Projectile Motion Inquiry Lab
Build predictions, conduct controlled virtual launches, and decide whether differences between theory and observation are explained by measurement uncertainty, air resistance, or systematic error.
1. The inquiry
Driving question: How accurately can we predict where a projectile will land, and how do uncertainties in launch conditions affect the quality of that prediction?
Before touching the simulator
Investigation constraint: change only one independent variable at a time. Run at least five repeated measured trials under the same settings before comparing two launch conditions.
2. Virtual projectile sandbox
Set launch conditions and measurement quality, enter a numerical prediction, then launch. The gold curve is the ideal trajectory with the same initial conditions; the blue curve includes the selected drag law.
Quadratic drag uses k = ρ Cd A/(2m), with units m⁻¹. It represents the combined effects of density, drag coefficient, area, and mass.
Optional physical drag coefficient
Use the quadratic drag mode to apply the calculated parameter.
Instrument model
Axes show position in meters; the projectile's speed is not visualized to real-time scale. Exact numerical solutions use fourth-order Runge–Kutta steps, with landing interpolated to ground level.
3. Record observations and compare conditions
Each launch uses the exact model trajectory plus a newly generated instrument reading. Hold settings fixed for five trials, then vary one quantity. The table preserves the physical settings for each launch.
| # | v₀ (m/s) | θ (°) | h (m) | Drag model / parameter | Prediction (m) | Model range (m) | Observed range (m) | Time (s) |
|---|---|---|---|---|---|---|---|---|
| No trials recorded. | ||||||||
Interpretation
4. Propagation of uncertainty
Specify uncertainty in the launch inputs. This is separate from the instrument's error in reading the landing location. The sensitivity calculation applies central finite differences to the currently selected motion model, including drag.
The drag uncertainty has units m⁻¹ (quadratic) or s⁻¹ (linear). When no drag is selected, its contribution is zero. Inputs are assumed independent and uncertainty values represent 1σ standard uncertainties. No covariance or model-discrepancy term is included.
| Input | ∂R / ∂input | σ(input) | Range contribution (m) |
|---|---|---|---|
| Run the calculation to see each component. | |||
Analytical check: equal-height, zero-drag case
∂R/∂v₀ = 2v₀ sin(2θ)/g
∂R/∂θ = 2v₀² cos(2θ)/g (θ in radians)
∂R/∂g = −v₀² sin(2θ)/g²
Convert angle uncertainty to radians when applying the analytical derivative. Near 45°, the first-order sensitivity of ideal range to angle is zero; larger angle uncertainties or large nonlinear effects require a higher-order or Monte Carlo treatment.
5. Analysis, error, and scientific argument
Final claim–evidence–reasoning
Submission checklist
Teacher implementation notes
Suggested duration: 55–75 minutes AP Physics C Calculus + uncertainty
- Prediction & ideal model (10 min): derive equal-height range and identify why 45° is optimal in that special case.
- Exploration & five repeated trials (15 min): set drag to zero, instrument noise nonzero; repeat without altering launch inputs.
- One-variable inquiry (15 min): vary launch angle or quadratic drag and distinguish numerical modeling from observed readings.
- Propagation (15 min): calculate sensitivity terms, identify the dominant input uncertainty, compare against instrument SEM.
- Conclusion (10 min): explain why better precision cannot resolve a systematically wrong model.
Numerics: state variables (x, y, vx, vy); acceleration ay = −g − k|v|vy and ax = −k|v|vx for quadratic drag; linear model uses −b vx and −b vy. 4th-order Runge–Kutta with steps no greater than 0.01 s. Flight termination at y=0 with linear interpolation; flight maximum computed from trajectory samples. For extremely high/low launch configurations, numerical results are approximate.
Next Step: Model Motion with Python
After finishing the lab, use your measured acceleration and time in a constant-acceleration model. Compare predicted velocity and displacement with your observations and explain the assumptions.
Predict first, build and run your program in the browser sandbox, then use the results to explain your model. Download your lab record when you finish.