
thermoprops
io.github.luisbedoia/thermoprops
Thermodynamic properties and cycles with CoolProp: fluid states, refrigeration, Rankine, Brayton.
Documentation
Thermoprops
Open Thermoprops → luisbedoia.github.io/thermoprops
Free calculator for thermodynamic properties of fluids and refrigerants. It runs entirely in the browser, on CoolProp compiled to WebAssembly.

Features
- 30 fluids and refrigerants: water, air, ammonia, CO₂, nitrogen, R134a, R410A, R32, R1234yf, propane…
- Fix any two independent properties (p, T, ρ, h, s, u, x) and get the full state, transport properties included.
- p–h, T–s, h–s, p–v, T–v and p–T diagrams, with the saturation dome, isolines and your states plotted together.
- °C, K or imperial units.
- A Worksheet that solves processes and whole cycles written as in a
notebook (
2: from 1, compressed to P = 1.2 MPa, eta = 80%). - Saturation and superheated tables for every fluid, in English and Spanish: steam tables, R134a…
- Shareable links: fluid, units and states are stored in the URL.
- Works on phones, offline once loaded, and can be installed as an app.
For AI agents: MCP server and API
Agents can compute properties with CoolProp instead of interpolating tables. Free, no key.
MCP server (Streamable HTTP): https://thermoprops-api.luisbedoia.workers.dev/mcp
fluid_state: one state from two properties ("T = -10 °C, sat vapor"), in SI or imperial units, with the reference state textbooks use.solve_worksheet: a whole Worksheet script, from a single process to a combined cycle.
Every answer carries a short link that opens the calculation solved, with
its diagrams, in the Worksheet. A short link not opened for a year may
expire; the Worksheet's own address (/worksheet/?s=<script>) never does.
Add it in Claude (Settings › Connectors › Add custom connector), ChatGPT
(developer mode), Cursor or VS Code with that address. It is listed in the
MCP Registry as
io.github.luisbedoia/thermoprops.
HTTP API: GET https://thermoprops-api.luisbedoia.workers.dev/solve?s=<script>
answers JSON. The address without a path
is its manual, and Link to a calculation
describes the script language.


Development
Properties come from
@luisbedoia/coolprop-rs-wasm,
published on GitHub Packages: installing it needs a token with
read:packages in NPM_TOKEN (.npmrc reads it).
NPM_TOKEN=<token> npm ci
npm run dev # development server
npm test # unit tests
npm run build # type check and production build
Pushing a v* tag deploys to GitHub Pages.
The API and MCP server (api/) are a Cloudflare Worker that runs
the Worksheet's own script language (src/dsl) on the same wasm:
cd api && npm install
npm run dev # local server on wrangler dev
npm run deploy # needs `npx wrangler login`
Credits
All property values are computed by CoolProp. If you use them in academic work, please cite:
Bell, I. H.; Wronski, J.; Quoilin, S.; Lemort, V. Pure and Pseudo-pure Fluid Thermophysical Property Evaluation and the Open-Source Thermophysical Property Library CoolProp. Ind. Eng. Chem. Res. 2014, 53 (6), 2498–2508. doi:10.1021/ie4033999
Developed by the Department of Mechanical Engineering, Universidad de Antioquia.
License
MIT
No installation packages available.