Super Guppy

Building the Super Guppy

Drawings

90 percent of the Super Guppy was built inside the computer before one piece of wood was cut. This series of drawings were needed to check for any interference between the cowl and the gearbox. First the motor and gearbox had to be measured and drawn. They were the put into place on the firewall, then the cowl was put in it's place. DesignCad 3d has an interference tool especially made for this situation. Some was found and the firewall was pushed back in the design stage before it was built. From a scratch-builders perspective, this has proved itself to be a well thought out design. Special attention was paid during the "kitting" of the model, from the measuring the size (a one inch square was added to all drawings and measured as it was coming out of the printer) of all plans and part sheets, to the cutting of all wood parts that were to become a Guppy.
(30K JPG image)

At this stage of constructing the virtual-Guppy, the outllines have been traced from the scanned bit-mapped image, and into vector format. This allows for the rotation and final placement of all outlines. All of the major bulkheads and ribs have been located in their proper location and orientation. The next job is applying the skin. By selecting the correct number of vertical and horizontal breaks, DesignCad creates the wing ribs in place and the correct size, ready to be traced.
(35K JPG image)

All of the jigs needed for proper alignment and decalage of the model were built in the same manner. First all unneeded layers were turned off, then an imaginary floor was temporarily put into place in the form of a plane. Then using the rib or bulkhead outline for the top of the jig, vertical lines were connected between the outline and the floor.
(38K JPG image)

This is Dan's magnificent 3-view of the Aero Spacelines 377SGT Super Guppy Turbine. Dan drew these after the model built, and from the original virtual-Guppy he built. An experienced builder could propably build their own Guppy from these drawings. Dan was very complete in his drawings. He included all bulkheads; this is not seen in many 3-view drawings.
(37K JPG image)

Pictures

The vertical fin and rudder has been pulled up from the "building board". This picture was taken before sanding the fin and rudder. The rudder was tapered to the trailing edge, sanded flat so as to minimize any "starved horse" look when covered, while the fin was left flat only rounding the the leading edge, with the dorsal fin tapering to a point at the front. The unusual construction of ribs was done to simulate the panel lines of the full size Super Guppy, and no sheeting was used in an effort to keep the model as light as possible.
(20K JPG image)

Both the vertical and horizontal tail surfaces were built using 1/2 inch thick contest balsa. The stabilizer and elevator were first tapered from 1/2 inch at the root to 1/4 inch at the tips (from 1/4 inch either side of the centerline to 1/8 inch either side of the centerline respectively). Then the elevator was tapered from the main stab spar to the trailing edge of the elevator. Again block sanding flat to minimize any fabric sag. The ribs were again placed on the scale panel line location.
(21K JPG image)

The conference table that was used for the construction of the Super Guppy measures approx. 4 ft. by 8 ft. The plans were laid down on the table and glass was laid down over the plans. Kitchen plastic wrap was put over the glass to keep the parts from sticking too well to the glass. The plastic wrap is the actual building surface. The top and bottom keels were glued into place first, then using a tall square, the bulkheads were one by one glued onto the keels. The side keels were dry fit into place first, helping to hold the bulkheads square and flat, then glued. All of the stringers first had to be spliced before being rag-soaked with water (at 8 ft. they were too long for the bathtub) to soften the wood for bending around the nose of the Guppy. As can be seen, the fuselage takes up the most of the table.
(19K JPG image)

3/16 square stock contest balsa was selcted for the stringers as the best compromise between weight to strength ratio. Before the fuselage was sheeted, it was sanded just enough to smooth all bulkhead to stringer joints. 1/32 contest balsa was used to sheet the fuselage. We decided to sheet the left half of the fuselage in an attempt to keep it as straight as possible after pulling the fuselage up from the building board. The very top and bottom bays and the tail section of the fuselage were left unsheeted until last in case any access was needed later. This actually worked very well, the fuselage didn't warp at all. It did however look like a odd shaped canoe when flipped open side up.
(20K JPG image)

The fuselage jig was also designed in the computer. The fuselage was first rotated 90 degrees along the longitudinal axis. At each bulkhead station, Dan drew a box using the bulkhead outlines for the fuselage saddle, down to virtual floor. Egg crate design was selected for it's simpicity. The jig was squared and glued directly to the glass to help hold it in alignment. The fuselage dropped right in like it was made for it. It was, but that's beside the point. We later found that the jig worked so well in fact, that gluing either the fuselage to the jig, or the jig to the table became unnecessary, and the jig was used as a craddle for working on the fuselage.
(29K JPG image)

The Balsa Canoe! Actually the fuselage was constructed using 3mm, 3 ft. by 7 ft. mohogany plywood doorskins for all keels and bulkheads. We decided to cut one piece keels rather than splice several smaller pieces together. The bulkheads were cut from the doorskins also. We initially purchased the plywood doorskins from a local HomeBase for about $7.00 each. Price was a consideration, because we found during parts layout for cutting, it would take at least 3 doorskins for the keels and bulkheads, not including the fuselage jig, that took another doorskin. Here the fuselage jig is glued down to the glass over the jig plans, and the fuselage has been put into place and glued, with the top and bottom keels flush with the top of the jig. The bulkheads were thinned from the intial design during cutting, as it was decided that less material on the inboard side (away from the stringers) would still retain enough strength, and would save weight.
(20K JPG image)

Here the fuselage is seen with the starboard side bulkheads, side keel and upper stringers. The lower stringers were left off temporarily for access to the inside of the fuselage. At this point, the fuselage can be pulled out of the jig and checked for warps in the fuselage. Some elements of the actual engineering, such as where to put the battery pack and how to mount it, the nosewheel mount and steering set-up, and the mounting of tail surface servos was not done in the computer, as we decided this type of work is more expediant if done in the real world.
(31K JPG image)

We initially purchased 32 Sanyo 1700 mah SCRC sub C-cell batteries, not knowing whether it would take 7 or 8 cells per motor to fly the Guppy. After motor and prop testing we decided that given the power to weight ratio, the extra 8.8 ounces in battery weight (2.2 ounces per cell), didn't provide enough additional thrust to push that weight through the air. The battery cells were first laid out on the bench on top of two strips of masking tape. The tape was used to hold the cells in alignment for soldering. Once we had two 14 cell sticks (actually, we built the battery pack with fourteen cells on one side, and eleven on the other, leaving one end of the battery pack open for later installation of either 3 or 7 cells, depending on the motor/prop test results), a four sided box was built around the battery pack using 1/8 inch balsa sheet. The masking tape was removed and the box sides were glued to the battery pack using silicon RTV. One hardwood rail was fitted to each side of the box. We decided to make the battery pack removable for charging, so a simple drawer design with a rear stop and front retaining pins was adopted.
(33K JPG image)

This is our low tech motor test stand. It looks crude, but it provided us with some very crucial data. Seen here starting with the stand itself, is the duel purpose motor mount, used for both the electric motors and an O.S. 15FP during baseline testing. The O.S. 15 with a 8x4 APC prop developed 1-1/2 lbs. of thrust at 11,000 RPM. This was the minium amount of thrust we thought would be needed to safely fly the Guppy. In order to achieve the most accurate results, we mounted all radio gear(receiver, airborne battery pack, switch harness, 7 or 8 cell 1700 mAmp motor battery pack, and speed controller w/fuse) on the stand. We also mounted a amp meter, and volt meter to the stand.

This left us with a free-rolling unit which we were able to test volts, amp draw, RPM, thrust and duration. Since the electric power is clean and quiet, all testing with electric motors was done in the author's garage on the glass covered table for the least amount of rolling resistance(about 2 grams start-up, and 1 gram constant under load). Also seen is the tachometer for measuring RPM, simple spring scale(gradiated in grams), and various props tested. The "tailwheel" has a safety line attached, with the other end tied to a heavy tool box.
(22K JPG image)

Because of space considerations during transport to and from the flying field, the Guppy's tail needed to be removable. Seen in this photo is the initial engineering for tail surface actuation. Note the cross members in the rear fuselage for the push-pull tubes. This idea was later scrapped in favor of free floating pushrods, again for the reasons of weight and simplicity. The support necessary for the push-pull tubes and inner pushrods didn't provide enough flexibility to lift the tail section up enough to connect and disconnect the connectors from the tail surfaces, and added too much extra weight.
(30K JPG image)

The fuselage at this point weights about 1-1/2 lbs. It took a while to get used to the lack of weight. We futher reduced the fuselage's weight by cutting lightening holes in the 5 wing saddle bulkheads. Experience has shown that to build a light model airplane, one needs to save a little weight every step along the way. Every little bit helps. Notice that the push-pull tube supports are gone. It was decided to keep the "girder" servo tray, we would be able to run the pushrods from their existing location, back to the tail.
(29K JPG image)

From this angle, the size and unusual shape of the fuselage is more apparent. Neighbors often wondered what type of blimp we were building. Luckily there was always a picture of the full-size Super Guppy on take-off hanging on the wall, to inspire us, and show the neighbors what we were building a model of. This began a trend that took both Dan and me by surprise, and continued the entire life of the Guppy. Everybody had a "Guppy Story". Or so it seemed. Many remember seeing the Pregnant Guppy flying in and out of Long Beach airport as children in the 60's. One guy's dad worked for Aero Spacelines, and he remembers as a kid 9 or 10 years old, playing on the Pregnant Guppy while they were building it. And we thought we'd picked a relatively obscure airplane to model.
(20K JPG image)

Saving weight a little here and a little there meant building up the filler block needed to complete fuselage.
(31K, 32K & 27K JPG images)

This series of pictures show the construction detail of the nose gear and cooling scoop incorporated into the nose gear doors as seen in the first picture. This frame work for the cooling ducting looks admittedly a bit strange, but it was built with weight and funtionality in mind. It had to direct the cooling air over the batteries in flight and be able to be removed without any tools to get the battery pack out of the airplane for charging. The third picture shows how the ducting was built around the nose gear steering pushrod. The last picture in this series shows the finished cooling duct with it's tissue paper covering. This was sealed with two coats of dope. Also seen is the nose wheel servo and battery rails and battery pack support. The nose, ducting and battery pack were all designed to be removed without tools. The lower scoop was also designed to be removable, but it was bolted into place using only one screw.
(40K, 52K, 48K & 46K JPG images)

The wing was built using the same glass over plans method. This stage of construction is very conventional. Contest balsa was used for the entire wing structure. As this was a prototype, certain changes would be made in future models, such as replacing the 3/8 inch square balsa spars with the same size in basswood or light spruce, and full length 1/64 or 1/32 plywood rib doublers at the landing gear station ribs instead of the half doublers used. The ribs are jigged up with a piece of 3/16 square strip stock, sighted from the tip to the root for alignment, and glued. The leading edge has been installed as well as the top spar. The wing is now ready for sheeting.
(24K JPG image)

The leading edge sheeting has been added, and the trailing edge core sheeting has been fit into place. The wing is built upright on the "building board", saving the bottom sheeting until after the landing gear blocks and wiring tubes for the servo extentions have been installed. Seen in the background, is are the other wing section's ribs and trailing edge sheeting. We felt that it would be lighter and easier in the long run to build up the ailerons rather than use commercially available solid balsa trailing edge stock. Because the design uses scale profiles throughout, including wing planform and reflex airfoil shape, sanding or planing the ailerons seemed like it would be more work. This proved to be true.
(25K JPG image)

The trailing edge sheeting has been added, as well as the inboard center sheeting and the outer sheeting planned for the base of the outboard engine nacelles. The wing is now ready for capstrips on the ribs. One of the benefits of electric powered models is the almost no vibration, this translates into being able to use much lighter structures on the airframe, since all "wet" power designs must be able to absorb the vibration of the engine.
(22K JPG image)

It looks a lot like a wing for a glider. Actually, the wing turned out light enough to be used on glider. We found ourselves constantly reassuring each other that "the wing area doesn't tell the whole story". It was a high-aspect, thick chord with a fair amount of reflex on the top portion of the airfoil. The B-29 was designed to carry a heavy load effiecently. We hoped that some of the characteristics would stay with the wing when scaled down. And we figured on some lift from the fuselage, though where and how much we had no way of guessing.
(25K JPG image)

Seen here is the bottom of the wing. The landing gear blocks have been installed, and both wing panels have been completely sheeted and sanded awaiting mounting of the engine nacelles. The wing center section still needs to have the shear webs installed and bottom sheeting fit and glued.
(18K JPG image)

The majority of the detail construction would be done before joining the wing panels. We began with the engine nacelles. The nacelles were constructed in much the same manner as the rest of the airplane, with plastic wrap over glass taped to the plans serving as the building surface. Because the inner and outer nacelles are different, only two sets of bulkheads could be cut at one time. The keels were cut individually as extra care was needed at the rear because of a very shallow taper. The bulkheads were all contest grade balsa and the keels were cut from the copious amounts of scrap plywood this project generated.
(26K JPG image)

Seen here are the wing and nacelle jigs. Missing from this photo are the triangle braces for the nacelle jigs. Having these jigs proved invaluable when it came time for the rebuilds after the crashes a prototype seems to endure. The same jigs were used throughout, and will be again for all future Guppys we build. As with all parts cut during the "kitting" phase of the project, care was taken to get consistant cuts as a test of the computer's rendition of Dan's design. And here, like all other parts, the fit was perfect.
(41K JPG image)

The plastic model of the Super Guppy was always near at hand during construction for reference. This model turned out to be surprisingly scale when compared to pictures of the full size Guppy. Most medium to high quality plastic models are a great source of scale documentation or ideas for the endless "next" airplane to build. The wing jigs were glued in place over the plans with the wing seated but not glued into place and a square was used to align the wing over the plans. This assured the proper alignment of the nacelles on the wing.
(24K JPG image)

The wing has now been glued into place on the jigs, and the nacelle jigs' placement has been finalized. It was necessary to round and final sand the leading edges of the wing for proper seating in the wing jigs. Also seen in the background are the nacelles ready for attachment, having been built in the previous construction step. We placed the jigs in their locations shown here for the purpose of supporting the wing in the area needed, towards the root, where the nacelles would be attached. No tip jigs were used at all for this step in construction.
(25K JPG image)

The nacelles were first dry-set into their jigs to check the fit. Where the rear of the keels sat on the wing was the determining factor in the fore-aft placement of the nacelles. After aligning the vertical centerline of the nacelles to plumb, they were tack-glued onto the jigs. We also were able to remove most of the minor warps from the plywood keels by carefully glueing the rear of the keels to the wing surface. Another plus for electric multi-engine models, is that no internal connecting structure is needed between the wing and nacelles. No vibration, so surface glueing is plenty strong enough.
(21K JPG image)

This is one of the steps we decided would be done more easily in the real world. Actually making working nacelles. We first stabilized the front portion of the nacelles with a flat plate of balsa plywood (three pieces of 1/32 balsa glued together with the grains of the wood running perpendicular) between the leading edge of the wing and rear of the second bulkhead. Two braces were also added at this time from the outer edges of the balsa plate and the top bulkhead-keel joint. This stabilized the nacelles enough for the next step, the rest of the bulkheads and stringers. We changed the design of the nacelles just after these pictures were taken. The original design called for direct-drive motors to be clamped to two dowels protruding from the front of the existing bulkheads. The Leasure gearboxes were made to mount to a firewall with screws or bolts.
(19K JPG image)

Our solution was simple, just tack on top and bottom keel extentions, add a couple of more bulkheads in front of the existing ones and add stringers. The firewall was cut out of scrap doorskin and glued to the front of the balsa firewall. The hole for the motor was cut and blind nuts were installed on the rear of the firewall before gluing the firewall into place. All cooling holes were cut in the remaining bulkheads before glueing in them into place. The white tube in the front of the nacelle is made from printer paper rolled into a one layer tube, and tack-glued into place with thin CA. The thin CA wicks into the paper very nicely and becomes almost like a very thin plywood. The rear cooling exhaust tube was made from black paper. The hollow bulkheads were necessary because the cooling air the motors were going to receive would come via a pressure cowl arraignment.
(44K JPG image)

This series of pictures was taken at a giant scale fun-fly in Lake Elsinore, California. We were very pleased with the way the Guppy was turning out. The warm reception we received at this fly-in helped fuel our desire to see the project through to the end. By comparing our Guppy to the plastic model, it was easy to see the lines were correct. Using scale outlines throughout the entire model was paying off. All of the angles of our Guppy matched the plastic model and pictures of the full size Super Guppy when viewed from any direction.
(37K JPG image)

Although the actual stringers and their slots weren't drawn in the computer, their locations were. Dan first superimposed all of the nacelle bulkheads on top of one another, once each for the inner and outer nacelles (the opposite side was mirrored as a last step in the process). Once the bulkheads were placed into their proper locations via a center mark, Dan was able to draw a radius of lines were the stringers would go. This gave the stringers a smooth, even flow from the front to the rear. The alignment marks were then transferred to the bulkheads and stringer slots were cut centered on the marks. The nacelle bulkheads' final placement was done via the trim and fit method, but a final tracing was done of all nacelle bulkheads before they were glued into place. With this job done, Dan invited me over to his house for a "stringering" party. We used 1/8 inch square contest grade balsa for all nacelle stringers.
(33K JPG image)

From this angle, it's easy to see the size disparity between the fuselage and the wing. The wing fairing and belly pan were constructed after the fuselage was fully sheeted. Many people we spoke with at later contests often remembered seeing the Guppy at this fly-in, and were glad to see that we finished the project. Everyone always expressed a desire to see the Guppy fly.
(32K JPG image)

Something that was always a novelty for us, was the people standing in the background staring at the model in disbelief. People would begin gathering as soon as the wing with it's four nacelles was pulled out of the back of the truck, followed by what at first glance, appears to be an odd shaped blimp. The tail feathers were removable for transportation, giving us three major sub-components ( the wing, fuselage, and tail feathers), and two minor ones (belly pan and nose cone) which when assembled, made up the Guppy. This is the Guppy's first day at the flying field being prepared for the test flight.
(32K JPG image)

The Super Guppy is enduring a brief photo session before test flight. Although the engine nacelles look small when compared to the fuselage, they are actually about a foot long with the circumference of your average 40 size R/C model airplane. If the nose of our model opened in a scale manner, four basketballs would fit inside the fuselage with almost enough room for a fifth! This is a good angle to see the world's simplest wing fairing. Also not seen in any photos, the cooling exaust hole, is the dark area on the bottom of the fuselage just as it angles up towards the tail.
(40K JPG image)

Dan's checking for the amount of ground clearence. Even increasing the ground clearence an inch by lengthening the landing gear struts and using larger than scale wheels, we still only had three inches of ground clearence total with the model at rest. The model would be limited to relatively flat take-offs without dragging the tail during rotation prior to take-off. The shape and height of the fuselage is seen very well here.
(24K JPG image)

If you can't see Dan's knees ashakin', it's because this photo was taken using special high speed photography. We were both pretty nervous at this point, almost two years of work about to take to the sky. We hoped. This is a good size comparison, the top of the rudder is 29 inches up from the ground, the top of the fuselage is almost two feet up from the ground! We called the Super Guppy our big small model.
(19K JPG image)

AARRGGHH!! Why didn't we follow our first instinct?!! At this point we were pretty discouraged. We had to sit back and take a look at what happened and why. Luckily we video taped the test flight and were able to go back and analize what went wrong that caused the model to crash. We inflicted much pain on ourselves by viewing the crash many times. We determined that the airplane was tail-heavy. Radically so. It had the characteristic excessive pitch sensitivity seen in the wild pitch up imediately after lift-off. After taking a second look at the actual damage, which was limited to the left wing panel from the battered outer nacelle to the wing tip, and partially crushed bulkheads on the front top portion of the fuselage. Some of the sheeting on the fuselage was of course broken and gouged.

Once the damaged sheeting was removed, it was a simple matter to repair the bulkheads, replace the broken stringers and resheet the fuselage. The wing from the right wing tip to the number one nacelle was perfect, we built another full left wing panel in anticipation of cutting off the left wing panel and cutting away the number two nacelle to be glued to the new wing panel and build a new number one nacelle. We later decided to replace only the broken portions of the wing. The new outer portion of the wing was butted glued to the rest of the wing and fiberglass reinforced. The joint was under the number one nacelle keeping it mostly hidden from view.
(39K JPG image)

After two months...Success!! In this recently declassified Russian spy photo, the Sooper Guppinski is seen in the skies over Moscow, no doubt ferrying parts of the Concordski. Actully the rebalanced model flew very smoothly. A little down trim was needed and the Guppy flew hands off straight and level. Some rudder was needed to co-ordinate the turns, with a small amount of rudder held in during the turns.

We came up with a unique solution to counteract any possible problems with p-factor or torque the motors produced(geared electric motors do produce a fair amount of torque). The motors were mounted with zero down and right thrusts. After testing and synching the motors, we mounted the strongest motor in the #1 position (starting from the left outboard nacelle), the second strongest in the #4 position, 3rd strogest in the #2 postion and the weakest in the #3 position. We figured that having slightly more thrust on the left side of the airplane would negate any tendancies for the torque of the four motors to pull the airplane to the left, and it did. The sight and sound of the Super Guppy flying above was most satisifying.
(8K JPG image)

These series of pictures were taken at the Western Regional Scale Masters Qualifier. We competed in the new Team Scale category. The guppy was an instant success. We were one of the first to fly Saturday morning and had the sky to ourselves. No one wanted to fly at the same time as the Guppy because as we later found out, they all wanted to see the Guppy fly, at least the first time. I spoke with one of the flight judges during our second round of flying, who was glad to be judging the Guppy because as he put it, "Last round, I had a hard time keeping an eye on the airplane I was supposed to be judging!"
(10K JPG image)

The Guppy is seen here on final approach prior to landing. Some power was needed for the approach, the drag from the fuselage and four propellers at idle was enormous, giving the Guppy the glide ratio of your car keys. No matter how many spectators were present, or how limited their knowledge of R/C, the flying field always became ghostly still for the Guppy's landing.
(14K JPG image)

Dan's about to "grease" on another landing. At about 2-3 feet the power is brought to idle and flare for landing. The ground handling on the Guppy proved to be impecable during both the take-off run and landing roll-out. Although Team Scale wasn't then a category in the Nationals, we were invinted to fly in the noontime demonstration. We were of course quite flattered by the offer. We felt a little guilty at first with the amount of interest the Guppy generated over other models present at the contest, both at the contest and in later magazine coverage of the contest. But then we came to the conclusion that "That's racing, you run what you brung." Our project was getting the amount of recognition it deserved for the amount of effort put into it.
(17K JPG image)

Once again the famiiar sight of bewildered onlookers. Seen in this photo taken at the Western Regional Scale Masters Qualifier from the left is; unknown Flight Judge, Dan, Author, unknown Flight Judge, Dick S., Dad Jan. The guys yukkin' it up have seen the Guppy fly before. Scale Model Research's Bob Banka's jaw almost hit the sidewalk when we told him the Guppy weighed thirteen pounds.Compare the size of the fuselage with the size of the wing's shadow on the ground.
(43K JPG image)

The Super Guppy is seen here on display at the 1995 Scale Masters Championships. The promoters of the contest thought it would add a nice touch to have the Guppy on display while the contestants' airplanes were being static judged. This was in addition to being asked to fly as a part of the lunchtime demonstration show. We felt doubly honored.We spent the most of the day at the static judging, talking with the other contestants. The Guppy proved to be a good conversation starter, and a great way to meet many new friends who share a common love for the hobby.
(27K JPG image)

Finally, it looks like a Super Guppy. The most recent additions include the nose section, cowls and spinners. The lightest way to build the removable nose section would be to construct a light wooden framework with a vacu-formed plastic nose "facade" glued to. This meant a plug would have to be made for pulling the plastic. We used a plywood frame with foam filler blocks carved to shape. The windows were 1/32" birch plywood cut to the proper size and glued into place. The plug was the glassed and sanded to a smooth finish. The cowls were carved from a plywood and balsa sandwich.

The cockpit plug was sanded smooth and glassed. We differed here for the vacu-forming plugs from the nose section. We waxed the plug and using an organic molding compound, made two female molds. It took a little experimenting to get the proper technique for making usable molds. In next step, we poured plaster of paris into our female molds, tapped for bubbles and waited for the plaster to harden. Plain plaster of paris works very well for vacu-forming as no mold release is needed. As always when molding plastic, the cleaner the mold is, the clearer the plastic will be. Boiling the plastic during the heating process, and junk on the mold during the actual molding, are what most junks up plastic.
(38K JPG image)

Movies

This video was shot of the Guppy's first successful test flight. Dan won the coin toss for the first test flight which resulted in the crash(through no fault of his own), so now it was my turn to try my hand at the helm. I let the Guppy build up speed on the take off roll... it lifted off by itself and began a gentle climb-out, with only minor trim adjustments, flying as if that's what it wanted to do all along. This solved the problem we had with limited rotation angle. It's a probably a good thing the Guppy flew as well as it did, because I was pretty nervous. The Free Flight guys flying across the hobby area of Mile Square Park certainly got quite a show that day.
(1655K QuickTime video)

The Guppy is seen during an overhead flyby. The sound of the geared electric motors, combined with the slow speed at which the airplane flew, was better than we could have hoped. The motors held their sync, never missing a beat. There was some question in out minds before this flight regarding the small size of the control surfaces, and how efective they would be. It turned out to be not a problem at all. The control in all axis was good without being sensitive. The Guppy took longer to build than we had anticipated, was more work than had been planned, caused much too many headaches, and reaped for both of us, more rewards personally, than all of the problems combined. The Guppy steals even the best Mustang's thunder.
(1354K QuickTime video)

Link here for Mac users having problems viewing movies. Use the Indeo Video 3.2 for Mac that installs Indeo version 3.22.24.09 for Mac 6.0.7 or later with QuickTime.