Open-FDD
Open-FDD is open-source fault detection and diagnostics (FDD) and retro-commissioning (RCx) for HVAC, packaged as a Normal Framework app. It runs on the building's NF edge box, reads the Haystack model and point history directly, and works on the local network or through the Normal Online portal.
- 77 SQL fault rules for air handlers, VAVs, plant, boilers, heat pumps, meters and sensors, run hourly inside NF. Results are written back as tagged NF points and can be served on NF's BACnet server.
- 60 retro-commissioning charts, CSV export, and an RDF model with a SPARQL query panel.
- All calculations in SI units; the display can be imperial or SI.
- No site-specific code: equipment and points are bound by Haystack tags and NF classes.
Editions
Open-FDD (nf-app/) |
Open-FDD Lite (lite/) |
|
|---|---|---|
| Hardware | 2 GB+ (4 GB+ for PGlite, TimescaleDB on larger hosts) | 1 GB, e.g. Raspberry Pi 3 |
| Includes | FDD engine, RCx charts, CSV export, SPARQL | Dashboard, charts, weather, BACnet mirror |
| Deploy | python3 scripts/nf/hooks/nf_app_sync.py --env <env> push nf-app/app.json |
... push lite/app.json |
Quick start
Try it without an NF box, using a synthetic two-site building:
node nf-app/dev/mock-nf.mjs
# open http://127.0.0.1:8765/api/v1/apps/static/open-fdd/
Install on an NF box (NF 3.10+): see the install guide (editions, one-command install, upgrade, troubleshooting). How API users read the results: FDD data access.
Run the tests:
node --test nf-app/tests/*.test.mjs lite/tests/*.test.mjs
(cd nf-app && npm ci && npm run typecheck)
python3 -m pytest tests -q
Repository
| Path | Contents |
|---|---|
nf-app/ |
Full edition: UI, hooks, rule packs, tests, mock NF |
lite/ |
Lite edition |
sql/fdd/ |
Fault rules (SQL), registry, role scales, tuning catalog |
open_fdd/ |
PyPI package: pandas reference rules and NF agent tools |
scripts/nf/ |
Build and deploy tooling |
deploy/nf-caddy/ |
Optional Caddy front door |
docs/ |
Documentation site |
openfdd_agent_spec/ |
Guidance for AI coding agents |
FDD rules in SQL and pandas
The fault rules also work on their own, without NF: as SQL on PostgreSQL / TimescaleDB, or as pandas in Python. Every rule runs in SI units. Imperial data (°F, inH₂O, cfm) is converted to SI before the rules run.
| SQL (PostgreSQL / TimescaleDB) | pandas (Python, used for testing) | |
|---|---|---|
| Rules | sql/fdd/postgres/rules/, one file per rule |
open_fdd/rules/, or pip install open-fdd |
| Cookbook | SQL cookbook: each rule's inputs, SI parameters and full SQL | pandas cookbook: source plus a runnable example per rule |
| Thresholds | sql/fdd/tuning_catalog.json: defaults for every rule |
SI defaults in each rule; any difference from SQL is listed in the parity report |
The two implementations are checked against each other in CI. The pandas vs SQL parity report shows which rules match and explains each known difference.
Python package
pip install open-fdd
Pandas versions of the fault rules plus command-line tools for working with NF data, reports and ECM analysis. The package is versioned separately from the NF app. See Python tools.
Support
Support this work
Donations fund ongoing development and independent security testing.
Questions and discussion: Discord.
License
MIT. See LICENSE and CHANGELOG.md.
1 maintainer
Metadata
1 maintainer
Release files for open-fdd 4.5.7
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| open_fdd-4.5.7.tar.gz | 530.7 kB | With provenance | Details for open_fdd-4.5.7.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| open_fdd-4.5.7-py3-none-any.whl | Python 3 | none | any | Details for open_fdd-4.5.7-py3-none-any.whl |
Total release size: 1.1 MB
Release files / open_fdd-4.5.7.tar.gz
| Download URL | open_fdd-4.5.7.tar.gz |
|---|---|
| Size | 530.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
529766beb15749c2ff7f0fad499d6919569026b46b0fdb059b286aead2bbc046
|
|
BLAKE2b-256 checksum How to use checksums |
5e7a89830da699a4836c135ead59ff7b6c4745f81112434da464e2e0999c15e5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 13, 2026.
Transparency logRelease files / open_fdd-4.5.7-py3-none-any.whl
| Download URL | open_fdd-4.5.7-py3-none-any.whl |
|---|---|
| Size | 525.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
4e369776fd268e1336756d0c2e652a3144ffc4a7435869d9cd6095003a789e77
|
|
BLAKE2b-256 checksum How to use checksums |
54079cddf6cd3dbb092ab314ad4bd7602a0e7a396faee74cce39fbd0ccd13835
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 13, 2026.
Transparency logSecurity for open-fdd 4.5.7
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for open_fdd-4.5.7.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 13, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
