How to Do Real Physics and Astronomy Research at Home: A High School Student’s Guide to Free Tools, Affordable Kits, and Public Databases

Physics used to mean a university basement full of optical tables, vacuum pumps, and beamlines. Today, the same questions can be attacked with a phone, a laptop, and a public archive that holds more data than any single lab could ever generate.

This guide is your starting point. It walks you through three things: the affordable home kit you can build on a hobbyist budget, the free professional software physicists actually use, and the public datasets from TESS, LIGO, CERN, and Gaia that you can analyze tonight.

Why this is possible now

Three shifts in the last decade quietly opened the door.

First, the world's biggest observatories went open. NASA's TESS and Kepler full-frame images, ESA's Gaia DR3 catalog of 1.8 billion stars, the LIGO Open Science Center strain data, and CERN's ATLAS and CMS Open Data releases are all free to download. The same files a postdoc uses are the same files you use.

Second, your phone became a laboratory instrument. A modern smartphone has a high-frame-rate camera, an accelerometer, a gyroscope, a magnetometer, and a microphone, all calibrated well enough for quantitative physics. Apps like Phyphox and Tracker turn that hardware into a sensor logger.

Third, professional simulation software became free. OpenFOAM for fluids, FEniCSx for finite elements, Quantum ESPRESSO for density functional theory, Geant4 for particle transport, Qiskit for quantum circuits, and REBOUND for N-body orbits all run on a laptop or a free Google Colab GPU.

A kitchen counter and a laptop are now enough to measure sub-nanometer thermal expansion, search for exoplanets in archival light curves, or simulate a parton shower at the LHC.

The physics and astronomy home kit

You do not need every item here. Pick the ones that match the kind of project you want to run.

Optics and lasers

• A 5 mW red, green, or blue laser pointer (under $10 each)
• A $10 to $20 diffraction grating sheet (1,000 lines/mm)
• A beam splitter cube and two front surface mirrors (about $25 together)
• A printed Ronchi grating or pinhole array
• A polarizer film set ($15)

Sensors and microelectronics

• Arduino Uno or ESP32 ($10 to $25)
• Raspberry Pi 4 or 5 with the HQ Camera or NoIR ($60 to $100)
• MPU-6050 or MPU-9250 IMU ($5 to $15)
• Hall effect sensor, photodiode with an op-amp, TMP117 thermistor (a few dollars each)
• AMG8833 IR thermal camera ($25)
• Strain-gauge load cell with HX711 amplifier ($10)

Radio and high-energy detection

• RTL-SDR dongle for radio astronomy ($30 to $40)
• Silicon photomultiplier (SiPM) module ($20 to $50)
• $40 NaI(Tl) scintillator crystal for gamma spectroscopy
• A USB DSO oscilloscope ($15 to $30)

Mechanics and materials

• A 3D printer or access to one at a library or makerspace
• Neodymium magnets, ferrofluid bottle, ball-chain
• Granite floor tile as a vibration-damped optical bench ($20)
• Piezo buzzers, stepper motors, and a $50 high-voltage flyback supply (with adult supervision)

Imaging

• Your smartphone (240 fps slow-motion is standard now)
• A $20 USB microscope
• A DSLR or mirrorless camera for astrophotography (optional)

A full kit covering most subcategories lands somewhere between $150 and $500, and most students start with under $100 of new gear.

The signature technique: smartphone-based quantitative measurement

If you take one tool away from this guide, take this one. Your phone is the most underrated physics instrument in your house. Here is the five-step workflow.

1. Pick the sensor. Camera for tracking motion or imaging fringes, accelerometer for vibration and ballistocardiography, magnetometer for field mapping, microphone for acoustics, gyroscope for rotation.
2. Calibrate against a known reference. Drop a known mass, point at a known star, or use a ruler in frame. Never trust raw sensor counts.
3. Capture data with Phyphox, Tracker, or sensor-logger. Phyphox exports CSV directly. Tracker handles video frame by frame.
4. Process in Python. Use NumPy and SciPy for filtering, fitting, and uncertainty propagation. Save every notebook.
5. Compare to a model. Fit your data to a closed-form equation or a simulation. The fit residuals are where the physics lives.

This workflow turns a phone into a refractometer, an interferometer readout, a seismometer, a photometer, or a pulse-wave-velocity probe.

The dry-lab side: free software you can install today

Numerical and symbolic computation

• Python with NumPy, SciPy, SymPy, and Matplotlib. The default stack for almost every physics calculation.
• Julia with DifferentialEquations.jl and DynamicalSystems.jl. Faster than Python for stiff ODEs and chaos.
• Wolfram Alpha and Mathematica student. Quick symbolic checks.

Fluids, finite elements, and continuum simulation

• OpenFOAM. Industry-grade computational fluid dynamics.
• SU2. Aerodynamics and compressible flow.
• FEniCSx, Elmer, MFEM, deal.II, FreeFEM++, Code_Aster. Finite element solvers for elasticity, electromagnetism, and heat.
• Basilisk. Multiphase flow and surface tension.

Molecular and materials simulation

• Quantum ESPRESSO, ABINIT, GPAW. Density functional theory on Colab.
• LAMMPS, LIGGGHTS, GROMACS. Molecular and granular dynamics.
• pymatgen with the Materials Project API. Pull computed properties of any known crystal.
• ASE (Atomic Simulation Environment). The glue between all of the above.

Astronomy

• Astropy, AstroQuery, photutils. The core Python astronomy stack.
• lightkurve. Open and detrend TESS and Kepler light curves with one line.
• exoplanet (PyMC), juliet, batman, emcee, dynesty. Bayesian transit and radial velocity modeling.
• REBOUND. N-body orbital dynamics.
• MESA. Stellar evolution from protostar to white dwarf.
• TOPCAT and DS9. Browse catalogs and FITS images interactively.
• Stellarium. Plan observations.

Particle, nuclear, and quantum

• Geant4 and Topas. Particle transport through matter.
• Pythia 8 and MadGraph 5. Parton-level and shower-level event generation.
• Rivet with HEPData. Compare your simulation to published LHC measurements.
• Qiskit, Cirq, PennyLane, OpenFermion. Quantum circuits and variational eigensolvers.

Machine learning for physics

• JAX, PyTorch, TensorFlow. All free, all GPU-ready on Colab.
• PyEMMA. Markov state models from MD trajectories.
• pyhf. Statistical inference for particle physics, used in real ATLAS papers.

Running professional tools yourself changes how research feels. You stop reading about physics and start doing it.

Public databases that count as real data

Exoplanets and stellar astronomy

• MAST. Hosts Kepler, K2, TESS, Hubble, and JWST data products.
• NASA Exoplanet Archive. Confirmed planets and candidate lists.
• Gaia DR3. Positions, parallaxes, and proper motions for 1.8 billion stars.

Time-domain and transient surveys

• ZTF public alerts. Optical transients from the Zwicky Transient Facility.
• ASAS-SN and AAVSO. Variable-star photometry going back decades.
• Pan-STARRS and SDSS. Multi-band imaging of half the sky.

Solar and space weather

• SDO, SOHO, and Parker Solar Probe public archives. Full-disk solar imaging and in-situ plasma data.
• GOES X-ray and NOAA DSCOVR. Flare timing and solar wind.
• INTERMAGNET. Ground geomagnetic observatory data.

Gravitational waves

• GWOSC. Every public LIGO and Virgo strain segment, plus event catalogs.

Particle and nuclear physics

• CMS Open Data and ATLAS Open Data. Real LHC collision events.
• MicroBooNE Open Data. Neutrino interactions from Fermilab.
• HEPData. Published measurements in machine-readable form.
• Particle Data Group. The reference table for every particle property.

Materials and chemistry

• Materials Project, AFLOW, NIST WebBook. Computed and measured material properties.

Small-body and minor-planet

• JPL Horizons. Ephemerides for every known solar system object.
• Minor Planet Center. Asteroid and comet observations.

High-energy astrophysics

• HEASARC. Chandra, XMM-Newton, Swift, NuSTAR, and Fermi archives.

A serious project can come entirely from re-analyzing data that already exists. Public-archive re-analysis is a legitimate research path, not a backup plan.

How to combine wet and dry: the strongest project shape

Pattern A: build, measure, simulate. You build a small physical apparatus (interferometer, wind tunnel, cloud chamber, gamma spectrometer) and collect data from it. Then you reproduce the data in a simulation (OpenFOAM, FEniCSx, Geant4) and study where the model and the measurement disagree. The disagreement is the science.

Pattern B: archive plus model. You pull a public dataset (TESS light curves, LIGO strain, ATLAS events, Gaia stars) and write a new analysis pipeline. You then run injection-and-recovery tests with simulated signals to characterize how well your pipeline works. The pipeline plus its calibration is the contribution.

Judges respond to this hybrid shape because it shows two skills at once: you can build or measure something real, and you can model what you measured.

Choosing a phenomenon that has not been done

Novelty is a process, not a fixed list.

1. Search Google Scholar for the keywords of your idea. Read the three most-cited results and the three most recent results. Look for a question they did not answer.
2. Search the Society for Science abstracts archive. This shows what other high school students have already submitted to ISEF and affiliated fairs. Find the angle that nobody has tried.
3. Search NASA ADS (for astronomy) or the arXiv preprint server (for physics) for the technical keyword. This is where current professional work lives. Find a method from a 2024 or 2025 paper that you can adapt to a smaller scale or to a public dataset.

If you find adjacent prior work, that is good news. It means the question is real, and you now know the language to describe your specific twist.

A realistic timeline

• One to two weeks. A focused replication or measurement. Build the apparatus, capture clean data, and write up the comparison to one model.
• One to two months. A full hybrid project for a regional fair. Pattern A or Pattern B end to end, with uncertainty analysis and a clear figure of merit.
• Full year. An ISEF-track project. Multiple iterations, a novel pipeline or apparatus, statistical rigor on par with a publishable paper, and a real poster.

If this is your first project, start with the one to two week version. Momentum beats ambition.

A starter checklist

1. A clean, well-lit workspace with a sturdy table and a place to leave equipment between sessions.
2. A free Google account for Colab GPU access and a free GitHub account for version control.
3. A local Python environment with NumPy, SciPy, Matplotlib, pandas, Astropy, and Jupyter installed.
4. The viewer or simulator for your subcategory installed (DS9 and lightkurve for astronomy, Qiskit for quantum, OpenFOAM or FEniCSx for fluids and mechanics, Geant4 for particle physics).
5. A bound lab notebook or a dated digital notebook for every experiment, with raw data filenames written down.
6. Phyphox or Tracker installed on your phone, and a known calibration target (ruler, tuning fork, known magnet) on hand.
7. A one-sentence research question written at the top of your notebook.

If you can check all seven, you are ready to start.

Where to go next

ISEF organizes Physics and Astronomy into six subcategories. Each one has its own MehtA+ project guide that fits the kit on this page. Pick the one that pulls you in.

• Atomic, Molecular, and Optical Physics (AMO). Lasers, interferometry, spectroscopy, single-photon statistics, and refractive index imaging.
• Astronomy and Cosmology (AST). Exoplanets, variable stars, transients, solar physics, cosmology, and N-body dynamics using public sky surveys.
• Biological Physics (BIP). Physics of living systems: fluid flow in plants, active matter, microrheology, biomechanics, and physiological signal processing.
• Condensed Matter and Materials (MAT). Electronic, mechanical, and thermal properties of solids, granular matter, and 2D materials, often through DFT or homemade probes.
• Mechanics (MEC). Classical and fluid mechanics, granular flows, aerodynamics, elastic instabilities, and bio-inspired structures.
• Nuclear and Particle Physics (NUC). Cosmic-ray detection, gamma spectroscopy, and re-analysis of CMS, ATLAS, ALICE, and MicroBooNE open data.
• Theoretical, Computational, and Quantum Physics (THE). Quantum algorithms, tensor networks, differentiable simulators, machine learning for physics, and lattice studies.
• Other (OTH). Cross-cutting projects: thermoacoustics, triboelectricity, gravimetry, radiative cooling, MHD, and sonoluminescence.

A kitchen counter and a laptop are enough to start. Pick the subcategory that interests you most and go.

Project ideas in this category (61)