Editor's Note: Below is a reproduction of Jed Marti's succesful NAR Level 3 documentation package. The package is meant to serve as a record fo the design and construction techniques used in the Level 3 vehicle's build. Jed has graciously allowed us to reproduce the document so that others can learn from it. Take some time and notice the detail of his notes. Thanks Jed!
First you have to unpack the 3 shipping boxes and make sure everything is there. This one’s a bit big for the office so it’s moved to the shop.
The motor mount is sanded and the center piece epoxied in position on the 75mm tube. This took 2 days and several layers of 60 minute JB Weld.
The fins were test fitted and we discovered that LOC cut the fin holes too short (for the wrong fin dimension they were). These were lengthened about .25” in the forward direction. The motor mount was epoxied just forward of the fins in the 7.5” tail piece.
To maintain alignment, the forward and rear bulkheads were put in place, but not glued. Both these bulkheads had eyelets to allow their removal. This was allowed to set overnight in an upright position to keep the epoxy from running all over.
A fin mounting stand was built so the aft tube would rest in a comfortable position and the fin be held at a right angle. The fins were sanded to an aerodynamic shape and roughed up on the inner side. My hands are too big to fit comfortably in the back to smooth the epoxy. Consequently the inside is a bit of a mess. 60 minute JB Weld is used everywhere except the first tacking of the fin to the motor tube where the 5 minute stuff is used.
Figure 2 First fin mounting
November 25, 2008
I’ve mounted the three fins and painted the nose cone. Most of the fin fillets are complete but I’m having trouble with the JB weld 60 minute stuff curing. It runs before it sets so only one fin surface at a time can be completed. Likewise, the aerodynamic fillets loose their shape (more later). Since I can’t reach my fat fingers all the way in, there’s quite a mess inside, the epoxy is smoothed and wedged into place with a long stick.
Figure 3 Three fins mounted, most fillets
The Rustoleum black high cost paint comes right off the nose cone. I stripped it off with a wood stick and plastic paint scraper and started repainting with Rustoleum designed to adhere to ABS plastic.
Figure 4 Scraping the paint off the nose cone
November 27, 2008
To glue the base support ring in place, I first cleaned off extra epoxy so that the ring fits snugly against the bottom fins.
Figure 5 Cleaning up the engine compartment.
To support the engine compartment in the upright position (so the epoxy will run into the joint, I cut a 2”x8” to easily fit and use the compartment’s weight to push the rear bulkhead against the fins.
Figure 6 Bulkhead support block test fit.
Finally, I applied epoxy to all the fins and internal surfaces and fit the rear bulkhead in removing the screw eyes and string used for the test fit.
Figure 7 Rear bulkhead glued in place.
I stood the engine compartment in the upright position covering the support block with a plastic bag to keep it from sticking.
November 28, 2008
I test fit one of the remaining 30” segments to the engine compartment top. It fits fine, but does not want to come off. After 20 minutes of hammer banging and damage to the segment, it finally came off. I will either purchase a new segment or remove the top ½” that was damaged. I put epoxy rings about ¼” before the forward bulkhead and top of the tube coupler and twisted the top section on. ½” before the tubes were together I ran an epoxy bead around the exposed coupler and then shoved the tubes together into their final position. The epoxy that squeezed out of the joint was removed and smoothed over. This is tube #7 of 60 minute JB Weld.
Figure 8 Abortive test fit and final engine compartment.
Figure 9 Top view of motor mount showing epoxy ring
After cleaning the nose cone with isopropyl alcohol I applied some layers of the Rustoleum plastic paint (black). Between each paint layer, the cone is smoothed with some 00 steel wool (Figure 10).
Figure 10 Nose cone after 3rd paint layer.
December 2, 2008
Sanding the exterior fillets has begun. The 60 minute epoxy has a tendency to slump and run off the ends. I’ve cleaned most of this mess up and started building up the fillets but with the 4 minute stuff to make a cleaner finish. These are carefully sanded to avoid gouging the G10 fins.
Figure 11 60 minute fillet (left) covered with 4 minute fillet (right)
Notice that the motor tube does not have an epoxy fillet – this will be added when the Slimline motor retention device arrives.
December 5, 2008
Payload bay arrived and assembly begun. Put two beads of epoxy around inner core.
Figure 12. Payload inner core with 4 minute JBWeld
I wrapped the sheath around this core and epoxied the large gap that showed up – all with 4 minute JB Weld. The piece was allowed to set for about 20 minutes.
Figure 13 Two layers epoxied.
The outer tube was coated with two substantial beads of 4 minute epoxy.
Figure 14. Outer tube with dual 4 minute JB Weld
The inner core was then shoved into the outer tube with even spacing (eyeballed) on both sides. Extra epoxy was removed from exposed shelves and the pieces clamped. Unfortunately, the inner piece has a very loose fit. Additional epoxy will be added to assure adequate adhesion.
Figure 15. Assembly complete.
The eyebolts were assembled per instructions and the 1” washers epoxied to the bulkheads with 60 minute JB Weld. Since the holes are much bigger than the ¼” bolt, the bolt was secured in a vice before the nut was tightened. The threads of both bolts were coated with 60 minute JB Weld before the nut was added.
Figure 16. Payload bulkhead eyebolt installation.
The bulkheads were sanded to fit the rest of the payload bay. The payload bay outside has been sanded considerably to fit the 30” phenolic tubes without binding (still not done).
December 5, 2008
Still working on sanding the payload bay to the right size. I discovered that there are lathe pieces that expand out to accommodate the bay from the inside. This allowed me to sand and smooth the bay for the proper fit.
Figure 17. Payload bay on the lathe.
When the bay fit both the engine compartment and top compartment, the 1” spacer was glued to the center with 60 minute JB Weld. The excess was wiped off and allowed to set.
December 6, 2008
I labeled the fins A, B, and C, and started keeping a log of their status for sanding, rough spots, and so on. One fin side is the letter A and the opposite side is A. For example the fillets on A and C are 60 minute JB Weld but rough. These were augmented with heavy layers of 4 minute JB Weld and allowed to set.
Figure 18. Fillets between A and C.
Needs fill in forward area. Final sanding/polish done except for forward.
Added 4 minute filler fillet, needs sanding.
Completed, finished with #00 steel wool.
Sanding in progress.
Added 4 minute filler fillet, needs sanding.
Needs finish coat of 60 minute after rough sanding exposed dells.
I primed the top section with spray on white primer. This will be covered with half white/black in the future.
Figure 19. Primer of top 30" section.
December 13, 2008
I began installing the slimline motor retainer. It does not fit the motor tube at all so all epoxy and cruft on the mount was removed with a dremel. The tube had to be reduced somewhat. Test fitting established that sanding needed to get near the back plate. Even so, about 0.125” remains between the rear flange and the rear bulkhead.
Figure 20. Sanded motor mount before epoxy and test fitting.
The motor tube was slathered with 60 minute JB weld as was the inside of the slimline retainer.
Figure 21. Epoxy motor tube and retainer for assembly.
The retainer was screwed on as far as possible causing epoxy to squeeze out the back and fill most of the area between the flange and the bulkhead.
Figure 22. Epoxied motor retainer.
We stood the motor section on its tip and noticed that the retainer was crooked. This was corrected with two levels, rubber mallet and a piece of wood.
Figure 23. Straightening the motor retainer.
December 15, 2008
The slimline flange to the rear bulkhead did not uniformly fill with 60 minute epoxy. A small 60 minute JB weld batch was made and the flange finished off.
Figure 24. Finishing the motor flange gap
We also declared victory (premature?) on the fin fillets.
Figure 25. Fillets between C and B, A and B, A and C.
Figure 26. Fillet on fin A.
Figure 27. Fillet on fin B.
Figure 28. Fillet on fin C.
We test fit the 75mm Aerotech casing in the engine tube and found it didn’t fit – some epoxy squeezed when installing the Slimline. This was removed with a Dremel tool and rotary sand paper.
Figure 29. After cleaning up Slimline and test fit Aerotech 75mm casing.
December 18, 2008
The top 1” of 7.5” tube has begun to wilt after too many contacts with the shop door and forceful test fittings of the instrument package. To minimize the risk of expanding the affected area and preserve structural integrity, we decided to inject low viscosity epoxy between the layers.
Figure 30. The Doctor is in.
Using xxx epoxy, we put about .5 cc in a small syringe and injected this up to ¾” into the wilted area.
Figure 31. Injecting epoxy into the wilted area.
We then covered the area with a plastic bag and clamped it together – this particularly high strength epoxy takes 24 hours to set..
Figure 32. Area is clamped for 24 hours.
December 21, 2008
I repeated the epoxy injection on a second small spot as above. I then completed design work and basic machining of the payload bay internal structure to supplement the cardboard tube and wooden structure provided by LOC. The basic structure has two round, flat ¼” aluminum plates supported by three internal and two external plates.
Figure 33. Payload support structure test assembly.
The support structure was machined from scrap ¼” aluminum plate on its way to the recycling center so it has occasional holes from previous uses. The incomplete structures weighs about 2 lbs which will increase as instrument and battery packages are added.
Figure 34. End plate detail.
January 1, 2009
The base tube has been primed and we’re adding the launch lugs for the 1.5” rail. First we draw the center lines as per instructions using a long wood stick.
Figure 35. Centering line for launch lugs.
Next holes for the #8 screws were drilled into the body and the lugs test fitted.
Figure 36. Test fit of rear launch lug.
Then both launch lug areas were sanded to the base cardboard for epoxy with 60 minute JBWeld.
Figure 37. Sanding in launch lug area.
I then screwed both launch lugs in place (you can’t reach the inside to put nuts on them) and test fit with an 8020 1.5” rail.
Figure 38. Test fit on 1.5" rail.
Both lugs were then glued in place with copious 60 minute JB Weld. The screws were coated as well and driven into the cardboard.
Figure 39. Forward and aft launch lug epoxy.
I placed the 8020 launch rail over the lugs and allowed them to set overnight. We then adjourned for the January 3rd Frozen Fingers launch.
Figure 40. Frozen Fingers launch.
We’re building a more rugged structure than that provided by the ¼” threaded rod, wood and cardboard provided by LOC. We’re attempting to increase the rocket’s weight to keep the maximum speed and altitude within bounds (10,000’ and less than mach .85). The internal structure is ¼” aluminum to fit within the cardboard payload section.
Main 4.5” Rack
The standard payload entity must fit a 4.5” rack. This has #4x40 threaded holes on ½” spacing in the center and at 2” from the center. Payload modules must fit this and provide appropriate 4x40 clear holes for secure mounting. Modules can be mounted on both sides and can be up to about 2” deep. For extra strength, the weight reducing holes are milled only 0.2” deep leaving approximately 0.05” aluminum backing. Large indents on the top and bottom leave space for the shock cord eyebolt nuts and washers.
Figure 41. Center plate dimensions.
Figure 42. Partially machined center plate.
These are milled out of ¼” aluminum as well, but the lightening holes are milled through to provide a place to tie off support wires. Two such plates were machined.
Figure 43. Side plate dimensions.
Figure 44. Machined side plate.
Experience with these indicates that the tie wraps need to be a good grade as the milled edges are moderately sharp. We will substitute FAA nylon tie wrap for the cheap ones from Harbor Freight.
Bottom and Top Support
The bottom and top support plates are ¼” aluminum plate 7” in diameter with two chord cuts at 2.5047” to allow passage of safe and arm switches during assembly. Lightening holes are milled to permit wire passage to the explosive charges. A large center hole provides clearance for the eyebolt nuts and washer. The bottom plate is screwed to the wooden LOC spacer with ¼” carriage bolts and nyloc nuts. The top plate has threaded holes for 5/16” bolts used for final assembly (since you can’t reach inside). These holes are not shown and are drilled to match the LOC spacer.
Figure 45. Bottom (and top) plate dimensions.
Since these were machined from scrap, there are a few extra random holes.
The forward plate uses two ¼” bolts with nylocks and is permanently attached to the support structure. The rear uses two 3/8” bolts and the corresponding holes are threaded. To assemble the system, these bolts are removed, the housing slides into the payload bay and the structure is reassembled. All wires from the exterior switches and the charge cup holder wires run to rugged 8 or 4 pin connectors.
Figure 46. Machined top plate.
A 9 volt battery is required for each Perfectflite altimeter both to fire the explosive charges and power the internal computer. Two are mounted in a heavy aluminum housing and wires run to the altimeter housings. The housing was first painted with zinc chromate primer and finished with bright yellow. The 1/8” cover has “9V” stamped on it. The sides are milled to lighten the structure. #4x40 1.5” flat head screws attach the housing to the 4.5” rack mount and the cover is secured with four #4x40 hex head screws. A 4 connector terminal strip is attached to the cover and wires run to the battery connectors inside.
Two housings are being constructed. The two batteries for the altimeters was first primed with zinc chromate and then painted bright yellow. The second contains two in series to run the second altimeter charges and is painted bright orange.
Figure 47. Dual 9v battery holder dimensions.
Figure 48. Dual 9v battery holder (one of two).
Perfectflite Altimeter Housing
The altimeter housing is also machined out of a solid aluminum block. The inside provides #4x40 tapped standoffs with more than the 1/32” clearance recommended by the Perfectflite specifications. Two exterior flanges provide .5” #4x40 clear holes to mount to the 4.5” rack. An aluminum cover with holes provides limited protection from any explosive gasses reaching this area.
Figure 49. Perfect flight altimeter housing dimensions.
Figure 50. Perfect flight altimeter housing.
The second altimeter is the RRC2 from Public Missiles and has a slightly different form factor. A similar housing was machined for it. Two standoffs were left in the base area to accommodate the mounting screws. This altimeter will use one of the batteries in the two pack described earlier.
Figure 51. RRC2 altimeter housing dimensions.
The cage is screwed together with #4x40 stainless flat head screws.
Figure 52. The electronics bay initial assembly.
The top wooden plate is attached with ¼” screws and nylocks. Later we added a steel support plate between the bolts for added structural integrity.
Figure 53. Forward plate is permanently attached.
The structure was cleaned up and painted with zinc chromate primer to avoid corrosion problems.
The #1 Perfectflight altimeter uses a single 9v battery to both power the altimeter and fire both main and drogue charges. The #2 RRC2 altimeter uses a single 9v to run the computer and two 9v’s in series to fire the main and drogue charges.
Figure 54. Core wiring diagram.
The sleeve has 4 switches. Two on one side control altimeter power on and two on the other are safe and arms for firing the charges. These are connected to 8 pin microphone plugs placed on internal bulkheads. The charge connectors run through the wood and are 4 pin microphone jacks.
Figure 55. External switch wiring.
All exterior connections go to 4 or 8 pin microphone connectors. The following shows the wiring of the altimeter switches that are mounted flush with the instrument bay ring. During assembly, the bottom wooden panel will be removed so the wires and switches don’t run into the support structure.
Figure 56. External switches connected to payload.
All internal wiring is #20 MIL Tefzel (MIL spec MIL-W-22759/16) unshielded white. This is not PVC coated and makes no toxic smoke in a fire. It is rated at 600 volts and temperatures to 105 centigrade. It has many more strands than common hookup wire and is thus much more robust.
The first half resembles the following:
Figure 57. Some initial wiring.
The initial PVC charge holders were deemed too small for the Bruiser. 4 new cups were machined from a 1” aluminum rod. These have a capacity of 2.6 cubic inches. The original cups were removed and replaced with these. In addition, the attaching rings were replaced with solid ones and bolted to the internal structure through a ¼” steel bar.
Figure 58. Charge cup holder design.
Figure 59. Charge cup holders (forward).
The entire payload is assembled by removing the bottom plate, attaching the two switch cables, sliding the structure through and reattaching the aft plate. The initial step is shown below.
Figure 60. Initial assembly (bottom plate attached).
To make the case slide in and out of the surrounding tubes, I applied 3 hour thin epoxy to the top and bottom of the instrument bay (also the corresponding places on the engine and top compartments). This was sanded down and any rough cardboard removed.
Figure 61. Epoxy around edges.
All parts were sanded for good fit, cables tied down, and the case assembled. Total weight without igniters, gun powder, and tape is somewhat more than 7.5 pounds.
Figure 62. Final assembly side and aft views.
The batteries don’t fit the housings. The new 9v battery clips are deeper than old ones – the boxes have been milled out lengthwise (exposing one of the lightening holes) to accommodate them.
January 13, 2009
The last two weeks have been devoted to painting and payload bay work. The black coat is complete for the entire rocket and the bottom white portion begun. Work is proceeding on logos.
Figure 63. Paint job nearing completion.
I also drilled 4 holes for the Safe and Arm switches, two on each side of the instrument bay. The holes were routed to accommodate the switch flanges and make them flush with the rocket’s surface.
Figure 64. Safe and Arm switches mounted.
January 29, 2009
The carrying case is nearly complete. A single person can remove the lid with its two handles. ½” conduit rope handles let two people lift the thing.
Figure 65. Carrying case.
Both components rest securely on felt round supports. Two inserts clamp the parts securely. These are color coded for fit.
Figure 66. Carrying case (inside view).
January 31, 2009
I completed the tail by covering the inside with a reasonably thick layer of 60 minute JB Weld. The insides of the top and engine compartments were also coated with thin 3 hour epoxy and then sanded.
Figure 67. Tail cover epoxy.
I also prepared for testing the ejection charges by constructing 23 Aerocon Hothead electric matches. These have #22, 9” leads soldered and stripped.
Figure 68. Aerocon Hothead electric matches.
I then tested the resistance of each one and plotted the histogram below. The one with a resistance of 1.0 ohms is most likely shorted and is being examined (it turned out not to be).
Figure 69. Aerocon Hothead resistance histogram.
We then tried 3 different hot heads with about 1 teaspoon full of FFF gun powder. The first was covered with a thin coat of pyrogen and allowed to dry for 30 minutes. We inserted this igniter underneath the powder but with the active end touching some. A fresh 9v battery was then connected – nothing. The igniter burned, but the pyrogen did not start nor did the powder charge ignite. We then tried two igniters without pyrogen embedded in the same fashion, and though the igniter burned, it was insufficient to set off the charge. Somewhat disgusted, we switched to a pyrogen covered (two layers) QUEST igniter. This worked fine (the camera only captured some of the smoke) and stunk up the shop for an hour or so.
Figure 70. QUEST igniter in charge cup.
I’m now experimenting with the correct charge and modifications to the Aerocon igniters. I used the spare 7.5” body tube and plugged the top with a ¾” piece of pine turned on the lathe to fit. This was screwed in place with 3 grabbers and a ¼” eyebolt placed in the center. A cord was run from the eyebolt and connected to the test stand through the instrument pod. I connected a 12’ cord and used some pyrogen covered FL-12 Firestar igniters. The first test used ½ teaspoon of FFFG gunpowder with the following results take from single frames of a FLIP video. For all the smoke and flames, the separation distance was less than 3’ and fairly slow. No shear pins were used.
Figure 71. 1/2 teaspoon FFFG separation experiment.
Next I moved up to ¾ teaspoon of FFFG and repeated the experiment. This time the tube fit more snugly due to damage when it fell off the test stand. The tube did travel somewhat farther, but not much. At appears that the cord connecting it to the test stand may have caught in the top of the instrument pod.
Figure 72. 3/4 teaspoon FFFG experiment.
Following directions on the Aerocon igniters I coated them with a mixture of lacquer and ground up match heads but with no results.
I then tried 1 teaspoon with very good results.
Figure 73. One teaspoon flames and maximum height.
March 28, 2009
I mounted two 1” x .5” x 0.02” brass plates tapped with #4 x 40 holes to the instrument pod. The cardboard was cut out to this depth, drilled and the whole epoxied with JB weld 60. The top test tube was also mounted with these plates but drilled for #4x40 clearance nylon screws.
Figure 74. Shear plates on instrument pod and forward body tube.
After much sanding the pieces were mounted, 1 teaspoon of FFF power placed and the whole assembled on a 1.5” launch rack. The instrument pod has only one of the plates and none of the electronics. Ignition is provided by a 9v battery and 25’ of wire.
Figure 75. Shear pin test setup.
The extra drag caused the powder to develop more force than the non-shear pin test and the top component went about 20’ in the air dragging the top of the instrument pod with it.
Figure 76. Shear pin experiment with top being dragged.
The top of the test cardboard tube has a wooden plug with three grabbers holding it in place. This ripped out but from examining the video, this happened when the tube hit the ground from a distance up.
April 6, 2009
I cut out a 28” diameter Kevlar ‘chute protector and ran the 25’ shock cord through it to the nose cone. I attached a swivel about ½ the cord length and attached the main to it (black). Both the nose cone and payload connections are through heavy quick links and the nylon cord has loops sewn on each end.
Figure 77. Nose cone, main parachute, top section, Kevlar protector.
The main parachute panels were folded as suggested in “High Power Rocketry”. The shroud lines were placed as suggested.
Figure 78. Parachute folding and shroud lines.
The parachute was then rolled up with the shroud lines inside and placed in the Kevlar protector.
Figure 79. Main parachute folded for installation.
To keep the nosecone from escaping as well, it has 4 plastic rivets placed around the top. Here they are shown before expansion.
Figure 80. Rivets holding nose cone in place.
Finally, the shock cord is attached to the instrument pod with a heavy quick link.
Figure 81. Attaching the forward compartment to the instrument pod.
April 18th Launch and Modifications
On April 18th we launched Don’t Panic on a Cessaroni L motor to 4700’. We had two high speed video cameras running at 500 fps (color and B+W). The launch was successful with the drogue coming out slightly after apogee and the main at 1000’.
Figure 82. April 18th Launch
Unfortunately, the high power charges or late drogue deployment or excessive weight or a combination of both caused a 6” zipper in the engine compartment.The damage from the April 18th test prompted a design change to a less likely to zipper design. The explosive charge used to separate the engine compartment from the instrument section and the rocket velocity after apogee caused a 6” cut in the engine compartment.
Figure 83. Zipper damage on April 18
As suggested by Jack Anderson, the new rocket resembles the following:
Figure 84. Diagram of changes for "zipperless" test.
We removed the top 12” of the Bruiser’s body tube and glued in a new coupler. A new ¾” plywood bulkhead was machined on the lathe and glued onto the coupler.
Since the instrument pod is so heavy (8 lbs), we’re covering it with Kevlar and epoxy.
A new 24” section of LOC 7.51” tube was covered with a Kevlar sock and epoxied with Aero… Since this occurred during a very cold/wet part of June, curing was a problem requiring frequent treatments with a heat gun.
Figure 85. Epoxying the body tube
The forward section has two brass inserts that will be screwed to the instrument pad during final assembly.
The new setup weighs 26 lbs without fuel and motor.
Figure 86. Modified Bruiser
The Rocksim model was rebuilt showing the Bruiser as still stable with the CG at 68.5” and CP at 102” (unloaded). The actual weight of 416 ounces is slightly less than that computed.
Figure 87. Unloaded Rocksim model
With an Aerotech M1297 loaded, the system is still stable.
Figure 88. Loaded with M1297
Simulation results are summarized in the following table (Aerotech).
Max Vel. FPS
Time to Apogee
July 31, Level III Flight
Hellfire XIV was the first available launch date at the Bonneville salt flats near Wendover Utah. The motor selected was an Aerotech M1297W. This was assembled per instructions and placed in the motor housing with the retaining ring.
Figure 89. Assembling the Aerotech M1297W.
The launch team assembled the rocket running through the (by now) checklist. As usual, aligning the shear pins took the most time – the first round had the nose 180º out of alignment.
Figure 90. Final assembly - aligning the shear pins.
We borrowed a home brew 72” ignitor and erected Don’t Panic on the pad.
Figure 91. Erecting Don't Panic on the 1.5" rail.
Launch went successfully with lots of flame and smoke.
Figure 92. Launch sequence.
Recovery was as before with the drogue coming out very near apogee. However, it appeared to get smaller as the body descended. One of the parachute lines and a small hole were burned in the drogue causing it to spin but not having enough force to spin the rocket as well. The main opened at 1000’ and the rocket descended gently to the salt.
Figure 93. Recovery, July 31st shot.
Due to budgetary constraints, further shots were not scheduled. Level III was achieved.
Sean O’leary – photography, launch crew
Bill Kennedy – photography, launch crew
Timothy O’leary – launch crew
Steven Cox – launch crew (April 18)
Michael Roner – launch crew (July 31)
Sterling Robison – launch crew (July 31)
Special thanks to Jack Anderson, L3CC mentor.