PML Intruder Level 1 Certification Flight

After months of preparation, I successfully completed my Tripoli Level 1 high-power rocketry certification on March 1, 2026, at the Tripoli The Netherlands launch site in Exloo. This post documents the build process and flight of my PML Intruder rocket.

The Goal

#

The Tripoli Level 1 certification allows flyers to use H and I-class motors (up to 640 Newton-seconds total impulse). To earn this certification, you must successfully build and fly a high-power rocket with an H motor, demonstrating safe construction and recovery.

The Rocket: PML Intruder

#

I chose the PML Intruder from Public Missiles Ltd, a classic 80mm diameter kit well-suited for Level 1 certification flights.

Here are the specifications:

Flight Simulation

#

Using OpenRocket, I simulated the expected flight characteristics with the Aerotech H219T-14A motor:

The stability margin of nearly 3 calibers provided excellent weathercocking resistance while remaining stable throughout the flight envelope.

Recovery System

#

The Intruder uses a single deployment configuration. I chose to put two parachutes on a single shock cord because of the extra weight of the avionics bay, and because I wanted a really slow touchdown. The rocket wasn’t flying high anyways.

• Main parachute: Stock PML 36-inch (915 mm) nylon chute
• Secondary parachute: An 850 mm white parachute recovered from an iMet-4 radiosonde, launched by the military weather service in Beauvechain

The deployment is handled by an Eggtimer Proton flight computer that I soldered myself from a kit:

• Main deployment: Dual redundant ejection charges fired at nose-over (approximately 1 second after apogee). Motor ejection as a backup, using the piston included in the kit.
• Drogue deployment: Dual apogee charges configured but tested in-flight only (igniters installed without charges to verify they would fire)

The dual-redundant main charge setup on the Proton, as well as the motor ejection as a backup, provides extra reliability for the critical deployment event. I configured the dual drogue setup as a test for the dual-deployment configuration I plan to use for my Level 2 certification flight.

Tracking

#

For flight tracking and recovery, I installed a Vaisala RS41 GPS tracker in the payload bay. The RS41 is actually a radiosonde, launched by our local weather service, that I recovered and reprogrammed to amateur radio frequencies. I track it using a TTGO device running the mysondygo firmware, with a companion mobile app that displays real-time position data.

Build Process

#

Avionics Bay

#

The avionics bay houses the Eggtimer Proton flight computer and RS41 tracker. Key features of the build:

• The Proton is mounted with “UP” labels pointing toward the nosecone for correct apogee detection
• Three 4 mm vent holes drilled in the AV bay for barometric pressure sensing
• All wiring properly secured to withstand flight G-forces
• 2S/7.4V LiPo battery (500 mAH) provides power for the Proton
• 3D printed screw-in charge wells to accommodate different sized ejection charges

The avionics bay was custom designed in Tinkercad and 3D printed in PLA.

Fin Alignment

#

Proper fin alignment is critical for stable flight and to avoid roll. I laser-cut a custom fin alignment jig to ensure the fins were perfectly straight and symmetric. The fins are attached through the body tube and onto the motor tube with epoxy, and filleted for maximum strength.

Paint and Finish

#

The rocket received the default PML paint scheme - a white base coat with red accents. A good plastic primer is essential for the paint to adhere to the Quantum body tube. I used automotive pinstriping tape to mask off the design.

The kit came without decals, so I redesigned the old PML logo in Inkscape. I applied the custom decals using waterslide decal paper and clearcoated over them for protection. The high-visibility color scheme helps with visual tracking during flight and locating the rocket after landing.

The Motor: Aerotech H219T-14A DMS

#

For the certification flight, I used an Aerotech H219T-14A Disposable Motor System (DMS). This is a 38mm diameter, 155mm long single-use motor:

• Total impulse: 233 Ns (H class)
• Average thrust: 222 N, peak 279 N
• Burn time: 1.05 seconds
• Propellant: 109 g of Blue Thunder
• Delay: 14 seconds

The DMS format eliminates the need for motor casing assembly and reloading, which suits certification flights. The less complexity of launch day the better.

Launch Day: March 1, 2026

#

Pre-Flight Checklist

#

Following my comprehensive launch checklist, I:

1. Verified all airframe components and recovery system
2. Confirmed Eggtimer Proton channel continuity (both igniters showing AQUA status)
3. Powered on and verified RS41 tracker transmission & reception
4. Completed RSO (Range Safety Officer) inspection
5. Installed motor, put the rocket on the rail and armed the Proton

The Flight

#

The countdown proceeded smoothly. At ignition, the H219T delivered a perfect boost, accelerating the 2.3 kg rocket off the rail. The flight appeared nominal throughout the burn phase.

At apogee, 253 metres (830 ft) up, the Eggtimer Proton fired the deployment charges.

Recovery

#

Only the larger main parachute deployed successfully. The smaller iMet-4 parachute got stuck in the body tube because an elastic band I used to wrap the shock cord got wedged between the piston and the body tube. Fortunately, the rocket descended in a stable manner and touched down unharmed.

The larger chute showed some signs of burning from the ejection charge - I need to use a nomex blanket next time to protect the parachute.

The RS41 tracker performed flawlessly, providing real-time GPS coordinates throughout the flight. The rocket landed safely in the recovery area, approximately 100 meters from the launch pad.

Post-Flight Analysis

#

After recovery, I inspected the rocket:

• Main parachute deployed correctly but showed minor burn damage
• Smaller parachute stuck due to elastic band interference
• Shock cord showed no signs of wear or heat damage
• Motor retainer remained secure with no gas blow-by
• All fins firmly attached with no damage
• Avionics bay intact with successful flight data recorded

What the flight log shows

#

The Proton recorded the whole flight. It took me until the analysis of the L2 log, half a year later, to read this one properly, but it turned out to be half the fun of the project. The chart below shows the altitude and acceleration profile; the time axis is corrected for the logger’s slow clock, explained below .

The boost lasts about 0.8 seconds and peaks at 12.6 G. Burnout happens at about 30 m (98 ft) up, doing roughly 75 m/s (246 ft/s, 270 km/h), the accelerometer says 83 m/s (272 ft/s) and the barometer-implied speed is about 70 m/s (230 ft/s), and I have no way to pick a winner between them. After a 6.5-second coast the rocket tops out at 253 m (830 ft), 7.3 seconds after lift-off. The Proton detects nose-over 1.1 seconds later and fires the two main charges half a second apart, at 249 m and 245 m (817 ft and 804 ft). The fourth channel, the in-flight test of the dual-deployment hardware, triggers on the way down at its 100 ft (30 m) setting, right on schedule. Touchdown comes 27.5 seconds after lift-off.

This table puts the OpenRocket prediction next to what the Proton recorded, on the corrected time axis:

The apogee came out 30% below the simulation. Two things in the log point at why. The reconstructed motor impulse is 13% below certification (see below ), and through the coast the two speed estimates disagree by 15–20%, with the barometer, which measures vertical speed, reading low. That is the pattern you get when the flight path is not perfectly vertical: with a stability margin of nearly 3 calibers, any wind at all makes the Intruder turn into it, while the simulation assumed no wind.

The logger’s slow clock

#

When I came back to this log, half a year after the flight, I already knew what to look for. The Proton log from the L2 flight in September , the same computer, the same settings, had revealed that the Proton’s timestamps are wrong: it counts samples and stamps each one as 1/33 s, but in its fast logging phase it only manages about 20 samples per second, so every time in that phase has to be multiplied by 1.67. Everything after nose-over is logged at a true 2 samples per second, but shifted by a fixed offset.

This flight’s raw log shows the same symptoms: a boost that seems to last only half a second, and barely 4.4 seconds from lift-off to apogee. With the same 1.67 correction applied, the boost lasts about 0.8 seconds, apogee moves to 7.3 seconds, and the reconstructed motor impulse lands within 13% of its certification, on the uncorrected axis it comes out at barely half. The post-nose-over samples carry the same fixed shift as on the L2 flight, here 3.4 seconds.

Reconstructing the thrust curve

#

Because the Proton logs acceleration, the motor’s thrust can be worked out from the flight data, the same way I did it for the L2 flight . During the burn, thrust equals mass times measured acceleration, plus drag. The mass starts at 2,364 g (5.21 lb), what the rocket weighed on the flight card with the motor installed, and drops by the 109 g (0.24 lb) of propellant as it burns. For the drag I used the deceleration right after burnout, when drag is the only force acting besides gravity: Cd ≈ 0.32, scaled with the square of the speed. The chart compares the result with AeroTech’s certification curve from thrustcurve.org .

The reconstruction tracks the certified curve through most of the burn: peak thrust of 314 N (71 lbf) at 0.10 seconds against 279 N (63 lbf) at 0.11 seconds certified. But the burn ends at about 0.8 seconds, a quarter of a second before the certified curve tails off, and that is what costs impulse: the area under the curve comes out 13% low. The ends of the burn are the least certain part of this reconstruction. At about 20 samples per second the logger takes a sample every 50 ms, and both the time correction and the drag estimate feed straight into the result. The L2 reconstruction had the same shape of shortfall, 8% on that flight. Individual motors vary a few percent; a 13% gap with a visibly shorter burn is more than that, but with a single data point I can’t say whether the motor or my reconstruction is to blame.

What else the data tells us

#

The drag coefficient is lower than the ≈ 0.55 the same body measured on the L2 flight , but that flight carried the camera fairing on the side, and the quantization noise on this log is large, so I would not read much into the difference.

The descent rate is worth a note: 13.9 m/s (46 ft/s) is brisk, and it is exactly why I had put a second parachute on the shock cord. With both chutes out it would have come down around 8–9 m/s (26–30 ft/s). The elastic band that jammed the smaller chute is the one item of flight prep I would do differently. And the main chute’s scorch marks are why there is a nomex blanket in the recovery kit now.

Certification Achieved

#

With the successful flight and recovery, I passed my Tripoli Level 1 certification! Many thanks to Tripoli The Netherlands for hosting the launch and to the RSOs for their guidance throughout the process.

What’s Next

#

With Level 1 certified, I planned three things: the Level 2 certification on a J motor, a dual-deployment configuration for higher flights, and onboard video. All three came together in September 2026, on the Level 2 certification flight of this same rocket.