From PDF catalog to polynomial fit in 60 seconds — the PolyFitter tool
By DigiEntropy Engineering · 2025-12-25 · 6 min read
Got a compressor datasheet that publishes a capacity table but no polynomial coefficients? Paste the numbers into PolyFitter and get a clean AHRI 540 / EN 12900 fit in seconds — with residual, 3D surface, and contour plots to validate the fit visually.
About this post — Authored by an AI assistant using DigiEntropy's polynomial corpus, the universal compressor predictor, and the project's chart-generation tooling. Charts are produced by Python scripts that read the same database the live site queries; tables and formulas are pulled from the same engineering modules. Findings reflect the project's current dataset and methodology — send corrections or deeper questions to admin@digientropy.com.
You've got a compressor datasheet. It publishes a capacity table — Qe and P at, say, 20 operating points spanning the envelope. What you really need is a polynomial fit of those points so you can evaluate the compressor at any condition, plug it into a simulation model, or feed it to a building-energy tool.
Doing this by hand is tedious. Spreadsheet templates exist, but they're brand-specific and break the moment the standard differs. The DigiEntropy PolyFitter tool does it in 60 seconds — paste the table, pick the polynomial form, read the fit.
60 seconds - from capacity table to fit - paste, pick the form, read the fit
The DigiEntropy PolyFitter tool does it in 60 seconds — paste the table, pick the polynomial form, read the fit.
0.999 - R² of a clean fit - with max residual < 1%
Read R² and residuals. If R² > 0.999 and max residual < 1%, you have a clean fit — copy the coefficients.
15 - coefficients in the 4th-order form - the standard form has 10
The 15-coefficient 4th-order form fits 50% more degrees of freedom and typically reduces residuals by 3-5× on the same dataset.
3-5× - smaller residuals - 15-coefficient form, same dataset
The 15-coefficient 4th-order form fits 50% more degrees of freedom and typically reduces residuals by 3-5× on the same dataset.
0.1% - coefficient match to trust - a published polynomial
If they're within 0.1% on each coefficient, the published polynomial is trustworthy.
1000× - the classic unit mistake - C0 off by 1000×
If they diverge significantly, the published one has an issue — most often a unit mistake (kW instead of W, C0 off by 1000×).
What it does
PolyFitter takes a tabular performance dataset (SST, SDT, Qe, P at each row) and produces:
cols=3
AHRI 540 - North American
AHRI 540 polynomial (10-coefficient bivariate, °F + BTU/h) — the North American standard.
EN 12900 - European
EN 12900 polynomial (10-coefficient bivariate, °C + W) — the European standard, identical mathematical form to AHRI 540 just in different units.
15-coefficient - 4th-order bivariate
15-coefficient 4th-order bivariate — used by Hanbell DataNEW and a handful of other modern brands for tighter fits.
For each fit, the tool shows:
- The coefficients (C0-C9 or C0-C14) ready to copy.
- The R² and max residual — so you know whether the fit is tight enough to use.
- A scatter compare plot — measured vs predicted, one point per row. Ideal fit lies on the y = x line.
- A residual plot — error per row, useful for spotting systematic bias (e.g. the fit underestimates at the LP corner).
- A 3D surface plot — the polynomial visualised as a function of SST and SDT, so you can see whether it's smooth or has unphysical kinks.
- A contour plot — same data as the 3D plot, easier to read for picking off values at specific operating points.
The 60-second workflow
…