2004 Discussion

CSXT - Civilian Space eXploration Team

The REST OF THE STORY first civilian amateurs fly the CSXT rocket to space

2004 Discussion - launch recollections with Tony cochran

At the CSXT Space Flight in 2004

by Tony Cochran

Here is my story on the part I played helping out on the successful 2004 Space Flight attempt by the Civilian Space eXploration Team (CSXT). While I did not build or design any part of the rocket, I was able to contribute in the ground support and wind balloon efforts. The times I spent out in the Black Rock Desert on all of the various launch attempts is a memory I still think back on frequently and am glad to have been on the team.

And I almost missed it all. In May 2004, I had been on vacation driving around the west coast of Ireland for several days seeing the sights. When I returned home and checked my emails for the first time in two weeks, there were several messages saying that the next CSXT launch attempt was a go for next week! The team had moved the launch window dates up a few months in an attempt to beat Burt Rutan and his SpaceshipOne as the first amateur team to launch a rocket into space. So, I made some hurried phone calls and plane reservations, and a few days later was on the playa.

I’ll say right up front that the flight was a great success and accomplished all we had set out to do. It seemed like everything came together that day perfectly to get the launch off. It was the first day of our launch window, the weather was great, and the playa was amazingly dry for May. After all the struggles in previous years, things just seemed to fall into place. This excerpt is from an email I wrote a few days after the launch. I was quite enthused:

Indeed it was a success, beyond all imagination! I just got back into

town from Black Rock, where Bruce Lee and myself helped out in the launch

of the first private, civilian, amateur, or whatever you want to call it,

rocket launch into space. The team performed exceptionally well together,

with many important contributions all around. The question on everyone's

mind is how high did it go? Jerry Larson, launch conductor, was working

late into last night downloading and calculating the exact altitude

reached. Some info I found on Rocketry Online today after getting home

says we reached 77 miles high, so Jerry must have been successful in

calculating the altitude and released that figure. Woo-hoo!! That number

exceeded our expectations and breaks all possible definitions of "space."

As a follow up, we later determined that 72 miles is the generally accepted and verified altitude that the rocket reached. Below I will go into a day-by-day summary of how we had such a successful campaign.

Friday, May 14: Bruce Lee and I arrived in Reno, and as one of our usual assigned duties we went to pick up the helium tanks used for the wind balloons, and the porta-potty. Two important items indeed. We met up with Ky Michaelson and Matt Murphy, who had driven up from Florida with the rocket motor, and had our annual all-you-can-eat sushi feast. After gorging ourselves, we stayed in Reno that night.

Saturday, May 15: Bruce, Ky, Matt and I drove up in two vehicles to Gerlach, NV on the edge of the Black Rock desert. We checked into Bruno's fine motel; it was the same as ever. Bruno himself was as cantankerous yet friendly as ever. The food in his restaurant was good this trip, they had a new cook. On leaving the restaurant we discovered a shredded tire on the trailer hauling the rocket motor. Somewhere in the 100 miles between Reno and Gerlach the tire had destroyed itself. A new tire was found and installed at the gas station. Also today the GoFast! RV arrived in town, they were one of the main sponsors for the project, and sent out their fancy, duded out RV and a helicopter for support.

Sunday, May 16: We unloaded the porta-potty and helium tanks at the launch site in the middle of the dry lake bed, not far from the eight-mile entrance. We also hooked up to the launch control trailer and hauled it out. The morning was spent in the job of putting together the launch tower, something that Bruce and I have gotten quite familiar with doing. The launch tower has to be bolted together piece by piece, the side legs with foot pads are slid onto the tower, a blast deflector is laid down, and two floor jacks are put out which will be used to adjust the angle of the launch tower for the winds.

The rest of the team was busy with putting the rocket payload, fin can, and motor sections together. One starts with the motor, the fin can is slid on and bolted to the motor, then the payload section with nose cone is loaded and attached to the top of the motor. After finishing assembly, we all then loaded the rocket into the tower and rotated it up to vertical. This involves a lot of people lifting the rocket with just their hands, sliding it slowly into the launch tower, then lifting the launch tower from horizontal to vertical, with the rocket inside. The rocket itself was quite good looking, red and black, with nice decals and red anodizing.

Monday, May17, launch day: This was the first day of our approved launch window. One of the requirements for the launch campaign was that we do a mission dress rehearsal of our launch procedures, so we did that first thing in the morning. We got out there before sunrise like we always did, in the cold, to start getting ready. The team went through a nominal dress rehearsal of the launch that morning, we went through the 45-minute countdown procedures and performed all the actions required leading up to an actual launch.

Bruce Lee was the Range Safety Officer of the team; he had to give a safety speech over the PA at the start of the countdown. He then had to monitor all vehicle traffic on the playa, and also coordinate with the BLM and two train-spotting teams located ten miles away, helicopter and surveillance airplane. My part was to fill and launch the balloons used to measure the upper winds just before launch, then be out at the launch pad to help Jimmy (Jim Hoffman PM) rotate and adjust the angle of the tower into the preferred position, based on the winds.

There were a few typical Burning Man people at the launch control site with us, along with quite a few other visitors. Tripoli Rocketry Association members included Chuck Rogers, Fred Brennion, Jeff Jakob, and a few others. Everything went fine in the dress rehearsal, except for a glitch receiving telemetry from the avionics unit. An hour or so was spent troubleshooting this, before it was resolved. The clouds were clearing and the winds were light, so we decided to go ahead with the actual launch, and started the countdown.

The trains cooperated by staying out of the area, and my part of launching the wind balloons went smoothly. As we approached the final two minutes of the countdown a couple of motorcycles appeared in the distance, so we had to hold for a minute or two as they drove up to our launch site and out of the way. There was another last minute concern when a truck with railroad wheels came rolling down the track right when we were getting close in the countdown, but he kept on driving south and left the area.

We resumed the countdown and at 11:11 am the button was pushed and the rocket took off! It had a deep roar and boosted straight. I was just watching and waiting to see if the rocket and motor were both going to stay together, and they performed well. There was a small chuff at the end of the burn that turned out to be benign. The rocket disappeared as soon as the motor burned out, and everyone knew that the flight was going to be successful.

There was much whooping and hollering. The trackers heard two separate pinging signals, indicating a successful separation of the payload section! There were two sonic booms several minutes later (from the two returning pieces) and we all knew the flight was a success. Flight data later confirmed this. The tracking team and the helicopter spent the rest of the day narrowing down the location that the payload section came to rest, but ran out of daylight. Everyone was feeling quite good, we made it to space!

Later that afternoon Bruce and I had the joy of taking down the launch tower in a raging wind and dust storm, in whiteout conditions. We tore apart the launch tower and threw the pieces into the trailer. Talk about being dust covered... That whiteout was one of the worst I have experienced, but not as bad as the June 2002 wind storm that CSXT experienced during our previous launch campaign, that wrecked the camp. One side note: At ignition the launch tower had shifted (luckily in the right direction) and the legs got bent but held together long enough for the rocket to leave. What happened is that after Jimmy and I had set the launch tower into its final position, there was a general rush to get everyone out of there and back to the launch trailer to continue the countdown. In this rush we neglected to screw the foot pads down to the ground, which would have helped support the load. So, there were only two floor jacks holding the tower at the correct angle, and at motor ignition the blast knocked one jack clean out and one corner of the tower hit the ground. But the rocket had already left.

Tuesday, May 18: The day after the flight was spent recovering the payload section and downloading data. The parachute lines had some bird trackers attached to the parachute, so we had a general idea where to look. The payload location had been narrowed down to be within a small valley/canyon in some rugged terrain, in the mountains north of the launch site. We drove out 20 miles or so, down a gravel road, up a side road for six miles, up a four-wheel drive road for a few more miles, and then had to walk up the valley. Bruce and I followed the recovery team out to the suspected landing zone and eventually started a search with about 15 other people.

There had been talk of organizing a grid search pattern, but the CSXT members just sort of scattered, went up the hills and started looking. After a few hours one of the team members found the payload almost at the top of a mountain in a rock field on a 45-degree slope. The climb up there was very tough, but several people made it up. Scott from the Pinnacle Video crew carried a full-sized video camera up there. The payload and nose cone were imbedded in the loose rocks, with very little damage, just some gouges in the sides from the rocks. The parachute was in good shape too.

We hauled it down, took it to Bruno's and opened it up. I now have a few small items that have been in space! Bruce recovered his credit card which also was in the payload section. Tripoli member Fred Brennion had put his driver’s license in the payload.

The above is a summary of my contribution to the project, other people have similar stories to tell of their parts. I had a great time, and was glad to be part of history. If the rocket ever ends up in a museum, I will be able to go look at it and know I was there when it happened.

Tony

2004 Discussion with motor designer/maker derek deville

This is the story of the final and successful launch with minor design modifications and a major motor change. This is the full article written by Derek Deville.

Putting the 'S' in CSXT

Making the Largest Successful Amateur Motor Ever

By Derek Deville

The Quest for Space

In 1995, with Ky Michaelson at the helm, the Civilian Space eXploration Team (CSXT) was formed with the sole mission of building and launching the first amateur rocket into space. In 1996, their flight of the Joe Boxer Rocket to about 66,000 feet primed the team and helped initiate some momentum.

In August 1997 the team made their first space-launch attempt with a 90-pound 6-inch-diameter two-stage vehicle. The first stage worked; however, after staging separation, the second stage failed to ignite. The abbreviated flight peaked at an estimated 77,000 feet, and the upper stage reentered supersonic and impacted the playa of the Black Rock Desert.

Only three months later, in November 1997 the Cheap Access To Space (CATS) prize was announced, promising $250,000 to the first private team to launch a 2kg payload into space, 200km or higher, using a privately developed launch vehicle. The flurry of activity surrounding launch-waiver applications complicated the process, and everyone was having trouble getting waivers, including CSXT. Aware of this, Jerry Larson, an aerospace engineer, offered to join the CSXT team and input his expertise in launch simulation. Jerry was immediately appointed the project manager, and he began work on the next vehicle.

By September 2000 the new vehicle was ready. It had lost the second stage but had grown significantly, into a single-stage 434-pound beast that was nearly 9 inches in diameter. With the rocket now powered by an R18,000 motor containing 228 pounds of propellant, space seemed within reach. At launch the rocket boosted well and was on its way to space when at 45,000 feet one of the fins succumbed to the high loads and broke free. Unstable, the vehicle tumbled and broke up. On a positive note, the team did set a new amateur speed record of 3,205 mph.

In November 2000 the CATS prize expired. Until this point Ky had primarily funded the CSXT project, but the expense was getting to be too great. A skeleton of the next generation rocket was displayed at a trade show, and new sponsorship was found. Primera, a leading manufacturer of CD/ DVD duplicating hardware, came to the table with funds to help put the team on track for another launch.

The 2001 vehicle had undergone a significant growth spurt. It now stood at 18 feet tall with an S20,000 power plant containing 300 pounds of solid propellant; the fins had been reinforced; and the design had been carefully reviewed. The launch was scheduled to coincide with BALLS at the end of September 2001. However, the 9/11 terrorist attacks caused the FAA to temporarily rescind all waivers. The launch would have to wait.

Rescheduled for June 2002, the 550-pound vehicle had its sights set on 62 nautical miles, the edge of space. As fate would have it though, this vehicle would not get the chance to prove itself. Winds of up to 60 mph (editors note: someone on site had an anemometer and measured just over 70 mph) pounded the camp for the three-day window of the difficult-to-acquire FAA space-launch waiver. To make matters worse, when returning from the playa, Ky found out that his mother had passed away. She had been one of his primary sources of inspiration and a key motivation for the space shot. With his spirits dampened, Ky nearly pulled the plug on the project. As time passed, Ky regained his desire to fly and was even more determined to succeed, to dedicate the flight to the memory of his mother.

In September 2002 the Primera rocket finally got its chance, but luck was not on CSXT's side: three seconds into the motor burn, hot combustion gases penetrated the casing, and the rocket broke up.

Ever determined, the CSXT team built an updated version of the Primera vehicle. The financial support was now coming from Go Fast Sports, the maker of the sports energy drink by that name. The Go Fast vehicle was ready in late 2003 for a flight attempt, but the propulsion was not. Rick Loehr, the long-time propulsion provider for CSXT, stepped down.

The team now had a vehicle without a motor and a propulsion system design they were unsure of. They began the search for new propulsion. CSXT was open to exploring other solutions. Hybrids were the other logical alternative. Ky had been friends with Korey Kline and had kept in contact with him since they had both been active participants in the CATS prize and subsequent forum discussions. Knowing that Korey Kline was the inventor of Hypertek, cofounder of Environmental Aeroscience Corporation (eAc), a hybrid-propulsion R&D firm, and considered by many to be the father of modern nitrous oxide hybrids, Ky had great faith in his abilities.

In November 2003 Ky and Jerry contacted Korey to express their interest in exploring a hybrid rocket-propulsion system to use in the next space shot. At that time, Korey and I were just finishing up our participation in the Burt Rutan SpaceShipOne program and had the necessary time to review the CSXT situation and do some design work. We began serious discussions with CSXT about how big the vehicle would have to be and what would be involved in developing the appropriate propulsion system. After several weeks of design simulations, we finally determined that a slightly downsized version of the Hyperion-2 (a 12-inch nitrous oxide hybrid) would be the best choice. We examined cost, schedule, risk levels, available hardware, previous development testing data, and oxidizer availability and storage. We would run motor simulations developing predicted performance data that Jerry would then input into flight-simulation software to model the flight.

The design began to become fleshed out. Avoiding exotic fuels, we stuck with standard well-known pre-characterized propellants. Burn time, thrust, and total impulse were narrowed accordingly. A 150,000 lb-sec design with a 20-second burn time began to surface.

Hybrid-to-Solid Switch

After several months of work, we discovered that the thrust to weight ratio, while high enough to be safe at liftoff, wasn't providing enough acceleration to offset weather cocking. The slow takeoff was allowing the winds to play too great a role in the trajectory. The possible downrange landing zone according to six-degrees-of-freedom (6DOF) trajectory analysis was wider than Black Rock could support. We were faced with a difficult decision: either continue with the hybrid design and find a new place to launch from, or change course altogether and develop a solid motor with a higher thrust that would overcome the wind sensitivity.

Jerry inquired as to whether Korey and I believed that we could make a solid motor big enough for the CSXT Go Fast rocket. We both had significant solid-propellant experience, but this was a whole new game. This was going to be the largest solid motor we had ever built, and, if successful, it would be the largest successful amateur motor ever. With this in mind we replied to Jerry with a timid yes, and so the design process began anew.

We agreed to participate with one caveat. There would have to be enough time and funding to do a full testing program, including sub-scale and full-scale static testing. Korey and I agreed that, by in large, many of the failed large amateur motors did so because they had not taken the time to go through the design process properly, with follow-through to full-scale testing.

The official change to a solid motor came in February 2004, and the anticipated flight was scheduled for May. Time was very short; we would have only three months to complete all of the design, propellant characterization, sub- and full-scale testing, and to make the flight motor.

Designing the S Motor

Jerry was handling the primary vehicle design. He would be getting the fin-can and nose cone machined, along with the payload section. We would be responsible for the motor-casing machining, nozzle design and fabrication, and propellant casting.

The first order of business was to determine vehicle sizing. I began by reviewing my existing propellant formulations, taking into account the average delivered density and specific impulse. From this I worked with Jerry to determine a preliminary required propellant mass estimate. Then we began a more detailed design. Running a number of flight simulations, we iterated the fuel load and inert masses until a believable picture came into focus. The design called for 425 pounds of propellant in a vehicle with a gross liftoff weight (GLOW) of 700 pounds.

Throughout this process we were scaling down large-motor technology rather than scaling up small-motor technology. This led us to a case-bonded monolithic grain design; the method most commonly used in military and space systems that contain over 100kg of propellant. A single continuous grain of propellant that is bonded straight to the casing wall offers several advantages. In this case, the primary benefit was the increase in structural strength to allow the grain to survive the high g-loads at launch and not collapse under its own weight.

We needed the motor to yield a neutral thrust trace. This can be accomplished in a single continuous grain motor by using port geometries. In a round port, as the propellant burns away, the port circle will grow larger in diameter and will expose more propellant, increasing the chamber pressure and thrust possibly to uncontrollable levels. We needed to have as nearly constant as possible amount of propellant surface exposed. This can be done with an irregular-shaped initial port that exposes a larger area, such as a star-shaped port. If done correctly, the change in area over the duration of the burn can be minimized. Examining the choices of port shapes that would work in this case, we narrowed in on what is known as a finocyl geometry. This is essentially a round port or cylinder that has rectangular or nearly rectangular fins around it, therefore being fins on a cylinder, or finocyl. This geometry can be easily fine tuned to the desired thrust trace and has the additional benefit of being fairly straightforward to fabricate.

The centered port geometry without exposed ends allows the propellant to act as an insulator for the casing for a large part of the burn. This would minimize the need for inert liners that reduce the available volume and add dead weight. A secondary benefit was the elimination of the need for custom-fabricated casting/grain liners, further reducing the cost and preserving a minimal timeline.

The final design called for 10,000 pounds of thrust, for a 10-second burn time, with an average chamber pressure of 800psi. This meant a total impulse of 100,000 lb-sec or 445,000 N-sec, a mid S motor.

Two parallel processes began at this point—the formulation and characterization of the propellant and the design and fabrication of the full-scale motor casing and nozzle.

Propellant Formulation and Characterization

On the propellant track, the designed burn time required that my best- characterized fuel be reformulated to slow down the burn rate. My standard DEAP 6 propellant has an average burn rate of .25 in/sec at 800 psi. Here we needed less than .2in/sec. You don't have to start worrying about the strength of the propellant until you get over 72% web fractions. In this case it was at 75%, so we had to be a little concerned. We reviewed the available literature on propellant strength and contacted a few propellant experts to determine what physical properties would best suit this application and how to optimize the propellant formula to this end.

The final formula had 78% solids loading. The binder system was composed primarily of HTPB with only 22% of the binder being plasticizer and 1% of the binder being bonding agent. This was done to ensure good tensile strength of the propellant and sure bonding to the casing, along with easy processing and pouring. The bonding agent was used due to the high level of large-particle AP that was necessary to slow the burn rate. The complete formula is shown in the attached table (located in the Derek Deville photo gallery).

The aspect of the motor design that had the greatest uncertainty for me was erosive burning. I had done significant research into the issue of erosive burning but had not yet developed any models or methods for predicting or accounting for the phenomenon. I refer to it as a phenomenon because it does seem to be a somewhat mysterious effect, with a kind of black magic feel to the lore that surrounds it.

Erosive burning is tricky to define clearly. Generally, it is defined as the increase in burn rate, usually in high aspect ratio (long and skinny) motors, which occurs at the beginning of the burn due to . . . (and here is where it gets tricky) . . . the high velocity of material through the port creating a scrubbing effect. Or maybe it's because of the high heat flux into the propellant. Or maybe it's because of the high mass flow rate. Or maybe, just maybe, it's a little bit of all of them. The one thing everyone agrees on is that it happens in long motors when their ports are the smallest, at the beginning of the burn.

Knowing this, I had to be on the lookout. The base grain was 9.5 inches in diameter at 165 inches long, yielding a grain L/D of 17:1. The port was even worse, with an area equivalent to a 5.2—inch-diameter circular port; the 165-inch length netted a nearly 32:1 port L/D.

Given that I have come to believe that erosive burning is a combination of all of the mentioned effects and that I was definitely going to be experiencing it, I still didn't know how to account for it. Enter Charles Rogers. I got permission from Ky and Jerry to bring Chuck on board. I relayed my situation to Chuck and told him of the design work that had been done to date and what my anticipated problem with erosive burning was. And let me tell you, I found out I'd gone to the right guy.

Chuck pulled together a number of resources and was able to sift through the technical language and boil things down to manageable terms. Chuck explained two easily calculated values that would define the onset of erosive burning and could also be used to assess the degree of increase in local burning rate. The two magic numbers were the local mach number and the local mass flux at the nozzle end of the grain. In its simplest terms, if either the velocity of the flow in the port exceeds 0.4 Mach or if the mass flux through the port exceeds 1 lb/sec/in2, there will be an increase in burn rate. The amount of increase is a part of the propellant's sensitivity to erosion, which ties into the propellant's physical characteristics, its chemical makeup, and its normal burning rate. Generally harder propellants composed of smaller particles and propellants with higher burn rates are less sensitive to erosive burning; however, this was just the opposite of the propellant I intended to use.

With Chuck's assistance, I was able to tailor the small-scale characterization motors to identify the burn rate at non-erosive conditions and to mimic the erosive conditions that we expected to have in the full-scale motor. This allowed us to get empirical data that could be correlated.

The first few series of tests were performed with 3-inch DPS hardware. This data was analyzed by Chuck, Jerry, and me and was very useful in determining the propellant burn rate coefficient and exponent, and the increases in burn rate due to erosion. As a side benefit, Chuck was able to use the data in his nozzle performance study, which was also used to optimize the nozzle for the CSXT motor.

The next step was to fire a 6-inch subscale version. This firing experienced a minor anomaly with a small portion of the nozzle's expansion cone separating during the test. The all-graphite nozzle used had been fired several times before and had seen better days. In spite of the separation, the data still proved to be very useful. The increase in burn rate as a function of erosion was clearly visible in the thrust trace and followed our predicted pattern. The burn rate, thrust, and chamber pressure were right on target. We were ready to build a full-size motor and move on to the full-scale static test.

The full-scale motor was designed with a baseline thrust trace that was progressive for the first few seconds of the burn. This was done so that the amount, by which the thrust level was reduced, due to reduced exposed propellant surface, was offset by the increase in burn rate due to erosion. The net effect of these things combined would lead to a neutral thrust profile.

Nozzle Design

The prior nozzle design which followed typical amateur methodology was a full-diameter billet of graphite that was machined into shape and insulated from the casing using a section of the phenolic liner that was previously used to cast grains into. There were two reasons why we chose not to reuse this design. First, given the fact that we were no longer using phenolic casting liners, there was no liner to borrow from. This was only a minor problem, given the fact that a small liner section could have been fabricated in short order if desired. The more important reason is the tendency for large pieces of graphite to fail under these conditions. Given the brittle nature of graphite, large thick sections of material must be left in place to bear the pressure loads on the casing. This creates the second and more likely to occur problem of differential heating. As the plasma flows through the nozzle the exposed surfaces are rapidly heated. The material begins to expand according to its coefficient of thermal expansion (CTE). However, the outside sections of the graphite are not experiencing the same expansion, and so large internal stresses build up. Eventually the stresses get too high, and the graphite fractures. This is likely one of the reasons for the fracture in the 6-inch sub-scale motor. I have also witnessed this happen in other 6 inch diameter motors, so in the case of a 10-inch motor, the risk was unacceptable.

NASA usually solves this problem by using thin layers of graphite or carbon-carbon that are supported by metal structures. That was out of the realm of possibility for this project for a number of reasons, primarily cost and time. The other option was to use composite structures to support a smaller graphite section.

I had done a lot of composites work in my hobby rockets. My structure of choice was carbon fiber with epoxy resin. Using this base knowledge, I began to explore the more advanced fabrication materials and methods. Through the use of vacuum bagging, autoclaves, and high-temperature cure resins, it is possible to make composite structures that can withstand the conditions that exist within a solid rocket motor. Using these specialized composite techniques, we could insert a heat stable graphite section that could survive the nozzle conditions, inside a composite section that would provide structural support and would extend aft as the expansion cone. This advanced nozzle design required the acquisition of new ablative composite materials. The basic design was scaled down to a 6-inch motor and was endurance tested in two long-burn hybrids. Once comfortable with the material fabrication and confident in its abilities, we chose to scale the design up to the 10-inch CSXT nozzle. A large male mandrel was fabricated and treated to withstand the heat and pressure of the curing cycle but also to release cleanly. To increase the effectiveness of the nozzle without adding length or weight, we chose to go to a bell-shaped exit cone. This geometry was integrated into the mandrel. The layup was performed in three steps, with short cures between layers, and a long final post cure. The nozzle was post machined to trim off the excess, and the final result was 11 inches long and weighed just 13.6 pounds, nearly half that of the similarly sized all-graphite nozzle.

Casting the Motors

So, how do you cast 425 pounds of propellant into a 15-foot-long tube with a finned hole in the center?

It's not easy! It starts with a 16-foot-long piece of foam CNC hot-wired into the desired port shape, but if the mandrel were all foam, it would easily bend and could possibly float right out. The trick was to have a hole hot-wired out of the center. The foam was then sheathed and bonded over a long aluminum support tube. This whole assembly was fitted with an adapter/sealing plug into the nozzle and supported at the forward end by a spoked hub affectionately called the spider. With the motor held vertically, we could then pour the propellant. The propellant was mixed in manageable 25-pound batches and poured into the motor. A time delay between batches would allow each batch to partially cure, providing the structure to hold the mandrel in place but remaining tacky enough to form a good bond to the next layer. Once the tube was full, the propellant was allowed to cure. While the motor was still in the vertical position, we poured acetone into the top of the motor, which melted out the foam and freed the mandrel guide tube to be removed. All told, it took the efforts of many people for several days to cast each motor.

Full-Scale Hardware

The initial case design was different from what we use as a standard at eAc and from what I normally use for my own large motors. The previous design used a 2 inch long section of threads into which a retainer was screwed. This supported a floating bulkhead on both the forward end and the nozzle aft. The thread design was a carryover from the previous CSXT motors. Although the design looked good on paper, it proved otherwise in the real world.

The primary motor-case tubing was extruded Alcoa 6061-T6511. The case outside diameter was 10 inches, with a 1/4-inch wall thickness. With most extruded tubes, there is some variation in wall thickness, OD/1D, and roundness. Fortunately for us, in this case the wall thickness was

fairly consistent varying by only about ±0.010 inch. The average diameters from end to end were varying by about ±0.020 inch, which only meant that we had to take care in identifying and maintaining the identities of the ends and their associated hardware. By far the biggest problem was roundness. The max and min diameters varied by ±0.040 inch from the average. The main issue with the roundness was how to cut the threads into an oval opening. Given the fact that we had .250-inch wall thickness, the answer was simple. Remove enough material to make the area of the case to be threaded perfectly round before cutting the threads. This is what was done.

With the female threads cut and the matching male retainers machined and test fit, we were ready to cast. The nozzle was installed, and the propellant was cast. The final motor assembly was done, followed by mounting the 600-pound motor to the giant 12-inch I beam at the eAc-donated test site. The 15-foot-long motor was secured to our 20,000-pound load cell, and the igniter was installed. With cameras rolling, the moment of truth had finally come.

The igniter popped, and flames grew from the nozzle, and WHAM! In only a fraction of a second, the motor came up to thrust, and then the end closure let go. The propellant burned out slowly without the added chamber pressure. A detailed post-mortem showed the failure to be a function of the retaining
ring threads. The threads were too fine for this diameter, and the ovalness of the tube stock exacerbated the problem when the motor came up to pressure.

But we had to have a full-scale static firing, and time was running short. Materials were ordered for another full-scale test, while the retaining ring design was changed. We defaulted to Korey's basic design, used on most of his motors, Frank-en-bolts as they are affectionately known. The radial bolt
design was calculated out to ensure the strength of the bolts and the material they were threaded into.

As fast as possible, a new case was fabricated and the retaining-ring design was put through its paces by way of a hydrostatic pressure test, which would have saved precious time, money, and effort if it had been done on the previous threaded design, an oversight that I will not make in the future. A new nozzle was fabricated and installed, and the propellant was cast. It was now the middle of April, and the launch window for our waiver was drawing close. If we missed this window, we wouldn't be able to attempt the launch again until September.

It took until May 7,2004, before we got the second full-scale static test motor cast. The waiver window was going to end on May 21. Time was passing quickly. We only had a few choices. 1) Do the static test before making the flight motor and wait until September for the flight. 2) Make the flight motor and ship it to Black Rock while we perform the static test on a second motor. 3) Fly the static-test motor.

There was also information coming to us that Burt Rutan and others were getting close to making space launch attempts. If we didn't fly now, we might lose the chance to be the first civilians in space. So waiting until September was out of the question. If we made the flight motor before the static test, and the test went well, everything was fine. However, if we made the flight motor before the static test, and the static-test motor failed, it would be too late to make any changes to the flight motor, and given that, by that time, the motor, rocket, and people would all be in the desert already, it was likely we would go ahead with the attempt with low confidence but high hopes. Reviewing this and all the testing that had been done to date, including the limited data from the first attempted full-scale static test, and after much careful deliberation, we made the difficult decision to use the second static-test motor for the flight. We understood that such a decision meant going into the flight with increased risk, but we felt that sufficient testing had been done to give us enough confidence to lay it all on the line. Indeed that was how it was, because Ky had previously let us know that if this attempt failed, he was done. Without him as a motivating force behind the project, this team was not likely to ever attempt another space shot.

The Flight

Final assembly of the motor was done on-site at the Black Rock desert. The launch tower with the rocket installed was raised the day before the launch. On the morning of May 17, 2004, the weather was good and all systems were go. At about 10 A.M. I installed the igniter. A small eyelet had been installed into the forward bulkhead insulator. I tied the igniter to the string that we had looped through this eyelet and pulled the igniter to the forward end of the motor. Everything was tied off and electrically connected. The area was then cleared.

The final fifteen minutes before launch was like no pressure I have ever felt. I looked around the desert and saw dozens of people and tons of equipment that represented years of efforts, and it was all riding on the motor that we had built. Without the full-scale static test I didn't have one hundred percent confidence that everything was going to work flawlessly. Had I mixed the propellant right? Was the propellant strong enough? Did we account for erosive burning properly? These and hundreds of other questions constantly raced through my mind. I couldn't sit down. I paced the desert, running through scenarios in my mind, making myself sick. The butterflies in my stomach that I normally enjoy before a launch had changed into carnivorous beasts from the netherworld.

At 11:12 A.M. the final ten-count commenced, and Ky and Jerry pressed the button. Time nearly stopped as seconds passed. There was a delay while the wireless ignition-control system confirmed activation and the igniter built up enough fire and pressure to get the motor going. Finally some flames began to lick out of the nozzle. My whole body tensed as I held my breath. A burst of flame blew out of the nozzle, and the rocket jumped. Visions of catos I had witnessed in the past flashed before my eyes, but that was not to be the fate of this rocket on this day. The Go Fast rocket held true to its name as it accelerated, pulling harder and harder balancing atop a thirty five foot-long flame. Every ounce of my being was willing the rocket onward. Inside, I prayed for just a few more seconds. The rocket roared. I could feel the vibrations through my chest. Simultaneously adrenaline surged through me. I was about to explode when finally my prayers were answered as the motor shut down smoothly, and the rocket began its long coast into space. I let out a yell and ran to hug Korey and Chuck. The dream of so many was coming true at that very moment. My eyes swelled with tears as my mind struggled with the reality of what was happening. It seemed like an eternity before Jerry called over the radio, "We should be hitting the edge of space now." We had done it.

Post-Flight Review

The CSXT motor had a final propellant mass fraction of 71.6%. As a reference Titan IV Solid Rocket Motors have a propellant mass fraction of 81.5%. Although we weren't quite at the space booster level of performance, we were far above the norm for the amateur community and obviously sufficient to get to space. The vehicle mass fraction of 60.1% was superb for an amateur vehicle. We could easily have done even better with a little effort; however, given the short burn time of the motor, any decrease in vehicle inert mass would have resulted in lower peak altitude, so we had no incentive to lighten the load.

The acceleration-derived thrust trace (see diagram in the Derek Deville Gallery photos) shows a slight step-function decrease in thrust at approximately five seconds into the burn. Before finding the booster section, we had theorized that the decrease in thrust might have been due to a partial loss of the nozzle expansion cone. This was confirmed with the discovery and examination of the booster section. As best we can tell, the graphite survived the entire boost phase intact and was only damaged as the booster impacted the ground. The graphite had weathered the high heat and pressure loads, and the composite structure that retained it had also performed flawlessly. There were no gas leaks between the nozzle and casing or between the components of the nozzle. There was however sufficient heat and plasma flow against the nozzle exit cone to cause it to become too thin and be ejected. The graphite handled the initial expansion, but without the exit cone, the flow was allowed to expand to contact the inside of the fin can, which functioned as a poorly shaped expansion cone for the remainder of the burn. The exit-cone material had demonstrated its ability to withstand rocket exhaust conditions in the sub-scale hybrid testing, but the exhaust species of the solid motor proved to be sufficiently more difficult to handle. However, we believe that a slight change in configuration along with application of additional material in critical areas would easily address this issue for future motors.

You can see from the high initial thrust peak (see diagram in the Derek Deville Gallery photos), we did experience some erosive burning. The degree of burn-rate increase was even higher than we had predicted and pushed the chamber pressure close to the upper limit. We believe that there were several factors that contributed to the increase in erosion. It is well documented that burn rate increases with acceleration and with rotation. In this case, we were pulling 24 g's off the pad, and the rocket was also spun to reduce in-flight dispersion. Another contributing effect was the geometry of the port. In the subscale testing, the ports were always round, unlike the finocyl that the flight motor had. This increase in complexity of the port cross-section can increase the motor's sensitivity to erosion. Also, the exposed corners of the finocyl shape have a tendency to round off and break away.

Final Thoughts

Talk about Go Fast: The vehicle's maximum velocity peaked at 3,420 mph. That converts to 5,016 feet per second, which is close to a mile per second. That's right, one mile every second. As a comparison, a .223 rifle bullet goes 3200-3800 feet per second, which means we easily beat a speeding bullet by over 30%. The acceleration was equally impressive. In one quarter of a second the rocket was going 104 mph. In only one second it was moving 399 mph at 296 feet altitude. It blew through the vertical quarter-mile in 2.29 seconds at 900 mph. In 10.5 seconds it had reached its maximum velocity at almost 35,000 feet. From there on, it was smooth sailing to space, an unbelievable 345,000 feet of vertical coast over the next 148 seconds.

Everyone who was there to witness this historic flight will remember it for a very long time. It was a fantastic effort by so many people, which came together thanks to the persistence of Ky Michaelson and the direction of Jerry Larson. For many of us it was the fulfillment of lifelong dreams. We are the first civilians to put a rocket into space, even beating Burt Rutan (SpaceShipOne). It just goes to show: persistence pays.

Performance Report Card

Final Motor Designation: S-50,150

Delivered Total impulse: 92,429 lb-seconds or 411,309 N-sec*

Delivered Specific Impulse: 212.5 seconds

Vacuum Specific Impulse: 225.8 seconds

Peak Initial Thrust: 15,800 lbs.

Average Thrust: 11,270 lbs.

Propellant Weight: 435 lbs.

Motor weight: 607 lbs.

Motor Mass fraction Vehicle: 71.6%

On-pad Weight: 724 lbs.

Vehicle Mass Fraction: 60.1%

Apogee Altitude: 379,900 feet (72 miles)

Max Velocity: 3,420 mph (Mach 5)

* Nearly half way to the maximum allowed without a launch license

SIDEBARS

Korey Kline and Derek Deville's History in Solids

At 16, Korey began making polyester resin and AP motors based on directions provided by Gary Rosenfield. He made hundreds of E, hundreds of F, and hundreds of G motors before moving into larger motors, testing various binders from epoxy and flexane, to the now-standard HTPB. In 1976 Korey started Ace Rockets. This company made nose cones and kits and would later sell single use rocket motors as Ace Aeronautics.

Years later, Gary Rosenfield and Korey worked together at Bermite under the tutelage of the famous Dr. Claude Merrill, working on Mod 9 reduced-smoke Sidewinders. Gary's work focused on maximum performance but resulted in motors with nearly clear flames with no smoke. Korey realized that the people interested in amateur rocketry, while concerned about performance, desired the visceral effects of fire and smoke.

As an example, black powder F-100s with very low Isp were loved specifically because of the fire, smoke, and noise. Korey did extensive testing of fuel additives, particularly related to the fireworks industry, looking for the best visual effects. As a result of the popularity of his special effect testing, Korey developed the Visijet line of motors, spanning from I to L class, primarily in the J class. The Visijet name followed the naming structure of the day where all motor names ended in jet (e.g. Enerjet, Plasmajet, Projet, etc.), tying in the visible nature of the exhaust. These were the precursor to all special effects motors in HPR (High Power Rocketry). The addition of zinc created billows of smoke, and copper carbonate changed the flame to a blueish purple, while coarse magnesium resulted in sparks and a throaty roar.

Bill Wood introduced the high-power altitude trophy that inspired Korey to join forces with Chuck Rogers to make over thirty attempts before successfully achieving a higher altitude than Goddard had ever reached, over 25,000 feet. After this Korey was inspired to build the largest motor possible for a Tripoli (Tripoli Rocketry Association) launch. This four-inch-diameter four foot long motor was at the time, and for many years after, the largest motor certified by Tripoli. Furthermore, this motor was the first certified motor to have a delivered Isp over 200 seconds. A total of ten N-1940s were made. As a result of the N-1940, the creators of Down Right Ignorant rocket, looked to Korey to provide propulsion for what was then the largest hobby rocket ever. Korey cast the propellant for the baby P-motor that was called a wimp-P. The volume of propellant Korey was processing became a concern for him, and his focus turned to hybrids. And the rest is history.

I have been involved in high-power rocketry since 1997. In late 1999, I got involved with Jim Mitchell of

Dynamic Propulsion Systems. At that time, I took the Thunderflame Propellant Course. The experimental bug bit hard and dug in. I was completely hooked. I began making 38mm motors faster than I could fly them. I graduated quickly through 54mm and spent most of my time working with 75mm hardware. I hit up Amazon and every university library I could find to satisfy my insatiable desire for more information. Books like Sutton's Rocket Propulsion Elements and Humble's Space Propulsion Analysis and Design became my bibles. I talked to anyone who would talk to me, to find out more about experimental rocketry. People like Jim Mitchell, Frank Kosdon, Paul Robinson, Jim Rossen, and John Johnston fanned the flames. Terry McCreary's book: Experimental Composite Propellant, was an invaluable resource. I was growing out of the five-quart Kitchen Aid mixer and graduated to a twenty-quart model. With this new tool, my motors began to grow. Six-inch hardware was on my Christmas list, but at that time there was no place like Loki Research to go to for such monsters. It was up to me to design the hardware, find the materials, and get things made. My 6-inch 0-motor came to life for the first time in February 2001. Since that time I have tested or flown six additional 0-motors, three P-motors, three Q-motors and, of course, this S.

2004 Discussion - souvenir by chuck rogers

Souvenir

By chuck Rogers

This story was posted to the newsgroup Rec.Models.Rockets by Chuck Rogers, one of the CSXT team members who launched the amateur rocket that reached space:

Fred Brennion and I were traveling back from the awesome flight to space of the CXST rocket. As we're heading back from Black Rock, being the yuppie that I am, I had the hankering for a raspberry mocha with soy milk, topped with whipped crème. Of course, heading back from Black Rock, once you're past Reno, you're traveling through the middle of nowhere. But low and behold as we went through Bishop, CA, we found a great coffee shop, the Kava Coffeehouse.

Well, Fred and I are in the Kava Coffeehouse, and as I order my raspberry mocha with soy milk and whipped crème, Fred takes note of a good looking young lady at the other end of the counter. I don't really notice her, being totally enamored with my lovely wife Brenda, but Fred moves on down the counter to introduce himself to her.

Well, Fred says "hi", and then says "I'll bet you'll never guess why I don't have my driver's license". The young lady looks at Fred like that's the lamest pick-up line she's ever heard, and she says "let me see, I bet you had a DUI". And then Fred says "No, I put my driver's license in a rocket that went into space, but they haven't found the rocket yet."

I wish I could have taken a picture of the young lady's face! It was a mix of incredulousness, but yet, a strange realization that Fred's comment was so completely off the wall, it probably was true!

Yes, in a strong vote of confidence that they'd recover the rocket, Fred asked the CXST team to put his driver's license in the CXST rocket payload bay. Bruce Lee also put one of his credit cards. These guys were confident that the CXST team was going to get that rocket back!

I told the nice young lady that the CXST flight was already on MSNBC.com (the Kava Coffeehouse had a couple Internet terminals, we checked Internet news sites and found it), and that she could check it out herself. Another incredulous, stunned look. If she caught it on TV later, she probably turned to her friends and said, "you're not going to believe this, but I talked to these two geeks in the Kava Coffeehouse...", etc., etc...

Needless to say, I drove the entire way home. Although if Fred was driving and we got pulled over, it would have been hilarious to watch him explain to the Highway Patrolman how he'd lost his driver's license. "You see officer, I flew my driver's license into space on a rocket, and they haven't found it yet". Yea, right buddy!

I'm sure the CXST team will be mailing Fred his driver's license. Fred's already getting a new one, because his old one has been to space and needs to be framed, or something!

Kudos' to the CXST team! Great flight! Fred knew he'd get his driver's license back! And if you're going through Bishop, stop in at the Kava Coffeehouse for a great mocha.

Word is that there was quite a collection of keepsakes and memento’s in the payload (also posted to Rec.Models.Rockets, from Pat G): You left out the part that Bruce Lee threw his credit card in (editor’s note: it was 2004 and a first available from BofA chip card, I worked in the credit card industry) , and after the launch and recovery, he used it at Bruno's, still worked, btw, after being in to space.