Flight #1 Analysis

Detailed analysis of CATS Vega telemetry data from the L1 certification flight.

Problem Statement

Simulation predicted 208m apogee, flight computer logged 141.58m — a 32% shortfall.

Flight Event Timeline

Data from CATS Vega log file fl001.cfl:

Event	Time (ms)	From Liftoff	Altitude (m)	Velocity (m/s)
Liftoff	278340	0.00s	0	0
Burnout	279670	1.33s	39.84	45.69
Apogee	284020	5.68s	141.58	0
Drogue	284320	5.98s	141.41	—
Main	284630	6.29s	141.43	—
Deployment (accel)	289145	10.80s	—	—

Burn time: 1.33s (matches H128W specification of ~1.4s)

Motor delay: 9.47s (from burnout to deployment shock in accelerometer)

Deployment after apogee: 5.12s

Telemetry Charts

Altitude, velocity, boost phase, and parachute deployment profiles from CATS Vega data.

Accelerometer Deployment Signature

The parachute deployment is clearly visible in the raw accelerometer data as a sudden deceleration shock:

Time (ms)	From Liftoff	Raw Accel X	Notes
289115	+10.78s	-29698	Pre-deployment baseline
289125	+10.79s	-27657	Shock begins
289135	+10.79s	-24535	Deceleration increasing
289145	+10.80s	-22717	Peak shock (~7000 LSB change)
289155	+10.81s	-29504	Recovery to descent

The ~7000 LSB change in 30ms represents the sudden deceleration as the parachute opens and catches air. This provides independent confirmation of deployment timing separate from the barometric data.

Note: The actual motor delay (9.47s) was longer than intended (8s), possibly due to cold weather (-5°C) slowing the delay grain burn rate.

Root Cause Analysis

1. Mass Discrepancy (Primary Cause)

Physics verification using burnout conditions:

• Burnout velocity: 45.69 m/s at 39.84m altitude
• Theoretical maximum (no drag): \(h = \frac{v^2}{2g} + h_0 = \frac{45.69^2}{2 \times 9.81} + 39.84 = 146\text{m}\)
• To reach 208m would require burnout velocity of ~58-60 m/s

Conclusion: Rocket was significantly heavier than simulation assumed. Additional mass from:

• Two flight computers instead of one
• Additional LiPo batteries
• Other small items not accounted for

2. Temperature Compensation Gap

The CATS Vega MS5607 barometric sensor has two temperature-related behaviors:

Component	Status	Notes
MS5607 sensor temperature compensation	✅ Working	Internal calibration
Ambient air temperature in altitude formula	❌ Hardcoded 15°C	TEMPERATURE_0 constant

Flight conditions: -5°C

The barometric altitude formula uses temperature to calculate air density. Using 15°C instead of -5°C causes approximately +7.5% altitude overestimate (~10m at this altitude).

Temperature-corrected apogee: ~132m true altitude

3. Venting Lag

The rocket body has no venting holes, causing pressure equalization lag during rapid altitude changes.

Evidence from telemetry:

• Altitude plateau at apogee: 141.4-141.6m sustained for ~1 second
• Pressure oscillated between 100680-100730 Pa for 1.5s around apogee
• Ascent velocities appear artificially low
• Only 4.6m apparent drag loss vs expected 15-25m for this rocket/velocity

The internal pressure cannot equalize fast enough, causing the barometer to "see" incorrect ambient pressure.

Timing Analysis

Parameter	Simulated	Actual	Difference
Time to apogee	6.0s	5.68s	-0.32s (5% faster)
Motor delay	8.0s	9.47s	+1.47s (18% slower)

The simulation was accurate on flight timing — the altitude shortfall is primarily from mass discrepancy, not aerodynamics. The longer motor delay is likely due to cold weather affecting the delay grain burn rate.

Recommendations

Hardware

1. Add venting holes to rocket body for accurate barometric readings
2. Verify actual flight mass vs simulation input before future flights
3. Account for cold weather when setting motor delays (add margin)

Firmware

Potential code improvement in CATS firmware — use MS5607 temperature reading in calculate_height() instead of hardcoded 15°C constant.

Relevant code locations (CATS firmware repository):

File	Purpose
flight_computer/src/sensors/ms5607.hpp	Barometer driver
flight_computer/src/control/data_processing.cpp	Height calculation
flight_computer/src/config/control_config.hpp	Config constants (TEMPERATURE_0)
flight_computer/src/control/kalman_filter.cpp	State estimation

Firmware repository: tonuonu/cats-embedded

Summary

Factor	Effect	Correctable
Mass discrepancy	Primary cause of altitude shortfall	Pre-flight weighing
Temperature compensation	+7.5% overestimate (~10m)	Firmware fix
Venting lag	Distorted pressure readings	Add vent holes
Cold weather delay	+18% longer burn time	Add safety margin

Despite the altitude discrepancy, the flight was successful and achieved L1 certification.