Fluid Systems

The final topic we'll cover in Spacecraft Systems is fluid systems. A fluid system is any system that uses a working fluid, liquid or gas, to do work.  Gasses and liquids are both fluids, they flow when subjected to a shear stress. 
Liquids are incompressible, their volume does not change under pressure.  This is because liquids have constant volume.  Gasses, on the other hand, are compressible.  Their density and volume change with their pressure.
Because fluids can flow under stress they will naturally move from areas of high pressure to areas of low pressure.  Because of this the pressure in a confined fluid is the same everywhere.  Known as Pascal's Law this property means that the pressure at one wall containing a fluid is the pressure at all the walls.
Pascal's law allows fluids to transmit force.  Consider the illustration below.  When piston one is pressed down pistons two and three raise up.

How far pistons two and three raise up depends on how far piston one is pressed and how wide the pistons are.  Remember that the volume of a liquid doesn't change under pressure.  So the volume that is compressed in piston one is the volume of pistons two and three that raises up.  Volume here is the cross sectional area of the piston times how far up or down it moves.
What this means is that we can construct a device that uses a fluid to take a force from one area and relay it to another.  We do so by increasing the pressure at one end of the fluid.  Remember that pressure is a force divided by the area.  So by creating a pressure at one end we create a force at the other end that equals the pressure times the area we're working over.
It also means that the working fluid can amplify our force.  If we push down on a cylinder with a small area we get a high pressure for our force.  If it's attached to another cylinder with a large area that pressure produces more force than we had to put in.  Similar to how a lever can amplify force by increasing the moment.
In a system that transmits force through a fluid we can use either a liquid or a gas as our working fluid.  Hydraulic systems use a liquid, often water or oil, as their working fluid.  Pneumatic systems use a gas, usually air or carbon-dioxide.  Spacecraft often use helium as a working fluid.  It is stored as a cryogenic liquid but expands into a gas when in use.
Pneumatic systems are usually simpler to design and control than hydraulic systems.  They also tend to be more reliable, compressible gas can absorb shocks reducing damage to machinery.  Pneumatic systems are also safer, the gasses used are nonflammable and the systems can typically fail-safe.
Hydraulic systems have their own advantages.  They are more efficient, precisely because gasses tend to absorb some energy while liquids transmit all of their applied load.  There's no need to bleed off excess pressure when disconnecting equipment from a hydraulic system.  But their biggest advantage is their carrying load.  Pneumatic systems can support about 80 - 100 psi.  Hydraulic systems typically support 1,000 - 5,000 psi and can be designed up to 10,000 psi.

Now let's consider a simple hydraulic/pneumatic system:

The various components of this system are connected by conductors.  In fluid systems a conductor is simply a pipe or hose that contains and transmits the fluid.  Conductors must be rated for the pressure that the working fluid is intended to be transmitted at.
At the start of the system is a pump or compressor.  These two devices serve the same function, they increase the pressure of the fluid.  Pumps add pressure to a liquid, while compressors add pressure to a gas.
Working fluid goes from the pump into an accumulator, a tank designed to hold pressurized gas.  The accumulator stores the fluid as the pump builds up pressure.  From the accumulator the fluid passes through a check valve.  If the pressure is correct the check valve allows fluid to pass to the loads, if the pressure grows to high the check valve sends fluid straight to the reservoir.
Loads are devices attached to the hydraulic system that do some work using the fluid pressure.  The fluid expends its pressure doing work and the now low pressure fluid passes into the reservoir.
The reservoir is a second, low pressure, holding tank that collects used fluid from the loads.  Here the working fluid is stored until it again passes through the pump and into the hydraulic system again.
Pneumatic systems using air don't have a reservoir attached.  The reservoir in this case is the atmosphere.  Air is released by the load into the atmosphere and the compressor pulls air straight from its surroundings.
Many systems are precharged, meaning that the accumulator is already pressurized.  In this kind of system there is no pump or reservoir, spent working fluid is vented overboard.  Such a system is much lighter and simpler than a full system with compressor, however they have a very limited operating life.  When the pressure of the accumulator is spent the system is no longer any use.  Many spacecraft use precharged systems.

Now let's take a closer look at the devices in a fluid system.  First consider the devices we already saw in the simple pneumatic system above.  It is composed of two pressure vessels, the reservoir and the accumulator; several valves; and a pump or compressor.
Pressure vessels are tanks designed to contain a fluid under pressure.  They are rated based on their wall thickness, which determines how much pressure they can hold.  Readily available tanks are also given in terms of standard sizes.  One design that increases the strength, and so the pressure rating, of a tank is a composite overwrap vessel.  In this design a metal liner is surrounded with adhesive and a layer of composite fiber.  This gives the tank additional strength with little increase in mass.

Valves are simple devices that either block fluid flow or divert it when closed.  They come in two types.  Type 1 valves remain in the position they were set to until they are again changed.  Type 2 valves are classifies as either normally open or normally closed.  Type 2 valves return to their normal position when force is removed.  Valves can also be one-way, allowing fluid to pass in one direction freely, but blocking all passage backwards.
Pumps and compressors take work from some outside source and convert it into fluid pressure.  The most common pumps used in fluid systems use turbomachinery, that is spinning parts, to push fluid.  The simplest example, there is probably one in the room, is a fan.  The fan's blades create an aerodynamic force which pushes on the air.  There are many different configurations of turbo- pumps and compressors.  Below is a gas compressor:

Not all pumps are turbomachines, however.  A simple bicycle pump uses a much simpler system.  There are two one-way valves, one allowing air in from the environment and one allowing it out through the hose.  Pulling on the piston creates a vacuum which draws in air, pushing it down forces the air out.
There is another type of turbomachine that does the opposite job.  A turbine uses fluid pressure to produce work.  Fluid passes through a sequence of turbine blades, imparting a force to them.  This causes a shaft to spin, which draws power from the fluid.  Energy from the fluid (which existed as pressure) is given to the shaft (as rotation) allowing us to use the fluid to produce useful power.
Pistons also remove pressure from the fluid and convert it into power.  In a piston pressure forces the plunger to move out.  The plunger is then pulled back, forcing the working fluid back out.  A full motion of the plunger up or down is called a single stroke. 
A two-stroke piston has two inlets, pressurized fluid pushes the cylinder out, expending its pressure.  Then pressurized fluid enters through the other inlet pushing the piston back in and forcing the spent working fluid on the other side of the plunger out.  This repeats.  Other designs have two or more pistons attached to a single rotating wheel, when one pushes out the wheel pushes the other back in.
In a hydraulic system turbines are used to produce rotation.  Pistons are used to produce a back and forth motion.  Pistons can produce rotation by attaching them to a crank shaft they can also produce rotation, as is done in care engines.  However turbines are far more efficient for pneumatic applications.
Heat exchangers add or remove heat from the working fluid.  Fluid passes through a series of thermal conductors, usually metal, that are connected to a source of heat or a cold sink.  Heat exchangers attached to points hotter than the fluid entering are called high temperature reservoirs.  Heat exchangers connected to points colder than the fluid are low temperature reservoirs.  Heat exchangers add or remove energy from the flow in the form of heat.
As we discussed when looking at rockets fluid devices called nozzles convert pressure into velocity.  Another device, called a diffuser, does the opposite job.  Diffusers slow down the working fluid in order to increase its pressure.  These two devices are very similar in shape and design.  For subsonic flow (which liquids must be and pneumatic systems generally are) a nozzle narrows and a diffuser expands.
Throttles also obstruct flow.  At the throttle flow suddenly narrows, drastically reducing pressure and trading it for velocity, unlike a nozzle where this change is smooth.  Throttles function much like valves, in that they open and close allowing some flow through. 

We've looked at fluid systems and devices in a very general sense so far.  But why are we even looking at them?  What are they used for on spacecraft that makes them worth considering?  Well, before looking at hydraulics on spacecraft lets consider two other kinds of fluid systems that we've already looked at.
Consider the rocket systems we saw before.  Rockets are fundamentally a fluid system.  The working fluid is the propellant.  The combustion chamber releases pressurized working fluid through a nozzle, which does work by converting pressure into velocity and causing a transfer of momentum. 
Liquid and hybrid rockets have a complex fluid system leading into the combustion chamber.  Cold propellant is sent through pumps into the combustion chamber.  In regenerative systems it passes though a heat exchanger, gaining energy from it.  The entire propulsion system is a fluid system.
Many thermal control systems are also fluid systems.  In this case the working fluid is coolant.  It passes through heat exchangers, taking heat away from systems that produce it.  The fluid is compressed and is forced to give up that heat, then passes through the system again.

Hydraulic systems are used on spacecraft because of their very high energy density.  Similar electrical systems would be needlessly bulky.  Below is a graphic of the hydraulic systems used aboard the Space Shuttle.

In general hydraulic systems are used any time a great deal of force is needed in a confined area.  Hydraulics are used for thrust vector control, shifting the orientation of heavy engines that are mounted on gimbals.  They move aerosurfaces, flaps and rudders that control the motion of launch vehicles against immense aerodynamic loads.
Hydraulic systems are also used to open propellant valves on large vehicles.  The amount of fluid and pressure involved in large engines is simply too great for a reasonably sized electric valve to be used.  In a sense the hydraulic valve amplifies the strength of the electric valve which is used to activate it.

Pyrotechnics

Pyrotechnic devices, or explosive ordnance, are used on spacecraft for many applications.  They have high energy density and are compact, easily stored, and controllable.  However they are only used reluctantly.  When another option is available generally it will be selected over pyros.
Pyros are single-use devices, once set-off there is no retrieving them and the system cannot be reset.  Pyros can only be used for a component that begins in one configuration and then changes to another and remains that way for the rest of the flight.
They produce very short, impulsive loads.  Despite their high energy density all that energy is expended at once.  Pyros don't release their energy in a controlled fashion and controlling the energy release involves additional structure and engineering.
A pyrotechnic device cannot be functionally tested, there is no way to be certain it works until it is used.  There are very few tests that can be done to ensure a charge functions before installing it.  Reliability of pyros is based on qualification.  If we get a box of 100 then we set of 20 of them.  If all of them go off we assume there's a good chance that the next one will when we use it.
If something does go wrong with a pyro during flight there's very little that can be done about it.  Failures can and do happen, and when they do the vehicle is already off the ground.  Redundancy can be used to mitigate failures, so that if one pyro doesn't go off another can accomplish the same job.
Pyro usage falls into three broad categories: operational, flight termination, and emergency use.  Operational devices are used as a standard part of the flight plan.  Emergency use devices are activated during contingencies, to regain control of the vehicle or attempt to salvage some critical system (like the crew).  Flight termination devices are used to safely abort failed flights.  An aborted flight is destroyed and/or sent into the ocean, to reduce the damage it does as much as possible.
Pyros are classified into two categories: category A, hazardous, and category B, nonhazardous.  Category B devices could go off in your hand without causing any injury.
With any explosive device we have two chief concerns: inadvertent firing and nonfiring.  Either of these could result in mission failure, since each step of the mission is time dependent.  Both are also serious range safety concerns.  An early firing could happen while technicians are on the range working.  A nonfire means a primed explosive is still sitting on the range, and no one knows when it might finally go off.
To minimize risk a number of requirements are in place for pyros.  They are always installed as late as possible during the assembly process.  After installation inhibitors are put in place to prevent the devices from going off, and are not removed until just before launch.
Once the pyros are in place radio broadcast at the range is limited so that the radio power at any device is less than 20dB under the no-fire power.  Electromagnetic waves can induce a current in a wire, so radio power must be kept low enough that the induced power is always below the lowest level that the devices will fire at.
During installation there is full radio silence on the range.  No other operations take place during pyro installation and all range personnel wear flame retardant, nonstatic coveralls.  All equipment and personnel must be grounded.  Pyros have faraday caps, metal covers that provide a voltage path around the charge.  All lines are checked for stray voltage prior to installation.  There must be a clear area of at least 10ft around any work with pyros.  Humidity must be at least 35% and there cannot be a storm within 5mi.  All these precautions are taken to ensure no stray spark accidentally triggers any explosive device during installation.
Pyrotechnic systems are designed to be completely separate from all other systems.  A pyro system has shielded circuits with a single common ground.  All pyros are built with metal connectors that cannot mismate.  The pyro system consists of the following, separate and independent, systems:

• power source - a dedicated battery or capacitor independent of the main power bus.
• firing circuit - circuit that connects the initiating device, or detonator, to the power source.  FIRE and ARM circuits are required to be separate, so that if either is accidentally triggered there is no ignition.
• control circuit - activates and deactivates the safety devices
• monitor circuit - monitors the firing circuit for stray voltage.  This must be completely independent of the control and firing circuits.

 The SAFE and ARM plugs are also required to be as close to the device as possible.  This reduces the number of locations for possible failures.

That gives you a rough idea of just how serious the safety concerns are when handling and installing pyrotechnic devices.  Now we'll look at a few of the ways they are used.
Initiators are used to trigger another device.  They can be used to ignite a solid rocket motor or to set off another pyrotechnic device.  The NASA Standard Initiator (NSI-1) is seen below, connected to the NSI-Detonator.  The detonator amplifies the explosive force of the standard initiator.

Frangible nuts are a category B ordnance used to uncouple structures.  They are ordinary nuts with breaking points added.  These breaking points are stuffed with explosives, so that when the pyro is triggered the nut breaks along the existing failure line.

Separation bolts serve a similar function.  They are hollow, with explosive charge built in and a designed failure point.  When the explosive goes off it pushes a pin, which breaks the bolt at the defined point.

Pyrotechnic linear actuators push a rod out.  The rod is contained in a hollow chamber with a small explosive charge.  When the charge goes off the pin is pushed out by the force.  Spring-loaded pins keep the rod in place once extended.

Pin pullers do the opposite job, they retract a rod.  In this case the rod ends in a plunger, with the explosive charge between it and the wall.  The explosive pushed the plunger back and the pin withdraws.

Pyrotechnic valves switch from open to close, or closed to open, once.  Normally open valves switch to closed, by pushing a plug into the flow.  Normally closed valves have a plug with an opening above in the flow, it is pushed so that the opening allows fluid through once activated.  Both of these pictures are normally closed valves

Shaped linear charges hold an explosive in a metal container, which expands when the explosive goes off.  Mild detonating chord contains a linear core of explosive material contained in a metal sheath.  Confined detonating fuse uses a combination of fiberglass outer sheath, rayon middle sheath, and lead inner sheath with layers of vinyl between.  CDF contains an explosion once the linear charge goes off.

Super*Zip is another method of severance.  Two sheet metal faces surround a core structure.  At points the core is replaced with CDF, as the CDF goes off its outer sheath expands.  This pushes the two plates out until they break, separating the structure.  It gets its name because it resembles a zipper being unzipped.

Structures

Before we look at how a spacecraft's structure is put together we need to consider how structures are built at all.  Most structures are built in one of two ways: they use a frame, a skeleton of members to give it shape and strength, or they use load bearing walls, in which the entire structure's face supports it.  Most spacecraft structures use a combination of the two.
A load is a force which a member is subjected to.  There are three forms of simple loads: tension, compression, and shear.  Tension loads pull against the member, compression loads push on it.  Shear loads push or pull two sides of the member in opposite directions.  Materials are generally stronger against compression and tension than they are against shear.

Imagine a block of wood supporting a load above it.  Naturally we know that a wider block can support more weight.  But all blocks of wood can support the same amount of stress.  Stress is defined as the load divided by the cross-sectional area of the member supporting it, the width and thickness of the block.  While a block with a larger area can support more load than a thinner one they can both support the same amount of stress.

When a member is stressed it changes its dimensions.  Think about a rubber band, when you pull on it it gets longer and thinner.  That's because there's still the same amount of material there was before.  So if we compress an object it should get wider and shorter.  The change in the length of a member under stress is called the strain.  Strain is defined as the change in length of the object divided by the original length:

When subjected to the same stress a given material will always experience the same strain.  This is a material constant, something that is the same for this material no matter the situation.  The property that relates stress and strain in a material is called Young's modulus (or the elastic modulus):

Most materials are elastic for a while, when deformed they spring back after the load is removed.  After a certain amount of stress, called the yield strength, they are no longer elastic and Young's modulus isn't useful anymore.  After the yield stress they are plastic, their deformation is permanent.  Eventually enough stress is applied that they break entirely, called the ultimate strength.  We usually don't want a material to yield, since we want it to keep whatever shape we put it in, so we need to make sure that members aren't subjected to loads above their yield strength.

In the elastic region stress and strain are related by the modulus.  After the material yields it enters the plastic region, where force simply goes into deforming it.  Eventually it deforms so much that it breaks.  Materials with a large plastic region are called ductile; we can shape them and they remain in their new shape and will resist changing from it.  Materials with a small, or no, plastic region are called brittle; once they yield they break.  Counter-intuitive as it may seem a rubber band is brittle, it has no plastic region and breaks as soon as it is over-stretched.
Many common materials have the same properties in all three dimensions: x, y, and z.  These materials are called isotropic.  Some materials do not have the same strength or stiffness in all directions, however, and are called non-isotropic.  An everyday example of a non-isotropic material is wood.  Wood is much stronger in the grain direction than against the grain, this is because the grain fibers are stronger than the sap that holds them together.  Consider also plywood, which is composed of thin sheets of wood glued together.  The glue is its weakest point, so the plywood has different properties in all three dimensions: the fiber is stronger than the sap, which is stronger than the glue.  Modern fiber reinforced composites are similar.  They are stronger in the grain direction than any other, and the matrix holding different plies together is the weakest point.

As simple as the three basic load conditions are, most materials won't actually fail in shear or compression. Under most situations when a material is subjected to a shear load it doesn't actually shear apart, it bends.  Bending occurs when part of a material is pushed in one direction and the rest is pushed in the other.  Usually shear happens when these two forces are very close together.

Observe that when the material bends its top half is in compression and the bottom half is in tension.  The bending stress in both is the same, but it is in opposite directions.  In the middle, the neutral axis, there is no stress.  It goes from negative that value on top to zero in the middle to positive on the bottom.  The bending stress depends on distance from the neutral axis, the further away from it the higher the stress.
This is the reason for I-beams.  Since there is little stress in the center there is no need for material there.  Most of the material is at the top and bottom, where the stress is highest.
Now, consider again the wood column with a load on top. Will it actually get thicker?  Usually not unless it is already very thick, most thin member will buckle when subjected to compression loads.  Buckling occurs when part of the material bows out.  It is like bending, except that the load pushes in along the axis.

We need to be concerned with both buckling and compressive failure when loading a column in compression.  Just as we need to be concerned with both shear and bending failure when subjected it to a shear load.  Bending and buckling depend on the moment of inertia of the structure, while compression and shear depend on the area.  This means that while the simple load cases (compression and shear) only depend on how much material there is, bending and buckling depend on how it is arranged.

Before we observed that elastic materials spring back into shape when deformed.  Picture a spring, when you push it down it bounces back.  But before it gets back to where it started it bounces back and forth for a while.  Materials that have deformed act exactly the same way, when they bounce back they vibrate slightly for a while.  All objects have a range of natural frequencies that they vibrate at when subjected to a force.  The lowest of these is called the fundamental frequency, it is the one which the object is most likely to vibrate at.  Think about a guitar string.  When you pluck the string it vibrates.  Shorten the string and the pitch gets higher, thicken the string and it gets lower:

The kinds of failure we looked at before come from static loads.  Static loads do not change with time.  There are also dynamic loads, which do change with time.  Dynamic loads can be transients, brief or rapidly changing loads like a ramp up/down or an impact, or cyclic.  Cyclic loads repeat themselves.  They subject the member to a force that is time varying, with some maximum, minimum, and average values, and repeats with some frequency:

Ordinarily we would only worry about the peak stress here being higher than the yield stress of a member.  However if the frequency of the stress is the same as the fundamental frequency of the member a situation called resonance occurs.  In resonance each time the material gets to the peak of its displacement it is pushed back by the force, pushing it a little further each time.  Picture a kid on a swing.  By pumping his legs just as gets to the back of his swing he can push the next swing a little further.  In a material this pushes the material's deformation further and further each cycle, until it fails.  Resonance can cause even small loads to build up to catastrophic displacements.  For an excellent reference consider the Mythbusters' attempts to recreate Nikola Tesla's claims to have invented an earthquake machine.

So that's what loads do to a member in a structure.  But where do the loads come from?  A vehicle in space experiences forces from several sources:

• propulsion - thrust is generally a static load applied by the spacecraft's engines.  Engine sputter can also cause small transient loads and the engine itself can vibrate causing small cyclic loads.
• deployment - extending booms or opening apertures creates transient loads on the vehicle.
• thermal - when a material heats it expands, when it cools it shrinks back, causing thermal strain.  If it's being held in place then a thermal stress results from holding it back.  This can be a static load if the material stays at the new temperature, cyclic if it heats and cools (like a spacecraft moving in and out of sunlight), or transient if the heating is brief.
• internal mechanics - mechanical systems by definition move, they subject some component to force.  They are also subjected to some force, as is whatever they are mounted to.  

While loads in space can be intense, the larger problem is often launch loading:

• thrust - launch thrust is much higher than the small thrust values used once in orbit.  Payload fairings are designed to lessen the g-loading payloads are subjected to, but cannot completely eliminate the force.
• aerodynamic forces - as the launch vehicle moves through the atmosphere it is buffeted by winds.  This is generally a transient load, gusts blow briefly, creating high force for a brief time.  By design little of this load reaches the payload.
• acoustic and vibration - the entire vehicle shakes during launch and produces extremely powerful sound waves.  These in turn cause the vehicle to vibrate.
• staging, separation, and injection - transient loads caused by the launch vehicle discarding unneeded parts during flight.

A spacecraft is composed of two forms of structure.  Primary structure gives the spacecraft its shape and holds all of the systems together.  Secondary structure provides support and mounting for the various systems and subsystems.  The secondary structure transmits loads from the systems to the primary structure and reduces loads going to systems.
Forces are carried by members in a structure and transmitted through it.  In a load bearing wall the wall carries forces through it, so that most of the wall is under the same load.  In a frame supported system each member in the frame carries a force through its length to the joint where it is connected to other members.  Consider the simple trusses below:

Each member caries a force through its length.  So a horizontal member carries horizontal loads, a vertical member carries vertical loads, and a diagonal member carries some vertical and some horizontal load.
On the ground our goal is to transmit the loads through the structure to the ground.  Each member carries the load in whichever direction it can until we get to the grounding points.  The ground provides the reaction force, it pushes back in the opposite direction so that our structure doesn't move.
In space there is no ground to bring the load to.  If the entire primary structure is subjected to the same force then the structure will move.  To prevent the spacecraft from moving we must either cancel or dampen the force.  Cancelling a force means applying another force of the same strength in the opposite direction.  Like the torque wrench example we talked about back in attitude control, the two up and downward forces cancel so the head doesn't move up or down.
Damping reduces or prevents motion.  Damping is usually caused by friction.  It happens inside all materials, internal friction happens when a material vibrates.  As it does the vibration also slows until it eventually stops.  Any time the structure deforms some of the energy that might have made it move is absorbed instead.
Viscous damping happens when a solid object move through a fluid, like air.  Though there is no air resistance in space we can add viscous dampers to reduce motion of components that should remain fixed.  The simplest viscous damper is a plunger in a piston of fluid, which resists the motion of the plunger. 
Shock absorbers and isolators are in essence simply springs with very high damping.  The spring absorbs the shock by compressing or expanding.  The damper prevents it from springing back quickly, so the load is transferred up very slowly.  Much of the load is also absorbed by the damper so there is less motion.
Mass also resists motion on its own, through inertia.  So when we apply a load to a structure there are three places it can go: into deforming material, into friction, or into accelerating the structure.  How much force is used deforming it depends on how far it is pushed, how much is spent against friction depends on how fast it moves, the rest tries to accelerate it:
 
So long as each member can support the load it is subjected to then it will act like a combination of a spring and a damper.  It will also try to move, but be stopped by the next member in the frame, which will experience whatever force wasn't absorbed.  This continues through the whole structure.  Eventually every member is under some stress.  When the load is removed everything springs back to how it was.  Whether or not the whole vehicle moves depends on whether every member is trying to accelerate in the same direction and if there is enough force to overcome the tendency of materials to act as springs.

Environmental Control

Every system and component on a spacecraft has its own range of operating temperatures.  Outside of this range it either breaks or simply can't work properly.  The duty of the thermal control system is to keep all the components within their allowable operating range.  An earth orbiting spacecraft can be subjected to skin temperatures from -270 to 2,000 degrees centigrade.  No component on board can handle that.  Reasonable operating ranges for spacecraft components:

• electronics: -10 to 45 deg
• batteries: 0 to 10 deg
• IR detectors: -269 to -173
• solid state particle detectors: -35 to 0
• motors: 0 to 50
• solar panels: -100 to 125

Each of these components produces heat, and heat comes into the spacecraft from outside.  Before we move on we must first discuss what we mean by heat and temperature, because they are not the same thing.
Temperature is a measure of how much energy is stored in a substance.  The atoms making up a substance are constantly in motion, in fluid they move around each other freely, in a solid they vibrate in place.  Temperature measures how much they move around, how quickly they flow past one another or vibrate.
Heat is a means of transferring that energy.  When something cools it releases heat it gives up some of that energy it contained.  When a substance is heated heat enters it and the motion increases.
An object can produce and release heat at the same time.  If it releases more heat than it produces then it cools, and temperature goes down.  If more heat comes in, or is produced, than leaves then the temperature goes up.  When the two are exactly equal the temperature remains constant, and we say the object is in thermal equilibrium.
That's where we want to keep our spacecraft systems, holding steady at their operating temperatures.  In addition to producing heat inside an object, heat can be transferred in three different ways.
Conduction occurs inside a substance.  The motion of atoms causes them to collide with each other, spreading their energy.  You can see this very clearly if you try to grab a metal bot handle.  Metals are good heat conductors, they transfer their energy easily.  Conduction occurs when one part of an object is hotter than another and stops when the whole object is the same temperature.  Conductivity is the measure of how easily an object transfers heat, heat transfer rate depends on that and the difference in temperature along the length of an object.  For a rod with a cross-section A:

Convection occurs when a fluid passes over a solid wall.  At the surface conduction happens at a tiny scale between the two, and also occurs in each substance away from that surface.  We call the whole effect together convection.  Convection depends on the temperature of the fluid and the wall, as well as the transfer constant between them:
 
Convection also happens inside a fluid when heat in one part of the fluid causes it to move.  Convection causes hot fluid to move up, or away from walls, because it is less dense and less viscous.
Radiation occurs in all substances, solid or fluid.  Atoms in a substance have energy from their temperature.  In addition to motion some of this energy excites the atom, changes the energy level of its electrons.  If the energy is so great they leave the atom it's called ionizing the atom and if that happens to the whole substance it becomes a plasma.  But more often the electrons are excited for a short time and then fall back to where they were.  When that happens they release electromagnetic radiation, or a photon.  At higher temperatures the energy of the photon emitted increases.  At first its invisible infra-red light, most heat stays here.  But as the heat increases it can raise to the visible spectrum.  If you've ever seen a piece of iron or steel in a forge or furnace you've seen this effect.  Red hot steel is hot enough that it radiates in the visible red spectrum.  As it becomes hotter it starts releasing yellow light in addition to the red, so it looks orange.  Even hotter and it releases higher frequencies too, which together makes it look white.  Iron is literally glowing hot, it's producing light, when it is red or white hot.
 
Radiation output of a substance depends on its surface area and temperature, obviously.  These are related to the radiative heat output by the Stefan-Boltzmann Constant (the Greek letter sigma) and a material property known as the emissivity (Greek letter epsilon).  Emissivity is the ratio of how much radiation a substance could put out, to how much it actually puts out.  It ranges from zero to one.  An object with an emissivity of one is called a black-body, a perfect radiator often used in physics as an ideal model.  At zero the body doesn't radiate at all, which also doesn't happen in real life, though some objects come close to one extreme or the other.
Bodies don't just emit heat radiation, they also absorb it.  The ratio of how much heat comes in to how much heat the substance absorbs is the absorptivity:

Besides absorbing the radiation the substance can also reflect some back into space and transmit some through it.  The ratio of how much transmitted through is called transmissivity, the ratio of how much is reflected is called reflectivity. 

Incoming can only do these three things, be absorbed by the material, reflected back by it, or transmitted through it.  So all three terms must add up to one:


A spacecraft usually isn't in contact with any other solid object, or moving through a fluid.  The only means of heat transfer between the spacecraft and its environment is radiative.  Environmental heat sources for a spacecraft in low earth orbit begin with the sun.  Direct sunlight is the greatest source of heat, averaging around 1358 watts per square meter of skin exposed to it.  Some sunlight is indirect.  When sunlight comes to earth some is absorbed and some reflected (it's too thick to transmit light well) earth's albedo reflects that light.  The amount of heat coming from earth's albedo varies drastically depending on position, a value used to design for is about 407 watts per square meter.  Finally some of the heat from earth has nothing to do with the sun.  Earth, after all, has temperature and radiates like any other body.  For a spacecraft in LEO earthshine accounts for 237 watts per square meter.  Unlike direct sunlight or albedo this is constant, earthshine doesn't depend on position.  The side facing earth always experiences this, while direct sunlight ceases whenever the spacecraft is eclipsed by the earth.  During eclipse one side is subjected to earthshine and the other has no incident heat.
So we need some method of limiting, or at least evening out, the environmental heating a spacecraft is exposed to.  We call means of controlling heat into the spacecraft external thermal control. 
One method is to simply prevent any one side from facing the source of heat for too long.  When the spacecraft is on the day side of the earth one side is facing the sun at over 1300 watts per square meter, the other side is exposed to less than a fifth from earthshine.  By simply rotating the spacecraft in roll which side is constantly changing.  This method, popularly known as the barbecue roll.  It takes time for the spacecraft to heat up when exposed to the sun, rotating the spacecraft prevents any one part from getting that time.  As it's facing a colder direction is is radiating that heat out.
A barbecue roll can slow the skin of the spacecraft heating up.  But we still worry about heat passing through the skin, to the more heat sensitive components.  Temperature sensitive components are covered by insulation to prevent heat from getting to them.  Multi-layer insulation (MLI) is the common method.  MLI is, as the name implies, composed of many thin layers of insulation protecting the spacecraft, each with low transmissivity.  Multiple layers are used because the radiation between layers is much slower than the conduction through a single thick layer.  Inner layers usually use Mylar and Kapton sheets, for their excellent insulating properties.  Outer layers are chosen for the expected operating range and durability.  Each layer is separated by a mesh or fine netting and coated in a fine layer of vacuum deposited aluminum, to increase its reflectivity and absorptivity. 
Like any blanket these not only keep heat out when the outside is too hot, they can also keep heat in when the outside is too cold.  Both these techniques keep the spacecraft temperature stable and even.  For many small spacecraft that's all that's necessary, preventing too much heat from getting in or out keeps the components in their operating ranges.  However often some components have very narrow ranges or wildly different ones.  When we worry about the heat produced inside the spacecraft being too much then we have to include internal thermal control systems.
When you think about cooling something many times the first thought is to use water or ice.  A variation on that idea is ablative cooling.  An ablative coating is one that rests on the component and melts or evaporates.  Instead of increasing the temperature of the component all the heat goes into melting or evaporating the coating. 
Coating that evaporates, of course, is gone and can only be used once.  But one closely related method is to use wax.  Heat melts the wax when the component is on, the wax radiates that heat away and cools.  When cool the wax settles and solidifies.  When the component turns back on the wax melts again, absorbing all the heat.  Obviously this is only useful when the component is turned on and off in a cycle, it's no use for always on components.The easiest way to remove heat from a component that produces too much is to simply add a large mass to absorb that heat.   Heat sinks are pieces of metal that conduct heat away from the source.  Instead of increasing the temperature of the component the heat goes into the heat sink.
Another kind of heat sink is a heat pipe.  Unlike the heat sink this time the method of moving the heat isn't a solid, but a fluid.  A heat pipe is a hollow tube filled with a fluid (called the working fluid) that is near its evaporation temperature.  Near the source of heat the fluid evaporates.  Gas is less dense and less viscous than liquid, so it is pushed into the center of the pipe.  The hot gas wants to expand, so it pushes its way through the tube to the cold end.  Here is gives up its heat and condenses.  The hot gas coming in pushes the liquid back towards the hot end.

Heat pipes are themselves a particular variant of a fluid loop.  Heat pipes are a passive fluid loop, generally when we refer to a fluid loop, though, we mean an active loop.  In this case it isn't evaporation pressure than causes the working fluid to move, but rather compressors and pumps.  The fluid starts out cold, it moves past the hot component and absorbs that heat.  Fluid is now hotter than it was, moving faster, with more energy.  It then passes through a cold dump, where it loses that energy.  At its very simplest that's all there is, just like the heat pipe.
Fluids used for spacecraft cooling vary.  Some use ammonia, or even common refrigerant used in air conditioners and refrigerators.  Spacecraft that have components with very low temperature requirements, like IR cameras, need something colder: liquid helium.  DeWar flasks contain liquid helium, which is produced by putting helium gas under pressure until it is forced into a liquid.  Condensing causes it to release its heat, cooling to very low temperature.  Old spacecraft released their helium as they used it, their lifespan limited by the size of the flasks.  As compressors have gotten smaller and more efficient new spacecraft can recycle helium into a fluid loop.  The system compresses the helium at the end of the loop to re liquify it, and the limit to its lifetime is the life of the compressor parts.
No matter which of these systems we use we have to do something with the heat once we've removed it from the systems.  We need to have some method of heat rejection.  Heat rejection is how we release heat from the vehicle into space, the opposite of environmental heating.
Again often on earth when we think about getting rid of heat we think of water.  Water evaporates taking the heat with it.  On a spacecraft we use a similar idea, called a flash evaporator.  We mentioned it before, one of the early method of using DeWar flasks.  Fluid moves past the heat source, taking heat with it.  Heated fluid is dumped overboard, removing heat from the spacecraft.  Though effective this obviously has a limited lifespan, and is primarily used to remove excessive heat for short periods of operation, not as a long-term method of heat rejection.
Just as the only way heat can enter the spacecraft is radiation, so too the only way a spacecraft can release heat is through radiation.  We use radiators to reject heat from a vehicle into space.  Radiators are just panels, either on the skin of the spacecraft or out from it, with large surface area exposed to space.  Most of our internal thermal control systems remove heat from a hot source to a cold source, the radiator is that cold source.  They can be the opposite end of the heat sink, heat pipe, or fluid loop.
Radiators have to emit heat, but we don't want them to absorb it.  We want our radiator to have a low absorptivity to emissivity ratio.  A way to improve that ratio is the use of a second surface reflector.  This is a two layer radiator.  The top layer is a high emissivity, but is also transparent.  The lower layer has a high reflectivity.  Incoming radiation is reflected by the second layer and passes back out.  Heat is passed from the second layer to the outer one by conduction, in essence we get the emissivity of one layer and the reflectivity of the other.
Now, radiators are good for removing heat from the spacecraft.  Sometimes, though, we don't want them to.  After all the heat coming in changes, sometimes we might not want to radiate heat.  Louvers allow us to control the emissivity of the radiator.  Louvers are rotating vanes that can be open or closed.  When open they expose a radiator below them to open space, when closed they reflect that heat back into the spacecraft.  Open louvers make a radiator with high emissivity, closed louvers make one with low emissivity.  If the louvers are closed the spacecraft retains more heat, if they are open it emits more heat.
So we can remove heat from a component and we can reject heat into space.  But we don't just have one component in a spacecraft, we have many.  And each has a different operating range, and each produces a different amount of heat.  Fluid loops don't just go from hot reservoir to cold reservoir, there are many heat sources.  So we start with the component that has the lowest operating regime and we take the fluid there first, when it is coldest.  We move fluid from place to place in order of how much heat it must remove.  If we have a component that must be heated in order to work we might add a heater to it, or we might add it after the working fluid has heated up.  If the fluid is warmer than the component then it will warm the component instead of cooling it.  Then the working fluid goes to a radiator and compressor and moves along the loop again.

For most spacecraft that is where environmental control ends.  The only important environmental factor is temperature.  However sometimes we don't just have electronic components and cameras on a spacecraft, sometimes we have people on board too.  Manned missions introduce a number of new challenges.
First the habitat, where the crew works and lives, must be in the proper temperature range.  The human comfort zone stretches from 15 to 20 degrees centigrade, though people can tolerate much greater extremes, most people enjoy a room temperature of right around 21 degrees.  The habitat also must provide all the necessary requirements for human life.
Let's consider a crew member a device which, in addition to doing work, converts input materials into output.  An average human being consumes 840 grams of oxygen, 3,520g of water, and 620g food each day.  Simultaneously he produces 1,000 grams of carbon dioxide, 2,280g of moisture, and 170g waste.  All these outputs must be removed or otherwise dealt with while we supply all the inputs.
Air is the most obvious.  For short duration flights we can often ignore food and water, waste, and other problems, but we can never ignore oxygen.  A person needs at least 14 kPa of oxygen (that is the pressure of pure oxygen, or contribution of oxygen to total pressure, partial pressure) to survive.  Below this, even in a pure oxygen environment, the body can't get enough and the person begins to asphyxiate.  At most a person can tolerate 48kPa of oxygen, any more leads to oxygen toxicity, a potentially lethal problem known to deep sea divers.
We can't use pure oxygen, however.  Oxygen doesn't just support life, it also supports fire.  The more oxygen is present the easier it becomes for fire to exist.  A pure oxygen environment contributed to the Apollo 1 fire, killing three astronauts during an early pad test.  On Earth we live in a low oxygen environment, only 20% is oxygen.  The rest is mainly nitrogen, so that's what we do in space.  Nitrogen is nonflammable and largely nonreactive, so we make our spacecraft atmosphere 80% nitrogen to reduce combustibility.
Some spacesuits use a pure oxygen feed, it reduces the pressure and makes them easier to move it.  However it also requires almost an hour of pre-breathing for astronauts before performing an extra-vehicular activity (EVA).  Using the same concentration as the habitat relieves this requirement, but reduces the astronaut's mobility in the suit.
For missions lasting more than a few hours we also have to worry about the crew getting food and water.  A human drinks 2 liters of water each day, and when we include things like washing and bathing that quickly increases to 20L.  Spent water can be recycled for washing, known as grey water, but drinking water must be fresh and clean.  Recall that fuel cells produce water as a by product, this is one possible source of clean water for the crew to drink.  Otherwise we have to carry it. 
Likewise we have to carry food.  Often astronaut food is freeze-dried to save space and to extend its shelf life.  With packaging the minimum amount of food can easily increase to 2kg per day per person that must be carried into orbit. 
Once the astronaut has completely finished with the food there is waste that must be dealt with.  At one time intimate contact devices were used.  This is a polite term for diapers and even more uncomfortable bags and tubes placed over or in the appropriate body parts.  Now suction toilets are used.  Suction doesn't work nearly as well as gravity and can be uncomfortable for all involved.  Typically the waste is jettisoned to burn up in the atmosphere once removed.  Sometimes it is collected and returned to Earth for analysis by flight medics.
That isn't the only form of waste produced by the human body.  People also produce moisture in the form of sweat and released by breath.  People produce heat.  These must be removed from the environment, as in space there's no reason for hot air to move as their is on Earth.  Fans are used to circulate air and force air through filters to remove hairs, skin cells, and moisture.
People also flood the air with carbon dioxide, which must be removed so they can continue to breathe.  CO2 scrubbers remove carbon dioxide from the air.  On the shuttle scrubbers known as LiOH canisters were used.  Lithium-hydroxide chemically reacts with carbon dioxide, removing it from the air.  After a while the canister becomes saturated and can no longer remove CO2 from the air, and the canister must be replaced.  The ISS uses the US made Carbon Dioxide Removal Assembly (CDRA) and the Russian made Vozdukh.  Both of these systems remove carbon dioxide from the air and vent it into space.
All the systems we've talked about just now are open loop systems.  In the beginning we input food, oxygen, and water; in the end we remove carbon dioxide and waste.  But we need to store or carry those inputs and the outputs are useless to us.  A closed source system would recycle the outputs back into new inputs.  There is a device here on Earth we're all familiar with that converts carbon dioxide into oxygen at the same time as it converts waste into food and filters water.  It's called a plant.  These amazing devices cannot yet be carried in space effectively and used as a life support mechanism.  It isn't possible yet to fully anticipate how complex an ecosystem needed so support large scale growing is, nor is there room in on the ISS, or any flight, for agriculture.  Though a simple and obvious solution, bioregnerative life support is still on the drawing board.

Electrical Power Systems

All the sensors, computers, and radio systems we talked about in the last few lessons require electrical power to work.  Since a spacecraft can't very well be plugged into the local power grid it must carry its own means of producing power on board.  We're going to look at some of the common ways that is done.

Solar arrays are the most common spacecraft power plant.  Solar cells produce electrical power when exposed to sunlight.  Two different semiconductor layers allow electrons to move when exposed to light.

Doping exposes otherwise pure crystal solids to defects, or gaps of other materials.  In the above image we see a p-doped structure with a layer of n-doped material over it.  The difference between the two is what the doping substance is.  The lower, p-doped layer produces "holes" of positive charge, which are filled by electrons from the n layer.  Eventually, at the boundary, enough electrons build up that no more can move through.  This creates a one-way barrier, negative on the p side and positive on the n side, called a p-n junction.  Mobile electrons on the p side are pulled by the positive side of this junction, once across they cannot move back.  This is also used to produce diodes, electrical components that allow current to flow in only one direction (p to n).
Photons strike electrons, knocking them free from their atoms and allowing them to move.  The freed electrons are drawn across the one-way barrier and move into the n layer.  This produces the negative lead, and the opposite side acts as a positive lead, hence the terms p and n layers.  Once a load is attached electrons want to move through the load from n to p.
So long as sunlight keeps impacting the cell they can keep moving.  The more light that hits the cell the more electrons are freed, and so the more current is produced.  Larger arrays mean more surface area for the sun to impact.  Area power density, J, is a statement of how much electrical power a solar cell can produce per unit area, given in mili-amps per square centimeter.  Most solar cells have area power density of 15 - 45 mA/cm2.
When no load is applied electrons build up until they can no longer move across the p-n junction.  This is the largest difference in charge between the positive and negative leads of the cell.  Open circuit voltage is the voltage difference that exists between the two leads at this point, the highest voltage the cell can possibly produce.
Conversely when there is an easy, no resistance, path between the two leads there is maximum electron flow between them.  Short circuit current is the highest current the cell can produce, when there is no resistance between the positive and negative poles.
When electrons move freely there is no voltage between the two poles, when electrons do not move at all there is no current.  For any real load the voltage and current will be less than these two maximum values but greater than zero.

Electrical power is the product of voltage and current, so during open or short circuit there is no power.  Somewhere between the two is peak power, some combination of voltage and current that produces maximum power output.  Peak power occurs at the "knee" of this graph.  The ratio of peak power to the product of open-circuit voltage and closed-circuit current (the upper limit of power production) is called fill factor.  A high fill factor means there is little resistance inside the solar cell itself, which means higher electrical efficiency.
Several factors can reduce the power output of a cell.  Over time radiation increases the internal series resistance of the cell, which degrades the fill factor.  Since the solar cell degrades over time when preparing for a space mission we can't just consider the power production at the beginning, we have to consider the end of life (EOL) production.  The degradation ratio, ratio of end of life to beginning of life (BOL) power production is usually about 0.7 to 0.9.
Temperature also reduces power efficiency.  Most solar cells give their efficiency at 28 degrees Celsius.  As temperature increases efficiency decreases at a rate of 0.025% - 0.075% per degree C.  Since spacecraft temperatures can vary dramatically this is a serious concern.
Area power density of a solar array assumes direct sunlight.  If the sunlight comes in at an angle less power is produced.  At a high angle of incidence no power at all is produced.  The rate of electrical power produced by a cell based on solar power incident:

Solar conversion efficiency, seen above, is the ratio of electrical power out to solar power in given best case conditions.  A single p-n layer cell has a theoretical maximum efficiency of 33.7%.  Multiple layers can improve the efficiency.  For an infinitely thick cell the theoretical maximum is 86%, called the Shockley-Queisser limit.  Commercial cells are beginning to approach the 33.7% limit.
A solar array is a collection of individual solar cells linked together to produce more power.  Connecting them in series (the negative lead of one attached to the positive lead of the next) produces higher voltage.  Connecting them in parallel (positive leads all connecting one end of the load, negative leads all connecting to the other) produces more current.
Remember, though, that a cell produces power based on its angle to the sun.  If there's no light on the cell it produces no power.  So how the cells are mounted on the vehicle matters. 
One method, called body mounting, is to simply cover the exterior of a (usually circular) spacecraft in solar cells.  Such a method is often combined with a dual-spin stabilization system.  Although body mounting ensures that the solar array produces the same power no matter where it points, it also means that much of the array isn't producing any power at all.  In fact on 1- pi-th (1/pi) of the array is producing electricity.  The rest is shaded or at two high an angle of incidence to contribute.  This means a lot of wasted mass on non-producing solar cells.
To reduce the mass of unneeded cells sun tracking systems are used.  Much like the sun-trackers we talked about for attitude control, these sensors follow the sun.  The sun tracker ensures the solar array is always pointing directly at the sun.  While this reduces the mass of arrays, is requires complex control and pointing systems and a large number of moving parts.  For small spacecraft this can be significantly higher than the extra mass of extra solar cells.

Solar cells have another limiting factor: the amount of sunlight available.  As a spacecraft moves closer to the sun the light becomes more intense and more power is produced.  Further away there is less power available.  Deep space probes exploring the outer planets cannot rely on solar power, another method is needed.  Radio-isotope thermal generators (RTG's) produce electricity using the decay of radioactive material.  They are used any time a long mission cannot rely on solar power, whether because of distance or because it is on the dark side of a body.
Radioactive materials decay, or break down, over time.  During decay a heavy atomic nucleus breaks down into two or more lighter ones.  The lighter nuclei move off with some energy.  As decay products collide they produce heat.  RTG's convert that heat into electricity.
RTG's use thermoelectric generators to convert heat into electricity.  A thermoelectric generator uses p and n doped semiconductors.  When heat is applied to an n-doped semiconductor it pushes electrons toward the cool side.  When the same thing is done to a p-doped semiconductor the positive "holes" move.  Linking the two together creates a voltage difference between the positive and negative sides:

A string of such connections in series increases the voltage produced.  They can also be connected in parallel to produce additional current, although this is rarely done in a single generator.  Rather multiple modules are connected together in series or parallel.
For an RTG the heat source above is a radioactive isotope.  During the mission the isotope decays into stable nuclei, which produce no heat.  Because of this there is a steady decline in the power output of the RTG, less radio-isotope means less heat means less power.  Fuel degradation is the primary reason for EOL power production being less that BOL for RTG's, although radiation damage of the thermoelectric generator is also a concern.
The General Purpose Heat Source was developed by NASA as the standard RTG heat source.  It uses plutonium-238 as fuel contained in iridium cladding for safety.  Plutonium-238 (not to be confused with plutonium-239, which is used in nuclear weapons and fission reactors) has a half-life of 88 years, which ensures that it decays at a rate that produces enough heat to produce power, but not so fast that it is spent before the mission ends.  Each GPHS module produces 250W of thermal power BOL.
Nowhere in the GPHS or in a thermoelectric generator are any moving parts, producing very high reliability since there's nothing to wear out.  However they aren't terribly efficient, producing almost nine times more heat than electricity.  Some proposals call, instead of thermoelectric generators, Stirling cycle engines for higher efficiency, but this is still under development.

Both solar cells and RTG's are used for very long missions.  Neither produces very high specific power, electrical power produced per unit weight (kW/kg).  To produce a large amount of power requires a very large mass of solar cells or RTG's.  Many short missions have high power needs which must be accommodated without requiring much mass or space.  Current space rated solar cells produce 77 W/kg, the Space Shuttle's fuel cells produced closer to 98 W/kg and required less space and less hardware.
Fuel cells are a type of electrochemical cell, similar to a battery (which we'll discuss next), which produce electricity from a chemical reaction.  In a fuel cell two different chemicals are used, fuel and oxidizer, just like the chemical rockets we looked at before. 

Fuel, usually gaseous hydrogen, passes through the anode into an electrolyte.  In the electrolyte it ionizes, it separates into hydrogen ions and electrons.  Electrons are forced to the anode while positive ions are allowed to pass through the electrolyte.  Electrons pass through the load and back to the cathode.  Here oxidizer, usually gaseous oxygen, reacts with hydrogen ions and electrons to produce water vapor, which is removed.
An individual fuel cell produces about 0.8 - 1.2 volts.  Multiple fuel cells connected in series to provide higher voltage is called a stack.  Fuel cell stacks are used to provide anywhere from 1-5 kW for periods of up to a month.  Since the fuel cell stack is designed to produce a specific amount of power they are easy to scale up.  Increase the size of the fuel tanks and you increase the operating life, no additional fuel cells or hardware is necessary.  Fuel cells also use the same reactants as many rocket systems, no extra tanks are necessary.  For manned missions fuel cells even produce drinkable water.  They were used on every manned NASA mission from Gemini to the Shuttle.

Fuel cells are one form of electrochemical cell.  Another that we're all familiar with is a battery.  In a battery two different metals are immersed in electrolyte, an acid. 

Anode and acid react to form positively charge ions and electrons, cathode and acid react to form negatively charged ions leaving positive "holes" in the cathode.  Ions move freely through the electrolyte from one electrode to the other.  The presence of electrons on one electrode and positive holes at the other produces a voltage difference.  When a load is connected between them the electrons move across it, creating current.  As the battery discharges the formation of ions reduces the acid content of the electrolyte, degrading the production of additional ions.
In primary batteries this flow is one way, they produce power only.  Secondary, or rechargeable, batteries can be reverse.  When a current is applied to the battery the poles are reversed and ions flow in the opposite direction, reforming acid.
Over time, even if no load is applied, there is some ionic motion which reduces the stored energy in a battery.  Self discharge limits the available charge in batteries that are not occasionally recharged, and can be an important factor in reducing the useful life of primary batteries.  It is not, however, usually a significant factor in the useful life of secondary batteries.
The lifetime of primary batteries is based on how much charge they contain.  Secondary batteries are measured by their cycle life, the number of charge-discharge cycles they can go through before they break down too much to be recharged again.
To extend the cycle life of a battery we can reduce the depth of discharge.  Depth of discharge is simply the percentage of stored charge that is expended during one discharge cycle.  High depth of discharge reduces the cycle life.  Because the most important factor for secondary batteries is cycle life usually they are made with higher capacity than necessary, to reduce the depth of discharge.  This trend causes secondary batteries to have a lower specific power than primary batteries.
Primary batteries are used for very short discharge, low power requirement tasks.  They're used for launch vehicles and pyrotechnics, and on some small satellites with very short missions or low power requirements.  However the short lifetimes and low power production of primary batteries means that they are rarely used as the main source of power for a large satellite.
Secondary batteries, however, are present on almost all spacecraft.  They're used as a supplement to the main power source.  Secondary batteries provide power when solar arrays fall into eclipse, when the earth blocks their view of the sun.  How long and how often a spacecraft is eclipsed depends on its altitude.  Low orbiting satellites have very short charge-discharge cycles.  Secondary batteries also provide power stability.  RTG's and solar arrays don't provide the exact same current constantly during use, it fluctuates slightly during use.  Batteries are used to absorb power spikes and substitute when power is low.

Another method of power storage is a flywheel.  Where batteries store power in the form of chemical energy, flywheels store power as rotational energy.  A flywheel is a rotating mass connected to a motor/generator.  It spins up to store power, and the generator taps that rotation to regain it.  Flywheel can double as momentum wheels, or two counter-rotating (rotating in opposite directions) flywheels can be used to prevent changing the vehicle's orientation.  Flywheels can potentially offer higher specific power than batteries, better efficiency, and prevent self-discharge losses.  However the presence of moving parts adds components that can break down and small frictional losses.

When power is stored some method is needed to connect the spacecraft loads, the energy storage, and the power source.  The power distribution system is responsible not just for connecting all the components, but also for power conditioning.  Power conditioning is the process of ensuring that the voltage and current is within the operating limits of the systems that require power.
When the power distribution system includes no active components it's called a direct energy transfer (DET) system.  DET systems use passive power controllers, either in series or parallel.  A shunt is a parallel controller, it bleeds off excess current.  A series controller controls voltage directly, but both accomplish the same task.
In an unregulated DET system the battery, shunt, and load are all connected in parallel.  In this kind of system the control is very simple.  If the voltage from the cell is greater than the voltage from the battery then it charges the battery.  If it is less then the battery discharges the difference.  In this kind of system the voltage can never be less than the voltage of the battery.  unregulated DET requires that most of the power conditioning is done at the loads themselves.  It's usually favored for systems with short charge-discharge cycles or on spacecraft with a small number of loads requiring a great deal of power.
Regulated DET systems have additional battery charge and discharge controllers.  These ensure that the battery only charges or discharges when called for.  Though more complex than unregulated systems they provide more flexibility and are preferred on most large spacecraft.
There are non-DET systems.  These systems have active controllers between the primary power source and the loads.  One method used on solar arrays is peak-power tracking.  Peak power tracking adjusts the voltage and current coming out of the solar cell so that the fill factor is always the highest it can be.  While this means the highest efficiency and power output, it means the voltage changes as the power available from the cell changes.  It speeds how fast batteries charge and reduces solar array area, but it increases complexity of the power distribution system.
When all systems accept the same input voltage and current the power conditioning is said to be centralized, only one power conditioning unit (PCU) is necessary.  Such a setup is ideal for unregulated DET systems.   However most off-the-shelf components have different requirements.  It's often easier to incorporate multiple PCU's at the loads, a decentralized system.

Command and Data Handling System

In the last lesson we looked at spacecraft communications, the command and telemetry system.  We saw how a spacecraft collects data from its various subsystems and how it relays information back to ground.  But we left out a very large part of that process, the computer system in the middle.  In this lesson we'll look at just that: spacecraft computer systems and how they're used.

This is a simple block diagram of the major components of a spacecraft computer system.  It's essentially the same as the personal computer you're looking at now.  At the center is the processor data bus, which transfers data between the various components.  There is also a spacecraft data bus, which performs the same task: it transfers data and instructions between the processor and the various spacecraft systems.  The two are connected by the bus interface, or bridge, which converts input into a form acceptable to the computer.  It acts as a bridge, hence the name, between the computer and the various systems.
The central processing unit (CPU) is what actually performs the tasks and carries out instructions.  It can perform a number of actions each second, defined by its clock rate, given in hertz (Hz).  CPU's also have an associated integer value, which defines the maximum value of integer it can handle.  An 8-bit processor can handle up to 8 binary digits (256 in decimal).  It also limits how many memory locations the CPU can access (again, for an 8-bit processor it is 256, that is it can keep track of at most 256 different variables).
Co-processors are additional processors added to assist the CPU by handling specific tasks.  The most common home example is a graphics co-processor, or graphics card.  This second processor handles all graphical tasks, reducing the load on the CPU.  Many muliprocessor (or multi-core) systems have extra processors that take up tasks as assigned by the CPU, rather than performing dedicated tasks.
Watchdogs are dedicated processors that monitor the CPU for faults.  A watchdog timer does this by looking for hang-ups, when the CPU has been locked into the same task for an unusually long time.  When it observes a hang-up the watchdog forces the system to reset.  It's necessary to have this on a spacecraft, since it must function with minimal human intervention.  There's no one around to hit the reset button, so we have a watchdog.
Memory comes in two main forms: random access memory (RAM) and read only memory (ROM).  RAM is volatile, it disappears when the system is shut off.  RAM is like short-term memory, it's where the computer stores what it's working on right now.  ROM is nonvolatile, it sticks around even when the computer is off.  It's like long-term memory, it's where the computer stores programs and instructions for later use.
Spacecraft can accumulate a great deal of data that must be stored until it can be transmitted.  Mass storage is a kind of large scale read only memory.  Magnetic disk drives (hard disk drives) are the more familiar mass storage device with magnetic disks that store information.  They are rapidly being replaced by solid state drives, which store information on collections of integrated circuits.  SSD's are smaller and denser, have no moving parts to generate heat or break down, and more durable.  The down side is that SSD's tend to consume more power when in use.
Input-Output (i/o) ports allow the bus to communicate with other devices.  They come in two varieties: serial and parallel.  Serial ports exchange one bit and a time, parallel ports exchange one word (a word being the number of bits the processor works with) at a time.  Since parallel ports must exchange on multiple wires (one per bit) serial architecture can have more connection.  Parallel can also become unweildly at high data rates.
Devices connected to i/o ports can be mapped (described to the processor) in two ways.  i/o mapping references a specific port location, separate from all other memory.  This dedicates that port to that particular function.  Memory mapping treats the port as a location in memory, which can be modified and altered through use.  Direct memory access allows devices to access memory without going through the CPU.  Rather than giving data to the CPU, which decides how to store it, the device places the data in memory for the CPU to access as necessary.

Faults occur when something goes wrong with the computer hardware.  Faults can be hard failures, they remain for the rest of the mission, or soft failures, they occur once.  One very important source of faults in spacecraft is ionizing radiation.  Radiation effects are cumulative, the total ionizing dose (TID) a component is exposed to limits its life span.  Over time the component breaks down, requiring more power to operate and slowing down its functions, until eventually it stops working entirely.
Because information is stored and transmitted in a computer by electrical charge ionizing radiation, which deposits or removes charge, can alter bits in memory.  This is known as a single event upset, and can be much more serious than it initially sounds.  A single flipped bit can convert a stored command into complete gibberish or, worse, another command.
Radiation induced noise can cause latch-up in CMOS (complementary metal oxide semiconductor) components.  Latch-up occurs when a path of low resistance forms over the semi-conductor, effectively a short circuit.  High power flow across it quickly destroys the component.
Faults prevention is possible through careful design of components, a process called radiation hardening.  Radiation hardened components are, however, expensive and bulky compared to their ordinary counterparts and little effort has been put into their development since the end of the Cold War.
Another method of prevention is shielding components.  Radiation loses energy, and thus capacity to do damage, as it passes through matter.  By covering sensitive components with high density materials the effect of radiation is diminished.  Shielding, however, requires excess mass be carried by the spacecraft and can interfere with cooling systems.
When something does go wrong fault detection becomes necessary.  One method is parity, using extra bits when storing data.  These extra bits record whether a word or collection of data is odd or even.  A common implementation of parity is called Hamming code.
Another method is triple modular redundancy.  Here three identical systems perform each task.  An error is quickly noted if one of the three disagrees, it can then be determined if it has a soft or hard failure.  Three isn't an upper limit, the Space Shuttle had four primary computers that performed each operation and a fifth that activated to resolve difference between the four.  Any error in one resulted in it being shutdown for the remainder of the mission.  While effective, redundancy requires significantly extra mass as it triples (or more) the number of components.
Multiple cores can be used as a form of redundancy.  In a master/slave (or watchdog) system two or more processors are used.  The slave processor performs all the same operations as the master, and checks the results. 
Instead of redundancy a watchdog can also look for specific signals that all is well.  An improper sequence detector has a checklist of results that it expects when operations are run.  If something doesn't occur, or occurs in the wrong order, it signals that a fault has occurred.
When a fault occurs the system has to repeat the last action to determine the nature of the fault.  Fault rollback records a rollback point each time control is transferred between tasks.  When faults occur it returns the system to what it was doing when the last task started.  This works best when the system transfers between many tasks quickly, since in that case the last task won't have been too long ago.

Older spacecraft used hardwired logic circuits, which perform specific operations based on their design.  Logic circuits have to be specifically designed based on their function, which cannot change during operation. 
Modern spacecraft use embedded software, electronically stored and implemented commands.  It's preferred because it's reprogrammable, programs can be modified and repaired by the ground crew during operation.  Programs can also be reused between systems and between missions.
Spacecraft software is designed in exactly the same way as the software on a personal computer.  Underlying everything is the operating system, which continuously runs to perform basic operations and schedule events.  When instructed, either by ground personnel or on a schedule, the OS executes programs, pre-written sets of instructions.
Computer programs are written using a programming language.  Programming languages work very much like spoke languages in that they convey information.  In this case they convey sets of instructions from the programmer to the computer.  Once written programs are compiled, or translated, from something a person can read to binary machine language the computer can read and implement.
Programs are built with functions, or subroutines.  A function is a self-contained collection of computer code that executes a specific task.  Functions are used for operations that must be performed many times or may be reused from one program to the next.  The computer knows what to do and when by following logical commands.
Computer logic is what allows the program to make decisions and act.  Logic falls into two basic types: selection and loops.  Selection allows a program to decide amongst a collection of alternatives.  The most basic form of selection is a simple if - then statement: if this, then that.  Additional options are added by an else statement: if this, then that, else this.  Loops tell a program to repeat an action.  It can be told to repeat the action while something is true, or until something happens.
That is essentially all the computer can do, respond to what happens based on what it has already been told to do.  An autonomous spacecraft must have its contingency procedures already written and must know when to implement them.  Remember: computer make decisions, they don't think.