Big Dumb Rockets
Attn: Lots 'o pictures (but worth it) - loads slowly !
Following an article I had in the BIS magazine SPACEFLIGHT, I got a number of enquiries from people who wanted to know more about the TRW motor and ohter low-cost efforts. Though not explicitly outspoken, I got the impression that a number of people feel that, if the motor was built then (25 years ago) for as low as $ 30.000,- , then we should be able to build one today for not much more. Perhaps even amateur groups might succeed. As a matter of fact, I feel this way, too, though I am trying to be realistic about the difficulties.
Rather than answer all enquiries individually, I have decided to put a summary on low-cost rocketry efforts in general onto our web site. Some of the information is from Arthur Schnitt's web site (see LINKS), but most of it actually comes from the US Air Force report by Lt. Col. John R. London IIIrd. I hope he doesn't mind. The Air Force has sent out copies of this report to interested parties free of charge (I don't think they have any left now). I find this the more noteworthy considering that there is a lot of criticism of Air Force policies in this report. In my opinion, the US Air Force deserves credit for this truely liberal policy of "Freedom of Speech", which truely identifies it as an instrument of a pluralistic and democratic society. Many thanks again for my copy!
The V2 lesson
Lt.Col. London starts his report with a review of German V2 efforts in the 2nd World War. I know the V2 is a touchy subject, in particular when a German writes about it and any report about the V2 not mentioning the pains and agony and dying of the inmates of KZ Dora is incomplete. However, this is not ment to be a historical V2 article. If you're interested, there is plenty of information in the Web, just look. (I would like to add a brief summary of V2 Historical Sites Worth Visiting). Here, only technical and statistical data shall be of interest.
The V2 is of interest because it was the first and to-date only complex large rocket which was truely mass-produced and launched. The V2 design team had to break a lot of ground and development efforts in terms of people, resources and money were probably of similar magnitude as the US development of the first atom bomb. Practically every bit was technical Terra Incognita. While designing the motor and turbo pump unit was probably difficult in its own right, building and testing a guidance and control system must have been an incredibly complex task. Tubes, coils, capacitors etc. were the only available electronics components. There was nothing like timing, measurement and programm control. Today, with computers and microchips around, this would be much, much easier.
Which brings us to the motor. The motor, interestingly, is a bundle of 18 smaller 1,5 to thrust motors, yielding 25 to of thrust in total. The concept is clearly visible on this schematic drawing and picture of an impacted V2:
The 1,5 to. motor was ready and fully tested from the intermediate A3 study rocket. Rather than designing and testing a much bigger motor from scratch, the team decided to just take a number of these small motors and put them together into a bigger one. There were even studies to carry this principle on and form 6 identical V2 motors into a truely big aggregate (by today's standards) of 180 to thrust to form the A9/10
Thus, we have a very early example of bundeling of identical motors for the purpose of reducing design cost.
The V2 was truely a big rocket in its time, but I am not trying to tell you it was dumb. In fact, a view into the open thrust unit lets one guess how complex it really was. Despite this, a V2 was only 1/10th the cost of a Heinkel bomber, which, on average, was lost after 3 missions. This clearly demonstrates the superior economics of throw-away over reusability in this particular arena. The V2 may be crude by today's standards, but it has all the elements of any modern-day rocket.
With a dry weight of appx. 3 tons and appx. 8 tons of fuel weight and a chamber pressure of only 12 bar (!), the V2 was well in the range of a much simpler pressure feed rocket. Its exaust velocity of only 2000 m/s was not that far away from a (then theoretically available) solid asphalt/KClO3 mixture. Thus, the Nazi's (and probably Saddam) clearly built the wrong type of rocket!
The lesson to be learned, according to Lt.Col. London is a different one, though:
The V2 was developed in difficult times. For a long time, the project was in low gear. Hitler was not interested and access to resources limited. In the production phase, conditions became truely adverse if not to say terrible. The V2 was built under ground by enslaved and mal-nutritioned laborers in a cold, damp and dirty environment where beatings, disease and threats of life were commonplace. Bombings took place. Shortages of material and critical components were the norm rather than the exception. The quality of material shimpments such as steel quality was constantly varying. Clearly, the KZ inmates of Dora had no interest in the success of their work and probably acively sabotaged the program whenever they safely could. Despite all this, between January 1944 and March 1945, a total of 5947 rockets were produced of which 3600 were launched "against the enemy" and 2890 actually reached their target. The V2 thus demonstrated an 80% reliability, a remarkably high percentage considering all the factors working against its success. Obviously, some of this reliability must have been design-inherent. Today, it should be possible to build a similar vehicle at a fraction of the original development cost, with much better performance and without overly pedantic precision and control standards and still achieve reliabilities well over 95%.
Private rocket experiments date back as far as the late 40ies. Here I will present only a few designs, the first having been built by the Pacific Rocket Society PRS of 1987 because it shows a truely streightforward and simple injector system which anybody with a lathe and skilled hands can easily replicate. Pictures show the engine and injector configuration as well as the injector plate.
The rocket itself worked on nitric acid and alcohol (a hypergolic combination) and flew to 3 km altitude.
In 1976, Dave Cristallini (Reaction Research Society RRS) built a LOX/Kerosine rocket which stood 5,6 m high and had a 400 kg thrust regeneratively cooled engine. The rocket flew to 3 km altitude and reached Mach 1 when the parachute prematurely deployed.
Recently, Cristallini has designed another rocket ment to reach a peak altitude of 56 km. Cristallini has done some static tests and has given the total cost of this project incl. the rocket, fuel, static tests, launch tower etc. with $ 6000,-. I am not aware if this rocket ever flew.
I also find Ken Mason's mobile test stand worth mentioning which he designed from surplus components. The test stand can accomodate engines of up to 2 to of thrust. It is mounted on a trailer and Mason supposedly takes it to "public gatherings" to perform tests. He also rents it to aerospace companies.
I must say that Mason has inspired me insofar as that today, at FAR, we also have put our test stand on a trailer, which has already saved us a lot of set-up time (and also get-away time because we have no permits !) when performing our tests.
Lt.Col London mentions a number of other experimenters as well. I have no ambition to be complete and I think the above examples will suffice to clearly demonstrate that the construction of liquid/liquid engines must neither be so complex that only large teams can do it nor must it be so incredibly expensive that only rich countries governments can afford it.
This was the first high-thrust motor ever, specifically built to demonstrate the feasability and reliability of low-cost engines. Even more noteworthy is that TRW is an established aerospace company which should have had an interest in expensive technology. TRW has built this motor as a demonstrator. Several tests were conducted and TRW tried (successfully) to pull other companies (such as Boeing) on board, too.
The background story of this motor is intreaguing. TRW, being an established aerospace company, did not set out to build a simple let alone cheap engine. The requirement came out of the Apollo Moon Landing project. When it came to designing the LM descent and ascent engines, NASA called for a maximum in safety and reliability. Obviously, the worst scenario would have been astronauts crashing "live" into the moon with the eyes of the world glued to their TV sets. Likewise, having astronauts stranded on the moon left to perish slowly because of some ignition failior with no ground crew etc. to help them was deemed unacceptable. Clearly, engines had to be extremely reliable, had to perform over a wide thrust range without combustion instabilities and had to have on/off capability. The choice fell to an engine concept using hypergolics (i.e. self-igniting fuels) naturally, a pressure-feed system and reduced part count. Note this is an extremely interesting point: when it comes to security, obviously less parts are considered the most secure option. Intuitively, we all agree that the fewer parts can break, the more reliable the overall system will be. This is also what the mathematics of fault analysis shows us. However, it is not the type of rockets we currently build for launching people into space!
As we all know, the Lunar Module TRW motors performed well. Key design element was the reduced part count of the engine (by a factor of 100 times fewer parts than in conventional engines). It turned out, quite naturally, that these simpler engines were also cheaper to build. TRW engineers wanted to find out wether one could carry this principle to bigger engines as well. It did indeed turn out that the size of the engine is irrelevant. This is an interesting finding. There is no upper limit to simplicity.
The engine was designed for 250.000 US pounds of thrust or apprx. 110 tons or 1,1 MN. It was built in 1966. The actual hardware job was given to a local pipe and boiler fabricator and the engine was built do "shipyard production tolerances". Only the injector plate was made in-house. The injector type used was a so-called coaxial pintle design. This injector type was pioneered by TRW and has never suffered a catastrophic failior from combustion instability. Its characteristic is a high degree of throttability without need for a corresponding injector pressure drop. This type of injector is not trivial, though.
The total hardware cost was given at $ 33.000. Total project cost i.e. incl. all tests etc. was $ 60.000, an unusually low figure for aerospace work. Later, ablative liners were added for another $ 60.000. To put all this into perspective: Ariane space has just spent more than 1,000,000,000 $ for an 80 to thrust engine for the Ariane 5. This is a much more sophisticated LOX Hydrogen engine, true, but who's interested? Back to TRW: C* (a measure for efficiency) was measured to 95 %. Throttlebility was also demonstrated.
The following paragraphs are taken from Arthur Schnitt's web site (see our LINKS).
"There was an amusing but instructive side to this program. TRW farmed-out the fabrication of the engine and its supporting structure, less the injector that they fabricated themselves, to a "job-shop" commercial steel fabricator located near their facility . The contract price was $ 8000. Two TRW executives visited the facility to observe the fabrication process. They found only one individual working on the hardware, and when queried, he did not know nor care that he was building an aerospace rocket engine."
Several tests of this engine took place, one of them at the Edwards AFB.
(Schnitt) " I had arrived late to witness the test, and only saw the firing. I was told by others who witnessed the entire test procedure that the engine was pulled out of outdoor storage where it lay unprotected against the elements. Before it was placed on the launch stand, the test crew dusted off the desert sand that had clung to it. This unplanned inlcusion of a bit of an environmental test also demonstrated hardware ruggedness of the kind no other liquid rocket eingine could approach."
As mentioned above, TRW also got a few other companies interested. Namely Boeing carried the idea to the next step: an engine doesn't make a stage. Rather, you need at least propellant tanks and a support (or thrust) structure. Could that be build at low cost as well? A commercial tank builder was assigned the job. The tank was sized to make a complete non-recoverable stage to fit the TRW 1,1 MN engine. Cost was in the order of a few $ per pound of tank weight (a number to remember for screeening calculations). This, obviously, including a commercial mark-up. Boeing people observed the manufacturing process, documented the techniques and tooling used and were able to produce a similar tank themselves in-house at even lower cost. The tank-engine configuration was demonstrated in static tests.
Interestingly, the two companies tried to get president Nixon to look at the stage on one of his West Coast visits, but were unsuccessful.
Other motors and concepts
Following the TRW motor experience, for a brief time, there must have been something like a craze about cheap rockets. Many companies presented their ideas, inluding Boeing's 3 stage double bubble design (picture) with a 15,5 to LEO payload capacity. Cost per kg was given at $ 930 per kg (i.e. less than 1/20 th of Space Shuttle cost). A complete double bubble tank was indeed constructed for demonstration purposes.
Other concept studies came from Chrysler (45 tons to LEO at $ 40 per KG), McDonell Douglas (same capacity, apprx. $ 80 per kg), North American Rockwell, Martin Marietta and others.
While it is certainly not worth presenting all those ideas in detail, it is noteworthy, that practically all aerospace companies with a reputation took on the idea of low cost rocketry. This is good evidence that they all thought the concept viable and essentially correct and that they wanted to be ready for a shift in NASA buying policy. It never came. NASA decided to build the Shuttle and that it would get all payloads. Hence, there would be no market for a low-cost vehicle and consequently all efforts in that direction died off. This was early in the 70ies. We know how the story continues. We don't know, however, how the story will evolve from here. The idea of low cost rockets and minimal cost design MCD is definitely undergoing a renaissance. Today, there is a true world market for payloads and launch services (which never existed before), there are private interest groups and there is an incredible proliferation of rocket technology. More than 30 countries strive to build them, and it is not just the sinister middle east dictator type. Countries like Brazil, India, Indonesia, China etc. are going for that lucrative future market and they are not going for the expensive high-end technology (perhaps, because they can't afford it). If they are successful with a poor man's rocket concept, they will beat us right out of the market.
The above picture (perhaps not very well recognizable) shows an interesting study motor of Rockwell International (who later built the Space Shuttle). It is a very modern concept, revived, though in higher sophistication, to power the X-33 next generation Space Plane (if it ever comes). The motor is a so-called linear expander. The exhaust expansion nozzle is, if you like, cut in half with one half practically non-existent (it is the atmosphere). The other half (the hardware half) is not the shape of a cone or any rotation-symetric body, rather, it takes the form of a linear curved surface. Such a surface is very easy to make (as you can guess by the picture), just flex some sheet metal. It weighs less and, most important, it always gives the motor the right expansion ratio at any altitude. In other words, it works at sea level and in vacuum without any efficiency loss, whereas classical engines are optimised to one altitude only! Thus, it is a truely modern concept.
The above motor was rated at 890 kN thrust (appx. 90 tons) and had the dimensions 3 x 3 x 2,5 m. The motor has two opposing rows of engine chambers. Actually, each row is made up of 10 individual chambers of rectangular shape (i.e. 0,3 x 0,3 m each). High thrust is achieved by bundeling the 20 in total together (any signs of V2 design ideas around here?). An individual chamber can easily be built and tested. If still higher thrust were needed, one might simply add more rows or chambers or both.
Clearly, if I had to design a cheap and simple LOX / Kerosine engine, this would be it!
The benefit and true value of the TRW work and other studies was that feasability of low-cost rocketry was actually demonstrated rather than just postulated on paper. It is work we should remember and that's why I write about it.
Recent efforts - Microcosm
After it became evident that the Shuttle would not meet the anticipated launch cost reductions, there were a few more half-hearted cost reduction studies. In 1980, for ex, the Air Force reapproached TRW to do a study for a 30 ton to LEO vehicle (i.e. roughly Space Shuttle capacity). Plans for the so-called Star Wars defense (offense ?) system gave theese studies a boost. TRW drew back on their original 1960ies work. As we all know, however, nothing was ever built.
Interestingly, there are a number of private groups (such as the Millenial Foundation, Space Launch Foundation etc.) expressing serious interest in designing their own vehicle (see our LINKS section). Their motives seem to be a mixture of commerce, live-free, create-a-better-society and what not. To my knowledge, no hardware was ever build. Then there are the true amateurs. If we could somehow bundle efforts .....
Among the more recent serious attempts I would like to mention Microcosm, of Torrance, CA. Microcosm is a semi-private group, i.e. they also receive funds from the government, essentially NASA. A quote of $ 18.000.000 spent so far was given, so this is perhaps not truely low-cost. Microcosm plans a family of low-cost vehicles (the Scorpius family) for a commercial market, starting with sounding rockets to a 200 km altitude up to LEO capability.
Microcosm has done impressive practical work. It has developed an engine in the 2 ton thrust range which, except for the injector plate, is made totally out of composites. Part count is by a factor of 30 lower than in traditional engine design and engines have held so far for a 110 sec fireing period "with very little erosion of throat and chamber walls" and total part count is given at 18! The engine is quoted to cost less than $ 5000,-. Microcosm says, there are no low-tolerance or precision parts.
The use of composites is also applied to the tanks, which reduces weight drastically (in fact, we use the same building method for our LOX tank, which has a very thin inner bubble of stainless steel with a fiber glass filament pressure case around it - see PICTURES).
Microcosm consequently employs the bundling principle, as seen in the picture. The strap on boosters and central unit are identical, thus takeing standardizing hardware to the max. Interestingly, fuel is consumed from two opposing strap-on's at a time and fed to all engines. When empty, those two strap-on's are jettisoned and so on. Microcosm thus achieves a high degree of stageing. If you have 6 strap-ons as in the above picture with 2 center stages, you actually have a total of 5 stages. This yields a very high mass ratio which can compare well with a turbo pump fed 2 stage'er.
Microcosm also has done work on a standardised guidance and control package which can migrate from vehicle to vehicle (another good idea - ignoring this one has killed the first Ariane 5!). All in all, I would say, of all groups I know, Microcosm is "most likely to succeed" (in High School year book terms). My latest information on their efforts dates back to mid 1997 and I am not aware if they have flown anything to-date. Check for yourself in our LINKS section.
The biggest rocket ever
In the early 1960ies and before the MCD criteria were formalized "in writing", Aerojet anticipated a low-cost and truely big dumb rocket: the Sea Dragon. It seems their gut feeling told them the way to go. To date, this is the biggest rocket ever seriously postulated. Sea Dragon was a two-stage reusable vehicle, the first stage using LOX and Kerosine, the second LOX and Hydrogen. The first stage would have made a parachute ocean landing. Both stages were single engine pressure fed (Aerojet's studies suggested that development of a single large engine would be simpler and reliably higher than that of a cluster of smaller engines). Sea Dragon was truely colossal. The mere numbers speak for themselfs: The Sea Dragon was to have had a payload capacity of 550 tons to LEO with a take-off weight in the range of 20.000 tons! The vehicle would have stood 168 m with 23 m in diameter. Twice those numbers and you have the TITANIC. No doubt these are ocean vessle dimensions rather than those of a flying machine (stressing my point that in rocketry we talk about ship building rather than airplane manufacturing techniques). Indeed, the Sea Dragon was to be built on the ocean front in a ship yard and then towed to the launch site. Launch site is actually saying too much, because the Sea Dragon would have launched, partially submerged, directly out of the ocean (this concept had evloved out of Navy experiments with submarine based missles). Ballast tanks would have positioned and trimmed the rocket properly before ignition.
Aerojet calculated the cost to $ 59 to $ 620 per kg. NASA had an interest in the Sea Dragon largely because of its large payload capacity. It had the cost calculations independently reassessed and they were largely confirmed. Immagine a $ 100 kg ticket price! In other words, we could fly into space for $ 10.000. Perhaps not cheap but well within range of affordability.
Too bad it didn't come true. As NASA's planetary ambitions shrank (to practically zero), the Sea Dragon was moth-balled and eventually forgotten.
(c) FAR 02/1998