Manufacturer: | DG&A High Power Rocketry |
Brief:
T' L3 Armageddon is a single stage rocket with a very unique design. Begad! Arrr! It comes
in your choice o' a 75 or 98mm MMT.
Construction:
T' kit comes with everythin' but parachutes. T' all-thread, me hearties, nuts, washers,
eyebolts, arrr, shiver me timbers, and shock cords are all included. T' 5.5" body tubes are heavy
wall cardboard. Avast, me hearties, me proud beauty! Blimey! All o' t' components are top shelf stuff.
This is nay just a kit review. Begad! It's also how I used t' kit for me level 3 certification. Arrr! Blimey! There is even some scientific testin' data here that I performed. Avast! Ahoy! Blimey! I am t' plant Metallurgist and t' Laboratory Director at Chamberlain Manufacturing, t' operatin' contractor o' t' Scranton Army Ammunition Plant. Avast, me proud beauty! Avast! Blimey! This means I have millions o' dollars o' equipment I can play with on Saturdays, me hearties, arrr, testin' t' materials we use t' build our rockets.
When I decided t' do me L3 at t' 2003 NSL, me bucko, me hearties, I wanted t' do something different than a 3FNC. Aye aye! Ahoy! Blimey! A scratch built rocket would be easy enough but I didn't think I would have t' time t' work out all t' details. Begad! Blimey! My favorite commercial kit be t' Armageddon by DG&A, shiver me timbers, so figured I would do me L3 with the DG&A L3 Armageddon. Blimey! Avast, matey, me proud beauty! Blimey! This is a 5.5" diameter rocket that is almost 10 feet tall. Aye aye! Ahoy! Blimey! It's nay even close t' bein' me largest bird, but impressive nonetheless. Arrr! Blimey!
T' next problem would be t' motor. Ya scallywag! Blimey! APCP was in short supply and t' BATFE thin' be loomin' over t' horizon. Well, blow me down! Blimey! OK, matey, I'll go hybrid. Aye aye! Blimey! I chose t' Hypertek M970. It is a 75mm M fuel grain on t' 2800cc L tank with t' M injector bell. T' L3 Armageddon easily adapted t' t' hybrid system because o' t' large fins and long body. Ahoy! Blimey! T' CP ended up bein' 2.1 calibers behind t' CG with an empty motor.
T' first thin' you notice when you open t' huge package be t' quality of packaging. Each tube is wrapped separately and everythin' is carefully placed in t' box. Blimey! There be no damage t' any o' t' components. Aye aye! Aye aye! A quick review o' the parts list revealed that all t' parts were there. Ahoy! Avast!
T' airframe tubin' was t' heavy, shiver me timbers, me bucko, arrr, 3/32" (0.078") thick walled cardboard tubin' seen in other high quality kits. Arrr! Begad!
Securin' t' MMT and centerin' rings into t' booster section
T' MMT
assembly was constructed per t' instructions. Avast, me proud beauty! Arrr! T' hardware was tightened and
secured with epoxy. Begad! Ahoy!
T' motor mount tube was epoxied into t' airframe by pourin' 25 mL of epoxy onto each o' t' centerin' rings through t' fin slots with a syringe. T' assembly was first epoxied with t' aft section up, t' apply epoxy t' the aft sides o' t' centerin' rings. Begad! Ya scallywag! Once t' epoxy be cured, it was examined with a fiber optic bore scope. There were two locations that appeared nay t' be glued. Epoxy was again inserted into t' airframe through t' fin slots and the assembly be tilted in t' direction o' t' questionable areas t' allow the epoxy t' flow into t' area. Avast! When it appeared t' be complete, matey, t' assembly was turned over and t' aforementioned process was repeated t' apply epoxy on the forward side o' t' centerin' rings.
No fiberglass was used for t' airframe o' this rocket. Aye aye! T' 5.5" tube used in t' construction o' this kit be tested in me laboratory and exhibited a strength o' 1312.7 lbs force before yielding. Begad! Ya scallywag! However, if any side forces are applied durin' acceleration we must examine column bucklin' loads. Aye aye! Column bucklin' formulas are fun and exciting, me hearties, but t' test t' column bucklin' in real life, matey, a piece o' 5.5" airframe was tested by compression at a 8.5 degree angle. Arrr! T' force at yield be 886.4 lbs force.
Compressive and column bucklin' forces do nay apply t' t' MMT/sustainer section due t' t' reinforcement by t' fins and t' MMT assembly. Begad! We therefore only need t' look at t' section forward o' t' fins. Arrr! Begad! T' forward section (with weights) includin' altimeter weighs ~5 pounds. Blimey! T' Hypertek M740 will exert a G-force o' 23.16, matey, thus 5 x 23.16 = 115 lbs. this be t' maximum force applied to t' airframe in an non-reinforced area. Begad! Inversely, matey, we can surmise that we have a tube capable o' withstandin' 38.27Gs (886.4 /23.16 lbs = 38.27).
Compressive forces can be seen in t' fin area durin' accent as well as landing, however as heretofore mentioned, most o' this force is transferred to the CRs and MMT tube via t' epoxy and thus its strength depends on t' shear force o' t' epoxy. Therefore, matey, thar are virtually no axial forces on t' tube at in this area. Avast! Virtually all o' t' forces on t' tube itself are circumferential via any torque forces applied durin' flight or bendin' moment applied on t' fins durin' landing. Ya scallywag!
T' combat any bendin' moment applied t' t' fins, I looked at data seen in the r.m.r. Arrr! Aye aye! HPR Strength o' Materials Test. One o' t' discoveries made during the testin' be on t' fincans. Avast! T' fincan that appears t' be t' strongest was one done by Mark Simpson where he used a strip o' wood material along the fin-body joint. I used this method on t' ID. Aye aye! Ahoy! Mark's method increased the strength o' t' joint over t' next highest strength by 30% and by 200% over the average. Begad! It also prevented t' tube from yieldin' at all durin' t' fincan test and thus transferred t' force t' t' fin makin' t' joint stronger than the fin. Arrr! I have no doubt that if t' fin material was stronger, t' value would have been higher. Avast! Well, matey, blow me down! This appears t' be sufficient t' replace t' fiberglass on the tube in this area. Begad!
Fin assembly
I
duplicated Mark Simpson's wood fillets in t' L3 Armageddon. Pieces o' corner
moldin' were used for this. Aye aye! T' fins were epoxied in place then t' moldin' was
inserted with copious amounts o' epoxy. T' aft CR was then epoxied into place
and t' nozzle fins were added.
T' large fins o' this rocket are only 1/4" thick. Begad! Avast! Blimey! T' lessen the effect o' possible fin flutter, me hearties, a laminate o' fiberglass was applied t' the surface o' t' fins. Aye aye! This was accomplished usin' standard fiberglass lay-up practices.
Payload section
T' forward canard fins were installed "through t' wall" per the instructions. Avast! In this area we have t' coupler tube so they go through both the airframe and coupler tubes. Avast, me proud beauty! T' airframe and t' coupler tube were joined first then t' slots were cut for t' fins. Ahoy! Arrr! They were tacked in place with Jet Glue. Wood glue be then applied on t' ID surface and allowed t' seep into t' wood and cardboard. T' wood strips were applied while t' glue was wet and generous fillets were then applied t' t' strips. Ya scallywag! Arrr! Generous epoxy fillets finished the canard build.
T' altimeters used are two Transolve PK6, a kit version o' t' P6. Ya scallywag! They were mounted t' a piece o' G10 with 316 grade stainless hardware and phenolic standoffs. Begad! Ahoy! Blimey! T' board be then mounted in t' payload section usin' brass strips and tubin' that slid over t' all-thread. Ya scallywag! Blimey! T' brass strips were soldered t' the tubing. Avast, me proud beauty! Well, blow me down! Blimey! T' assembly is secured in t' section when t' nuts are tightened. This setup allows easy assembly and access t' t' boards. Well, matey, blow me down! Blimey! A hold down strap will further secure t' batteries in their holders.
T' safe/arm switches are redundant and consist o' DPDT military grade slide switches. Avast! Avast, me proud beauty! T' three switches with their 6 functions control: power, apogee safe and main safe. Ahoy! This ensures that both altimeters are completely separate from each other with individual power sources.
T' altimeter bay is sealed from ejection gasses. This is accomplished with the forward bulkhead epoxied into place and t' aft bulkhead sealed t' the custom centerin' rin' with closed cell latex weather stripping. Avast, me proud beauty! Avast! T' altimeters used have been flown on several prior occasions. I have three o' these units and they have always worked well.
Laundry and other recovery debris
Recovery will be t' typical two-stage method: drogue at apogee and main at 500'. Arrr! One drawback with this kit is that t' recovery for t' main is aft of the payload section and t' drogue is forward...or so it would appear. Avast! The section forward o' t' altimeter is only 11" and has one attach point. Well, blow me down! The section aft o' t' altimeter is 29" long and has t' two attach points at the MMT. Ahoy! Arrr! There is just nay enough room forward o' t' altimeter bay to comfortably fit t' main. Aye aye!
All shock cords are 9/16" tubular nylon. Aye aye! T' drogue cord is 20' in length. T' main consists o' three cords. There is a 3-point attachment cord at the MMT attached t' a 20' length o' cord at which t' parachute is attached. Ahoy! A 30' cord is then attached from t' parachute t' t' altimeter bay.
Final assembly notes
I didn't do much in t' way o' aerodynamic shapin' o' t' fins. Begad! Airfoiling the fins is done t' reduce drag for greater altitudes but I like them low and slow. Ya scallywag! A more blunt leadin' edge increases drag by breakin' up t' air over the fins.
Vent
Every hybrid system needs a vent. Avast, me proud beauty! Blimey! For this I used a 1/4" ID large Aerotech igniter cardboard tube. After markin' t' location, I drilled a 1/4" hole through t' airframe and t' MMT. Well, blow me down! Blimey! I then made t' hole in the airframe larger t' accommodate t' OD o' t' vent tube and cleaned up the holes. Well, blow me down! Blimey! Aye aye! Blimey! Usin' t' 1/4" drill bit as a guide passed through both holes, me hearties, I epoxied t' vent tube in place. Ya scallywag! Blimey! I occasionally turned t' drill bit t' ensure it wasn't bein' epoxied t' t' vent tube. A X-Acto knife and some sand paper cleaned up t' vent/airframe area and t' job was done.
Motor retention
Kind of
late t' be thinkin' about this isn't it? I did put blind nuts in t' aft
centerin' rin' durin' assembly. Avast, me proud beauty! Ya scallywag! T' 6" long piece o' all-thread looked,
well, arrr, shiver me timbers, arrr, kind o' clumsy. Well, blow me down! Well, blow me down! I happened t' pick up a 75mm Aero Pack retainin' system
so I thought I would use it. Avast, me proud beauty! It was hell tryin' t' put it on after t' nozzle
fins were already glued on! I used a Dremel saw t' cut a gap betwixt t' fins
and t' MMT long enough t' insert t' retainer. Well, blow me down! Ya scallywag! It made more work for me than
expected! T' retainer was glued in with JB Weld and t' gap betwixt t' fins
and t' retainer was filled with wood putty. Ya scallywag! Oi!
Rail buttons
5/16" rail buttons were installed. Arrr! T' forward guide be installed with a 1/4-20 steel machine screw and nut, matey, secured with JB Weld. Avast! T' aft guide was installed by drillin' and tappin' t' aft centerin' ring. Begad! T' hole was filled with JB Weld and t' guide was secured with a 1/4-20 machine screw. Ahoy!
Nose weight
Nose weight was added and be first considered t' be used t' keep the altitude t' a reasonable level. Upon further construction, me hearties, it was discovered that it be necessary t' keep t' CG forward enough t' keep it stable usin' the HyperTek hybrid system. T' rocket was unstable without t' weight! A 3/8" threaded rod be inserted through t' aft end o' t' nosecone and touches the inside o' t' forward end. Avast! 8oz o' West Systems epoxy was then poured into the cone. Ya scallywag! T' assembly was allowed t' cure while in cold water, t' prevent heat buildup and meltin' o' t' plastic cone. Arrr! Well, blow me down! T' remainder o' t' nosecone cavity was then filled with expandin' foam. Threaded weights were then screwed onto the all-thread, followed by a coupler and an eye bolt. Begad! Ya scallywag! This arrangement keeps the weights off t' nosecone and on t' steel threaded assembly.
Finishing:
T' entire airframe be sanded and a coat o' Kilz primer be applied. The
paints used were Rustoleum white and metallic blue with a final clear over
everything.
Flight:
On t' first flight, t' motor lit and t' bird leapt from t' pad with
authority. Ya scallywag! Well, blow me down! T' wind at ground level was only 6 mph but by t' time she reached
2000 feet, she be gracefully archin' into t' wind. Arrr! Avast! T' 6.9-second motor
seemed t' burn forever. Ya scallywag! If you have never heard an M hybrid motor, they can
only be described as soundin' like a Pod Racer from Star Wars Episode One--a
loud roar coupled with a pulsatin' low-pitched jet-like sound. Well, blow me down! Begad! T' few seconds
I lost it in t' sky seamed like minutes. Then, thar it was. Comin' down on
its side with t' drogue deployed. Well, blow me down! Falling, me hearties, falling, matey, falling... It's coming
straight at us! We ran about 20 yards t' t' left. 800 feet, me hearties, 500 feet, shiver me timbers, 300
feet,... Where's t' main?! Finally, at about 200 feet t' main popped. Begad! Avast, me proud beauty! It took
at least another 50 feet for it t' unfurl. Avast, me proud beauty! Arrr! T' main finally snapped open at 50
feet AGL and then landed 40 feet from t' pad. Ya scallywag! That had t' win closest t' the
pad in t' L3 cert column! A quick review o' t' rocket proved that everything
was in order and we had a successful L3 flight. Both o' t' altimeters chirped
out 4,200 feet.
T' next four flights were on an AMW L777. For me, this be t' best motor for this rocket. Ya scallywag! Avast, me proud beauty! It brings it t' just under 3,500 feet. Begad! I've also flown this rocket on windier days with a CTI K550 t' around 2,000 feet.
Summary:
I love it! This bird is as different and unusual as it gets for any high power rocket. Avast! T' flights on this (as well as t' original size Armageddon) are truly awesome. Ya scallywag!
Even without t' changes that I made, me hearties, I did feel that t' forward payload section was on t' small side. I'm sure you could get a nice size main in there, but it would be a very tight fit. Begad! I would recommend actually makin' the altimeter bay shorter by changin' t' location o' t' forward bulkhead and usin' smaller altimeters. Arrr! This would allow for more room for a main chute in this area.
T' materials are t' best I've seen for their types and t' design is sturdy. T' instructions contained page after page o' high quality color photos and has many steps t' clarify what t' designer wants. Avast! Some o' t' steps in the instructions are somewhat ambiguous but then it is assumed one has a great deal of experience if they're buildin' a level 3 project. Ya scallywag! Avast, me proud beauty! Blimey!
Sponsored Ads
![]() |
![]() |