1/12 Liberty Bell 7 Mercury Redstone MR8 - NARAM 44 (2002)
The 1/12 scale model which won Team Sport Scale at NARAM 44 (2002) is an all new second-generation version of that flown two years earlier at NARAM 2000 and utilized many improvements in its overall design as a result of lessons learned.
This model was a true engineering project with several design parameters established both by myself and by the rules of competition. I knew from the 1/12 scale Mercury Redstone from the previous competition that it would be easy to meet the 3.3 pound weight restriction.
Click on images below to enlarge.
Scale Projects-1/12 Mercury Redstone
Even though I was going to build "lighter" another requirement was that the overall model would also be stronger. The NARAM 2000 Mercury Redstone model from two years earlier was only my second venture into very large model structures using very thin plywood construction...the 1/10 Vanguard model from 1999 was the first. Thought the results were very satisfactory in both models I had learned much from them and, being a rabid "experimentalist", had developed and refined many of the plywood construction techniques by building structures almost exclusively from 1/64" (.6mm) thick plywood and testing them to destruction. For example, where my previous "plywood" models had utilized 1/64" ply skins over rather substantial 1/8" aircraft plywood internal rings, this model used only 1/64" ply in ALL of its wood structures.
Since I was going to utilize an autopilot system that gimbaled a relatively low thrust motor to achieve a nearly vertical powered flight regardless of wind conditions, the thrust structure would be entirely different and non-conventional. I decided that the whole tail section would be a non-thrust bearing structure (which would allow me to make it lighter) and that all of the thrust would be directed into the aft "tank" section utilizing a tubular carbon-fiber structure which evolved after a number of experimental thrust structures. What resulted was a visually (and structurally) light structure of surprising strength, withstanding a 100 lb. Load during development and testing.
There are a lot of details in this photo (click image to enlarge). I took photos of several Redstone tail sections to arrive at the proper location and "count" of the rivet details. The rivet detailing on the 2002 model is much more accurate and "subtle" than than of the 2000 model. The rivet patterns were developed in CAD software which was converted to G-Code and then a CNC Mill was used to emboss the rivet details onto the back of pressure-sensitive adhesive-backed Super Monokote trim material.
The paint pattern was acheived not by painting but by using black or white Monokote where appropriate. The color demarcations are butt-joints between the black and white areas...no overlapping of Monokote. This requires a fair amount of precision but is not really difficult to do, using a "wet" application of the Monokote.
Near the top of the photo you will see one of the several "weld lines" on the tanks of the model. These weld lines were done by first applying the embossed tank panel (made from Trim Monokote) to the lower end of the tank and then applying iron-on Monokote to the next tank section with about a 1/16" overlap of the iron-on Monokote over the lower panel. This results in a very subtle and uniform simulated "weld." All other tank sections were done in the same way using white iron-on Monokote.
Here is a view of the "other side" of the tail section clearly showing one of the two scale access hatches. The hatch detail was made entirely of white trim Monokote. The "base layer" of the hatch detail is actually three thicknesses of trim Monokote with the outermost layer being the embossed detail layer. Note that there is a step in thickness near the top of the panel right at the top edge of the hatch door. The hatch door is from two layers of trim Monokote. The "hinge" is an Evergreen styrene rod.
With the exception of the fueling ports and the recessed area of the Peroxide Fill, everything you see is either black or white trim monokote and not paint. The areas at the base of the fins are from Chrome trim Monokote.
The fins are made from expandable urethane phone made in rubber molds. A single coat of gelcoat resin was applied to the fin mold and then the two-part foam poured in and allowed to expand. The cured fin is removed from the mold and trimmed and cleaned up and sanded at the root to fit the contour of the tail cylinder. The embosed Trim Monokote of the appropriate colore is applied the appropriate fin. Finally, the fins are attached to the tail section (which has already had its embossed Trim Monokote applied) via double-faced tape and two nylon screws (for each fin) from the inside of the tail section.
The fueling ports are CNC milled from a material called Renshape. The curved rain gutter over the hatch is from "L" shaped Evergreen structural shape with rivet detailing.
The entire tail section of the model "unbolts" from the tank forward of the gap between the tail section and the tanks. There are eight small screws in the gap that attach the tail to the rest of the model.
The instrument section is the most heavily detailed portion of the Redstone booster with many hatches, antennas, and gobs of rivet detailing. Again, the rivet detailing was done by creating patterns in CAD and then converting to G-Code and finally applying the rivet details by embossing the backside of Trim Monokote.
The various hatches are double-layers of Trim Monokote with the larger "rivet looking" details around the periphery of each hatch being round-head brass nails.
The antennas at the aft end of the instrument section are made from styrene rod and are removable. Each antenna simply inserts through two holes in the instrument section with the forward and middle antenna legs being inserted into the holes. Like the tail section, the instrument section "bolts to the forward tanks of the Redstone booster via four small socket-head screws cleverly hidden (I think) in full view. Look at the lowermost leg of the antennas and you will see the small silver "dome" of the screw.
Not prominent but just visible is the black "gap" between the instrument section and the Redstone booster. In this gap are hidden the optical ports that the autopilot uses to "see" the horizon. You can just barely see one of the ports in the gap near the lower left corner of the second-from-left white roll pattern detail.
The spacecraft adapter ring features almost a hundred bolts and nuts, scale Marmon clamp cover retractors, and the not-so-correct fairings located at three points about the periphery of the red clamp covers.
The roll pattern on the instrument section is black paint over the white Monokote.
The capsule is a thin resin component which is cast flat in a rubber mold and then rolled into the conical shape. The cylindrical recovery compartment at the forward end of the cabin cone is also a resin casting which is a thin-wall centrifugal casting.
The master for making the rubber mold was machined from Renshape using a Sherline CNC mill. The "screws and washers" around the corrugated shingle panels are each individually applied. The washers are of three different sizes and are hand-punched from thin sticky-back vinyl while the screws are tiny dollhouse nails. The red Marmon clamp cover is built up from sheet styrene and the conical adapter between the Marmon clamp ring and the mounting flange is .01" styrene.
The adapter mounting ring is also of styrene and has nearly one hundred tiny brass hex bolts and nuts about its periphery.
Cable detailing at the base of the tower (right) consisted of different diameters of solder.
The escape motor and escape tower were major projects in themselves.
The original prototype tower was fabricated from machined tubular aluminum and was tremendously strong but I feared objections to such a metal structure on a "model rocket." I duplicated the tower using styrene tubing and rod which you see here.
Almost the entire motor, the forward closure, tunnels, nozzles, jettison motor, aft closure and other details are resin castings. The cylindrical portion of the motor is a layer of .01" thick styrene sheet over a paper tube.
The Grand Central Rocket Company logos and lettering on the motor, the nozzles, and the labeling on the small jettison motor which is nested between the three main escape motor nozzles are all decals which were printed on an ALPS MD5000 printer. Even the smallest lettering is readable. Notice that on the main motor markings (right) that the required NAR Team identification, as required by NAR rules, is integrated into the labeling.
Preparing the model for flight was involved and took a bit of elbow room. Fortunately, "Scale Day" does not involve the volume of models that other competition days do. Since mine was to be the last official flight of the competition, I had a bit of extra time. Still, I had to make sure that both sets of batteries were fully charged, that the RC systems were functional. The horizon sensing system had to be calibrated and zeroed. The ejection system had to be prepped. Ditto with the motor. The two rather large recovery parachutes (a 36" canopy for the capsule and a 60" canopy for the Redstone booster) had to be folded to fit into the 1.6" diameter recovery department. Lastly, a fresh Grade A large hen's egg was bagged and nesteled into the Mercury capsule. This would count toward additional flight points, simulating a delicate living payload, if successful.
The booster was detailed right down to the umbilical connectors and turbine exhaust on the tail of the model. There were even replicas of the graphite exhaust vanes in the scale locations around the model rocket motor.
In flight, the ejection charge of the model rocket motor simply vented into the interior of the booster body which was not a sealed unit and was not linked to the recovery system (the presure could vent aft around the model rocket motor nested in the scale engine nozzle). Instead, the ejection module which was a 13mm motor casing with the appropriate amount of black powder was plugged into the aft end of the 6" long by 1.6" recovery tube at the fore end of the instrument module. A radio command from the ground triggered a pair of redundant ejection igniters to fire the ejection charge.
The image on the left shows the Redstone booster and the bulkhead on which the electronics for the model will rest. Everything you see here, with the exception of the paper tubes that will mount the photosensors, is 1/64" thick aircraft plywood. I created a tank section that was "scale" with respect to each section of the real thing between welds. Scale weld lines would be located over the joints between tank sections.
The image at the right shows the installation of the photosensors and their circuitboatd. This circuitboard plugs into the autopilot controller. The hole in the center of the bulkhead is for passage of the control wiring leading back to the engine gimbal as well as passage for the Kevlar shock cord which attaches to the thrust ring of the gimbal mount.
The tube which passes through the bulkhead at about the 8:30 position is the internal launch lug which extends back to the tail of the model.
The covering and lettering have been applied to the model at this point. Also, the wooden tabs serve as a secure point for the screws which hold the instrument section in place to thread into.
The "loaded" model electronics bay. It looks like a wiring mess but everything was designed to retain the wiring and connectors that were purchased with each electronic component. This makes for easy component replacement or repositioning. Everything here is a ready-to-fly off-the-shelf commercially available component for the model RC airplane market.
At the 10:00 o'clock position are a photo sensor and the base loaded antenna. Proceeding clockwise is the main blue system battery (a NiMH unit) which provides all electrical power to the model with the exception of the ejection charge ignition system (after this photo was taken a charging jack was added to allow external charging). Next to the battery is the four-channel RC receiver. At about the 12:30 position is the signal generator which provides a constant signal to the autopilot system. Unlike a "standard" RC autopilot system which uses two channels of the RC receiver for the two axis' they would normally control, the two gimbal channels are controlled by a constant signal from the signal generator. This de-links the autopilot from the RC system and centers the engine gimbal and avoids any unintended gimbal movement due to spurious RC signals. Between the signal generator and the yellow ejection batteries is a mercury switch which, as long as the model is upright and on the ground, prevents accidental recovery system triggering. Finally, the black module at the 3:00 o'clock position is the autopilot controller.
The bare thrust structure was primarily six tubular carbon fiber struts which transmitted thrust into the thrust ring at four points and into four longerons on the inside of the aftmost Redstone tank section. A load of 100 lbs was applied to structure without problems.
The thrust structure with servos and control linkages (right) are installed and ready. Over the four years prior to this configuration my engine gimbals had grown simpler and simpler. Early designs were complicated by my attempts to make the geometry of the mechanism to allow motion that was as linear as possible no matter how extreme the gimbal motion. It became apparent that such attempts were not necessary.
The engine gimbal itself is a very simple structure made from a section of Plastruct tubing (grey). When combined with the geometry of the linkages and servos it has a motion of about 6 degrees from center and can move fully through any range in about .2 seconds.
I toyed with the idea of building an even larger Mercury Redstone in 1/10 scale to go with the scale of my Vanguard. But, having a ready supply of my 1/12 Mercury capsule kits, I resisted reinventing the wheel at a larger scale. At this time, the MRC Atomic-City 1/12 had not appeared on the market (not that I would have used a "commercial" kit not of my own making...call me biased). Anyway, I ruled that the model would not exceed 3 lbs. And, while the NARAM 2000 model utilized radio-control to activate the recovery system, this model would utilize RC recovery in a different manner. Also, whereas the NARAM 2000 model did not utilize an "autopilot" to assure a vertical trajectory regardless of wind conditions, this one would. Though I knew the model would fly very well without the autopilot, my experience two years earlier where a storm was passing over with erratic winds almost caused me to voluntarily abort my flight attempt (and therefore be disqualified from the competition). After all, a sudden gust of wind during my competition flight of the 1/72 Saturn V in Arizona five years earlier had resulted in the severe weathercocking of the model and its subsequent crash.
Tom Beach is taking a look at some of the detailing on the model just prior to launch
This video by James Duffy was of the contest winning flight. If you watch closely, you can see several course corrections under thrust of the G25 motor which gives the model a nice steady ride.
You will also notice during recovery that there is a small parachute above the main canopy of the booster. This drogue deploys and then releases the main chute once the shock cord is withdrawn to full length.
Click on image (above right) to play WMV video of the model in flight.
Model Rocketry and Spacemodeling?
Drop us a note!
If you find the information or images here of value, please consider a donation* to keep the Accur8 website going! Donations will be handled securely via PayPal.
*Due to PayPal fees, no donations less than $2.00, please.