INTRODUCTION
This project started because I wanted t' do somethin' different. Avast! Arrr! I had been doin' camera rockets exclusively for over 12 years and wanted a different challenge. Avast! When I considered an upscale o' a model rocket, matey, I knew t' only choice for me was t' Mars Snooper II (Photo 1). Arrr! It was first released by Estes in 1975 and ran through 1980. Because o' its futuristic design I thought it would be a challenge and an attention getter. It proved t' be both.
I had a Mars Snooper II a long time ago but I never cared for its paint scheme. Ya scallywag! T' original was painted almost entirely red with only white decals. Ahoy! Well, blow me down! I painted mine red but I added black for t' fins, reducers and nose cone. Avast! Blimey! For the upscale I wanted a multiple color scheme so I opted for a white body tube with black fins and reducers and then trimmed with silver and red (Photo 2). Also it is nay a true upscale because o' t' coolin' fins. T' original had only 8 and I wanted nine for symmetry.
Construction
I decided t' make mine out o' fiberglass. Avast! T' 4 times upscale showed that it would be 8 feet tall (Photo 3) and t' fin span would be about 3 feet, and I knew I needed somethin' substantial t' withstand t' forces o' flight. I designed it around a 75 mm motor mount so it could be capable of (theoretically) Level 3 flights. Well, blow me down! Because o' these differences I decided t' call mine t' Mars Snooper III. T' first challenge was t' general design o' the rocket. Arrr! With a 75 mm motor mount thar wasn't enough room in t' 4" body tube for both t' longer motor and a parachute. Begad! Begad! T' parachute would have t' go into t' upper 3" body tube. How would they be connected t' get the parachute attached t' t' lower section and still have access t' install the electronics for ejection?
I designed t' motor mount tube t' go all t' way up t' t' top o' the 4" tube and use a long coupler that t' 3" upper section would slide over. Ya scallywag! T' coupler would have t' be long enough for screws t' be installed above the upper transition. These screws would prevent t' upper body tube from comin' off at ejection. Ahoy! T' electronics would be installed via a hole in the coupler and then slid down into t' 4" body tube and attached.
Photo 4 shows t' motor mount with t' fins before t' 4" body tube is installed. Well, blow me down! Blimey! T' motor mount extends t' t' top t' t' outer body tube and t' coupler extends 12"s beyond it for t' upper body tube t' attach. Ahoy! Blimey! T' openin' in t' coupler tube allows t' electronics t' be installed and t' lower openin' is so it can be placed next t' t' outer 4" body tube. Avast! Blimey! Ya scallywag! Blimey! This allows t' outer body tube to be freefrom a door t' access t' electronics.
T' design also proved t' be a challenge because I knew t' fins had t' be attached t' t' motor mount very well.
T' fins would be heavy and t' acceleration could easily rip them off if they were nay attached well. Avast, me proud beauty! T' shock when t' rocket landed would be even greater. Blimey! Ya scallywag! T' fins were constructed from 1/8" G-10 fiberglass laminated with 3/16" plywood. Ya scallywag! T' entire thickness would be a true scale 1/2 thick. For t' fins outside t' pods (Photo 5), I used slightly thinner 1/8" plywood so they are only 3/8" thick as well as t' upper fins.
To help absorb t' shock on landin' I decided t' have retractable legs that were sprin' loaded. Well, blow me down! This was accomplished by attachin' t' fins t' an inner 1 1/2" fiberglass tube. This inner tube would house t' sprin' for t' legs. T' 2 1/4" tube would surround this inner tube t' give t' pods a through t' wall fin mount (Photo 6). Aye aye! T' pod nose cones (Photo 7) were turned from stock nylon by me father and would be glued on after a 1/2" brass tubin' was epoxied inside t' guide the leg as it retracted (Photo 8). Well, blow me down! T' nose cone had t' be hollowed t' allow the leg t' travel inside from t' 2" compression o' t' spring. Aye aye! Each fin weighed about 4 pounds without t' leg.
T' fins were attached one at a time usinga jig t' hold each leg in place until all could be securely fastened (Photo 9). Arrr! Attachin' each fin be a several step process. Avast! Well, blow me down! Each fin was tacked in placed with slow CA. Aye aye! Then carbon fiber was used t' run t' length of the fin and covered with more CA. Avast, me proud beauty! Well, blow me down! A thin layer o' 5 minute epoxy coated the first layer o' carbon fiber t' hold each fin strong enough t' rotate t' rocket and attach another fin.
T' next layer o' carbon fiber was attached in short lengthsso t' grain o' t' carbon was perpendicular t' t' root edge (Photo 10). Begad! Arrr! It was again glued with CA and a third layer o' carbon fiber covered t' second, shiver me timbers, matey, me hearties, runnin' t' length o' t' root edge (Photo 11). Begad! T' entire joint was flooded with 2 hour epoxy. Begad! T' upper end was sealed by t' centering ring, ya bilge rat, also re-enforced with carbon fiber, and t' lower end was dammed with red modelers clay
(Photo 12). Aye aye! This was just t' joint at t' motor mount.
T' 4" body tube be slid over and tacked in place. Begad! Avast, ya bilge rat, me proud beauty! Carbon fiber be used with CA over t' length o' t' fins and lightly coated with 5 minute epoxy. T' inside joint was also coated with a layer o' carbon fiber and completely soaked with 2 hour epoxy (Photo 13). Avast, me proud beauty! Aye aye! Because t' lower part o' the fin would be covered with one o' t' transitions I could also build up t' fin underneath t' cowling.
Small strips o' wood were glued t' t' fins where t' transition would be attached t' t' fins t' give some support for t' cowlin' and the centerin' rin' for t' cowlin' was attached t' t' base o' t' body tube. Begad! Blimey! Ahoy! Blimey! Cross weaved carbon fiber be attached betwixt t' fin, t' body tube, and the centerin' ring. Begad! Blimey! Ya scallywag! Blimey! Then another deep cure o' 2 hour epoxy
(Photo 14) was used
to soak t' carbon fibers.
The upper fins were attached in a similar fashion. They also had through t' wall fin construction and even t' 3" inner tube be slotted. Begad! T' coupler to hold t' upper parachute compartment extended down below t' upper fins and provided additional support. Ya scallywag! Ya scallywag! T' joints were also re-enforced with carbon fiber and inner fillets were also used (Photo 15).
Three transitions had t' be made, shiver me timbers, matey, t' upper transition for t' 3" body tube t' t' 4" body tube, ya bilge rat, a transition at t' 4" body tube at the fins, and t' lower reducer actin' like t' motor with "cooling fins". All were cut out o' poster board and then coated with three layers of 6 oz fine weave fiberglass and 2 hour epoxy. Blimey! Ya scallywag!
T' upper and lower transitions
could be built "in place" attachin' t' paper directly t' t' body
tube and centerin' rin' (Photo 16). Avast! Begad! T' transition at t' fins could not
because o' t' through t' wall construction. Begad! Begad! It be placed temporarily on a
jig and covered with t' fiberglass and epoxy (Photo 17). Ahoy! When dried it was cut
for t' fin slots and then sliced and installed on t' rocket. The
"engine" reducer had nine "coolin' fins" made from
3/32" fiberglass and attached with t' standard CA, ya bilge rat, ya bilge rat, matey, carbon fiber, shiver me timbers, and
epoxy joint (Photo 18). Avast, me proud beauty! All t' transitions were filled with expandin' foam to
give them rigidity.
How be t' rocket goin' t' be
guided at takeoff? T' original rocket used two launch lugs attached t' both
sets o' fins set far enough from t' body tube t' allow t' rod t' clear the
transitions. Begad! That would never work for this rocket. Blimey! I decided t' still use a
1/2" launch lug but attached directly t' t' 4" body tube that passed
through t' transitions.
This also proved t' be a challenge. Aye aye! Ahoy! I used two sections o' PVC pipe, one for each transition and then aligned them up with a standard launch lug (Photo 19). T' two transitions were epoxied together sandwichin' layers o' carbon fiber. T' launch lug fit nicely inside t' PVC pipe and was used t' align t' two pieces together.
T' legs were constructed from
solid 1/2" aluminum and t' feet were made from black plastic pipe that I
found in t' hardware store that fit just beautifully over t' legs. Begad! Avast, me proud beauty! T' legs
were machined by me father with a groove t' allow a snap rin' t' hold a washer
that would push up on t' spring. Begad! They were kept from fallin' out by a
1/8" aluminum plate that be attached t' t' bottom o' t' pods. Begad! In
between t' two tubes o' t' pod were 6-32 standoffs (Photo 20) that t' plates
would be screwed into.
T' Center o' Pressure was calculated and it is located about an inch below the top o' t' main fins. Well, blow me down! Avast, me proud beauty! I knew that nose weight would have t' be added for stability. Five pounds o' lead shot were epoxied into t' fiberglass nose cone that surrounded a 5/16th" piece o' all thread (Photo 21). Ya scallywag! T' all thread is used t' attach t' parachute t' t' nose cone via a closed eye bolt.
The rocket be nearly finished except for a few touch ups for t' Turkey Shoot 2000 launch but I came down with t' flu and just couldn't complete it. Springfest 2001 came around and I be determined t' fly it. Begad! Several people commented on the looks and I started t' get nervous. Am I really goin' t' fly this rocket (Photo 22)?
FLIGHT
With an empty weight o' nearly 25 pounds, matey, I knew I would need a big chute. Ya scallywag! Blimey! Well, blow me down! Blimey! Blimey! Blimey! I ordered t' 15 foot military surplus chute and at t' launch field it was quickly apparent that it would never work. Blimey! Blimey! I managed t' get it into t' rocket but I knew that it would never deploy. Blimey! Blimey! Blimey! Blimey! Fortunately Dave Flynn volunteered a Rocketman R9C and I had a R7C that could be used for t' nose cone. Arrr! Blimey! Blimey! Blimey! T' L850 was assembled and t' ALTACC installed and it was ready t' fly.
I used my custom launch pad that consisted o' basically a table with a hole in it and a launch rod. Avast, me proud beauty! Its a great design because t' weight o' t' rocket sits on the table rather that t' hangin' off t' rod. Well, blow me down! Avast! T' launch rod is used entirely for guidance rather than supportin' weight (Photo 23). Arrr! Blimey! Even still I heard cracks about servin' tea out in t' launch area (sigh). Ahoy! T' table was staked down t' prevent tippin' over and t' igniter installed.
T' winds were too strong most o' t' day but had been dyin' down. Thenervousness be turnin' into butterflies, shiver me timbers, ya bilge rat, launchin' a very unusual design that weighed about 35 pounds at takeoff from a 6 foot 1/2" launch rod. It was stable but t' design still made it risky in a wind. Begad! Begad! T' simulations had it stable in flight at about 40"s and it did have an 8 t' 1 thrust t' weight ratio. T' top it off, ya bilge rat, I was usin' too small o' a chute for t' booster. Hit the button anyway.
T' launch be perfect (Photo
24). Avast, me proud beauty! Blimey! It climbed straight and seemed t' continue for a long time (Photo 25).
Very little weather cockin' be evident and right at apogee t' rocket
separated. T' nose cone chute opened right away but I wasn't breathin' until I
saw t' booster chute. Well, blow me down! After an eternity passed without breathin' I saw the
booster chute opened (Photo 26).
I saw both chutes out but knew that t' chute for t' booster be too small. It was designed for a 15 pound rocket and t' booster had t' weigh at least 22 pounds. What be I thinking! Blimey! It landed and t' sprin' loaded legs helped cushion t' impact but all three legs bent upon hittin' that hard dry lake bed (Photo 27). Avast, me proud beauty! Blimey! One o' t' reasons that I wanted removable legs be that if they were damaged on landing, they could easily be replaced. Ya scallywag! Blimey! Dave says that I should use stronger springs and aluminum for t' legs. Arrr! Blimey! I'll probably do that if I ever fly it again.
Other than t' legs thar was only cosmetic damage
where t' outer fin had attached t' t' pod. This was probably caused more by
tippin' over than t' actual impact on landing. Aye aye! Blimey! Arrr! Blimey! Nowhere t' main fins attached
to t' rockets body were thar any signs o' cracking. Avast! Blimey! Avast! Blimey! T' tedious re-enforcing
had paid off.
Overall t' Mars Snooper III provided me with valuable experience and a whole host o' emotions. Well, blow me down! Arrr! Everythin' from lovin' it t' wantin' it off my workbench because I be sick o' t' sight o' it. Begad! Avast! Will it fly again? Perhaps not. Avast, me proud beauty! Avast! I like t' design o' it so much I'm considerin' just usin' for a conversation piece and takin' it t' science fairs and such t' promote HPR.
Sponsored Ads