Thursday, December 28, 2017

History of the Delta rocket Pt. 2: Delta-B to Delta-N

Last time we stopped our review at the first of the Delta rockets proper, the Delta-A.  The Delta-B was the next variant of the Delta family, and was very similar, with a lengthened second stage and an upgraded second stage engine, the AJ-10-118D. The Delta-A used the AJ-10-118.

The AJ-10-118D burns unsymmetrical dimethylhydrazine (UDMH) as a fuel, and inhibited red fuming nitric acid (IRFNA) as an oxidizer, instead of UDMH and white fuming nitric acid (WFNA), which is what the AJ-10-118 used. Variants of the AJ-10 have been used on the Apollo program, the Space Shuttle program, and are planned to be used on the Orion program.

Fuming nitric acid is more concentrated than concentrated nitric acid (>86% vs ~68%), and WFNA is nearly pure nitric acid. It makes lab gloves burst into flames.
WFNA and IRFNA are both hypergolic with UDMH. IRFNA has slightly higher performance than WFNA, however it is also considerably more dangerous, as in addition to being corrosive to almost everything, it is also more toxic and gives off nitrogen dioxide fumes. IRFNA has an inhibitor added to prevent it from being quite as corrosive. If you want to read more about this kind of thing, I can't recommend Ignition! highly enough.

Delta-B launched nine times, with one failure. The Delta-C increased the fairing size, and used an upgraded 3rd stage. It launched 13 times, with one failure.

The Delta-D, aka the Thrust Augmented Delta, added 3 strap-on Castor I solid boosters to the first stage. The Delta-E was known as the Thrust Augmented Improved Delta, with Castor 2 solid boosters, and increased the thrust of the first stage engine, the MB-3 (in this case, MB-3-III), which is part of the LR-79 family. Some sources say that the upgrade to the MB-3-III was on the Delta-D, but most say Delta-E. The second stage was made restartable, and was enlarged, along with the fairing. The third stage was changed again, and another third stage was available as an option, with which it was known as the Delta-E1.

Delta-F would have been similar to the Delta-E, but without the solid boosters, but was never built. Delta-G was a one-off, built for just two launches, Biosatellite 1 and 2, and lacking the third stage. Delta-H was similar to the Delta-G, but without the solid boosters, but was never built. Delta-I was never built, likely to avoid confusion with a possible future Delta One. Delta-J had yet another third stage, and launched just once. Delta-K was a design for a Delta with a liquid oxygen/liquid hydrogen upper stage, and was never built.

Delta-L introduced the Extended Long Tank first stage, which was longer, and not tapered. Delta-M and -N were very similar, but with different third stages. There were variants of the Delta-M and -N, known as Delta-M6 and -N6, which had six, rather than three solid boosters.

In 1972, Delta numbering systems changed from the old letter system to a four-digit numbering system. Next time, I'll cover everything under that system.

Wednesday, November 29, 2017

History of Delta rocket Pt. 1

I waited too long too start writing this post, so I'm going to divide it into multiple parts.

The history of the Delta rocket begins with the PGM-17 Thor, which was an intermediate range ballistic missile (IRBM) with a range of 2,400km. It was designed to be able to hit Moscow from a launch site in the UK, and first flew in 25 January 1957, seven months after Douglas Aircraft was contracted to build it.

The missile used an LR-79 engine for its first (and only) stage, along with two LR-101 Vernier engines, for roll control. Its propellants were LOX and RP-1.
However, the missile was modified many times between this and the Delta, including changing the engine, and I won't go over them all.

The engine of missile 101 failed immediately after lifting off the pad, and the rocket fell back to the pad and exploded. It was determined that the failure was due to debris in the engine, from a LOX filling line that crews dragged over a patch of sand.
Missile 102 was erroneously destroyed by range safety (footage later in the above video), missile 103 exploded four minutes before the planned launch, and 104 broke up due to an electronics failure.
105 finally succeeded, but the Thor missile remained unreliable for many flights, mostly due to its turbopump design, which was eventually fixed.
The only other notable event in the Thor's history was the launch Bluegill Prime, on July 26th 1962, as part of Operation Fishbowl, which was a series of upper atmosphere nuclear weapons tests.

The rocket exploded on the pad, destroying the nuclear warhead and contaminating the pad with plutonium.

Thor was modified with several upper stages for use as a orbital launch vehicle. This first was Thor-Able (Able is the name of the upper stage, but was named Able because it is the first modification of Thor, and the first in the allied military phonetic alphabet).

The next variant was not Thor-Bravo, but Thor-Agena. Agena was an already existing upper stage/satellite bus, with stabilization, communications, and power built in. When those were removed, it was known as an Ascent Agena. It used Bell 8048, 8081, and 8096 engines burning JP-4 and IRFNA, on Agena -A, -B, and -C respectively. Attitude control was provided by nitrogen-freon cold gas thrusters. 

Then the Thor-Ablestar was designed, which was the same as Thor-Able, but with a larger Able upper stage.

After this, the Thor-Delta was built. This is the first Delta rocket, which will be henceforth referred to as the Delta-A. The first stage used a Rocketdyne MB-3 engine, and the second stage (derived from the able upper stage) used an AJ-10-118 engine burning hydrazine and nitric acid. It first flew on May 13 1960, with a solid third stage. The first flight was a failure, but the second flight, in August, successfully launched NASA's first communications satellite, Echo 1A, ("A" due to the previous launch being a failure) into orbit.

Monday, October 30, 2017

ITS/BFR design updates

At the last IAC, Elon Musk presented the latest changes to the BFR.

Throughout this post, I will be referring to what was previously the ITS as the BFR, as that appears to be what SpaceX now refers to it as externally now. The interplanetary spaceship will be refered to as the BFS, and the booster will be called the BFB.

Some of the numbers were changed between the presentation and the PDF being posted. One of these changes was a factual error, about the internal cabin volume of the BFS, and the other was a change of the total BFB thrust, and the removal of the total BFR mass. For this post, I will mostly use the numbers from the presentation, as they are more complete, but I will note where the numbers differ or are wrong.

The reusable payload to LEO has been cut by about half, from 300t to 150t. The payload to Mars has gone from 450t to 150t, and the goal for the number of people per flight has gone from 100+ to 100.


The Raptor engine is further along in development than last time, and has been slightly downscaled to match the smaller BFR. This is because it is harder to make engines throttle to lower percentages of full thrust, which they would need to do to land safely, and still have redundancy.
"The engine thrust dropped roughly in proportion to the vehicle mass reduction from the first IAC talk" (source)

Elon Musk said in the presentation that the Isp had the potential to be increased by 5-10 seconds, and the chamber pressure by 50 bar, which would make the 2017 Raptor have the same stats as the 2016 one.

250bar vs 300bar
Deep throttle to 20%/ 20%


Exit diameter 2.4m/ expansion ratio 200
thrust 1,900kn/ 3,500kn
isp 375s/ 382s


exit dia 1.3m/ expansion ratio 40
Thrust (SL) 1,700kn/ (SL) 3,050kn
Isp (SL) 330s, (Vac) 356s/ Isp (SL) 334s, (Vac) ???


58m x 9m/ 77.5m x 12m

Mass is tricky, as the video of the presentation gives exact values, while the PDF of the slides give a different exact value for thrust, and no numbers for mass. For this post, I'll use the numbers described in the presentation. Dry and prop masses are based on ratios from the 2016 BFB.

Mass (dry, prop, wet, prop mass fraction[prop/wet])
275t, 6700t, 6975t, ~0.96
~126.75t, ~3,088.25t, 3215t, ~0.96

31/ 42
(Apparently the important scientific and fictional reasons weren't that important, however the PDF does not say how many engines it has, which means that it is likely changing.)

This is weird. The talk showed a thrust of 5400 tons, which is equal to 48040.79kn, which is how much thrust 28 and a quarter SL raptors produce ASL. It's possible that this is a mistake, as there was also a mistake in the interior volume of the BFS, but it's also possible that it was from a different version of raptor.
In the PDF, this was changed to 52,700kn, which is exactly what 31 SL raptors produce ASL, lending weight to the theory that this was a typo. I'm going to assume that it was a typo.

52,700kn/ 128mn (SL) 138mn (Vac)

11288.2m/s / ???
10463.8m/s / 10590.47m/s
(note: this seems odd, as the thrust and fuel mass are roughly double in the 2016 BFR, but remember that Delta-V is a function of propellant ratio and Isp, and I assume the propellant ratio to be the same. Basically, scaling up or down a spaceship design just changes how much extra payload affects Delta-V)


The spaceship's largest change, aside from size, is in shape. The 2016 IAC spaceship had a much more complex body shape, with 3 fins that blended into the overall shape of the ship. The 2017 ship is much simpler, with a simple cylindrical shape and a single set of delta wings. This was done to avoid building a "box in a box" (source)

The BFS now refuels via a connection at the end of the ship, rather than at the side.

48m x 9m/ 49.5m x 17m (max) 12m (base)

Using the masses from the interplanetary ship in the 2016 numbers.
(2017)(2016)(dry, prop, wet, prop mass fraction[prop/wet])
85t, 1,100t, 1,185, ~.93 / 150t, 1,950t, 2100t, ~0.93
(In the presentation, Elon Musk said that the current design has the dry mass as 75t, but mass growth would likely occur)

(All vac)
9689.63m/s / 9886.28 m/s

4 Vac, 2 SL/ 6 Vac, 3 SL

7,600kn/ 21,000kn

3,400kn/ 9,150kn

However, Elon Musk said here that a 3rd medium area Raptor was added to the BFS since the IAC. I don't know what form this would take, or how this would fit on the BFS.

Sunday, July 30, 2017

Interesting space videos

Today I'm taking a break from the alternate ITS missions series, as I'm not yet finished with it. So I've going to post a couple of videos about spaceflight I've found.

First, a video from Scott Manley, about reused spacecraft.

This video is a timelapse of the night sky, but with the motion of the Earth corrected, showing how the Earth moves relative to the stars. Bonus video.

This shows six daylight Falcon 9 landings, timed so they all land at the same time.

And this, from Tom Scott, is a video on how spaceflight hardware and zero-g experiments that do not need enough time in zero-g to require a parabolic flight are tested.

Sunday, July 2, 2017

Alternate ITS missions part 3

This post got too long again, so I'm splitting it into yet another section. This will cover how to calculate interplanetary trajectories.

All orbits are conic sections. circles, ellipses, parabolas, or hyperbolas. Circles and ellipses are the orbits that a spacecraft is in when it is in a stable orbit around a planet. Parabolas are when the spacecraft is going at exactly escape velocity, and hyperbolas are when the spacecraft is going faster than escape velocity.

For ellipses (assuming circles are a type of ellipse) the planet the spacecraft is orbiting (or the primary) is always at one of the two foci. The major and minor axes are the longest and the shortest lines that can be drawn through the ellipse, respectively. The semi-major or minor axes are half of the major or minor axes.

The periapsis is the lowest point on an orbit, and the apoapsis is the highest. They are always 180 degrees apart. The closer to the periapsis, the faster the satellite travels, and the closer to apoapsis, the slower.

A quick aside on nodes: nodes are the points at which the orbit passes the plane of the equator or ecliptic. The ascending node is the one the spacecraft passes travelling from south to north, and the descending node is the one the spacecraft passes travelling from north to south.

Eccentricity is the distance between the foci divided by the length of the major axis (always between 0 and 1 except for hyperbolas), and is essentially a measure of how "squished" the ellipse is. 0 is a circle, 1 is parabola, > 1 is a hyperbola.

Inclination is how tilted the orbit is from the equatorial plane (or the ecliptic) of the primary. An inclination of less than 90 degrees indicates a orbit in the same direction as the spin of the primary, or prograde, an inclination of greater than 90 degrees means that the orbit is in the opposite direction of the spin, or retrograde.

I'm ignoring how to calculate these, as that would make this post much longer.

So, let's imagine you want to go from one circular orbit to another. What is the most efficient way of doing this? The answer is a Hohmann transfer.

Image from Leafnode

A Hohmann transfer is the most efficient way to go from a coplanar orbit to another. The image shows a transfer with two circular orbits, however it is the most efficient for any two coplanar orbits where one cannot be reached from the other with only a single burn. For example, in the image, 1 and 3 require a hohmann transfer, while 1 and 2 or 2 and 3 only need a single burn.

The Delta-V required for a hohmann transfer is the sum of the Delta-V required for the burn to go from 1 to 2, and the Delta-V required to go from 2 to 3. It is the same in the opposite direction (3 to 1 = 3 to 2 + 2 to 1), and the burns are the same, except with the spacecraft burning its engines in the opposite direction.

The transfer orbit between two planets is the same thing, with the transfer orbit going from one planet to another, except for the fact that you start in orbit around one of those planets.

Transfer windows are times when the orbits of the planets align such that a spacecraft on a transfer orbit reaches apoapsis (or periapsis) at the same time as the destination planet moves through the part of the orbit that intersects the transfer orbit.

It isn't very important to calculate transfer windows for this post, as they do occur regularly for most planets.

Delta-V for an interplanetary transfer is less than you might expect, as the Oberth effect comes into play on the escape burn. The Oberth effect is, in short, an effect that makes Delta-V count for more when near a large mass. It is why the Delta-V required for a hohmann transfer to GEO is more than the Delta-V required for escape velocity.

The Oberth effect is the effect of the extra velocity of a circular orbit around earth has compared to a circular orbit around the sun at earth's altitude. As the satellite orbits around the earth, it traces out a spiral pattern relative to the sun. When it is orbiting in the same direction around the earth that the earth is orbiting around the sun, the velocity relative to the sun is increased. It isn't as simple as adding the velocities, but we'll get into that later.

On arrival at the destination planet, the Delta-V can be further reduced with aerobraking or aerocapture. This is where the you use the atmosphere of the planet you are arriving at to slow down or speed up without using any fuel. This only works on planets with atmospheres, though, and you cannot capture into a circular orbit without a burn, although it is much less with most of the velocity bled off already.

A very useful equation is the Vis-visa equation:
v^2 = GM(2/r - 1/a)
Where v is the speed of the spacecraft at any point in a orbit, G is the gravitational constant, M is the mass of the primary, r is the distance between the spacecraft and the primary at the point at which you wish to know velocity, and a is the semi-major axis.

The calculations for hohmann transfer orbits assume instantaneous impulses, which lessens delta-v requirements. To compensate, I will multiply the estimate by 1.1.

To calculate Delta-V requirements, you simply need to get the difference in velocities between the orbits at the correct points. This is why the Vis-visa equation is so useful.

We apply this to a hohmann transfer from one planet to another, getting the velocities of the transfer at apoapsis and periapsis. Then, we calculate the velocity of the starting orbit, and calculate the required Delta-V to go to the transfer orbit, accounting for the Oberth effect.

Reversing the rocket equation will then let us calculate payload.

Earth to Venus example

Earth has a mass of 5.9723 * 10^24 kg. It has a semi-major axis of 149.6 * 10^9 meters. Its orbital radius goes from 147.09 * 10^9 m to 152.1 * 10^9 m.

Venus has a mass of 4.8675 * 10^24 kg. It has a semi-major axis of 108.21 * 10^9 m. Its orbital radius goes from 107.48 * 10^9 m to 108.94 * 10^9 m.

The sun has a mass of 1988500 * 10^24 kg.

A transfer orbit between them depends on the position of the planets at departure and arrival. For this example, the position of Venus does not matter, as aerobraking will work at either of those speeds.

Earth should be at apoapsis, because in this case we want to lose, rather than gain, velocity as Venus is in a lower orbit than Earth.

Let's begin. First, the velocity of Earth around the Sun at apoapsis. We don't really need to calculate this, as it is well known: 29290 m/s.

The transfer orbit is an elliptical orbit with its apoapsis at Earth, and its periapsis at Venus. This gives a semi-major axis of:
(Radius of Earth's orbit at departure + Radius of Venus' orbit at arrival)/2.

Or (152.1 * 10^9 m + 108.94 * 10^9 m)/2 = 130.52 * 10^9 m.

Velocity at apoapsis:
v^2 = 1.327 * 10^20 (2/152.1 * 10^9 - 1/130.52 * 10^9)
v = 26985 m/s

At periapsis:
v^2 = 1.327 * 10^20 (2/108.94 * 10^9 - 1/130.52 * 10^9)
v = 37676 m/s

The velocity of the spaceship in Earth orbit can be calculated from the altitude, and the altitude is as low as possible to take maximum advantage of the Oberth effect. I'm going to guess around 200 km.

v^2 = 3.986 * 10^14 (2/6578000 - 1/6578000)
v = 7784 m/s.

To figure out how much of an advantage the Oberth effect gives us, we use something called hyperbolic excess velocity. This is how much extra velocity the spaceship will have when it reaches infinite distance on a hyperbolic escape orbit. If you reach exactly escape velocity, HEV will be zero. The desired HEV is equal to the orbital velocity of the primary - the velocity of the transfer orbit at the point where it begins so when the spacecraft escapes the primary's gravity, the remaining velocity puts it on the transfer orbit.

The equation for HEV is HEV^2 = v^2 - ve^2

Our desired HEV is 29290 - 26985 = 2305.

Rearranging the equation:
v^2 =2305^2 + 11008^2.
(as escape velocity is equal to the velocity of a circular orbit at that altitude times 2)

So v = 11246 m/s.

11246 - 7784 = 3462.

Here you can see the clear advantage of this over a escape burn (3224) + a burn outside of Earth's sphere of influence to put the spacecraft on a transfer orbit (2305), which would be 2067 m/s more.

Venus is inclined 3.39 degrees off of Earth, however this amounts to at most a few m/s if it is corrected for at the very start of the transfer. For this kind of loose approximation, we can ignore this.

So, multiplying by 1.1, we have a Delta-V of 3808 m/s.

Next post, we will calculate transfers for other missions, including approximate landing Delta-V, and use the reverse rocket equation to figure out how much payload can be carried to them.

Wednesday, May 31, 2017

Alternate ITS missions part 2

Unfortunately, this post will have to be broken into three parts, since I started looking into how life support worked, and the post got really long. This part will mostly be about life support, and the next one will have details on actual missions.

There are two primary ways to support humans in space with oxygen, food, and water: ecological, and chemical. Ecological uses plants to reprocess human waste, carbon dioxide, and trace nutrients into food, and humans to process food and oxygen into what the plants need, making a closed loop, as long as you have sunlight.
The largest problem with this is that the amount of plants required to support one person's oxygen needs would produce about half the required food for that person. If you have enough plants to provide enough food, half of them die off due to lack of carbon dioxide.

Chemical life support uses chemical reactors to crack carbon dioxide into oxygen, and extract usable water from sweat and urine. The problem with this is that you must pack all the food you will need.

Storage of oxygen is simple, as you can use boiloff from the main oxidizer tanks. 

The air on Earth at sea level is at 101kpa. The atmosphere is 21% oxygen, so the partial pressure of oxygen is 21.21kpa. 

Partial pressure, in a mix of gases, is how much pressure one of the gases would be under if all other gases were removed.
The minimum safe partial pressure of oxygen is 16kpa, and the maximum is somewhere between 50 and 100 kpa for short periods of time (a few hours). For long term breathing atmosphere, you want to stay under 50 kpa, ideally at about 21 kpa. 

A high pressure pure oxygen environment is very dangerous because of fire, the Apollo 1 fire happened in a pure oxygen environment at 115kpa, this is what wood burning in a 50 kpa partial pressure environment looks like: After that all US manned spacecraft used a 79% nitrogen/21% oxygen environment at about 101kpa.

So why was the Apollo 1 command module pressurized to 115 kpa? Because it was designed to hold pressure in, not out. During launch, the pressure would have gradually been reduced to 34 kpa. I don't know whether the ITS will use a nitrogen/oxygen mix, or pure oxygen which is always kept at safe pressures, but either way it will take about 0.85kg/day of oxygen, as nitrogen is not consumed when inhaled.

A larger problem than oxygen is carbon dioxide. Every astronaut exhales about 1 kg of it a day, and if it rises above 0.5% concentration it can become a serious danger. However, 0.273 kg of that is carbon, so most of it can be breathed again if it is cracked back into carbon and oxygen. 

There are two options for dealing with carbon dioxide, cracking and scrubbing. Scrubbing uses a catalyst, which removes carbon dioxide from the air, producing water. It can be cleaned and used again by blowing hot air through it for 10 hours, however, you lose all that oxygen locked in the carbon dioxide.

A chemical reactor would use the sabatier reaction, like the ISS. 
CO2 + 4H2 → CH4 + 2H2O
Exhaled CO2 would be processed with hydrogen to produce CH4 (methane) and water. The water can be electrolyzed into H2 and O2, and the CH4 can be pyrolized into C and H2. The hydrogen outputs can be fed back into the sabatier reactor, closing the loop. It would probably not be perfectly efficient, but since you only need about 180 grams of H2 for every kg or CO2 processed, even at 90% efficiency you would only need to add 18 grams of hydrogen a day.

However, every crew member exhales 1 kg of carbon dioxide/day, but only 0.727 kg of that is oxygen. That means that you have to add 108 grams of oxygen every day.

Astronauts need to drink about 3.9 kg of water every day, some mixed in with their food, if it isn't freeze-dried, and about 26 kg/day of water for personal hygiene. Most of the waste water (grey water, human waste, sweat) can be distilled and filtered. I estimate that about 0.1 kg of water would be lost in recovery every day.

Astronauts on the ISS eat about 2.5 kg of food every day, which consists of a mix of different kinds of food, mostly freeze-dried or otherwise stabilized. Some of it is fresh.

For the purposes of this, we can assume that all non-reusable waste (packaging, filters, solid human waste) is non-existent, as it can be dumped overboard between burns. If you wished to prevent the build up of trash in solar orbit you could dump it only while on a collision course with a planet, however that would likely only be a problem on trips that would be taken frequently.

Weights required:
5 tons for storage, processing equipment, etc.100 grams of water x mission length in days18 grams of H2 x mission length in days108 grams of oxygen x mission length in days x crew members2.5 kg of food x mission length in days x crew members

In a CELSS (Closed (or Controlled) ecological life support system), rather than using chemical reactors, the life support loop is fully closed, like Earth's ecosystem. 

If you could get a CELSS working properly, it has the potential to be much more mass efficient than chemical life support system, with the crossover point for efficiency at around 2 years, according to Rocketpunk Manifesto. However, in their current state, they have some problems. For many plants, the amount of plants needed to provide food for astronauts is different for the amount need to provide oxygen, so the imbalance would cause an excess of oxygen, which would kill some of the plants, and then there wouldn't be enough food.

One of the most promising crops is Spirulina, a type of blue-green algae that, Marshall T. Savage claims in The Millennial Project, could close the life support loop almost fully with only 6 liters of algae per person (about 6.6 kg). (Note: I haven't read that book, this is taken from Atomic Rockets)
Even if you needed twice that much per person it would be very impressive.

For the purposes of this analysis, however, I will stick to existing methods of food production, as the goal is to use the fact that the ITS would be flying regularly to study a mission that could be much cheaper than a spacecraft designed only for the purpose of that mission. 

In the next part we will finally look at the actual mission profiles.

Sunday, April 30, 2017

Alternate ITS missions part 1

The Interplanetary Transport System is a spacecraft concept developed by SpaceX and designed to colonize Mars. While it is designed for that specific goal, it could feasibly be adapted to other missions. This post, part 1, talks about the performance of the ITS, and the second part will talk about what missions it is capable of.

The ITS has two configurations, the ship, and the tanker. The ship is designed to carry 450t or 100 people to Mars, while the tanker is designed to refuel the ship in Earth orbit.

Let's take a closer look at the performance.

First of all, the maximum delta-v.
Using the Tsiolkovsky rocket equation (DV = Ve*ln(minitial/mfinal)) like so:

DV =  3,747.42*ln(2100/150) = 9889 m/s

This is an impressively high delta-v, however it assumes no payload, with the standard ITS configuration (not the tanker).

With the tanker configuration:

DV = 3,747.42*ln(2590/90)  = 12589 m/s

This also assumes no payload.

You can calculate how much delta-v you need for transfers between planets with a porkchop plot, or you could just use a handy delta-v map of the solar system that assumes reasonably efficient transfers.

(Chart from Atomic Rockets)

We can reverse the rocket equation to find out how much payload we can carry, given a specific required delta-v. Wolfram|alpha can do this for us easily, and the answer is mfinal = minitial*-DV/VE
= Euler's constant, or ~2.71.

Payload equals mfinal - empty mass
Empty mass differs between the tanker and the standard ITS configuration by 60t. It's hard to tell what of that is necessary for a much smaller crew, as might be used on these missions. A safe bet is probably around 120t, halfway between the two numbers.

Propellant mass also differs, with 1950t for the ship and 2500t for the tanker. In this case, to keep costs low, the ship value of 1950t would be used, as there is no space for crew in the tanker, and changing the size and shape of the composite tanks is expensive.

Note that while SpaceX uses a value of 6km/s from low earth orbit (LEO) to trans-Mars injection (TMI), more efficient (but slower) transfers use on the order of 3-4km/s. Presumably, this is done to minimize life support costs; both consumables and radiation shielding. However, if you're sending 5 people instead of 100, consumable costs per day only scale 1/20 as fast per unit time, so almost any other missions would sacrifice travel speed for payload.

The ITS has 3 sea level Raptor engines, and 6 vacuum Raptors. Basically, in a vacuum you could use the sea level engines in addition to the vacuum engines, but it would just hurt our delta-v. In an atmosphere, you can't use the vacuum engines due to flow separation, which happens when your engine is overexpanded. (flow separation is an extreme example of what causes the space shuttle main engines to wobble as they start up, as they have to be efficient throughout the entire atmosphere:

So the ITS has different thrust-to-weight ratios (TWR) for different situations.

3*SL Raptor = 9150kn =  933 metric tons-force = empty TWR of 7.775
= loaded TWR of 0.45

6*VAC Raptor = 21000kn =  2141 metric tons-force = empty TWR of 17.84
= loaded TWR of 1.03

Max possible thrust in VAC = 3074 metric tons-force = empty TWR of 25.62
= loaded TWR of 1.49

These are all assuming you're on Earth, with no payload.

The ITS can withstand 10-15gs, and a minimum of 1,700 degrees C.

In the next part we will go over what missions the ITS could perform now that we have determined its performance.

Friday, March 31, 2017


This is not the Space Shuttle. This is Buran. Buran was a soviet project from 1974, to build a craft that could counter a perceived military threat from the US Space Shuttle. It was believed by the designers of Buran that the shuttle's entire purpose was military due to its large cargo capacity.

(Left to right: Soyuz, Space Shuttle, Buran)

Surprisingly, it appears that no espionage was involved in the development of Buran other than external photos, as there are are many differences between the two spacecraft.

Most notably, it did not have an external tank which had no engines on it like the space shuttle, rather it launched piggyback on a Energia rocket which had its own engines. Also, rather than using two solid strap-on boosters, it had four liquid oxygen-kerosene strap-on boosters.

It also used two liquid oxygen-kerosene engines for maneuvers in orbit, rather than MMH/N2O4 engines.

Buran made its first and last flight in 15 Novemeber 1988. It was entirely unmanned for the entire flight, and worked perfectly, despite landing in crosswinds of 38 mph (61.2 kph). After that the project was put on hold, and finally cancelled in 1993. The only Buran to have flown, OK-1K1, was destroyed in 2002 when the roof of the hangar it was stored in collapsed.

There were four other flight articles (craft that were built to fly), and a handful of test articles for various systems. Wikipedia has a full list here.

Ralph Mirebs took pictures of two of the remaining flight articles, which you can see here.

Tuesday, February 28, 2017

Nuclear rockets

One of the most important measures of rocket engines is specific impulse, which is just how efficiently a engine burns fuel. It is almost the same thing as exhaust velocity, which is how fast the exhaust stream is going when it leaves the engine.

Most rockets use chemical engines which combine a fuel and an oxidizer to accelerate the exhaust products and provide thrust. Isps of chemical rockets max out at about 450 seconds. Nuclear thermal rockets get Isp from 900 seconds on up. This is because their exhaust velocity is not limited by the energy of a chemical reaction, as they use the heat from a nuclear reaction to accelerate hydrogen by expanding it inside the core of a reactor.

This is the simplest type of nuclear thermal propulsion, and NASA did some tests on it in the '50s and '60s. The problem is that it's performance is limited by the melting point of the reactor. There are a couple of types of nuclear thermal propulsion that avoid this problem in various ways:

Twisted ribbon, pebble bed

By changing the shape of the fuel rods in the reactor, you can increase the heat transfer to the propellant. Twisted ribbon looks like this:

Pebble bed reactors are made out of "pebbles" of fuel that look like this;

Pulsed solid-core

This type of engine works by pulsing the power at which the reactor runs at from its normal power level to far beyond what the materials that it is made could withstand. As the temperatures are momentary, the cooling system can keep it from melting down. This is even better than it sounds, as neutrons generated while it is at maximum power will heat the propellant even further.

Liquid core

Solving the problem of preventing your reactor from melting down by designing it to run while molten! Outside-the-box thinking.
Problem include keeping your molten core from escaping with the propellent.

Isp 1,500 seconds.

Gas core

The logical extension of the liquid core design by allowing the uranium to turn not just liquid but to gas.

The problem of preventing the uranium plasma from escaping is exacerbated with gas core.

Isps 1,800 seconds to 7,000 seconds with advanced designs.

Nuclear lightbulb

It works like an easybake oven, except the lightbulb is a uranium plasma contained by a quartz vessel that doesn't melt because quartz is transparent to the majority of the radiation that is emitted, and the cake is hydrogen travelling at 20 km/s. Ages 10 and up.

Isps between 2,000 and 3,000 seconds.

Fission fragment

This skips a separate propellant entirely and uses the split atoms from the fission as propellant. The original concept used discs coated with the fuel, but a more efficient design uses ground-up fuel (around 100 nanometers) which are magnetically contained before they fission.

Isp 100,000 to 1,000,000 seconds.

Nuclear pulse

Uses nuclear bombs as propulsion, by dropping them behind a strong plate that dampens the shock and heat enough that anyone on board doesn't die. It apparently would work, and was even tested with conventional explosives. Development stopped mostly due to the nuclear test ban treaty and that this is best suited to lift spacecraft from the ground to orbit.
[This] is not nuts, [this] is super-nuts
-Richard Courant on viewing an nuclear pulse engine test

Isp 4,000 to 7,000 (with fusion bombs)

Nuclear salt water

This is one of the most crazy concepts for a rocket engine ever made. Its fuel is 20% enriched uranium mixed at 2% with water. It is stored in tubes coated with some kind of moderator which prevents it from reaching critical mass inside the tanks. When it is injected into the combustion chamber, the uranium reaches critical mass and begins a nuclear detonation. This can be sustained with more of the fuel being injected into the chamber. This has the advantage of the nuclear pulse engine's high efficiency, with continuous thrust and the ability to work at small scales.

Monday, January 23, 2017

Cause of AMOS-6 failure

On September 1 last year, a SpaceX Falcon 9 rocket exploded during fueling for a engine test that SpaceX does before every launch.

Since then, SpaceX has been analyzing the data that they gathered before and during the anomaly.  Recently they posted this report detailing what happened.

The report says that the second stage exploded due to heat or pressure coming from something called COPV.  COPV stands for composite overwrapped pressure vessel, which is a kind of tank that, rather than being made out of a strong material, uses tightly wrapped carbon fiber wrapped around a thin internal aluminum tank.  This is simply the lightest way of making high pressure tanks (in this case, around 350 bar, or ~5000 psi), and was used extensively on the space shuttle.

This is what a cutaway of one looks like:

Here's a video of one being wrapped:

This kind is filament wound, which is what the Falcon 9 has.  Braided ones are stronger and don't fail as explosively (not that it would have mattered) and are even more fun to watch being made (this is just carbon fiber being braided, not over a tank):

SpaceX uses these on their Falcon 9 in the second stage to hold helium that is used to fill up the tank as the fuel is pumped out, as the tank is airtight, it would crumple if pressure was not maintained.

There are three tanks in the second stage, visible in this image:

Here's a size comparison of one of those which survived reentry:

So, the report says that liquid oxygen got between the liner and the overwrap.  Normally this wouldn't be a problem, but somehow a buckle formed in the liner, like the way you can pop in the side of a plastic water bottle.  When the liquid oxygen pooled in that buckle, it may have also frozen, because SpaceX uses liquid oxygen that is not just liquid, but much colder and closer to solid oxygen (the kerosene fuel is also much colder than usual), which increases the amount of propellant that can be stored in a tank of the same size, and the performance of the engines, which is improved because more propellant can flow into the engine with pipes of the same size.

SpaceX will first change the fueling procedure, which worked perfectly on Iridium NEXT, and later they will change the COPV design.

Some articles have said that SpaceX's failures have been caused by a more silicone valley iteration based mentality, however I think that they have more to do with the fact that SpaceX is developing the first new rocket technology since the '70s.  If you look at failure rates from rockets from that time, you see even higher failure rates.  The Falcon 1 (the first SpaceX rocket) had a failure rate of 60% (3 out of 5 launches over 3 years) and the Falcon 9 has a 6.897% launch failure rate (1 out of 28, over 7 years)  not including partial failures.  The AMOS-6 failure is counted here despite not being a launch failure, as the payload was destroyed.

The total failure rate for all Falcon rockets is 14.71%, 5 out of 34, over 11 years.
Now, compare that to the Atlas ICBM variant D which flew 67 times in its ICBM (non Mercury-Atlas) configuration over 4 years (1959-1962), and had a 35.82% failure rate.  In Febuary of 1962, after 68 flights with 36.76% or 25 of those resulting in failure, NASA actually began launching this rocket with humans aboard.  This isn't as crazy as it sounds, as the launch escape system in use at the time had never failed.

Most of this information came from this image:

So, in short, I think that SpaceX's higher failure rate are caused by flying the most recently developed rocket.  There's actually a term for how new technologies fail more often while new, called the bathtub curve.