Fins (3 per stage) are multiple sandwich layers of CF cloth over aircraft plywood, with beveled G10 edges on 3 sides, surface-mounted to the airframe with Kevlar & CF cloth at the joint and large-radius formed fillets reinforced with chopped CF and fiberglass. All fins are reinforced after mounting and filleting with CF cloth and FG cloth, with 5 layers tip-to-tip, rotated strand orientation. They are 1/8" thick at the tip and 3/8" thick at the root where the fillet begins, with airfoil cross-section.

All load-bearing separation joints (sustainer drogue at apogee, booster main chute at apogee, and staging separation) use 8"-long same-materials couplers (FG to FG) which are close enough in tolerance to avoid airframe wobble, but with enough clearance to avoid binding under temperature variation. All separation joints were lubricated with graphite lubricant and shear-pinned (as well as vented) to avoid premature separation. All coupler tubes are reinforced internally with CF lamination and aircraft plywood bulkheads.

The entire 17-ft airframe is very strong, with very little weight penalty (Aerosleeves CF). I can suspend the 17' airframe horizontally at the booster fins and sustainer nosecone and bounce up and down on it with all my weight: the sustainer flexes slightly, and the booster doesn't flex at all. The fins are completely rigid and do not flex at all, even when I stand on the fin can. All 3 electronics bays slide into their respective body tubes without a joint or break in the body tube, which reduces wobble and improves strength.

The airframe is designed for Mach+ flight, using high-temperature materials: fin lamination is done with high-temp epoxy, and the conical Hawk Mountain nose cone has a sharp 5"-long tip which I machined out of aluminum. The paint is another story (see below).

Recovery harnesses are 60 ft long x 1" wide tubular nylon, attached to 1/4" steel U-bolts and 1/2" aircraft plywood bulkheads.

Parachutes: the sustainer has a Rocketman drogue and a Skyangle main. The booster uses a single Skyangle main. All chutes are protected by Nomex "burrito" wraps. Here's a good one: the sustainer used a single FG piston for the main. Yes, a piston. :-) The booster main had no reason to use one, nor did the sustainer drogue chute.

Deployment: The sustainer uses a Rouse-Tech CO₂ system for high-altitude apogee, with dual potted ematches. Main chutes use BP charges (2 gm ea). For deployment electronics, the sustainer uses a Blacksky Altacc2C and an Adept Alts60k, with a Timer2 for a "hail-mary" ballistic re-entry backup charge. The booster uses an Altacc and Timer2 for backup. All deployment systems have full redundancy (batteries, ematches, etc.)

Tracking electronics: two BeeLine transmitters, one in each stage.

Staging is active, rather than using drag separation, to allow the rocket to slow down to well below Mach turbulence velocity before attemping separation. I used a Rouse-Tech CO₂ system and shear pins, with staging separation initiated from an Adept staging timer in the sustainer, and timer backup in the booster. The CO₂ system is located in the forward end of the booster airframe. The staging coupler was lubricated with graphite lubricant.

Sustainer motor ignition is controlled by a Blacksky Timer2 in the sustainer. Wiring for separation and ignition runs down the outside of the sustainer airframe, using tape wire laminated into the body tube, and routes through a pair of thin channels passing through the staging coupler. Both staging separation and motor ignition circuits have "pull before flight" shunts on the sustainer. I did several ground tests of the CO₂ staging system to make sure that it worked reliably.  For example, I had to upsize the CO₂ cartridge to achieve repeatable results.

I used a CTI N2500 in the booster, and an AT N2000 in the sustainer. Motor retention was by bolting each motor to its forward bulkhead and recovery hardware, and each motor also has a snap-ring retainer at the aft end (98mm Slimline)

For igniters, I used a couple of Blacksky ematches dipped in Firestar pyrogen, and inserted into small slugs of white lighting propellant. For the booster, the ignitor was held in place by a thin dowel, and on the sustainer, I used a toothpick inserted across the core of the top propellant grain to retain the ignitor wire at the top of the motor.

Guidance off the pad was from three Blacksky extreme rail buttons mounted to the booster at airframe hard-points, a couple near the aft end and one at the center of gravity.

Booster ignition was rapid, and liftoff from JohnL's A-frame pad at the 1500' away cell was smooth and straight up. Burnout was about a mile up. Staging was perfect: right on time, clean, and straight. It had some drama because most people are used to drag separation, so some watchers thought that something was wrong when it coasted for 4 seconds after booster motor burnout. Sustainer ignition happened a second after separation, and the sustainer just took off. It was hard to tell at that altitude, but the smoke trail of the sustainer looked like there was Mach turbulence, some wind shear at high altitude, and possibly some coning. We saw the booster arcing over and reaching apogee, but we lost sight of the sustainer after the N2000 burn. The BeeLine in the sustainer was still indicating "up", so things looked OK. Finally we spotted the sustainer in the binoculars about 5 minutes later. The only flight anomaly was that the main chute was out at apogee. The sustainer took about a half-hour to fall from apogee to the ground.

Alex and Dave found the booster about 1.5 miles out, and brought it back to camp while we were still watching the sustainer come down. The whole gang (GregC, RickC, my friends Randy and Gary, our two boys, and DaveB at the wheel and Alex riding shotgun) piled in the truck and we drove Mad-Max style across the playa, passing Jane, who had followed it until the drift velocity exceeded her vehicle velocity. We finally found the sustainer about 5.5 miles from the pad. The rocket had no damage, except for Mach heating damage to the paint (leading edges of fins, little warts/bubbles on and around wiring fairings, etc.), and the upper body tube had lots a scraping damage to the paint from being dragged around the playa by the chute.

The Rocksim simulation predicted 44,000 to 48,000 ft on the two N motors, but actual was 40,098 ft, off the Adept Alts60k. Unfortunately, the Altacc data was garbage, so I don't have the full flight profile for the sustainer. Bummer. I did get good flight data off the Altacc in the booster, and it recorded a booster apogee of 13k ft.

For construction, I'm making some changes. First of all, I like the Perf Rocketry G10 tubing a lot better, and I want to eliminate some drag (no wiring fairings, better paint (or no paint), smaller staging coupler OD, etc.) and some weight (downsize chutes, less tubular nylon, less body tube lamination, etc.).

Motors: for 2-stage flight, I get another mile of altitude by boosting on an N2000 and staging to an N1100. I would also use a spirit level to make sure the rocket was pointing exactly up at liftoff. For XPRS'05, I angled the rocket away from the flight line a bit. I need to figure out the coning thing too.

Future record attempts: I want to try 3-stage, using commercial motors and the techniques I developed for the two-stage version. I plan to re-use the booster, but build a new 4" dia 2nd booster and a 3" dia sustainer. With a pad weight of 110 lbs, RockSim predicts just a little under 100k ft, by flying long-burn motors (N to N to M, for 93% of full O impulse). The trajectory certainly won't be perfectly vertical, so I'm sure I'll lose a couple of miles off the altitude, but I should be able to double my current record, while staying in the same motor class.

Additional challenges: lighting up a sustainer motor at 35k ft (burst diaphragm), reducing rocket length (new ideas for avionics bays & recovery harness), 3rd-stage timing (smarter timer & pull-connector interlocks), optimum sustainer fin size (TBD), arming a 24-ft tall rocket on the pad (huge ladder?), launch pad (TBD, but if Jumbo were available?...) better tracking (GPS, plus recruiting some helpers stationed several miles downwind), video (light-weight on-board video camera, maybe some telemetry).

When? Who knows... Building what amounts to two high-perf rockets (incl one with video), in half a year is a stretch for me, cuz I don't get much rocket-building time these days. But my goal would be Sept'06 at XPRS. After that? Who knows? Big EX motor in a larger 1st booster, with spin-stabilized sustainer??  :-)