• A 10 x 10 x .3 (in)
• B - .5 x .8492 x 8.25 (in)
• C - .4822 x .4822 x .25 (in)
• D - .12 x .25 x 1.697 (in)
• E 1.9418 x .12 x .25 (in)
• F - .12 x .25 x 1.9423 (in)
• G - .12 x .25 x 1.1709 (in)
• H 11.78 x 1.311 x .25 (in)

NADIR ISOGRID (Side 1, Depth 3)
12
Node Identification Continued

• Total Nodes
• Completed
• - Nadir Frame (111 Nodes)
• - Side Panels (462 Nodes)
• - Zenith Panel (35 Nodes)
• - MLI Nodes (56)
• Incomplete
• - Tip Mass (About 400 nodes)
• - Components (About 75 nodes)
• Percent Complete 58

13
Conductors

• There are two kinds of conductors, which connect
  nodes and explain heat transfer between them.
• - Conduction based and Radiation based

14
Conduction Conductors

• The most basic conductor is the conduction based
  conductor. This represents heat that transfers
  between nodes via conduction.
• The equation is C kA/L
• k material conductivity
• A - is the interface area between nodes
• L - length between the centers of each node
• ERROR POTENTIAL This assumes a tight interface
  between nodes, which becomes an issue in cases
  where nodes are connected by pressure (screws,
  MLI-side panels, etc.)

15
Radiation Conductors

• Radiation conductors are a bit more complex.
  Blame Ludwig Boltzmann for this. And for this
  reason, rather than hand calculating each
  conductor we use an algorithm developed by Bob
  Poley at Ball Aerospace.

16
Poley-gorithm

• Radiation heat transfer is a result of a
  temperature gradient existing over a space. So,
  we need to determine which nodes see each
  other, how much they see of each other, and the
  heat flux between these nodes within sight of
  each other.
• This process is accomplished by use of four
  programs Supview, FindB6, Albedo, and Reflect

17
Poley-gorithm (Supview)

• The purpose of Supview is to input the
  orientation of all nodes modeled and to determine
  how much of one another they see. In this
  case, the nodes are treated as surfaces defined
  by Cartesian coordinates with respect to the
  exact center of DINO. The output is a collection
  of view factors to be taken as the interface area
  between the nodes.

18
Poley-gorithm (FindB6)

• FindB6 determines orbital characteristics. We
  can find when the satellite enters and exits the
  shadow of the earth. Also we can establish a
  cold case as well as a hot case, based on the
  highest and lowest possible solar fluxes DINO
  could encounter.
• Inputs
• Starting/Ending Days (3/21/6-3/22/7)
• Universal Time (6AM or 3600 seconds)
• Apogee/Perigee (6728 km)
• Inclination (51.6 degrees)
• Beta Angles (Hot-75.1 degrees, Cold-0 degrees)
• Solar Flux (Hot-1428 w/m2, Cold-1316 w/m2)
• Earthshine IR (Hot-227 w/m2, Cold-175 w/m2)
• Albedo (Hot-56, Cold-37)
• Outputs
• Enter Shadow (Hot-never, Cold-198.6 degrees)
• Exit Shadow (Hot-always, Cold-61.4 degrees)

19
Poley- gorithm (Albedo)

• Albedo finds how much radiation hits each node.
  This does not determine how much is reflected or
  absorbed, but simply how much is incident. The
  inputs for this program include the view factors
  each node has with the sun and earth determined
  by Supview, as well as the positions given by
  FindB6 that we are interested in solving for.
  This file is run for both hot and cold cases.

20
Poley-gorithm (Reflect)

• Finally the conductors between each node are
  calculated with the Reflect Program. Given the
  incident radiation energy determined by Albedo,
  and the radiation properties of the material, a
  conductor can be created between all nodes and
  cold space. Furthermore, conductors are
  determined based on thermal properties alone for
  radiation transfer between all nodes that see
  each other.

21
MLI

• Purpose of Multi-Layer Insulation
• Keeps our satellite as close to adiabatic as
  possible as warm interior is insulated from cold
  exterior surfaces.
• Secondary benefits include atomic oxygen and
  micrometeoroid protection, as well as protection
  of electronics from direct radiation.

22
MLI continued

• MLIs are typically constructed by encasing
  multiple layers of Dacron netting between double
  aluminized Mylar.
• In space the MLI blanket puffs out, as would a
  marshmallow in a vacuum. The result is multiple
  layers of material that can only transfer heat
  via radiation, as they will not make contact with
  each other as is necessary for conduction.
• The Mylar pillowcase has extremely low
  emmissivity and absorptivity values, slowing heat
  transfer due to radiation.
• The Dacron pillow is a netting, so in the event
  that layers make contact and result in
  conduction, the interface area is a minimal.

Mylar Pillow case
Dacron net spacers
23
MLI continued

• MLI construction
• It was hinted at recently that perhaps we
  could get pre-assembled MLI blankets from Ball.
  I have not been able to follow up on that at this
  time, but it is a possibility. Otherwise, we can
  simply purchase the materials and sow it together
  ourselves. We have met with an advisor at Ball,
  Leslie Buchanan, who is willing to aid us in the
  process.

24
MLI continued

• MLI selection
• - Materials. If we do end up purchasing the
  materials, we can select them based on which
  emmissivity and absorptivity values are required
  for our system to work. For example we can
  select other metallized finishes.
• - Inner Layer Optimization. We need to
  optimize how many layers of Dacron netting are
  needed.
• - Size Optimization. Currently we are
  assuming MLIs will surround all surfaces with
  the exception of the zenith panel and aerofins.
  However, we must also determine if its possible
  to wedge MLI blankets between the frame and solar
  panels as well as between Lightband and the Nadir
  Plate.

25
Model Completion

• Several variations of our thermal model will need
  to be made for the following cases to have an
  accurate final model
• Tip Mass, Aerofins, FITS system undeployed
• Electronics failures
• Remodel for Temperature Gradients exposed by
  TAK

26
Testing

• The norm for testing the thermal model of a
  satellite is by placing it in a thermal vacuum,
  dropping the temperature to around 5 degrees
  Kelvin, and monitoring the temperature at several
  pre-selected locations using thermocouples.

27
Tip-Mass

• Modeling Plan
• Because of similarities between the main
  satellite and the tip-mass, a model can be made
  by shrinking the main satellite.
• The model can then be added to the main satellite
  to create a Dino model.

28
Tip-Mass (Cont.)

• Problems
• The tip-mass will need to be modeled multiple
  times for different deployment scenarios.
• First as a part of the satellite
• After it has been deployed

29
Thermal Desktop

• Thermal Desktop can be given to us for free if we
  want it and have AutoCAD.
• AutoCAD will cost us about 100 per year for the
  license.

30
Thermal Desktop vs. TAK

• Advantages of Switching
• Could be a lot faster to model and change models
• Creating different scenarios could be easier
• Thermal Desktop has a Graphical User Interface
• Gives us the ability to look at what we are
  modeling
• TAK only has numbers for inputs and outputs and
  is difficult to troubleshoot

31
Thermal Desktop vs. TAK

• Disadvantages
• AutoCAD will cost us money
• We dont have anybody that knows how to use
  Thermal Desktop
• Still Looking
• All of the structures files are in Solidworks and
  we need them in AutoCAD
• AutoCAD is difficult to use to model 3D objects
• We would need to start all of the modeling over