Inside the LEO Doghouse: The Art of Expander Cycle Engines

If you go back several generations on my mother’s side of the family, you will find a famous artist named Charles Frederick Kimball.  Also on my mother’s side of the family, in a different branch, a couple of generations later, there was a professional commercial artist.  On my father’s side, my grandmother was a wonderful artist who painted mostly landscapes of the Mohawk and Hudson River valleys in upstate New York.  And, of course, I’m married to an extremely talented artist.  You would think with those bloodlines and that much exposure, I’d have a just bit of artistic ability myself.  You would be wrong.  I love art.  I just can’t make it.


The closest thing that I come to visual expression is confined to Microsoft PowerPoint creations.  However, within that narrow arena, particularly when it comes to engineering subjects, there is still fun to be had.  What we’re going to do for this article is undertake one of my favorite pseudo-artistic hobbies and play with expander cycle engine schematics.

So, let’s start with a simple, happy little cycle called the Closed Expander Cycle.  Most of what you need to know about this cycle is in the name.  First, it is closed.  That means that all of the propellants that come into the engine leave by going through the throat of the main combustion chamber thereby yielding the greatest chemical efficiency available.  Later, we’ll see that the opposite of “closed” is “open.”  Second, it is an expander.  That means that turbomachinery is driven by propellants that picked up heat energy from cooling circuits in the main combustion chamber and nozzle.  Typically, expander cycle engines use cryogenic propellants so that when these propellants are heated they change from liquid-like fluids to gas-like fluids.  Turbines very efficiently make use of gas-like drive fluids.  (Note that I keep referring to “fluids” rather than simply liquids and gases.  That’s because it’s usually a good idea to deal with supercritical fluids in cooling tubes or channels.  Phase changes can be unpredictable and lead to some odd pressure profiles.)


Above is a Microsoft PowerPoint masterpiece illustrating the Closed Expander Cycle rocket engine.  Fuel and oxidizer come in from the stage and are put through pumps to raise their pressure.  On the fuel side, the pump discharge is routed through the main fuel valve (MFV) to the nozzle and the main combustion chamber (MCC) cooling jackets.  I’ve not shown the actual routing here.  Typically, the MCC is cooled first and then, the now warmer fuel is used to cool the nozzle.  The heat loads in the MCC are significantly higher than those in the nozzle.  But whatever is the exact routing of the cooling fluid, the discharge, now full of energy picked up from the process of cooling, is fed into the turbines.  The oxidizer turbine bypass valve (OTBV) shown in the diagram is a means for controlling mixture ratio by moderating the power to the oxidizer turbine.  In some cases, if you have only one mixture ratio setting for the engine, you might be able to put an orifice here rather than a valve.  The turbines are driven by the warm fuel and then the discharge of the turbines is fed through to the main injector and then into the combustion zone.  On the oxidizer side, the routing is much simpler.  The oxidizer pump discharge is plumbed through the main oxidizer valve (MOV) directly into the main injector.  Within the MCC, you have the combustion of your propellants, the resultant release of energy, the generation of high-velocity combustion products, and the expulsion of these products through the sonic MCC throat and out the supersonic nozzle.  Ta-da, thrust is made!

The closed expander is one of the most simple engine cycles that has ever been imagined.  The venerable RL10 engine first developed in the 1950s and still flying today is based on this cycle (with the slight twist that there is only one turbine and the pumps are connected through a gear box – thereby eliminating the need for the OTBV).  This simplicity is both the strength of the cycle and also it’s limiting feature.  Consider the fact that all of the fuel – hydrogen in the case of most expanders – gets pushed all of the way through the engine to finally end up getting injected into the combustion chamber.  All that pushing translates to pressure drops.  It means that the turbines don’t have that much pressure ratio to deal with in terms of making power for the pumps.  In other words, the downstream side of the turbine is the lowest pressure point in the cycle and that’s the combustion chamber.  The result is that your chamber pressure can’t be very high.  That means that the throat of your MCC is relatively large and then that means the expansion ratio of your nozzle and nozzle extension start to get limited simply by size and structural weight.


Also, note that all of the power to drive the entire cycle is provided by the heat picked up by the fuel in the MCC and nozzle cooling channels.  This then becomes a limiting factor in terms of the overall power and thrust-class of the engine.  As an engine gets bigger, at a given chamber pressure, the thrust level increases to the second power of the characteristic throat diameter, but the available surface area to be used to pick up heat to power the cycle only increases by that characteristic diameter to the first power.  In other words, thrust is proportional to “D-squared” but, to a first order, turbine power is proportional to “D.”  Thus, you can only get so big before you can’t get enough power to run the cycle.  One means for overcoming this is to make the combustion chamber longer just to give yourself more heat transfer surface area.  The European engine called the Vinci follows this approach.  But even this approach is limiting if taken too far since a chamber that is too long makes for less efficient combustion and, of course, a longer combustion chamber also starts to get awfully darn heavy.

So, how big can a closed expander cycle rocket engine be?  Well, that’s a point of recurring dispute and debate.  I can only give my opinion.  I would say that the closed expanding cycle engine most useful and most practical when kept to a thrust level of less than approximately 35,000 pounds-force.

Getting back to the notion of artistic expression, what then are the possible variations on the theme of the expander cycle engine?  Well, the themes and variations are used to explore and potentially overcome perceived shortfalls in the Closed Expander Cycle.  The first in this series is the Closed Split Expander, the portrait of which is below:


The shortfall being addressed here is the fact that in the Closed Expander Cycle all of the fuel was pushed all over the engine resulting in large pressure losses.  In this case, some – usually most – of the fuel is pumped to a lower pressure through a first stage in the pump and then another portion is pumped to a higher pressure.  Thus, the fuel supply is “split” and that’s the origin of the name.  It is this higher pressure stream, routed through the fuel coolant control valve (FCCV) that is pushed all over the engine to cool the MCC and nozzle and to drive the turbines.  The lower pressure stream is plumbed directly into the main injector.  The theory is that by not requiring all of the fuel to be pumped up to the highest pressure, you relieve the power requirements for the fuel turbine.  It is always the hydrogen turbopump that eats up the biggest fraction of the power generated in the cycle so this is an important notion.

Does this cycle help?  Yes, some.  Maybe.  The balance of how much to split, what that split does to the efficiency of the heat transfer (less flow means possibly lower fluid velocities, lower velocities means lower heat transfer, lower heat transfer means less power…) makes it not always clear that you gain a whole lot from the effort of making the cycle more complex.  The portrait, however, is nice, don’t you think?  It has a realistic flair, a mid-century industrialist-utilitarian feel.

Next, wishing to express yourself, you can address the age-old issue of the intermediate seal in the oxidizer turbopump.  Take a good look at the first two schematics presented here.  You will see that the oxidizer pump is being driven by a turbine using fuel as a working fluid.  This is a very typical situation with rocket engines, whether they’re expander cycle engine or other cycles.  For example, this is the situation that you have in the RS-25 staged-combustion cycle engine and in the J-2X gas-generator cycle engine.  What that situation sets up, however, is a potential catastrophic failure.  You have fuel and oxygen in the same machine along with spinning metal parts.  If the two fluids mix and anything rubs, then BOOM, you have a bad day.  So, inside oxidizer pumps you usually have a complex sealing arrangement that includes a continuous helium barrier purge to keep the two fluids separate.  For the next expander cycle schematic, however, we can eliminate the need for this complex, purged seal.


This is a Closed Dual Expander Cycle.  It is still “closed” in that everything that comes into the engine leaves through the MCC throat.  The new part is that it is “dual” in that we now not only use the fuel to cool, but we also use the oxidizer.  Thus, we use heated fuel to drive the fuel turbopump and heated oxidizer to drive the oxidizer turbopump.  For this sketch, I’ve used a split configuration on the oxidizer side with a portion of the flow being pumped to a lower pressure and routed directly to the main injector and another portion pumped to a higher pressure, routed through the oxidizer coolant control valve (OCCV), to be pushed through the regeneratively cooled nozzle jacket and then through the oxidizer turbopump turbine.  I’ve done this since you’re likely running the engine at a mixture ratio (hydrogen/oxygen) of between 5 and 6.  You wouldn’t want to push that much oxidizer through the nozzle cooling channels or tubes.  Now, if you’re designing an expander with something like methane as your fuel so your mixture ratio lower, then maybe you can consider a non-split oxidizer side.

Note that with the dual expander approach I’ve gotten rid of the need for the purged seal package in the oxidizer pump and thus I’ve eliminated a potential catastrophic scenario (in the event of seal package failure).  However, I’ve accomplished that at the cost of some cycle complexity.  Also, cooling with oxidizer does not always make everyone happy.  Whenever you have a cooling jacket (either smooth wall or tubes), you always have the potential for cracking and leaking.  If you’re cooling with hydrogen, then a little leakage of extra hydrogen into a fuel-rich environment is a relatively benign situation.  It happens all of the time.  But what if you leak oxidizer into that fuel-rich combustion product environment?  Well, some studies have suggested that you’ll be fine, but it makes me just a little uneasy.  Then, also, you’re using heated oxidizer to drive your turbine.  It can be done, but using something like oxygen to drive spinning metal parts requires great care.  Under the wrong circumstances, a pure oxidizer environment can burn with just about anything as fuel, including most metals.  So, for all your effort to eliminate the seal package in the oxidizer turbopump, it’s not clear to me that you’ve made the situation that much safer.  However, despite these potential drawback, the schematic portrait itself has a certain baroque feel to it with the oxidizer side being positively rococo.

So, you’ve gone this far.  Why not take the final plunge?  Introducing the “Closed Dual Split Expander:”


By now, having stepped through the progression, you understand how it is “closed,” how it is “dual,” and how it is “split” (on both sides this time).  It’s not practical in terms of being a recipe for a successful rocket engine design for a variety of reasons balancing complexity versus intended advantages, but it’s an impressive schematic.  To me, it has a gothic feel, almost like a medieval cathedral with glorious flying buttresses and cascading ornamentation that just leaves you dazzled with details.

So, we’ve wondered off and into the weeds of making expander cycle portraits for the sake of their beauty rather than necessarily their useful practicality.  Let’s return to the more practical realm and question that which has been common to every cycle thus far presented.  It’s been the word “closed.”  Does an expander cycle engine have to be a closed cycle?  Of course not!  Once we’ve made that observation, we come to a very practical option.  Introducing the “Open Expander Cycle:”


 This biggest difference between this and every other previous schematic is the fact that the working fluid driving the turbines is dumped into the downstream portion of the nozzle.  This is a much lower pressure point than the main combustion zone.  The first thing that most people think when they see this cycle is that it must be a lower performance engine.  After all, you’re dumping propellant downstream of the MCC throat.  And, yes, that is an inherent inefficiency within this cycle.  Whenever you expel propellants in some way bypassing the primary combustion, you lose efficiency.  However, here is what you gain:  lots and lots of margin on your pressure budget.  Because I don’t have to try to stuff the turbine bypass into the combustion chamber, I can make my chamber pressure much higher.  In a practical sense, I can make it two or three times higher than in a simple closed expander cycle engine.  What that allows me to do is make the throat very small and that, in turn, provides for the opportunity for a very high nozzle expansion ratio within reasonable size and structural weight limits.  The very high expansion ratio means more exhaust acceleration and, in this way, I can get almost all of the way back to the same kind of performance numbers as a closed cycle despite the propellant dump.


Here, however, is the really cool part of the open expander cycle: I can leverage the high pressure ratio across the turbines such that I can get more power out of a given heat transfer level in the cooling jackets.  Up above, earlier in this article, I suggested that there was a practical thrust limit for closed expanders of approximately 35,000 pounds-force (my opinion) and this was due to the geometric relationships between thrust and heat transfer surface area.  For an open expander, I can design high-pressure-ratio turbines for which I don’t need as much heat pick up to drive the pumps.  Thus, I can make a higher thrust engine.  How high?  Well, my good friends from Mitsubishi Heavy Industries (MHI) and the Japanese Space Exploration Agency (JAXA) have designed a version of this cycle that gets up to 60,000 pounds-force of thrust and I’ve seen other conceptual designs that go even higher.  The folks in Japan already fly a smaller version of this cycle in the LE-5B engine that generates 32,500 pounds-force.  Note that they often refer to this cycle by another name that is very common in the literature and that’s “expander bleed cycle,” with the “bleed” portion describing the overboard dump into the nozzle.  I prefer the designation of “open” since it clearly distinguishes it from the “closed” cycles illustrated earlier.

We have just about reached the end of this article but we have not reached the end of possibilities with expander cycle engine schematics.  That’s what makes them fun and, in my mind, kind of like playing with art.  You can come up with all kinds of combinations and additions.  For example, what if you took an expander cycle and added a little burner?  Over and over I’ve said that the limiting factor for a closed expander is the amount of heat that you pick up in the cooling jackets.  Well, okay then, let’s add a small burner that has no other purpose than to make the turbine drive gas hotter.  The result looks something like this:


This cycle has a gas generator but is not a gas generator cycle since the combustion products from that GG are not used to drive the turbines directly.  Rather, the GG exhaust is piped through a heat exchanger and then dumped overboard.  Yes, you lose a little of your performance efficiency because it’s no longer a closed cycle, but the GG flows can be small and what you get out of it is a boost in available turbomachinery power and therefore potential thrust.  That’s my own little piece of artwork to demonstrate and anyone can do it.


Remember Bob Ross from Public Broadcasting?  I loved watching his show and, as I’ve said, I can’t paint worth a lick.  But his show was relaxing to watch and listen to and he was always so relentlessly supportive.  There never were any mistakes.  Everything could be made all right in the end.  And anyone could make pretty mountains and happy little trees.  I’d like to suggest that the same is true about my little hobby of assembling happy little expander cycle schematics.  No, most will probably never be built or fly and the schematic portraits will probably never grace the walls of MOMA, but that’s okay.  My artist grandmother used to tell me that sometimes the purpose of doing art was not necessarily found in the end product, but instead as part of the journey of creation.

Inside the LEO Doghouse: Light My Fire!

This article is the second part of the story focused on how we start rocket engines.  In the last article, we discussed the matter of delivering propellants – oxidizer and fuel – into the combustion zone.  In this article, we will discuss how these propellants become fire and smoke (…or steam).  Of course, the musical reference for which you’re waiting ought to be based on the title of this article and the song by the Doors.  Right?  Well, with all due respect to The Lizard King, I would prefer to reference here the immortal writings of The Boss:

I will now be so bold as to translate Mr. Springsteen’s words into functional advice for rocket engines.  Sitting around and crying or worrying about the world are both passive, energy-draining activities.  The only way to start a fire is to add energy, e.g., a spark, to the situation.  He’s absolutely right.  And I would just bet that you never knew that The Boss was a rocket scientist.

In an article about combustion instabilities many months ago, I used the image below to illustrate a situation of limited stability.  The ball sitting on top of the hill will sit there forever unless or until something disturbs it.  Give it a little bump, i.e., an insertion of energy, and the whole scenario rapidly changes with the ball speeding down the hill.


This is also how I think about the process of ignition for typical, non-hypergolic (see previous article) propellants.  You can have fuel and oxidizer sitting around, intermixing, but until it gets that bump of an insertion of energy, there is no combustion.  No combustion; no high-energy gas production.  No high-energy gases; no propulsion.

Let’s start from the other end.  For a moment, think about a fire in your fireplace.  Once you’ve got a good fire up and going as in the picture below, you don’t have to re-start the fire each time that you add a log.  The existing fire sustains itself so that the energy produced by the combustion in one moment is sufficient to continue the fire into the next moment using additional fuel (the wood) and oxidizer (from the air).


This is generally the case for rocket propellants as well.  Once the fire is lit (i.e., once the ball is rolling downhill), the process is self-sustaining.  So, the whole issue about making a fire really does come down to the start of the process.

How many different ways can you start a fire?  One way is to use another fire.  Think about the folks running around the countryside with the Olympic torch before the games.  They use that torch to light another torch to light another torch, and on and on, all of the way until they light the big torch in the stadium.  Another way to start a fire is to use heat.  That, effectively, is how I lit a cigar the other evening.  I used friction to generate heat to ignite a match.  Then, holding the match like an Olympic torch, I used that fire to light the fuel of the cigar tobacco.  This model of a cascading series of larger and larger fires is used over and over in different forms.  Thus, when we talk about starting a fire, we often have to discuss not only the small initial energy bump, but then also the chain of events leading to the complete, steady state process.

So, first we have the initiation, or as The Boss said, “the spark.”  Off the top of my head, I can think of four ways that we’ve practically implemented on rocket engine systems to provide that initial energy boost and one other way that, to date, remains somewhat experimental.  There may be others, but these are the ones that are most obvious and frequently used in different forms.

The first method is exactly what The Boss calls for, an electrical spark.  In most cases when lighting liquid propellants directly, the components on rocket engines used to make electrical sparks are not a whole lot different than higher-energy, more robust, and more reliable versions of the spark plugs that you’ve got in your automobile.  They use a high-voltage electrical circuit to make a spark jump across a gap thereby exposing whatever is around that gap, namely vaporized propellants, to ionizing electrical energy.


The second method also uses electrical energy but in this case rather than making a spark, you use it to make heat.  Think about an incandescent light bulb (i.e., the bulbs rapidly becoming old fashioned these days).  The intent of the wire filament is to produce light.  And it does.  But is also produces heat.  What if you apply that heat directly to a combustible mixture?  Depending upon the mixture, that’s all you need.  I’ll explain more below when we talk about the cascade.

These first two methods rely on electrical energy and that’s always convenient since wires are easy to run.  While it’s true that the ultimate power source can be heavy for the vehicle (batteries for example), the rest of the system is relatively light and easy.  The third method for providing that initial energy bump is not quite so clean.  Rather than relying on transferring electrical energy into a chemical reaction, it uses a transfer of energy from one chemical reaction to kick off another chemical reaction.  In the previous article we discussed hypergolic propellants.  These are propellants that combust spontaneously when they come into contact with each other.  They don’t need any energy boost to start reacting.  Well, what if you had a fluid that did that when it came into contact with your primary fuel or primary oxidizer?  You could squirt in some of this spontaneously combusting stuff, light off a small bit of your fuel or oxidizer, and then the energy for that small fire could light off the rest of the propellants.  This is a common means for starting kerosene (also called RP-1) engines.  The massive F-1 engine used as part of the Saturn V vehicle was lit by a hypergolic ignition system for the main combustion chamber.  The most common hypergols for this purpose are triethylborane (a.k.a., triethylboron), triethylaluminum, or some mixture of the two.


The fourth and last method that I can think of for supplying that initial energy bump again starts with electricity, but instead of generating a localized spark or heat, you transform the electrical energy into a laser.  I will not even begin to pretend that I know much about lasers other than the fact that they can provide a very focused, directed beam of energy, photon energy in this case, to exactly where you want to put it.  You can use that energy to make heat for ignition or – and now I’m way beyond my knowledge base – you can tune the wavelength to excite the propellant molecules directly.  I have a friend in Germany who has experimented with using lasers for rocket engine ignition.  Thus far, I know of no fielded rocket systems where this ignition method is used (although I’ve been told that the Russians have demonstrated it on a full-scale engine), but it offers some very interesting possibilities.


So, we’re done, right?  After all, you’ve got your spark (or some other energy boost) so you’re lit and ready to go.  Well, not always.  For the most convenient ignition source, specifically the electrically-flavored ones, our bump in energy, our spark or heat, is usually very localized.  Rocket propellants are usually highly energetic and that’s why they’re rocket propellants.  But that also means that you have to light the fire well.  I struggle with how to explain this in a positive sense, so I’ll explain it in the negative, i.e., tell you what you do not what to do.

In your combustion zone, you do not want to ignite just one small space, i.e., one corner, and let the fire spread unevenly.  A fire on one side of a combustion zone but not the other could allow unburned propellants to momentarily “pool” in the one region.  This could lead to detonation and/or conflagration pressure waves bouncing around your chamber until everything evens out.  That can be extremely dangerous to the point of tearing apart the engine.  Or maybe, because of these pooled, unburnt propellants, you get mixture ratios that cause hot streaks.  Most practical combustion chambers are not built to accommodate stoichiometric or oxidizer-rich combustion (unless it is specifically an ox-rich preburner where it is should be very ox-rich to avoid this same issue).  A localized phenomenon of a slight ox-rich ignition could burn a hole right through a combustion chamber wall.  Or, if you’re talking about a gas generator or a preburner, you could get hot streaks that damage turbine components.  I have seen the kind of damage that can be done in a turbine due to ox-rich hot streaks for just fractions of a second.


Ideally, what you want is for your propellants to arrive and, blammo, everything it lit.  That “blammo” can be difficult to achieve with a localized energy into like a spark or a small electrical heat source, especially for larger engines.  To overcome this issue, we turn back to the simple analogy of the fireplace.  There, we go from the localized effect of the match, to perhaps a ball or two of crumpled newspaper, to shavings or kindling, to larger sticks, to eventually the logs.  So there is a cascade of events from small and localized to large and generalized.  I will give you two examples of how we apply this concept in rocket engines.

The J-2X gas generator has a pyrotechnic ignition system.  It’s quite easy to tell people that we ignite the GG with little, solid propellant charges.  Okay, but is that the whole story?  No, it’s not.  The solid propellant charge (think about little Estes® rocket motors) is just the fire-lighting-the-fire end of the process.


It all starts with electrical current running through an igniter wire.  The electrical resistance of the igniter wire causes heat as the current passes through.  That heat is enough energy to push what’s called the “pryogen” into ignition.  You can think of the pryogen as being like the stuff on the head of a match.  Other flammable substances are often used but the idea is still the same.  That little fire in the initiator ignites the solid propellant and the solid propellant then shoots hot gases into the GG during the engine start sequence to ignite the hydrogen and oxygen just as they arrive.  Pyrotechnic igniters like this are highly reliable.  If that electrical current arrives, everything beyond that is pure chemical chain reaction that produces a powerful blast of ignition energy.  On the negative side, such an igniter can only be used once.  I guess that you could inspect and refurbish elements of the piece, but considering the trauma of the process it experiences, it is easier and cheaper to simply replace the whole thing.

Another example of the concept of using an ignition cascade can be found on the J-2X in the form of the torch igniter used for the main injector.  Here’s an interesting little piece of history (as it’s been told to me).  The J-2 engine, back in the 1960’s was a pioneering effort.  While the RL10 was already flying, the use of hydrogen as a propellant was still something relatively novel.  For the J-2 main injector they developed a torch igniter system.  That system was later adopted and modified slightly for use as the ignition system for the Space Shuttle Main Engine main injector and both preburner injectors.  When we came to the development of J-2X, we started with our many years of successful experience with the SSME torch ignition system, made some modifications and, through a dedicated test program at the igniter level, effectively revalidated and expanded upon the pioneering efforts of the 1960’s.  It’s good to be part of another small step in that long and successful history.


The torch igniter concept starts with an electrical spark from what really looks like your ordinary automobile spark plug.  But such a spark is very small, very localized.  So what you do is swirl into that localized area just a little hydrogen and oxygen.  This is the kindling.  The electrical potential across the gap of the spark plug causes the gasified propellants to ionize and become very hot, hot enough to start to spread the fire and, from that, thereby creating a flame front.  That flame is then directed into the combustion zone just as the rest of the propellants are reaching the injector.  The whole igniter system is effectively a torch ejecting a flame into the combustion zone.  In the J-2X (and in the SSME and in the J-2), the torch is right in the center of the injector face.


Okay, so there you have it, in two articles, how to get a liquid rocket engine up and going.  First, you have to get the propellants moving to the right places and, second, once they’re there, you’ve got to light the fire.  For large rocket engines, the whole process from the receipt of the start command from the vehicle until the engine is functioning at full power level takes anywhere from about three to six seconds.  During that time, pumps have to start spinning, valves have to open, propellants have to reach their destinations in the correct proportions, and the ignition source has to try to light the fire not too early and not too late.  It really is quite an orchestration of events across a brief period of time.  And the more complex the engine, the more difficult it is to get the orchestration right.

Looking into the database for SSME history, the very first test was conducted on 10 May 1975 with development engine #1 on test stand A1.  It was not until the forty-second test of the test series, nearly ten months later, that they eclipsed five seconds of firing duration and reached true mainstage operation.  So, it was not easy making that orchestration work.  Over the years, I’ve had the opportunity to meet and work with a handful of the folks who were there figuring out how to make the SSME work.  They were all very impressive engineers and thank goodness since we are still benefitting from their efforts.  And with that final historical note, we end this article with some more words of wisdom from The Boss:


J-2X Extra: Human-Rated Chili

I enjoy cooking.  Most people think that when I say that, it’s because I’m an engineer by training, that I like cooking for the structured notion of a recipe and exactly measuring things out and the chemical precision of mixing that with this, at this speed, under these conditions, with these implements, and then forming it all together with a specified heat input over a given time using appropriately sized and shaped pots and pans optimized for uniform heat transfer, blah, blah, blah, blah…

But, that’s way, way off from the truth. 

Actually, I like to cook things that allow for, let’s say, “significant organic creativity.”  I make a mean vegetarian chili, but you can be sure that it will be different every single time that I make it since it’s always from memory and my memory ain’t what it used to be.  I wing it.  And that’s fun.  And even though it’s fun and even though the details vary slightly, it’s been good every time (so far).  The worst side effect that I could attribute about any particular version might be a bit of heartburn (properly mellowing and blending habanero peppers is an imprecise art form I have not yet consistently mastered).

So, what does my free-form chili cooking this have to do with J-2X?  Believe it or not, I want to talk about one of the adjectives that we frequently apply to the J-2X engine: “human-rated.”  What does that mean?  We use that term (or the older, less politically-correct formerly used term “man-rated”) all of the time and, for the most part, those of us within our little clique understand the general context of its meaning.  But if you asked any of us to explain, you’d likely get a wide variety of different, complex, and mostly correct yet often partial answers.  I am no genius and, despite all odds, I will do my best to provide a reasonably complete framework for a definition so as to help you better understand the J-2X engine. 

And, it will come back to my cooking analogy.  Really.

First, we need to recognize that there is really no such thing as a “human-rated rocket engine.”  That is shorthand terminology that ought to be written out as: “a rocket engine that could be suitable as part of an overall, human-rated launch system.”  Think of it this way:  Let’s say that you had a total junker of a car but you installed one perfectly pristine, top-quality piston.  Do you now have a good car or do you still have a junker?  You’d still have a junker, of course.  Or, let’s say that you had a really nice car but all of the spark plugs were corroded, eroded, and barely functional.  Do you still have a nice car?  Well, maybe the paint job is pretty and the stereo sound is clear, but it’s not going to get anywhere quickly, reliably, or efficiently with bad plugs.  The point is that no single element of something as familiar as an automobile makes it complete and good and, in an analogous manner, no single element of something as large as a launch architecture is, in itself, human rated.  The whole system is rated for human spaceflight because the system as a whole, as well as its constituents such as the J-2X, meet certain standards and processes that we’ll discuss below.  We call the J-2X “human-rated” as a shorthand way of saying that it could be part of a human rated architecture consisting of the rest of the vehicle, ground operations, mission control, and exceptionally well trained ground and flight crews, etc.

Second, let’s think about the adjective term “human-rated” itself and its definition.  What does that mean?  It means simply this: the estimated risk is acceptably low so that we can responsibly decide to put human beings into the vehicle for launch.  Again, we can relate this to automobiles.  When you drove to work today, you took a risk.  Unfortunately, auto accidents happen on the roads and highways and, more unfortunately, despite all of the protective apparatus built into our cars, people do sometimes get hurt in these accidents, or worse.  But you accepted that risk and drove to work anyway.  You judged your auto to be sufficiently safe.  You judged that the roads were well paved and properly marked, that the police were properly monitoring bad and endangering behavior on the roads, and that the weather was clear enough to allow for safe operation of your vehicle.  Thus, your “drive-to-work system” was, today, according to your judgment, “human-rated” for you.  You weighed the risks — consciously or subconsciously — and decided to accept these risks and make the trip.

Spaceflight is ten thousand times more complex than driving to work, but the rationale is entirely analogous.  The “fly-to-space system” (note again it’s a “system” not just a vehicle) is  “human-rated” when we judge the risk to be acceptable in light of the potential rewards.  The important and fundamental point is that, in the end, it is a judgment.  Sometimes, for example, we accept more risk because we judge that the potential rewards are that much more significant.  Think back to the early days of human spaceflight.  I can guarantee that there is no way in heck that we would today put an astronaut into some of those early vehicles.  We would not today consider those early systems to be human-rated by our current standards.  But at that time, we as a nation accepted the risk and, by the way, achieved extraordinary milestones.  Today, our objectives and potential rewards are different and so our judgments with regards to risk are accordingly different.

So, if it’s all just a matter of judgment, then doesn’t that mean that there really is no such thing as “human-rated”?  No, I would strongly disagree. 

Here is where I get back to my cooking analogy.  While my chili may have slightly different constituents each time that it’s made, and while it might taste a bit different each time, there is no question as to whether it is chili.  I use my expert cooking judgment to combine the essential ingredients into a recognizable and tasty product (with or without subsequent heartburn).  When we talk about an engine being  “human-rated,” we too are not basing that judgment upon a fixed recipe.  We are basing it upon a combination of essential ingredients and expert judgment.

If you’re wondering whether NASA maintains some kind of formal recipe for human rating, I refer you to NASA Procedural Requirements (NPR) 8705.2, revision B (effective May 2008), “Human-Rating Requirements for Space Systems.”  While this document is helpful, in a general sense, with regards to what technical and programmatic areas to consider, it is written at a very high level, i.e., at the “fly-to-space system” level.  As such, it does not offer a great deal of rocket-engine-specific information.  This, in my opinion, is exactly as it should be.  The actual making of the chili should be left to the expert cooks.  Even NPR 8705.2 makes it quite clear that the intent of the document is only to establish a framework within which “human rating” takes place.  It is not intended to be a step-by-step recipe book for the many, many diverse parts of a human spaceflight system.

What then are the essential ingredients for a human-rated engine?  Not surprisingly, the answer can be thought of as somewhat following the life cycle of an engine development project.

Design and Development
Specific technical requirements — There is a small handful of specific technical requirements that effectively flow down from NPR 8705.2B and impact the engine design.  One is the requirement that, where appropriate and where it can be shown to increase reliability and safety, we should use redundant systems.  On the J-2X, the clearest manifestation of this is the use of an engine controller with two channels.  Should one channel fail (as even heavy-duty computer systems sometimes can), the other channel can take over and continue safe operation.  Another specific requirement at the system level is that there exist abort systems that allow the crew to escape from a bad situation on the vehicle.  This requirement decomposes to a requirement on the J-2X for a redline health monitoring system that shuts down the engine in the event of an imminent failure and notifies the vehicle of this shutdown.  This thereby allows the crew the opportunity to perform an abort.

Design, construction, workmanship standards — Not surprisingly, we don’t start from scratch every time that we sit down to design something.  We know how to do things.  We have lessons learned.  We have rules of thumb.  And, at the top of the list, we have standards.  These are specialized requirements documents that focus on specific, narrow technical areas.  For example, NASA-STD-5012 tells you what you should do for the structural design of a rocket engine.  It lays out the essential analyses to perform, the way that the environments should be evaluated, and what factors of safety are appropriate.  For J-2X, we had over thirty different standards that were (and are) part of the requirements imposed upon the engine design details, design processes, fabrication processes, and testing scope and procedures. 

Even here, however, after you impose a standard you have to acknowledge the fact that there can exist more than one way to do things and do them safely.  For example, on J-2X we imposed a structural design standard that, at a lower level, imposed a standard for how fasteners (i.e., bolts and nuts) are properly lubed and torqued.  In order to investigate this issue, we set up a mini-test program to better understand the results from the different methods.  It kind of sounds silly, but fastener torque is extremely important in high-pressure systems and proving that the contractor process was equivalent and safe could save us money in the long run since it is a standard procedure for them.  So, we had a guy follow the procedures several times and we measured the strain induced into a series of bolts by the applied torquing method.  The measured strain was converted to applied force and this thereby validated the procedure.  Across the spectrum, we had a number of similar examples where we interpreted the technical intent and purpose of a detailed requirement and, working with our contractor, found the best way to comply.

System safety program — As an engineer, the question foremost in your mind is always, “How can I make this thing work?”  Without that mindset, we would never get anywhere.  However, when dealing with something as complex and as potentially dangerous spaceflight, you must go beyond this level of thinking and must also continuously ask yourself, “What could go wrong with this thing and how do I mitigate that potential as much as possible?”  In the most basic sense, this is the motivation for developing a system safety program.  As part of the engine design and development process, you look at this issue from two directions. 

First, you look at the piece-part level and ask, “What could break, how or why, and what would be the effects?”  That’s a reliability analysis.  You look at all of the pieces and figure out what circumstances could result in something not working as intended.  Could the design be mistaken because we didn’t understand the loads?  Could the loads go off nominal because of some unusual flight situation?  Could the manufacturing of that piece go awry so that you don’t have the intended design margins in the actual, physical part?  And, for all of these questions, you have to provide answers as to how best to ensure that the part won’t actually break during operation.

Second, you start from the other end.  You start with the grim notion that you’ve failed and that the crew didn’t make it.  From there you work backwards and figure out how and why that situation could take place.  This process grows into a tree of circumstances and possibilities and is called a hazards analysis.  Was it an explosion?  If so, where did the fuel and oxidizer and ignition source come from?  If the fuel came from tank, then how did it escape?  Was it instead something having to do with navigation?  Or maybe there was a weather-related issue, perhaps, say, lightning? 

Obviously, in many places these two assessments eventually meet in the middle.  The one starts at the bottom and works upwards.  The other starts at the top and work downwards.  When they meet, then you know where throughout your system are your critical points.  In some cases this drives design features, special inspection requirements, or, for example, in the case of lightning protection, the design and construction of a launch pad system for dealing with the hazard.  This overall effort allows you to prioritize your efforts to ensure safety and, in the operational phase, potentially apply greater attention prior to committing to launch. 

Test and Evaluation
Structured verification planning and reporting — Believe it or not, we don’t march into an engine test program all willy-nilly and make a bunch of smoke and fire just for the sake of impressing our friends.  We do it to generate and collect data.  The data that we collect largely goes towards the systems engineering endeavor known as requirements verification.  Verification is defined as the process of demonstrating that the product design — in our case an engine — is in compliance with imposed requirements.  Verification can, and does, take a number of forms.  Testing is one form.  Analysis and inspection are others. 

Note that the “structured” part of the “structured verification” title above is a key consideration.  You must lay out plans saying, “Here is my requirement and here is what I plan to do to prove that I meet it.”  Then, based upon peer review of experts, this plan can be approved or modified.  This is an essential part of the whole judgment aspect of human rating.  If I demonstrate that I meet the requirement with one engine on one test, is that good enough?  If not, how many engines or tests do I need?  Or, if it’s verification by analysis, do you agree with the analysis methodology that we propose to use?  Do you concur with the assumptions and the simplifications inherent in any analysis method?  The whole process, when properly approached, has the flavor of the classic scientific method.  The hypothesis is that the product meets the requirement and then you set out to prove that hypothesis.

Smart people with backgrounds in mathematics inevitably jump into the conversation here and declare the supremacy of statistics.  Using statistical analysis, we can determine how many samples and tests are necessary to achieve a mean and variability assessment at a given confidence level.  Unfortunately, as good as those methods might be, we can never come close to affording the kinds of programs that a purely statistically based assessment would suggest.  Maybe back in the day we could afford to build and test 100 engines before we’re ready to fly, but today our constraints are to accomplish the same level of risk mitigation with an order of magnitude fewer samples.  We have to be wiser and more efficient, and yet still have sufficient confidence to declare that the design meets its requirements.

Test, test, test, and then test some more — Now, after having discussed a fundamental motivation for testing engines, i.e., requirements verification, you have to get down to the nuts and bolts of the issue.  You must test and you must do it a lot.  Yes, “a lot” is not what you’d call a scientific term, but it can be decomposed.  “A lot” means that you cover your verification plans in terms of samples and repeat examples.  It means that you push things beyond normal operation to prove margins.  You test longer — both single run and cumulative on a given engine, both starts and seconds — than any flight engine could possibly ever see.  And throughout this process, you continuously learn things that you didn’t know that you didn’t know.  While it is theoretically possible that we could design an engine, put it into test, and find that we’d properly characterized every environment and every engine response to those environments, but I’ve never seen such a case and nobody that I know have ever heard of such a thing.  Engine testing is always an education.

The other aspect of testing that is sometimes categorized separately is teardown and detailed inspection of the hardware afterwards.  If you predicted that something wasn’t going to crack and, upon teardown, you find a crack where it shouldn’t be, then you’re not as smart as you thought you were (a phrase I’ve used before).  If you tear down and find that something was rubbing in a valve or a turbopump, then that might be an issue.  Or, instead, it might have been planned that way.  You look for discoloration that might suggest unexpected operational conditions or potential changes in material properties.  You check dimensions of everything to make sure that you didn’t deform pieces or possibly lose material that was consumed by the engine.  Thus, while you collect lots and lots of data during the engine tests, it is also the data that you collect after the testing is complete that contributes substantially to your understanding of the design and its safe operation. 

Quality processes — Twenty-some years ago, the Ford Motor Company had a motto that they used in advertising: “Quality is Job One.”  With all due respect to that venerable motor company, those of us in the rocket world have known this for a long, long time. 

When we certify an engine design and say that it is “human-rated,” that is a contingent description.  It is contingent upon future flight engines being produced in the same manner and to the same detailed workmanship standards as the design that you certified.  That means that the fabrication and testing processes are the same, the materials are the same, the people doing the work on the pieces have had the appropriate training, and that the finished parts have been scrutinized to the same inspections and inspection standards.  And, if things can’t be exactly the same (for example, vendors can change over time), then you must have a process in place to assure equivalence between what you had before and what you’re going to use new. 

Also, should something go awry during the manufacturing or assembly of any part — and things always go awry to some degree at some point — you need to have processes in place to identify what went wrong, how to avoid that issue in the future, and what to do with any hardware that was exposed to the issue.  Can you fix it and still meet your requirements and drawing specifications?  Or, do you have to scrap the part because it can’t be saved? 

These considerations are all part of a good, solid quality system.

Configuration management — The first cousin of quality assurance is configuration management.  While it sounds like a simple premise, this discipline deals with making sure that the exact, particular pieces on the vehicle are the exact, particular pieces that you intended to put on the vehicle.  This means, for example, that every bolt on the engine is suitable for a flight engine.  No, not every bolt has a serialized part number, but they are segregated by lots.  Lots intended for flight usage are subjected to a stringent quality processes and must, therefore, be kept separate from any similar-looking bolts that might not meet the high standards for flight.  Plus, of course, we track throughout their lives the history of our serialized assemblies like turbopumps, combustion chambers, nozzles, ducts, lines, controllers, valves, etc., along with their associated documentation.  And engine is composed of thousands of parts and, one way or another, we track them all. 

The combination of a good quality assurance system and a good configuration management system guarantees that what you have delivered and put on the launch vehicle is exactly what it is advertised and intended (and needs) to be.

That’s it.  Those are, in my opinion, the key ingredients for human rating.

So, getting back to cooking.  In order to make vegetarian chili, you need tomatoes, beans, and chili powder.  That’s it.  But chili made with just these ingredients would be terrible.  I add peppers (of multiple varieties) and onions and garlic and other spices.  Corn can add a nice sweetness.  Sometimes I sauté chopped portabella mushrooms and toss them in.  Beyond that, I’ve been known to add all kinds of oddball stuff including, once, green beans.  And, in the end, it’s good.  I promise.  That’s because I’ve made it probably thirty or forty times over the years and therefore I am a subject matter expert (within my tiny culinary world).  Solid, well-defined ingredients and expert judgment inform my chili.

In order to have a “human-rated” rocket engine, all of the topics that I mention above represent the key, essential ingredients: (1) a few, specific human-rating design requirements, (2) a set of established design, construction, and workmanship standards, (3) a thorough safety program, (4) a structured verification process, (5) system testing campaign, (6) a solid quality assurance system, and (7) a reliable configuration management system.  They are all necessary.  And certain bounds, limits, or standards can be established (and are documented) for all these various disciplines and undertakings, but an exact, repeatable, or universal, step-by-step recipe is extremely difficult to conjure up.  Just like my chili, the details of how, when, and why an engine is “human rated” fall within purview having good key ingredients and then applying expert judgment.


Welcome to the J-2X Doghouse: All a Matter of Balance — and Power

One of the most important analytical tools used in development of a rocket engine is called a “power balance.”  A power balance is, stated simply, a simulation of the steady-state, internal conditions and functioning of the engine.  It can, on one extreme, be accomplished with a spreadsheet or, on the other extreme, take the form of a complex computer program with hundreds of theoretical calculations bolstered by dozens upon dozens of embedded, empirical relationships customized for a particular hardware configuration.  But first of all, let’s talk about what a power balance is from a purely conceptual point of view.  You start with a schematic of the engine:


       MCC = Main Combustion Chamber
      GG = Gas Generator
      MFV = Main Fuel Valve
      MOV = Main Oxidizer Valve
      GGFV = Gas Generator Fuel Valve
      GGOV = Gas Generator Oxidizer Valve
      OTBV = Oxidizer Turbine Bypass Valve

Next, you break down these pieces of the engine, the components, into descriptions with regards to how they relate to power, pressure, and temperature:

  • Pumps:  Convert shaft power into fluid power in the form of elevated pressure
  • Turbines:  Extract power from the turbine drive gases and converts it to shaft power
  • Gas Generator and Main Combustion Chamber:  Generate power from combustion
  • This power is in the form of elevated temperature (combustion = fire = hot)
  • Ducts, Valves, and Injectors:  Control fluid movement in order to get propellants and combustion products where they need to be, i.e., plumbing.  Each of these items reduces pressure in the fluids flowing through them
  • Cooling Jackets:  Here too pressure is lost as the fluid flows through the cooling passages, but temperatures are elevated as heat carried away (i.e., as cooling takes place)

Thus, in terms of the most significant power considerations, here is what is going on with the rocket engine:

You’ll note that all of the power stuff happening in the engine is happening up on the top portion of the original schematic (and I’ve chopped away everything else).  In other words, the major power transfer stuff happening in the components that make up what we call our “powerpack” testing.  See?  That’s why and that’s where the name comes from.  Pretty clever, huh?  The whole idea is to get power to pumps so that they can makes lots and lots of fluid pressure so that they can push lots and lots of propellants through the system and into the combustion chamber.  That’s the whole point of the rocket engine, push stuff to the combustion chamber to make thrust. 

So, how much pressure do you need?  That’s a matter of how much stuff you’ve got to push the propellants through and how much pressure you want in the chamber at the end.  I sometimes think of it like that great old board game Monopoly ®.  You pass “Go” and get $200.  Remember that? 

Well, in a rocket engine, your pump is “Go” and at that point you get an allotment of pressure.  Then, as the fluid goes through the system, from component to component — ducts, valves, cooling jackets, injectors — you have to pay rent in the form of a loss of pressure.  That’s like landing on the various squares around the board.  Paying all that rent is just fine.  You can’t really avoid it.  But you have to make sure that you save enough money to stay at the hotel on Boardwalk in the end without going bankrupt.  In other words, you need to get your propellants into the chamber at the residual pressure that you desire.  Here’s a representation of this pressure management process within the J-2X on both the fuel side and on the oxidizer side:


The explosion-looking symbols in that diagram represent combustion zones.  One is the gas generator, where you make the power to drive the pumps, and the other is, of course, the main combustion chamber, where you make your thrust.  The gray lines represent combustion products coming out of those combustion zones.

One last question that needs to be considered is this: How much combustion chamber pressure do you want (and/or need)?  In other words, when your propellants arrive at the main combustion zone, at what residual pressure do you want that combustion to take place?  Sounds like a simple question, right?  Well, of course, you want it to happen at the “optimal” pressure.  But what does that mean?  That is not an easy question to answer.  In terms of energy release, within certain bounds, the chamber pressure does not much matter (or, at most, it’s a secondary factor).  What it really comes down to, believe it or not, is engine size and weight and a handful of manufacturing considerations. 

In the drawing above, I have tried to show two combustion chamber and nozzle combinations where the one on top has a throat diameter and nozzle exit diameter twice as large as the respective measurements in the lower version.  Thus, both engines using these combustion chambers and nozzles would have the same ratio of nozzle exits area to throat area.  It’s just that the one on the top would have a throat with four times as much area (area being proportional to the square of the diameter).  Would it surprise you to learn that these two engines could generate the same thrust if the one on the bottom had four times as much chamber pressure as compared to the one on top?  Yep, it’s true.  If the top engine has, say, 500 psi (pounds per square inch) chamber pressure and the bottom one has 2,000 psi, then these two rockets are — to first order estimates — operating at the same performance level. 

What does that mean?  That means that you could have a great big, bulky rocket engine or you could have a small, “tight” one.  It would seem that the small one feels more efficient except that with that high chamber pressure you have to generate all that extra pressure in your pumps.  That takes a lot of pump power and therefore turbine power.  And containing all of that pressure throughout the engine system means thicker walls on your ducts and valves and everything else.  Thicker walls mean heavier pieces.  So maybe that “tight” engine is really more wasteful.  So instead, maybe the big bulky engine sounds like a good idea since it’s easier on your turbopumps.  Except then you realize that it’s too big to fit on your vehicle and, by the way, that monstrously big nozzle weighs a ton and nobody has machining tools large enough to produce the thing.  So maybe the bulky one isn’t right either.  Blah, blah, blah…  It’s enough to give you a headache!  But those kinds of discussions back and forth are what are known as trade studies and they are the foundation for what your engine will eventually become.  There is rarely a simple, obvious answer since everything has impacts on everything else.

So, how does all of this get back to the power balance?  Well, you take all of those notions discussed above and start applying the following:

  • Calculations that describe how much energy is released by the combustion of your propellants.
  • Calculations that relate pump speed and pump design features to fluid pressure increases.
  • Calculations that relate turbine-drive gas conditions and turbine design features to power extraction.
  • Calculations that describe pressure losses for fluid flowing through ducts, valves, cooling jackets, and injectors.
  • Calculations that relate fluid flow and fluid conditions to heat transfer processes in cooling jackets

Once you have all of these relationships, then you can perform a power balance.  You use your power balance to inform your trade studies.  Bigger or smaller?  Faster or slower?  You just have to realize in using it that you can’t get anything for free.  The power that you generate in your gas generator uses up some of your propellants (for a gas generator cycle engine) so they can’t go through main injector with the purpose of generating thrust.  You cannot perfectly extract the power from the turbine drive gases.  And, you also cannot pump with perfect efficiency.  These considerations all have to be taken into account in your calculations.  But the result will be an analytical model that can tell you the pressure and temperature of the propellants throughout their journey through the engine.  It will tell you shaft speeds of the turbopumps.  And it will give you overall performance of your rocket engine.


So, let’s say that you’ve been given the job of designing an engine from scratch.  You have a thrust requirement and a specific impulse requirement.  Let’s say, further, that you know what your propellants are supposed to be and let’s even go so far to say that you’ve been told that it ought to be a gas generator cycle engine.  Okay, so now what do you do?

Here’s one approach (…one of many, many possible):

  • Pick a chamber pressure.
  • Because of your thrust requirement and specific impulse requirement, you can start with a pretty good guess as to your propellant flow rates.
  • Next, generate your schematic layout of the engine and the various components and piece together your simulation of the system.
  • Then, figure out how much pressure your pumps need to generate and, therefore, how much power you need your gas generator to create.
  • Balance that pump power needed with turbine power to be extracted; you’ve now set your gas generator conditions.
  • Based upon how much propellant that you’re “losing” down the gas generator / turbine drive leg, you can figure out how much nozzle expansion ratio you need to get to your specific impulse requirement.
  • You’ll probably go around a bit in circles with the previous few steps — also known as iterating — until you get a completely self-consistent set of answers (It’s essentially a process of making educated guesses, seeing if everything balances out, making new guesses based upon any lack of balance, and again seeing if everything balances.  With a good solution scheme, you’ll eventually arrive at a place where all your guesses work and your system is balanced.)
  • You now have a rocket engine design.

But, is it what you want?  Can you build it?  Does it fit with the vehicle?  Will it be too heavy?  Are the component performance factors within reasonable expectations (i.e., rules of thumb carried around by the various component experts)?  Is the design close enough to a legacy design so that you might be able to leverage previous, related experience?  Or, perhaps, is the design all so new and different that the necessary development program will be quite extensive (and therefore expensive)?  It may be that there are a whole bunch of reasons why your design, frankly, stinks so you need to go through the whole process again.  In the end, after several cycles through, you almost never come up with a design that makes everyone happy from every perspective, but you come up with one that is sufficient, acceptable, and reasonable.  So that’s the design that you go and design, develop, and test.

Hopefully, I’ve shown you that a power balance, an analytical simulation of the internal workings of an engine, is an integral tool in the conceptual design of a rocket engine.  Once you’ve got some general idea of some key parameters you need, the power balance fills in the details, sets the necessary parameters for your turbopumps, captures your fluid splits and conditions, and establishes the general sizing for your main combustion chamber and nozzle.  It uses physics and physics-based empirical relationships — combining the disciplines of fluid dynamics, heat transfer, combustion science, and hardware mechanics — for all of the major components of the engine to balance the power generated against the power used and, in so doing, describes conditions throughout the engine.

(This, by the way, is my favorite kind of analytical modeling simply because it combines so many different disciplines and yields such a broad and useful tool.  I was lucky enough to be assigned to power balance modeling activities for the Space Shuttle Main Engine when I started working.  And that experience has informed everything else I’ve done for the last 20+ years.)

Welcome to the J-2X Doghouse: Old Dogs, New Tricks

A couple of articles back, I asked the following question:

“The whole orange-flame thing is not something I entirely understand…Any ideas from anyone else?”

I was talking about the flame stack during a night test at the NASA Stennis Space Center.  It was a legitimate question.  Combustion chemistry is really not my specialty.  Lots of things are not my specialty.  Try as I might, I’ve found that I can’t know everything about everything.  Indeed, considering the many brilliant and knowledgeable people with whom I have the privilege of working here at NASA, I’ve come to accept the conclusion that there is a lot more stuff to know than can ever be learned.  But that can never stop you from learning something new.  And so I have with this.

In response to my blog question, we received a number of comments on the blog and those are posted.  Thank you for your inputs and interest. 

However, behind the scenes (so to speak), a coworker of mine, Robin Osborne, who does have experience with this kind of stuff read the blog and starting poking around amongst her notes and amongst her fellow experts in the field of flame spectroscopy.  Below is a picture taken of igniter testing at MSFC using a gaseous hydrogen-oxygen mixture.  Here too you can see a red-orange flame although it takes a distance for that colored portion to show itself.

According to Dr. Robert Pitz from Vanderbilt University, “Pure hydrogen (with no sodium) — air flames will glow red in a dark room due to the water vapor emission lines.”  Both Dr. Joseph Wehrmeyer working in support of the Air Force and Richard Eskridge from NASA concur, noting that water vapor generates an orange-red-infrared continuum in such flames.  However, all of these individuals also noted that there is a strong orange coloration in such flames due to sodium contamination within the hydrogen.  The sodium is present as sodium hydride within the liquid hydrogen which decomposes at high temperatures to generate the vibrant color.   The sodium contamination is a byproduct of how large, industrial quantities of hydrogen are made for uses such as, for example, flying the Space Shuttle.  Dr. Christopher Dobbin of NASA noted that in the 1990 timeframe he was engaged in an analysis of the flame plumes ejecting from the Space Shuttle Main Engines (SSME).  He said, “The (time) average sodium concentration we measured in the SSME exit plane was 0.091 parts per billion.”  That doesn’t sound like much, and it’s not enough to impact engine performance or operation, but it’s still enough to measure based upon spectral analysis of the plume.  Another possible contaminant, according to Richard Eskridge, is potassium and that can further contribute red emissions.

So, there you go.  It’s a matter of water vapor at the right temperature and pressure (and therefore density) and a couple of key contaminants in the fuel.  It’s “common knowledge” around here amongst us Datadogs that the plume of a Lox/Hydrogen rocket engine is clear.  But that’s not entirely correct.  It’s nearly clear.  It still has the characteristic red-orange tint, but it’s at a density where the emission is too low to see.  On the other hand, for the flame stacks at the test facility — the origin of this whole discussion — we’re talking combustion at atmospheric pressure so the water vapor products are denser as are the relative contamination levels since it’s a fuel-rich environment.  And that’s why they show up at those brilliant colors in the nighttime pictures.

See, even old Datadogs can learn new tricks.  Thank you to everyone who added their two cents, but especially to Robin Osborne for her inputs and insight.

Welcome to the J-2X Doghouse: You Dropped a Bomb on Me, Baby!

The Gap Band, 1982!  Yep, just recalling those wise lyrics and the electrofunk music of my youth and I’m up and dancing and strutting around in my office like a goofball.  Luckily, nobody is watching.  So, is there a point to this opening other than imposing on you the painful image of a middle-aged bureaucrat getting down and getting funky in his office?  Yep, there is.  Here it goes:  The next J-2X Engine 10001 will be a bomb test.

You read that correctly: Bomb Test.

Now, I could probably weave a complex and fanciful tale explaining how “bomb test” is really just a creative government euphemism, but we’ve been straight with each other before, right?  So, the truth is that for this next test we will be mounting into the main combustion chamber a 100% genuine bomb, a small explosive device.  And, yes, we will detonate that device to set off an explosion.

Other than proving once again that we’ve got a cool job and that we’re really like a bunch of 14-year-olds who like to make loud smoke and fire, there is actually a technical reason for doing this.  Way, way back in February, some 20 articles ago (“”-2X Extra: Shiny Metal Pieces”), I briefly mentioned the possibility of combustion instabilities in the gas-generator.  I likened them to the melodious sounds from a pipe organ although combustion instabilities in rocket engines are far, far from melodious.  Indeed, they can be dangerous and destructive.  Our bomb test is a means for characterizing the combustion stability of the J-2X engine.

In order to understand combustion instabilities in a very general way, we have to take about a dozen steps backwards and get to some really basic physics.  The issue comes down to one of natural frequencies and resonance. 

Have you ever tried to push a kid on a swing?  Almost immediately, instinctively, you know that there is a particular rhythm with the swing and if you push in concert with that rhythm, the magnitude of the swinging motion will be increased.  If you try to push at a different rhythm, the kid just kind of sits there in the middle, bobbing around and getting annoyed.  The swing has a natural frequency and if you match that frequency, you get a big response.  That’s resonance.  If you fight against it, your energies are dissipated without much else happening.  Everyone can picture that, right?  That’s a simple, common, shared experience.  Good.

Next, we’re still going to use a rope, but in a little different manner.  In the diagram below, you see someone – i.e., you – holding onto a rope that is secured to a wall.  If you jiggle the rope with no rhythm (i.e., if you jiggle it kind of like how I dance to The Gap Band), not much happens.  It will wiggle and move, but with no organized pattern.  But very quickly, again almost instinctively, you can find the rhythm to make the rope define, back and forth, a single, graceful arc of movement.  With a little practice, you can then jiggle the rope at exactly twice that first speed and get it to make the shape shown in the middle.  And, if you’re really good, you can jiggle it at three times the speed of the first one and get the shapes shown on the bottom. 

If you do this, you will have demonstrated the first, second, and third natural modes of the standing waves for that length and properties of rope.  You cannot get those shapes by inputting any random jiggle on the end.  You have to input a specific forcing function, tuned to a specific frequency, and you will get the desired results.  Your forcing function must resonate with the mode.  Just like with the kid on the swing, if you input the wrong forcing function, nothing much happens.

Another point to consider — in addition to considering the forcing function — is that the rope has particular characteristics that define its natural modes.  But anyone who has ever picked up a guitar knows this, right?  Each string is a different thickness, each is pulled tight to a particular tension, and by putting your fingers on different frets, you alter the effective length of the string.  So, each string, when made the correct length and plucked, vibrates in its first natural mode to yield a particular note.  Because the guitar string is fixed on both ends, what you get when you pluck it is like the top picture in the jiggled rope discussion, the first natural mode.

Now, we’re going to make the jump from wave shapes in ropes to pressure waves in air.  Imagine rather than a string showing wave patterns, pressure variations in air.  Can’t image that?  Okay, then imagine someone talking to you.  Sound travels via fluctuations of pressure in the air.  When you talk, you tighten or loosen your vocal chords, make them vibrate like a guitar string, and sound emanates from that physical vibration turned into pressure variations in the air.  This is the jump from structural vibrations to acoustics but it’s still dependent on the notion of waves and frequencies. 

Okay, so rather than thinking about your vocal chords, imagine playing a trumpet.  You make your lips vibrate in the mouthpiece and, at certain particular frequencies, you can make the trumpet sing clear, bright notes.  You supply the forcing function; the forcing function matches a natural frequency of the column of air within the trumpet; and you create standing pressure waves that sound like music to everyone in the vicinity.  Quite simple.  And different instruments have different natural frequencies.  A tuba will never sound like a trumpet.  Why?  Because it has a different natural frequency and responds differently to different forcing functions. 

So, we started with a kid on a swing and now we’re talking about trumpets and tubas.  And what does any of that have to do with a bomb in a rocket engine?

Let’s review.  In our discussion we’ve learned that things have natural frequencies.  A rope has a natural frequency.  The space within a tuba has a natural frequency.  Everything around you has natural frequencies.  And we’ve learned that if you input a correct forcing function to a system, we can get organized results by working in conjunction with the natural frequencies.  We can get resonance.  A child swings high on the swing.  A rope makes neato patterns when jiggled.  A trumpet blares a high-C over the cheering crowd.  Okay, so here’s the kicker: all this stuff that we’ve discussed is exactly what we DON’T want to happen in the rocket engine combustion chamber.

Just like any other semi-enclosed space, a combustion chamber has natural frequencies.  These frequencies relate to all three dimensions in space (longitudinal, radial, and circumferential since chambers are typically cylindrical) and to the “stuff” in the space (for a tuba that “stuff” is air, in a combustion chamber it’s the propellants and, primarily, the combustion products).  These characteristics together define many, many potential natural frequencies, or modes, where the chamber could “sing.”  The forcing function is the combustion itself, which makes lots and lots of noise at many, many different frequencies and at extraordinarily great magnitudes all jumbled together.  So, you have lots of potential modes and lots of very powerful, very high-energy forcing functions.  Should these combine such that one feeds the other, then you could get resonance.  Again, resonance is what happens when you push the kid in the swing at the right rhythm.  But, taken to the extreme, a situation of resonance in an environment like a combustion chamber can continue to grow out of control until it becomes destructive.

Everyone’s favorite example of destructive resonance in practice was the Tacoma Narrows Bridge collapse in 1940.  The bridge was a suspension bridge, which means that it was kind of like a long, heavy, hanging piece of rope made of concrete and steel.  Well, it turns out that when the wind blew across the bridge at the right speed, it excited a natural mode of the hanging roadway.  As the forcing function blew, the bridge oscillated in response, more and more, quite violently, until, ultimately, the structure crumbled into Puget Sound.  These were not tornado-like winds.  They weren’t even unusually high winds.  They just happened to tune into the natural frequency of the bridge and the bridge responded by tearing itself apart.

Lesson:  Big vibrations can be destructive.  The bridge was fine when it was built.  It was fine for several months thereafter.  But when the right forcing function came along, it was disaster. 

This is as true in combustion chambers as it is for suspension bridges.  In order to avoid this, we build into combustion chambers such things as acoustic cavities, which are sized cavities that are  tuned so as to damp known natural frequency vibrations should they arise.  We also use physical barriers across the faceplate of the injector so as to disrupt the establishment of radial or circumferential pressure wave patterns.  These are features that we build into the design to help ensure that the space within the combustion chamber is not excited into any organized pattern that could build up to destructive levels.  We simply don’t want the chamber to sing. 

And this, finally, is where the bomb test comes in.  We use this kind of test to help prove that the features we’ve included in the combustion chamber do indeed suppress the formation of destructive oscillations.  During the test, we will set off the bomb.  It will act as a broad spectrum forcing function with sudden input of energy.  We need something as extreme as a bomb explosion to perform this energy input because there’s already so much energy being released in the combustion chamber.  It’s not like we could toot a horn at it and try to find some particular frequency.  That would be like trying to whisper to the person next to you while sitting in the fifth row of a rock concert.  It ain’t gonna get through.  So, we set off the bomb and if we have a mode lurking in the chamber that is not sufficiently suppressed by our design features, it ought to poke its head out of the noisy response that follows the explosion.  We will analyze the pressure oscillation and structural vibration data and look for notes that might “sing.”  We don’t expect any to be destructive based upon many years of design experience, but even if we identify any that don’t die down quickly we will have cause for further assessment. 

To wrap this up, I will use one more image that helps me whenever we talk about “instabilities.”  Through the whole discussion here, I’ve talked about vibrations and oscillations and natural modes and forcing functions and resonance, but what does that have to do with stability or instability? 

In the image above, I have drawn two situations of stability.  In both cases there is a ball sitting at rest between two hills.  For the ball on the left, if you perturb the ball slightly to the left or to the right, it will roll back to the middle and sit there peacefully.  However, for the ball on the right, if I perturb it much in either direction, the ball will crest one of the hills and fall into oblivion.  Thus, both situations shown are stable, but the one on the left is more intrinsically stable than the one on the right.

Similarly, the Tacoma Narrows Bridge was stable when it was built.  But given the right perturbation, it was knocked out of valley of stability and became destructively unstable.  Also, we know that the engine is stable.  We’ve already run several tests and it’s been fine.  While we don’t expect anything dramatic or destructive to happen on our upcoming test, this notion of “how stable” is what we’re examining with the bomb.  We are using the bomb to knock the ball off the center of the valley and then we’ll measure how quickly it comes back to rest, i.e., how deep and steep is the valley of our stability.  This is a measure and demonstration of the robustness of our engine design.

J-2X Progress: Mission-Duration Test

Five hundred seconds is exactly eight minutes and twenty seconds.  Nope, that’s not rocket science.  But that was what I had to keep in mind as I watched the stopwatch application on my smart phone during the last J-2X test.  Eight minutes and twenty seconds.  That seems like a really long time when you’re counting every second.

Let me set the scene.

At the NASA Stennis Space center you have collected the directors from seven of the ten NASA field centers around the country.  You have representatives from the NASA headquarters in Washington, DC.  You have a live feed being picked up by NASA TV and broadcast into the living rooms of thousands or millions of dedicated NASA TV junkies.  You have dignitaries in suits and technicians and test conductors in jeans and Hawaiian shirts (test-day tradition), reporters with notepads and cameras from every paper and television station in the greater New Orleans and southern Mississippi area, and, sitting in his ceremonial throne, the Grand High Exalted Mystic Ruler of the International Order of Friendly Sons of the Raccoons.

Well, okay, that last part about the Exalted Mystic Ruler is just fictional (bonus points to anyone who gets the 20th-century cultural allusion without Google help), but that’s the way that it felt.  This was test A2J008, the seventh planned hot-fire test of the very first development engine and it was time to play show-and-tell.

Does everyone remember show-and-tell in elementary school?  You bring in something that you think is neato or special and, by getting up in front of class and talking about it you reveal something about yourself and you accidentally practice public speaking and presentation.  Once, when I was seven years old, I brought in my new baby brother, or, well, my mother did so at my behest.  I wish that I could remember what I said about him.  I imagine it was something like, “He’s short, cranky, and smells funny.”  Today, at least he can say, he’s taller than me.

J-2X is our new baby brother — of a sort to carry forward the analogy — and we’re showing him off to the world.  Through the first six hot fire tests of engine E10001, we accumulated a total of 225 seconds of test time.  For test A2J008, on November 9th, our show-and-tell for the world, we scheduled a test lasting 500 seconds, which is the mission duration requirement for the engine.  Here is what I saw during the test, while holding the stopwatch, standing out in the field in front of the test control center:

Can’t see anything?  Okay, I’ll expand the picture in pieces starting on the left.

This is the hydrogen burn stack.  All of the excess hydrogen coming from the facility or from the engine before, during, and after the test needs to be burned off.  This is all bleed flows and waste flows that you cannot avoid when dealing with a cryogenic propellant.  If you let hydrogen accumulate anywhere around the facility, then “BOOM” you’re eventually going to have an explosion.  Talk to the guys who work out in the test areas and they’ll tell you plenty of tales of such things.  What is amazing as you’re standing out in that field to watch the test is the radiation heat coming off that thing.  It was a chilly day and yet you almost feel like you’re going to end up with a sun-tanned face.  It feels like the sun while you’re on the beach except that as warm is it makes your front side, your back side is still chilly from the blustery November breeze.  Kind of an odd sensation being both overheated and chilly at the same time.

In the middle of the picture is a sign for anyone who was born without that instinctual reflex for self-preservation.  While it would seem obvious to me to not walk in front of a roaring rocket engine throwing out a plume reaching hundreds of feet in the air, the fact that they have a sign like this suggests to me that for someone, somewhere, at some time, this was not so obvious.  An unfortunate thought…

And, on the right-hand side of the picture, in the distance, is test stand A-2 with the engine firing.  In the middle of the picture below, there is a tiny, very white spot in the middle of the test stand.  That’s the flame coming directly out of the engine nozzle.  In the bottom right corner of the picture below you can just see the edge of one of the liquid hydrogen barges.  For both liquid oxygen and liquid hydrogen, for extended duration tests, the propellant tanks on the test stand are not quite big enough to hold all of the necessary propellant.  So, during the test you actually transfer propellants from these barges into the test stand tanks.  So, the engine is draining the test stand tanks while you are simultaneously re-filling them from the barges.  With all of this going on, you start to appreciate the coordination necessary to pull off one of these tests.


When you’re standing there being halfway cooked by the burn stack, several hundred feet away, the roar from the engine is nearly deafening.  Many people wear hearing protection.  Others of us are aging rock-n-roll fans.  I honestly don’t think that anyone gets a complete sense of how powerful these machines are until they see, hear, and feel one of these tests.  All of the performance numbers in the world simply do not have the same visceral impact as when the engine lights and the initial sound wave runs over, around, and through you and you watch the flame bucket fill with billowing, thick, white steam.  Even twenty years after having seen my first test in person, I still cannot help but stand there like a bedazzled goof and say to myself, “Wow.”

Here, below, is a picture of the test from the other side of the stand.  Why is this important (other than the sign on the fence clearly advertising what you’re looking at)?  Because this is the side from which all of the non-NASA folks, some of the NASA dignitaries as well – including an astronaut representative – and the local press corps watched the test.  

So as to not keep you in any more suspense, the test came off perfectly.  The full, planned duration of 500 seconds was achieved thereby effectively tripling out test experience to date.  The coverage on NASA TV was good.  The bigwigs clapped and cheered with infectious excitement right along the rest of us.  And the press corps wore out their thesauruses trying to capture just a slice of the actual experience.  It was a complete success on all fronts.

Congratulations to the Pratt & Whitney Rocketdyne J-2X development team, the NASA SSC test crew, and the NASA Marshall Space Flight Center project management team.  While I had every bit of confidence that we’d be successful, with so many people watching our show-and-tell exercise, those 500 seconds — eight minutes and twenty seconds — ticking away on my stopwatch seemed like a whole lot longer.  Whew! and Yahoo!



J-2X Extras: Here Comes the Bride — Vehicle Integration

I just got back into the office a few days ago after a long weekend in Philadelphia. My wife’s niece got married. It was a beautiful venue and a moving service and good food, fabulous band, great party, and celebratory beverages flowed freely. Our niece looked gorgeous and her new husband was suitably handsome. A good time was had by all! Congratulations Ashley and Carmello!

And in the midst of all these festivities, an analogy came to mind regarding J-2X (well okay, perhaps not truly in the midst of the festivities, but certainly as part of the next-day hangover).  It’s not a perfect analogy, but it kind of works on a couple of levels.  I am referring to engine-to-vehicle integration.  Here, follow my thinking…

The wedding itself is a great big project.  Everything needs to be figured out, from the biggest stuff (Where?  When?  Who to invite?) to the finest details (What food is to be served at the cocktail hour?  What are the different lighting schemes for the service and for the dinner?).  So too is the development and launch of a great big launch vehicle.  When the engine and the stage come together and the mission comes off as planned, it’s beautiful.  Launch day is just like a well planned, well coordinated wedding.

Also, beyond just the singular event of the wedding day, there it the issue of everything that follows, i.e., the marriage.  And that is a matter of compatibility.  The most spectacular venue for the service and the best food for dinner and the grooviest band for the reception doesn’t guarantee happily ever after.  Things have to work together on many levels in order for success to be found in a match, whether that’s two people married or the engine and the stage coming together and successfully fulfilling a mission.

(Okay, so how’s that analogy working for me?  Not bad, huh?)

So what’s “vehicle integration”?  Well, it’s lots of stuff.  On the one hand, it’s the basic engine requirements.  After all, who says that J-2X ought to generate 294,000 pounds-force thrust at vacuum conditions?  It’s not as if us engine folks get to randomly pick a power level requirement out of thin air.  It comes from an integrated, comprehensive mission analysis of the vehicle.  While we like to think that the engine folks run the world, the truth is that without a vehicle and a mission to dictate requirements, we’d be nothing more than an expensive science project.

But beyond this, how do we interact with the stage?  I would suggest that there are four essential categories of interaction:
• Integrated analysis
• Boundary conditions
• Induced environments
• Operations

The first area, I’ve already discussed in part.  Integrated vehicle/mission analysis is used to establish the basic requirements for the engine.  Beyond that, though, you have other analyses such as contingency and hazards analyses that examine what happens if something goes wrong.  How should the vehicle respond if there is an issue with the engine or the stage or with how the engine and stage interact with each other?  So, in addition to defining upfront what the pieces should do, integrated analysis looks at how the actual, designed parts will interact under different circumstance.  Note that an output of integrated analysis often leads to the category of induced environments discussed below.

Next, you have boundary conditions and these are the most straightforward consideration.  In order to figure out what you need here, all you have to do is draw a box around the engine and see what stuff has to go into or out of the box to make the engine-vehicle combination work.  In fact, that’s basically how we started in creating the Interface Control Document (ICD) for J-2X.  That’s where you capture all of the agreements between the engine and the stage.  Here’s a piece of that “what’s crossing the box” diagram:

This diagram shows the fluids (liquids, gases) that cross the interface with the stage.  You have, of course, the propellant flows of liquid hydrogen and liquid oxygen, but then you also have the propellant tank pressurization flows that are used by the stage to keep the tanks pressurized during flight.  There are also gases used for pneumatic control of the valves and to perform purges through different phases of the flight.  There is a dedicated line that handles high-pressure helium for spin-starting the engine.  And then there are drain flows back to the stage for disposal of excess hydrogen and oxygen.  This latter category is necessary for safety reasons since, for an upper stage engine, it’s usually enclosed within the vehicle for much of the mission and you don’t want to build up an explosive mixture of fuel and oxidizer in the intertank area.

For each of these interfaces, we have to define throughout the different phases of the mission acceptable pressure ranges, temperature ranges, flowrates, and fluid qualities (purity, particulate contamination, etc.).  Both sides have to agree that these values at this interface will happen during the mission or else someone might make an erroneous assumption and either the engine or the stage could fail to perform.  Sometimes, we need to specify even more detail to ensure mission success such as the two-dimensional velocity profile of the propellants as they enter the engine.  Something like this can drive significant design effort on one side or the other (or both) so such details are rarely trivial. 

Now, add to this one set of interface just for fluids additional interfaces for electrical power, control and data transmissions, and then the actual physical connections (including not just the forces and moments applied to these connections but the actual physical designs themselves in terms of dimensions and materials, bolt-hole patterns, and seal configurations).  After you’ve done all that — fully negotiated and agreed to by both sides — you then have an ICD, one of the bedrock documents in the life of any engine.  It’s like a really, really detailed marriage license that goes on and on between the engine and the stage: who cuts the grass, who does the laundry, who sleeps on what side of the bed, who cleans the litter boxes, who opens the pickle jars, who has the remote control come football season…

The next area of consideration with regards to vehicle integration is induced loads.  In truth, these are really just another boundary condition, but we often break them out separately for convenience of tracking and documentation.  What we’re talking about here are loads: structural dynamics, acoustics, and thermal loads.  Rocket engines and launch vehicles make lots of rumbling, roaring noise and lots of smoke and fire.  That’s part of what makes them kinda cool (right?!).  But it’s also the kind of stuff that can cause damage if not properly accounted for in the design. 

Above is the output from an integrated analysis looking at thermal conditions of the engine during a stage separation event.  In this case, depending upon the design of the stage separation system, there were situations where the engine was getting exposed to damaging thermal loads.  In other words, the stage was imposing a load on the engine that jeopardized mission success, so the stage design was altered.  All of the elements of the vehicle have to live with the environments created by everyone else.  So, this is not too much unlike figuring out how to live together after getting married.  You learn, for example, that the combined environment of stogie smoke, an overgrown lawn, and blaring NASCAR on television apparently do not constitute the most congenial, constructive induced environment at home…

The last category in my simplified breakdown of vehicle integration is that of operations.  This comes down to who does what, when, and how.  Bringing together a whole vehicle requires quite a detailed set of instructions.  It’s a lot more than “Insert tab A into slot B.”  And the pieces that you’re assembling come from several different project office and different contractors located all over the country.  So, on the one side of the issue is the technical matter of how you do the whole thing, but on the other side, just as importantly, you have the issue of who is responsible for performing the tasks.  With tasks come manpower, roles and responsibilities for facilities and tooling and, before you know it, meaningful expenses.  Thus, (ta-da!) you’ve got more negotiations and agreements and documentation.

So, engine-to-vehicle integration is, in the end, like a long, complex, heavily negotiated, analyzed, and documented marriage.  Perhaps then, other than the documentation part, it’s probably like most successful marriages over the long haul.