ISRA

Research

We think rocketry propulsion knowledge should be accessible to student teams everywhere. So we publish our engineering as we go: test records, thrust data, firmware and designs, failures included.

The repositories

Each one takes questions and ideas through GitHub Discussions. For anything touching propellant formulations or safety, please open a Discussion before a pull request.

  • Propulsion

    Motor test records, thrust data, scanned log sheets, test videos, the analysis script and the thrust-rig firmware.

    MIT. Updated 4 Oct 2026.

  • Electronics

    Flight-computer firmware, sensor code, wiring and circuit diagrams, a flight simulator and bench-test logs.

    MIT. Updated 27 Sept 2026.

  • Recovery

    Parachute and ejection-system records. Nothing committed yet.

    CC BY 4.0. Updated 18 Oct 2026.

  • CAD-exports

    STEP and STL exports of our structural parts. The live designs are in Onshape. Nothing committed yet.

    CC BY 4.0. Updated 18 Oct 2026.

  • Simulations

    OpenRocket designs, pre-flight predictions and post-flight comparisons. Nothing committed yet.

    MIT. Updated 18 Oct 2026.

Static fires

Every static fire for Homi Shatam II with thrust data. Compare up to four to see how the motor has changed between tests.

Pick up to four tests to compare. The last one you pick is highlighted.

T014: thrust rose to 10.45 N 0.70 s after ignition; the recording stops at 0.80 s, and the motor exploded at 3.62 s.
Data cut outMotor exploded
Peak thrust
10.45 N
Data cut out
0.80 s after ignition
Motor exploded
3.62 s after ignition

Thrust is weight-corrected as in our analysis script: tared against the first 49 readings, then corrected for propellant burned. We draw only what was measured, not the extrapolation in the team's plot. Raw data: T014.csv.

Every test

  1. T00126 Jun 2026

    Changed: First test with the thrust rig.

    Result: The rig didn't record, so there's no thrust data.

  2. T00229 Jun 2026

    Changed: Same motor as T001, on the Electronics team's improved rig.

    Result: Our first full thrust curve.

  3. T0033 Jul 2026

    Changed: A repeat of T002 with a smaller nozzle.

    Result: The rig dropped samples; we filled the gaps. Thrust came in lower than T002.

  4. T0046 Jul 2026

    Changed: A change to the propellant and how it was prepared.

    Result: The nozzle failed and the motor exploded. No data.

  5. T00510 Jul 2026

    Changed: A CNC-machined steel nozzle.

    Result: The casing failed and the motor exploded. No data.

  6. T0069 Aug 2026

    Changed: A new propellant formulation and a cardboard casing.

    Result: The nozzle blew out. No data.

  7. T00713 Aug 2026

    Changed: A thicker nozzle; everything else as T006.

    Result: The nozzle blew out again. No data.

  8. T008Uploading soon
  9. T009Uploading soon

    Changed: Uploading soon.

    Result: A full burn of about 2.1 s.

    Peak 12.16 N, burn 2.10 s, total impulse 7.37 N·s. Figures from the team's analysis plot. The committed data file doesn't reproduce them yet, so we don't draw its curve.

  10. T010Uploading soon
  11. T011Uploading soon
  12. T012Uploading soon
  13. T013Uploading soon
  14. T014Uploading soon

    Changed: Uploading soon.

    Result: The motor exploded 3.62 s after ignition. Data cut out at 0.8 s, just after peak thrust.

Safety

All propellant work is done under the direct supervision of qualified advisors, with institutional clearance. Every session follows our propellant safety protocol, PS-01: full protective gear, never fewer than two people, doors open, and only as much material as the batch needs.

Several of our motors have exploded. The thrust rig radios its readings to a ground station, so nobody has to stand near a motor to record it.

Our first motor used potassium dichromate. We’ve stopped, and we urge others not to use it: it’s a toxic, carcinogenic hexavalent-chromium compound.

We publish what we did so others can understand it, not as instructions. Don’t attempt propellant work without the expertise, facilities, supervision and legal clearance it needs where you are.

Read PS-01 (PDF on GitHub)

  • Preparing propellant in full protective gear, as PS-01 requires.

  • Preparing for an ejection charge, in masks and gloves.

How the numbers are computed

Every number on this site comes from the raw thrust data, computed exactly as our analysis script computes it, so it matches the team’s own plots.

Tare
The mean of the first readings, before ignition, is subtracted from every sample.
Weight correction
The load cell also feels the motor getting lighter. We add back the weight of propellant burned, assuming it burns at a constant rate.
Burn time
From ignition to burnout, both read off the raw curve by the team.
Peak thrust
The highest corrected sample during the burn.
Total impulse
The area under the corrected curve from ignition to burnout (trapezoidal rule).
Average thrust
Total impulse divided by burn time.

The rig samples at 10 Hz. T003’s recording dropped samples; its gaps were filled using T002’s curve as a reference, so those spans are drawn dashed. T014’s data cuts out 0.8 s after ignition; we draw only what was measured. T009’s committed data doesn’t yet reproduce its published figures, so we show the figures without the curve. Questions about any of this belong in the Propulsion Discussions.