Unfortunately, I didn't have the forethought to take pictures, and follow the progress of this major project from the beginning, but I will be sure to make that effort on future projects. For now, you will have to either piece it together through the limited number of pictures and files here, or use your imagination.

Otherwise, just bear with your connection to download the thumbnails, and enjoy the ride. Click on any thumbnail to get a larger version of the picture.

The 'Daft Dream' is a three stage rocket designed to be very fast, and fly very high. This project started out as a friendly competition between myself and my brother, and this is actually the second rocket created for this purpose. Neither of them have yet flown, and while the launch date has not yet been determined, it is long overdue, and I am very far behind my anticipated schedule.

'Daft Dream' Specs:

• Three stage, Minimum Diameter design

• Must weigh less than one pound. This design is estimated to come in right around 420g

• Must be propelled by less than 113g of propellant. I found two motor combinations, either for high altitude (G40, F20, D13; 102.2g) or for maximum speed (G80, F50, D24; 104.4g) both combinations fit the requirements.

• 'Daft Dream' is made almost entirely out of Carbon Fiber. It's not the standard woven Carbon Fiber, it is called 'random weave' or 'veil' and has no directional pattern or strength bias. It is also very thin (0.003") so it can conform to smaller shapes, and allows for the optimum 50:50 ratio of Carbon Fiber & Resin in smaller objects.

I started the design for this rocket innocently enough. I have learned a lot, both about design, and fabrication along the way, and even now, I look back at some things that I had done on this project, and I can think of several ways to do things differently.

The biggest challenge to me, was not so much being able to think of a way around a problem, or even to envision how things could work in my head, for some reason I have always been real good at that. It was how to actually show it to someone else so that I could get their take on it. There are a lot of people that played a vital role in shaping this rocket from what it 'could' be to what it physically possible, or structurally sound, or slightly easier to fabricate. There are many details of many articles that had to be somehow changed from my original concept in order to be easily manufacturable, or cost effective, etc., and understanding all of those constraints was a bit of a challenge, but once all of the limitations were understood, I was able to figure out a way around the problem.

One such example of this is actually staging a minimum-diameter rocket. Everyone has staged Estes motors, and the principal there seems to be get the motors as close as you can to each other, and the ejection charge from one ignites the next, and so on. As I understand it, this doesn't function well for composite motors, and black powder motors can't possibly give enough punch to get much altitude out of them, so something else had to be figured out. At the time I was researching this, the electronics that I could find for rockets was rather large, and typically needed a 9v battery in order to function, much less ignite a motor. Everything I knew pointed to a custom application, and some sort of custom electronics device, so I embarked on a side-project. It's design was specifically to fit in this rocket, and control up to four events (for this rocket, I would need two stage ignitions, an apogee charge, and a main charge for recovery) and record altitude. My Brother was of great assistance on this, as was some of the software that I use at work. He designed a schematic, and defined all of the components, and I did the layout, and packaging of all of the components. We had PCBs made, and made our own altimeter.

Not all design challenges were so difficult, but some did depend on others. I did a lot of calculating and simulating to try to figure out the optimal motor configuration for this rocket. My initial selection was to use three Aerotech F-72s, but I was having difficulty finding any at all. As I have gotten into the hobby a bit more, I now know several people that have a couple, so it is possible to be able to find some, but likely not enough. And, as Aerotech has not begun producing these again since the tragic fire, I doubt they will be an option at any point. Would still like to try one though. My original design was going to use three 24mm motors. One configuration that I had discovered used as many as five 24mm Aerotech Reloadable motors. The problem with that, is that the body tube's O.D. was the same as the O.D. on the flange on the bottom of the reloadable motors. Since my Brother and I decided that we were going to use un-modified, commercially available motors for this competition, turning down the flange on the reload casings was out of the question, even though that would be the easiest way to fix the problem. And believe me I wanted to.. one of the motor configurations that used those reloads was simmed to 25,000'! So I had to find another way. I tend to get a sort of tunnel vision once a design feature is set in my mind. My problem was that I was thinking in a 24mm world. I didn't explore larger and smaller motors. So I broke out of that mold, and began to look at other methods of reaching 113 grams of propellant. I spent a lot of time with Excel, and the motor specifications of every motor manufacturer that I could think of that actually made 18, 24 and 29mm motors (and could ship them) and found the two motor combinations listed above. Now all I needed to do was design a rocket to put them in.

After reading Stein's book, I made it a point to learn very well, the relationship between Center of Gravity (CG) and Center of Pressure (CP) and I can't count the number of rockets that I've created in RockSim to test this. Two such articles proved to actually be.. well, less than stable (see Rock-014, and Rock-011) though I think more a result of not carefully building them to their design, rather than faulty design, though material weights, and consistencies might have entered into it a little bit. Everyone knows that a rocket is supposed to have a nose cone, and a body length, and some fins, etc. but I had a string of rockets that I drew in AutoCAD, and then converted to RockSim files, that actually were stable from the get-go. No modification, no fiddling, no adding nose-weight, etc. A far cry from what I was capable of only a year prior.

One of the more difficult issues that I had to deal with as recovery. As one of the guidelines for our competition was that points would be awarded for landing closest to the launch pad, I didn't want to put a giant parachute on it, and have it float back home. Of course, from 13,000 feet, it was going to drift quite a ways under even a small parachute. In my simulations, I figured what I thought to be a decent descent rate, and designed a parachute to match. I managed to do this for each of the stages individually. Now, not only am I making rockets, and altimeters, but I am also fabricating my own parachutes! I figured on a 12" chute for the main, which gave a descent rate of approximately 12 mph. This was going to be soft enough to not kill the rocket on landing, but to keep it coming down fairly quickly. You must remember, that while the sustainer of this rocket isn't really much larger than say, an Estes Yankee, maybe twice as long, it's weight density is much greater because it carries a 2 cell 6v Camera battery in the nose-cone, and an altimeter. There isn't any extra space on this craft. Everything that is on it or in it has a designed purpose, and I have trimmed as much fat as I think I possibly can, so it's density is much greater than that of a normal rocket. But still, looking at it, you wouldn't think that it could weigh less than a pound, especially given all that has gone into it.

In my fiddlilng with Rip-Stop Nylon, and Kevlar yarn, I managed to turn out three fairly decent parachutes, and a rather large handful of not-so-good ones. Many of my early shapes ended up having a 'nipple' on the top of the dome, which was simply a waste, because the weight of the material wasn't contributing to creating drag as much as it should. Through a little bit of research, and a lot of trial and error, and some practice with a sewing machine I finally managed to get it. I did have to change my design just a little bit, but I think it worked out for the better in the end. Instead of bringing all of my gores together to an attempted point, I replaced that area with a circle of nylon, and sewed it in place. Not only did this smooth out the shape of the canopy considerably, but it drastically lightened the canopy by removing all of the excess material in the top of the canopy. If you think about it, every gore was sewn to another gore, and while the sewing technique I used is designed for sewing raw edges (surging they call it, I think) it still retains a small fold of material all the way to the end of the material, which at some point becomes excess weight (and reduced volume inside the canopy) rather than adding to the performance of the chute. So simply replacing this with a fabric circle flattens this out, and removes the excess. All of the chutes use Kevlar 'yarn' of various sizes and lengths. I would like to thank Richard Nakka and Richard Nakka's Experimental Rocketry site for some of my original calculations, and designs. Anthony Thyssen for his assistance and input into clarifying some of the pattern making, and sewing techniques that I used, and for lending me the resources in his really cool Kite Site, and for the adventures of Tuffy and Tuffy's Hemispheric Parachute Design.

I did fiddle with making a drogue for apogee deployment, but I wasn't having much luck. After asking around a bit, I was given the name of Bruce Feaver of Anchor Parachutes and was told to contact him with my needs. I did a little bit of research, and figured that if I could get my drogue to simply control the attitude of the rocket on the way down, then I can have an orderly main deployment, which would allow me to deploy at a lower altitude, and therefore minimize drift. It takes a long time to get to the ground when you're traveling at 12fps from 13,000', so I was thinking of something more along the lines of 90mph as the descent rate of the drogue. This is almost like a very late deployment, except the rocket won't be traveling at it's terminal velocity (which I figured to be about 350mph on the way down) for ejection. I actually made a jig with a sort of spring on it, and made marks for the extension of the jig using the descent weights of all three stages, and my Brother helped me by hanging this jig out the window of my car, and noting the speed at which a particular parachute hit the mark for a particular stage just to be sure. I know it's not the real thing, but it's the closest test that I could think of being able to perform. I made a 4" dia parachute with 4 gores, and tested it, and sure enough, it pulled the right amount of weight at 90mph (don't ask me how I know) but it was horribly unstable. I would imagine that some of that is due to the turbulence of the car it was being tested in, but I'm certain that a lot of it was a result of my less-than-accurate sewing. So Bruce came through for me. I sent him my RockSim file, and my desired descent rate, and he did the rest. In a couple of days, I had a brand new parachute, custom made specifically for this purpose. it is a 4" 4 gore chute with a rather large spill hole. The quality is top-notch, and ever stitch is tight, with no loose threads. Even the shroud lines are gathered, and woven into each other to form a loop to which I can attach my shock cord. A very fine product indeed. Even better still, it fits in the space I had allocated for it with just a little bit of room to spare. Thanks Bruce

Because I knew that this rocket would see extreme altitudes for it's size, as well as extreme speeds, (one of my early sims resulted in a maximum velocity of Mach 2.5) I knew that I couldn't very well make this out of cardboard Estes tubes. For my first attempt, I obtained a 24mm mandrel from a machinist friend of my Dad's in order to form the body tubes on. The mandrel had an 8" working length, so the unfortunate part, is that no section could be longer than 8" without the use of a coupler. I did lots of research on how to work with epoxy & cloth materials, and finally decided that using a carbon fiber veil would yield the best results for the size that this rocket was going to be. I intentionally 'over-bought' as far as materials go, because I didn't know how much I was going to need to use, nor really what the yield would be from a given amount of material because of my lack of experience with forming. Fortunately, my Dad had done several projects in the aerospace industry, and was able to lead me in the right direction for supplies, materials, and processes. It was because of this knowledge, hands-down that I was able to get anywhere near as far as I did. Certainly, there is always room to grow and learn.

It took us a couple of tries to get our first tube, and looking back, it was very likely far too thin, but again, the goal was not to make an indestructible tube, but rather to simply exceed the strength of paper tubing, with the same or less weight. Sounds easy, right? Not at first. Initially, wrapping the cloth around the mandrel was a three-man job. We used an old wood lathe to hold on to the mandrel, and so that we could easily rotate it. If you are doing this for the first time I suggest three things: 1) Find someone to help you, 2) prepare all of your materials beforehand, including pre-heating your oven to the manufacturer's suggested cure temperature (if it will fit), and cutting all of your fabrics, liners, bleeders, etc. before you even think about breaking out your epoxy. 3) Practice. There is certainly a finesse to making a good tube, and it doesn't get better just by staring at it. We had many tubes that split during the extraction process, and quickly learned to treat our tooling better. I have a respect now for the cost and complication of forming composite parts, and I enjoy it immensely.

Once we were able to get fairly consistent results with extraction, it was on to getting consistent results in the finish of the tube. We used heat-shrinkable tape with perforations in it for curing. Very simply, the mandrel was highly polished, and then waxed (we used a sprayable mold release) and then slathered with epoxy. Once we got the CF cloth lined up, it was simply a matter of applying epoxy evenly to the mandrel as we rolled the CF veil onto it, and making sure that everything was saturated (but not dripping) with epoxy. Then we very quickly wrapped it from one end to the other with the shrink-tape, and secured it (usually with a couple of extra wraps over itself at both ends) and stuck it in the oven. Fortunately, the parts we were forming were not too large to fit in an oven, but I can see how that could quickly become not possible on a larger rocket. In the oven, the shrink-tape constricted around the mandrel, squeezing out all of the excess epoxy. Additionally, the heat helped expand the aluminum mandrel from the inside, further compressing the cloth against the tape. The finished part would end up a couple of thousandths larger than the mandrel at best, but it would certainly make a difference. After it was cured, but while still warm, we went back out to the lathe, and removed the shrink-tape, and did our best to remove the 'steps' in the surface caused by the overlapping shrink tape. Typically, a little bit of sanding did the trick (use a sanding block so you only hit the high-spots. Continue until the surface is uniform in thickness and all of the steps are gone) and then we would trim the ends, and remove the excess. Then it's off to the freezer for an hour. This helps shrink the aluminum mandrel to help use get the tube off of it without it breaking. If you hold it in your hands just enough to get the chill off of the carbon fiber, then it usually takes a couple of good whacks with a BFH to break the tube free, and then it simply slides off.. that easy.

Or so you would think.

The 24mm mandrel wasn't cared for very well, and will likely not ever form another part again. We used wood lathe tools to cut the ends of the tubing while the lathe is spinning (so that the cut is straight) and that usually scarred the surface of the mandrel, and eventually, we had to resurface it so many times (LOTS of sanding, polishing with lacquer thinner/mineral spirits, and cleaning with acetone) that 24mm motor cases just simply would not fit in the tubes anymore, so the tool was retired. The tools linked above incorporated a groove in both ends to allow for the cutting of the tubes, and each end is very slightly smaller than the forming surface (a verbal request to the machinist) so as not to inhibit part removal. Also, I specified to the machinist, that if there has to be a taper, that it be consistent, and go in a certain direction so that I could actually remove the parts from the tool. On the larger mandrel, I discovered through destroying several parts trying to extract them, that the diameter on the curved transition was larger than the base's diameter, by only 0.001", but it seemed enough to start a fracture, and split the part. After this was solved, forming was relatively easy.

After I had all of my tubes formed, it was time to start laying out cuts, and tubing lengths, and to determine which section would be used for which part. Most of the tubes were fairly uniform in finish, but some of them were sub-par in some areas, having wrinkles, or an odd texture or something, so sections were chosen carefully based on the intended internal structure, and neighboring parts. For example, the joint between the upper sustainer and the mid-section of the sustainer that houses the main chute, there is a slight diameter difference. The coupler went on the upper portion so that I could make it slightly thinner to match the mid-section's diameter, and not lose strength. Most portions were made from carefully matched sets, or facing parts of the same cut for this reason. It makes sense that the surface needs to be as smooth as possible, but the better off you start with, the easier it is to attain that finish when you're done.

All of my fins and bulkheads were laser-cut on the machine we have at work, and are made from several layers of CF veil, and basswood laminated together. You would be surprised how strong that combination is. I formed the sheets by using two pieces of aluminum, and polishing one face of each piece. A little bit of brasso, a lot of mineral spirits, a lot of acetone, and an amount of elbow grease that would wear out a solo porn star. The two faces were smooth enough, that without any lubrication, they would stick together simply with suction. The sustainer fins & bulkheads, as well as the engine block (motor ejection is not used) are made from 4 layers of CF veil, then a layer of 1/32" thick basswood, and another 4 layers of CF veil. Both booster's fins, and all associated bulkheads are made using 5 layers of CF veil, a layer of 1/32" basswood, 3 layers of CF veil, another cross-grain sheet of 1/32" basswood, and a final 5 layers of CF veil. All of these panels were laminated using peel-ply on both sides (so the polishing on the aluminum seems pointless, but I did make some pure CF sheets with them, and they didn't stick) and the thicker sheets were vacuum bagged onto the aluminum plate using a food-saver. Once all of these were formed, I laid out my shapes on the panel, and then handed it over to the laser department at work, and when I was done, I had exactly matching sets of fins, with no further trimming, or sanding to do in order to ensure that they were all the same. Using a round dremel tool, I cut a groove in the leading edge, and epoxied a round CF rod in the groove to increase strength, and give me a good start for the airfoil shape I was going to try to attain. Once trimmed, all of the fins were again matched to ensure consistency. I set up a jig on a drum sander so that I could bevel all of the exposed edges at a 10° angle. The beveled edges were touched up with sand paper, but as far as I could tell were identical to their respective sets.

I needed a method of transferring power from the nose, where the battery was, to the electronics bay where the altimeter was, which actually proved quite easy. But after that, I was on my own. And knowing that my parachute bays were going to be a very tight fit as it was, I had to use a little bit of my creativity to get around this problem. Walking through Fry's one day, I found some chip sockets for PC Boards that had machined pins, rather than stamped. They were in nicely molded plastic housings that it looked like I could trim down, and make several sets of pins. My original intent was to use these to make ejection charge 'pellets' that simply 'plugged in' to a matching pin epoxied to the bulkhead. Using an Estes BT-5 section, about 1/2" long, I would solder nichrome into the pins, and then coat that with a quick dip of quick-flash compound, with a very light coating of BP & Mg bound in NC Lacquer on top of that. The igniter would then be held in a jig at the bottom of the BT-5, and a small bit of epoxy would be poured in along the wall of the BT-5 to fill the base, and retain the plug in it's place. This would then be filled with BP, and capped with a thin layer of hot glue. Having to make these ahead of time could prove troublesome, if not simply time consuming, but having a pre-made ejection charge that you simply 'plug in' seemed to me like a fantastic idea, so long as they are properly stored, and kept away from any static shock. Use this idea at your own risk, it is still just an idea, and has NOT been tested. You've been warned, don't sue me.

So to transfer power from the business end of the altimeter to each of the event locations was going to be fun. The altimeter has two 4-pin plugs on one end, and it is actually attached to the rocket by these plugs. Two holes at the base of the altimeter allow air to get in, and a special tool to be inserted in order to release the altimeter to get it out. Not to worry, it can be programmed using IR communications while it is inside the rocket! So, now that I have my wires throughout the altimeter bulkhead, I just need to get them all to their respective locations. Cat-5 cable has some nifty wire for this purpose, and I happened to have some handy, so a stripping I went. Not to worry, this is PG. Here's where the previously mentioned chip sockets come in. Cat-5 wires just so happen to snugly fit into these sockets with enough resistance (physical, not electrical) to ensure a good connection, but not so much that a good BP charge couldn't yank the connection apart. So, the sockets were mounted in an accessible location near the opening of each separation, and a loom of Cat-5 wires were attached to the other end. Very simply, as I assemble the rocket, I need to connect the Cat-5 wires to the appropriate location. I made sure that each socket had the correct color, so all I have to do is match color to color in the plug I insert it into, and my events will happen as they should, and not in some other, more entertaining sequence.

Here's the sequence: The first booster is lit on the ground, by the LCO. The launch activates the altimeter. As soon as it detect deceleration, the second stage is lit. The nozzle points into a nomex-protected, sealed chamber that the igniter is connected to, which kicks off the first booster. The first booster is then left to it's own apogee, and motor ejection of a 14" chute. The second stage burns, and again, the altimeter detects deceleration, and ignites the sustainer. The sustainer's nozzle is also in a sealed, nomex-protected chamber that houses the igniter. The second booster is left to motor ejection as well as it's own apogee. The sustainer will then hit it's apogee, and deploy the tiny 4" drogue. The sustainer's motor will not have any BP in it, and it's open-end will be vented to prevent any kind of over-pressure in that chamber, or blowing the casing out of the rocket. The altimeter will detect a set altitude, and blow the main, which will slow the sustainer from 90mph down to 12mph. If I calculate it correctly, the sustainer may land before either booster touches down! First Stage burnout is at nearly 3000' with either motor combination.

Well, then how do I get power to the igniters? Not only am I crossing a tube coupling, but an entire rocket coupling, and on a minimum diameter airframe to boot! I could tell you... but then I'd have to kill you.

Okay, I won't kill you, but I will dish out a thorough tongue lashing. As this rocket has NOT yet flown, try this at your own risk! Prior to mounting the fins to the body tube, I routed all of my wires inside the rocket. One of the interesting limitations of electronically staging a minimum-diameter airframe, is that where the motor is, there is NO ROOM to pass wires through. I knew that this was going to be an issue from the beginning, but I didn't know how I was going to figure it out until I had a revelation. I had looked at every kind of connector that I could think of that would fit in those locations, without causing too much drag elsewhere. Basically, the connector needed to hide behind the airflow at the root of the fin. Then I came across a blade fuse, and it struck me. If I could embed a 'blade' of metal (I used a brass plate) inside the root of the fin, then I could use something that would accept this type of connection to pick it up on the next stage. The disruption of airflow would be very minimal, and I would simply need a small housing on the outside of the rocket, which could be streamlined for aerodynamics. I did manage to find one type of connector that fit this bill, but as it turned out, the connector was very weak, and easily susceptible to damage, so I decided to try something else. The idea was there, but my execution is questionable, since it wasn't my idea from the beginning, but I will try again, and see if I can come up with a better design. Another issue that I ran into was the conductivity of Carbon Fiber. It WILL conduct electricity. I did a little bit of testing, and fortunately enough, it's not enough to cause a problem in this instance, but be fore-warned, it can conduct electricity. What I ended up doing was adding to my 'blades' a female amp pin, and a male on the other side. These pins actually have in place a friction-fit via a pressed 'spring' that pushed on the male pin to ensure contact. Brilliant idea, but a bit too strong, I thought, so I made sure to crimp the male connectors slightly to reduce this spring's tension on the male pin. I wanted to transfer electricity, but also release quickly and evenly. I can still pick up the whole assembly, fully loaded, just by the nose cone, and everything stays together, so I don't think I'll run into any drag-separation issues (I hope) but things are just loose enough I don't need to use 10g of black powder to separate them. Remember, I need to stay under one pound.

So now that I have my power from one stage to the next, I can finish everything up. I used the 'tip to tip' technique to add strength to my fins on all stages. All of my fins were mounted with 3M 467 2-part epoxy, which oven cures very nicely at only 150° for about half an hour. The 3m 467 comes in a dual tube, and is designed to function with a gun-like applicator, and many nozzles are available, including mixing nozzles of different lengths, and syringe-like tips that can be disposed of after use. I didn't bother with the tips, I just mixed it all by hand on a piece of wax paper, or cut-up cereal box. The good thing about it, is that it is not as runny as the gallon & quart Jeffco epoxy that I used to do the CF lay-up, and would stay where I put it. The 467 has the consistency of thick honey, so it can still be spread fairly well with minimum run. Pop it in the oven for half an hour, and you've got a bond you won't likely be able to break. For fin fillets, I used my Jeffco epoxy with some 'carbospheres' as filler. I mixed in enough to get the consistency of creamy peanut butter, and slathered it on, and then I used a large punch as a radius gauge, and simply ran it from one end of the fin to the other, and carefully removed any excess with my finger. That was a black, sticky mess, let me tell you. It cleaned up fairly well with acetone, but it was interesting having black fingers while doing this. The fillets came out fairly well, I think and with the exception of needing to sand it a bit here and there, it was really clean, and smoothed out really well. In fact, many parts of the fillets were too smooth, and were hit with a bit of 60 grit paper prior to laying up the tip-to-tip re-enforcement. I did my best at every stage to get the best finish that I could, so that when I was done, I wouldn't have a lot of finishing to do before I could prime and paint it.

The tip-to-tip job was fairly easy once I got started. I cut all of my pieces before hand, and got my bags ready. I don't know if it was necessary or not, but I did plug the open cavities of the tubes with a spent motor casing, or a good motor, both wrapped in wax paper prior to putting them in the tube. I didn't want to have permanent motor retention, and I didn't want my food-saver to be the death of any of my tubes. Every face got two additional layers of CF veil added to it with epoxy. Everything was oversized, and all of the excess simply spilled off the edge of the fins. I used peel ply, and bleeder cloth to soak up the excess epoxy, and as quickly as I could, arranged everything in the food-saver bag for vacuuming. I only had one mishap that wasn't discovered until after it was cured, where both the peel-ply, and the bleeder somehow slipped off the corner of a fin on the first booster, so I got the nice waffle pattern on the face of the fin, which took a LOT of priming, and sanding to remove. I was very impressed with the result of the tip-to-tip lay-up, as it was the first time I had used such a technique, though it did leave quite a bit of cleanup, and sanding to get it back to being smooth and clean, but you can certainly tell the difference in strength, now that everything is in one piece, and everything is formed together. I think the only way to make it stronger would be to do all of this while all of the epoxies are in a "B-stage" or "green" cure. Then, I wouldn't only have a mechanical bond, but a molecular bond as well.

I had made a maple nose cone for my other attempt at this Carbon Fiber rocket, and while I liked the idea, and it finished very nicely, I don't think that I was able to get as much out of a nose cone as I could. For weight distribution, I wanted to put the heaviest component in my electronics as far forward as I possibly could, and that meant putting two 17mm batteries in the nose cone. This would be very hard to accomplish with an 18mm O.D. piece of wood, so I figured forming a nose cone would be my best bet. Now, mind you all the while the techniques and materials that I am using and studying are put into effect on MUCH larger rockets than what I was making. It is perfectly conceivable to make a male plug, form a mold from that, and then lay-up in two mold halves, and lay them up together, using additional material on the seam for a nosecone that is 11" in diameter or better, but on something that you can easily fit in your pocket, it seems less than feasible. The other way to do it is to actually make a very costly tool, and make a three-piece mold, place your material, and then inject your epoxy under pressure, and then cure it. Well, I didn't have that kind of budget, so I did the next best thing. I had my machinist buddy cut a piece of steel in the profile that I wanted to use on my nose-cone, and then grind down opposing faces as much as possible. The idea was to essentially make a 'paddle' drill bit out of this tool, and drill into a wooden mold. The wooden mold consisted of two pieces of hard maple that had been planed, and screwed together. Then I turned them down to round, and drilled a series of pilot holes in it to help keep the paddle bit on center. My Dad was key in both concept, and execution of this tool, and I think that it came out very nicely, and both of us learned a thing or two. Once the female was drilled to the right depth, we turned a male plug that would go inside of that. The first one was made from more hard maple, and turned out very nicely, but ended up sticking in the first nose cone that we attempted. The reason for this is not quite clear, as all surfaces that were going to come into contact with epoxy were sanded with 600 grit paper, and then was given a polyurethane coating, and heavily waxed before use, but my suspicion, is that through sliding the CF veil on the male, and the 'churning' of the epoxy inside the female cavity to get thorough epoxy wet-out might have removed the mold release from the surface of the male tool. But we didn't make the same mistake twice. A quick trip to Fry Steel, and several hours on the (wood) lathe and a metal file resulted in the right shape for a male tool. This one was undersized slightly to make forming easier, at the expense of a less than 50:50 ratio of epoxy to CF veil, but you gotta do what you gotta do. Then, after a good polish, the tool was ready to try again. This time, I also changed my strategy on the cutting of the material that would go inside of the tool, scalloping the 'coned' edge of the material to allow it to collapse inside the female tool to the point that I wanted. Some quick, careful work of getting the material inside the tool, and inserting the male plug, and epoxy, and holding everything together with a clamp, and putting it in the oven produced a fairly decent nose cone.

I was fairly disappointed with one part of the process. Apparently it is pretty difficult to get enough material to the tip of the cone, so the end result has a slightly larger radius than I really wanted, but I really can't complain that much. After the cone was formed, it stayed on the male tool, and was popped form the two female halves, and it went back on the lathe to get a shoulder on it, and do finish it's OD to that of the tube it was going to be used on. In the end, I have a carbon fiber nose cone that holds the batteries for my altimeter as far forward as possible. When you can move your CG forward without adding weight, it's a good thing.

Forming the second booster's transition was the one thing I was dreading most, because I knew that it was going to be a pain. I had done a little bit of probing before I had the tools made, and what I came up with was that I could very easily over-form it, and shape it to what I wanted afterwards, but I didn't do a very thorough job on all of my dimensions. The shoulder for the transition was only 0.013" thick, which wasn't very much at all. Fortunately, I had intended some internal structure for it to support the sustainer squarely, but in the end, I would have liked to have a thicker shoulder on it. I might have made the shoulder longer as well, just for that much more stability. After forming the first transition, I learned something: It's not very easy to form a complex, conical object with flat material that isn't very carefully cut. Admittedly, this was probably the sloppiest part that we made for this rocket. The first one was too thin, and shattered coming off the tool. For the second one, I cut the material a little differently than before, in the hopes that perhaps some structure would help make it strong enough to get it off of the tool. When the second one split coming off the tool, I figured that there must be something wrong, so I measured the tool, and discovered that the base of the cone was 0.001" larger than the shoulder, and the part could only come off in one direction. Again, with the metal file, we re-shaped the tool, and then polished it up again. For my third attempt (Third time's a charm, right?) I cut the material very slightly differently from my second attempt, and I used more of it. Again, coming off the tool, the part split. I tried to drill the end of the crack again, but missed it by less than 0.007". I decided to go with it, and try it anyways. I epoxied the fracture together, and cured it, and then began forming it. Unfortunately, during the forming process, it again split, and so it made our shaping rather interesting, using tape to hold it together as it was being spun on the lathe. After I got it formed, I immediately filled it with internal structure, and epoxied everything together, because I feared it breaking. It turned out alright, I think, but I could do some things different to get a better part, I think.