 |
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
Flash GordonAt Mudroc in June 2001, Jason Blatzheim, the motor developer for AeroTech, suggested that we do a joint BALLS project. This sounded like a good idea, but I needed something to make it unique. Most BALLS project are just big motors in heavy rockets, with the emphasis on how impressive the motor is. It took me a while to come up with a good idea, but when I did it was a great one! Jason makes a great sparky propellant with white smoke. How about a model of the Flash Gordon rocket ship? After all, Flash Gordon's ship had four sparky, smoky motors which propelled it. 
The Flash Gordon rocket ship is a pretty simple shape, a hemispherical nose and a hemi-elliptical body. The fins are very large, but run well forward to the back of the nose cone. The engine nacelles are well forward on the rocket. In the serials, they seemed to fire straight back (although it's pretty hard to tell exactly what they are doing). The rest is just surface detail. The Pictures
LDRS XXI, Amarillo TexasThe 1/5 size model was flown at LDRS XXI to test the stability of the design. This first model was built with prototypical fins and most of the detail planned for the larger model. (You can see the rocket ready to fly in the picture above.) I chose an AeroTech H123, since it was the lowest average impulse 38mm motor I could find. With this motor, the lift-off weight was 5.0#. The H123 pushes with about 34# of thrust for the first second, giving a thrust-to-weight ratio of about 7:1. The measured C.G. was 10¼" from the cannon tip (7¾" from the tip of the nose). My estimated C.P. was 13½" (11" from nose tip), giving about ½ caliber of stability. (See the Design section below for more on this.) 
The picture above, taken by Rick Clapp, shows me prepping the small Flash Gordon model for its first flight. Even at this small size, the rocket is quite fun!
The sequence above (down the left) shows four stills of the liftoff and flight. Click on the smaller images to display a larger image on the right side. The rocket wobbled a lot on the way up, which isn't shown well by a sequence of still photos. 
On the right you can see the two sections as they were recovered. (The ALTS24 beeped out 646'.) Two of the fins broke off cleanly and there was no other damage. I'm not sure why two fins broke off like this (the rocket was recovered by children before I could find it), but my guess is it came down hard on two fin tips and sheared them off. At any rate, this damage is no big deal since the next version will be built with larger fins anyway. I was pleased (and somewhat surprised) that the cannon made it safely. I had assumed that this would break off in landing, but it didn't even seem to stick into the ground. (The nose must have come down on its side...) Aeronaut 2002, Black Rock DesertAt the August 2002 AERO-PAC launch (Aeronaut 2002), I flew the small version again, this time with slightly larger fins. (See the Fifth-size Model section below for more info.) 
The LDRS flight was not stable enough to make me comfortable, so I added .3" to the semi-span of the fins and made the aft end of the tail a bit more bulbous. To the right you can see Renée holding the second version at out in the Black Rock Desert. Despite the large fins, the model still looks good. In fact, the larger fins give a little nicer shape since the tail doesn't seem to be so small compared to the nose. (The semi-span of the fins is still smaller than the radius of the nose cross section.) This second flight was also on an H123 for comparison purposes. The weight of the rocket did not change significantly so the only real change was the move in C.G. due to the larger fins.
(Photos above by Renée.) This flight was a bit wobbly, but much better than the LDRS flight. I think the rocket still needs a bit more stability, but the design is certainly workable! The model flew relatively straight and up to 810' according to the Adept ALTS25.
The sequence of photos were taken at about ½ second intervals. The flight was a little wobbly, but it went more-or-less up. Click on the smaller images to see them in larger size on the right. The DesignFirst of all, the challenge was to research the rocket. The old Flash Gordon serials have been collected and made available in DVD format. I bought Flash Gordon Conquers the Universe (12 Startling Chapters) and watched the first few segments to get a better idea of the ship.
To get the dimensions, I found a spot in the movie where the ship filled the screen with a side view and paused it. Then, I was able to take measurements directly off the screen. (The measurements were scaled to result in a 10' long rocket.) 
This rocket is too unusual to model in RockSim (or any of the other common rocket simulators). The shape also may not match the assumptions in the Barrowman equations very well either. As a result, I was unable to determine a property center of pressure. However, I used a neat tool to check how the airflow over the body would be affected: MicroTunnel which simulates a 2D wind tunnel. The program is designed for airfoil cross-sections and thus you can't model fins or any other 3-dimensional shapes. I entered the nose and tail sections into the program and ran the simulation. The display you see below shows the pressure flowing over the rocket body. I also overlaid the outline of the fins on the chart for reference. (Click on the graphic for the full chart with axes and summary info.)
Note that while there is an extreme low pressure (dark blue) around most of the tail section, the aft end of the body is at ambient pressure (green) and much of the fin surface area exists in this region. So, there will be air flowing over the fins at the aft end, but the fins will be much reduced in effectiveness (perhaps down to a third).
This allows us to do a RockSim analysis, with a straight body tube and small fins placed aft. This is an approximation (hopefully an underestimate) of the effectiveness of the fins. RockSim places the C.P. 54% back from the forward end. Fifth-size ModelJason was concerned about the large nose shadowing the fins, rendering them less, or even not, effective. I thought it would be OK, since we were keeping the speed low. But, when I couldn't even model the shape properly in RockSim, I agreed that a smaller model should be built to test the stability. I decided on a one-fifth size model (2' overall length) as an easy-to-build size. The main advantage of the smaller size is that the body and nose can be turned out of balsa, making the construction much easier.
Above, you can see the plans I drew for the smaller version as well as the turnings that came back from San Francisco Wood Turning.
Of course, the turnings just supplied the overall shape as solid bodies. I next had to create a hollow in the nose for the recovery electronics. Also, I had to drill through the tail for the motor mount tube and create a hollow at the front for the parachute and bridle. (This took most of the afternoon of Saturday June 29, but I wanted to fly the the model at LDRS XXI two weeks later.) 
In the picture above, you can see how the nose and tail mate together to form the whole body and where they open for recovery. This is more or less how the full-size one will open, although it will use several tubes and/or pins instead of a single 3.9" tube and coupler. Note also how the nose section (on the right) includes an electronics bay. (A cover plate screws on, sitting inside the coupler and on top of the tubing section.) 
Next up was forming the fins. I decided to use 1/16" plywood laminated with carbon fiber and 4oz. fiberglass on each side. This is a quick and easy layup that produces a very strong and stiff fin without becoming too thick. (The fins on the rocket are much longer than on most rockets.) Above you can see the two fin blanks (to make two fins each) under vacuum pressure using my trusty FoodSaver.
Above you can see the two body sections complete. Now all that remains is setting up the recovery system and a quick paint job.
Here is how the electronics bay works in the nose section. An Adept ALTS25 is mounted to the electronics bay cover plate. wires to arm the ALTS25 are run out through one of the side vents and the electric match wires are run out through the plate. The picture on the right shows the nose section ready to attach to the bridle. 
And on Monday, July 8 the small version was ready to fly just in time for LDRS XXI! As detailed in the Pictures section above, the flight at LDRS wasn't stable enough to make me comfortable. So, I decided to replace the fins on the model with slightly larger ones. 
In the drawing above, you can see the larger fins relative to the scale fins. Note that the larger fins have about 4½ sq. in. more area than the old ones, even though they are only 0.35" larger in semi-span. Also, the extra area is farther away from the body and at the aft end with the larger tip. Tail SectionThe full-sized rocket was built out of fiberglass shells. This is the only practical way to make a rocket this large and with this unusual a shape. Nothing standard comes even close for the basic body shape.
The start of building was to create a quarter-section plug out of wood. First two pieces of ¾" plywood were joined at a 90° angle to create the flat edges of the quarter slice. Then quarter-circular ribs and curved-top rectangular stringers were cut out of ¼" plywood and built inside.
In the pictures above, you can see a batch of stringers stacked up with the paper template on top, ready to be cut out on the band saw. On the right, you can see the first three quarter-circle ribs laid out with the end stringers in between. 
Once all the ribs and end stringers had been installed, the excess plywood was cut away and the edges sanded to a nice slope to create the outline of the quarter plug. 
In the picture on the right, you can see the quarter plug cut and sanded to shape. Note that all the circular ribs are in place, separated by the outside stringers. More stringers will be added between the ribs to maintain the full shape and support the balsa sheeting. Each of the ribs had to be cut a different radius and its curved edge beveled to match the angle of the stringers at that point. Each set of stringers was a rectangle with a curved top to match the appropriate edge profile between those two ribs. Everything was fastened together with white glue and screws and small legs were attached to allow the plug to sit upright.
Once the structure was complete (above left) with all stringers, balsa sheeting was applied to create the surface (above right). 
Finally, two layers of 9oz. fiberglass tape (6" at the large radius and 4" toward the tip) were applied to create a hard surface and give some body for sanding and smoothing. 
My wife Renée took the pictures above of me laying up the fiberglass. This was a pretty easy layup since it's a large piece and the fiberglass tape is easy to work with.
Once the fiberglass was cured, it was time to sand it smooth (always a nasty job). Then I trimmed away the excess balsa sheeting and fiberglass to create the final shape of the plug body (above left). Also, the shape for the engine nacelle was created from a 6" body tube and plywood sheeting (above right).
The final steps for the plug were to attach the nacelle and finish the surface. Above left is the plug with the nacelle attached and all major surface holes filled. Upper right shows what it looked like after several days of priming, filling and sanding. (This is the last coat of primer; whew!) 
After the last coat of primer comes more sanding! (Isn't there always more sanding?) This sanding isn't that bad, though, since it done wet. The primed plug was wet sanded with 150, 220, 320 and 400 grit until it was completely smooth. (Photo by Renée Coker.) 
And above you can see the plug sanded and ready to create the mold. That was a lot of work and I haven't even made a part that will be on the rocket yet! Here there was a four-year interlude where I worked on other projects. In July 2008 I resumed work where I'd left off: making the mold. Because this was such a large surface, I sanded this plug all the way down (up?) to 800 grit. Then, I used a small buffer to apply mold release wax to the plug surface, twice.
Once the plug was ready, I applied a gel coat layer. In this case, I just used thickened epoxy with a coloring agent added. Once the gel coat had partially hardened, I applied the fiberglass cloth over it, starting with 4oz. and working up to 10oz. Because of the complex shape, I had to use pieces about 12" square and overlap them in various ways. The part around the engine nacelle required particular care.
After curing over night, it was time to part the piece from the mold. I had been worried that this would be difficult, since there is so much surface area, but it actually popped right out once wedges were worked in around the perimeter. 
The picture above shows the freshly-parted mold. It turned out very well, with only one small void to repair. I probably could have let the gel coat cure a bit longer as there were a few areas of cloth print-through, but it doesn't really matter for a mold. (The mold looks greenish because of the epoxy and fiberglass underneath the gel coat.) Nose SectionSince there was only a single nose piece, I decided to make it one-off using Styrofoam with a fiberglass shell. (We did a lot of this on the N1 project.) Fins & StructureThe fins are not attached just outside the shell. The shell is in four pieces so the fins create a X-shaped structure inside and the shell quarters fit between them. This makes the fins/structure the real frame of the rocket, with the fiberglass pieces just a shell. Recovery SystemSince I didn't know exactly how much the rocket would weigh, I needed to keep the recovery system flexible. (Jason and I planned on 50-60lbs. for motor design purposes, but I thought it would be lighter when finished.) Partly for this reason, and partly because I like multiple 'chutes, I decided to use three or four 8' circular parachutes, bringing the nose down on a separate parachute. The MoviesJust for fun, below are some more stills from the movie.
These old serials are really awful, but they are entertaining in their own way. 1940 was a very different time and even though there were dramatic tensions in the world, these serials still show a naivete and optimism which you don't find today. Pick them up on DVD; they are worth watching. Tom Aldridge has put together a detailed analysis of the Flash Gordon serials for those who are interested in the series itself: Flash Gordon and His Universal Serial Compatriots: A Critical and Sentimental Perspective.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
