TL;DR
After fermentation begins, the ethanol in your sample changes the refractive index of the liquid. A refractometer will read significantly higher than the actual gravity — a reading of 8 °Brix might correspond to an actual FG of just 1.005, not 1.032. To get accurate final gravity from a refractometer, you must apply a correction formula using both the original Brix (pre-fermentation) and the current Brix (post-fermentation). The Terrill linear formula is the most widely used and reliable method. This guide explains the science, provides the formulas, includes a comprehensive correction table, and walks you through the Wort Correction Factor.
The Problem: Ethanol Fools Your Refractometer
A refractometer measures the refractive index of a liquid — how much a beam of light bends as it passes through the sample. In a pure sugar-water solution, the refractive index correlates directly with sugar concentration. This is why refractometers work perfectly for measuring wort gravity before fermentation.
But fermentation changes the composition of the liquid. Yeast converts sugar (high refractive index) into ethanol (lower refractive index than water) and CO₂ (which leaves the solution). The resulting liquid is no longer a simple sugar-water solution — it is a complex mixture of residual sugar, ethanol, water, proteins, and other compounds.
Ethanol has a refractive index of 1.3611 at 20 °C, compared to water’s 1.3330 and a typical wort’s 1.3400–1.3550. The refractometer “sees” the ethanol and interprets it as sugar, reading higher than the actual dissolved sugar content.
How Big Is the Error?
| Scenario (OG 1.050 beer) | Hydrometer FG | Refractometer Raw Brix | Apparent FG from Raw Brix | Error |
|---|---|---|---|---|
| Well-attenuated (80%) | 1.010 | 7.2 °Bx | 1.029 | +19 points |
| Fully attenuated (90%) | 1.005 | 6.0 °Bx | 1.024 | +19 points |
| Stuck fermentation (50%) | 1.025 | 9.5 °Bx | 1.038 | +13 points |
The error grows larger as more alcohol is produced. For a high-gravity beer (OG 1.090) that finishes at 1.012, the raw refractometer reading might suggest an FG of 1.042 — a staggering 30-point error that would make you think fermentation has barely started.
This is the single most common measurement error in homebrewing, and it leads to unnecessary stress about “stuck fermentations” that are actually complete.
ABV CalculatorCalculate your alcohol by volume from gravity readings
The Science of Refraction in Ethanol-Sugar Solutions
The refractive index (n) of a solution depends on the concentration and refractive properties of all dissolved components. For a simple sucrose solution:
n = 1.3330 + 0.00142 × C
Where C is the concentration in grams per 100 ml.
For a fermented solution containing both sugar and ethanol, the relationship becomes:
n_measured = f(sugar_remaining, ethanol_produced, other_solutes)
This is not a simple linear relationship. The ethanol contribution partially offsets the sugar contribution, but not in a 1:1 ratio. This is why we need correction formulas that account for both the original gravity (which tells us how much sugar was available) and the current refractometer reading (which is the combined signal).
Correction Formulas
The Terrill Linear Formula
Developed by Sean Terrill through empirical calibration against hydrometer readings, this is the most widely used correction. It calculates the corrected Final Gravity directly:
FG = 1.0000 − 0.0044993 × OBrix + 0.011774 × FBrix + 0.00027581 × OBrix² − 0.0012717 × FBrix²
Where: - OBrix = Original Brix reading (pre-fermentation, corrected for WCF) - FBrix = Final Brix reading (post-fermentation, raw from refractometer) - FG = Corrected Final Gravity in SG
Worked Example
| Parameter | Value |
|---|---|
| OBrix (pre-fermentation) | 12.5 °Bx |
| FBrix (post-fermentation, raw) | 7.1 °Bx |
Step-by-step: 1. −0.0044993 × 12.5 = −0.05624 2. 0.011774 × 7.1 = 0.08359 3. 0.00027581 × 12.5² = 0.04310 4. −0.0012717 × 7.1² = −0.06410 5. FG = 1.0000 − 0.05624 + 0.08359 + 0.04310 − 0.06410 6. FG = 1.0064 ≈ 1.006
Compare this to the naive conversion of 7.1 °Bx, which would give an apparent FG of approximately 1.028. The correction moved us from 1.028 to 1.006 — a difference of 22 gravity points.
The Terrill Cubic Formula
For improved accuracy at the extremes (very high or very low attenuation), Terrill also published a cubic version:
FG = 1.0000 − 0.0044993 × OBrix + 0.011774 × FBrix + 0.00027581 × OBrix² − 0.0012717 × FBrix² − 0.0000072800 × OBrix³ + 0.000063293 × FBrix³
The additional cubic terms provide a slightly better fit at OG values above 1.080 (≈19 °Bx) or when attenuation is below 60% or above 85%.
When to Use Linear vs Cubic
| Scenario | Recommended Formula |
|---|---|
| Standard beer (OG < 1.070, attenuation 70–85%) | Linear — simple, accurate |
| High-gravity beer (OG > 1.080) | Cubic — better at extremes |
| Wine or mead (OG > 1.090) | Cubic — or use hydrometer for FG |
| Low attenuation (< 65%) | Cubic — linear can over-correct |
| Very high attenuation (> 90%) | Cubic — linear can under-correct |
The Bonham Formula (Alternative)
Louis Bonham published an alternative correction in Brewing Techniques (1997) based on a different calibration dataset:
FG_Plato = −676.0842 + 1286.4830 × SG_ri − 800.8179 × SG_ri² + 190.7410 × SG_ri³
Where SG_ri is the specific gravity calculated from the refractometer index. This formula is less commonly used in modern brewing software but remains valid. Its main drawback is that it requires converting between SG and refractive index rather than working directly with Brix values.
Comprehensive Correction Table
The following table provides corrected FG values for common OG/FBrix combinations, using the Terrill linear formula. Find your original Brix along the top and your current (post-fermentation) Brix reading down the left side.
For OBrix = 10.0 (OG ≈ 1.040)
| FBrix (raw) | Corrected FG | Approx. ABV |
|---|---|---|
| 4.0 | 0.998 | 5.5% |
| 5.0 | 1.002 | 5.0% |
| 6.0 | 1.006 | 4.5% |
| 7.0 | 1.011 | 3.8% |
| 8.0 | 1.017 | 3.0% |
| 9.0 | 1.024 | 2.1% |
For OBrix = 12.5 (OG ≈ 1.050)
| FBrix (raw) | Corrected FG | Approx. ABV |
|---|---|---|
| 5.0 | 0.999 | 6.7% |
| 6.0 | 1.003 | 6.2% |
| 7.0 | 1.006 | 5.8% |
| 7.5 | 1.009 | 5.4% |
| 8.0 | 1.011 | 5.1% |
| 9.0 | 1.017 | 4.3% |
| 10.0 | 1.024 | 3.4% |
For OBrix = 15.0 (OG ≈ 1.061)
| FBrix (raw) | Corrected FG | Approx. ABV |
|---|---|---|
| 5.5 | 0.997 | 8.4% |
| 6.5 | 1.001 | 7.9% |
| 7.5 | 1.005 | 7.4% |
| 8.5 | 1.010 | 6.7% |
| 9.5 | 1.017 | 5.8% |
| 10.5 | 1.024 | 4.9% |
| 11.5 | 1.033 | 3.7% |
For OBrix = 18.0 (OG ≈ 1.074)
| FBrix (raw) | Corrected FG | Approx. ABV |
|---|---|---|
| 6.5 | 0.997 | 10.1% |
| 7.5 | 1.001 | 9.6% |
| 8.5 | 1.006 | 8.9% |
| 9.5 | 1.012 | 8.1% |
| 10.5 | 1.019 | 7.2% |
| 12.0 | 1.031 | 5.6% |
| 13.5 | 1.045 | 3.8% |
For OBrix = 22.0 (OG ≈ 1.092)
| FBrix (raw) | Corrected FG | Approx. ABV |
|---|---|---|
| 7.0 | 0.993 | 13.0% |
| 8.5 | 1.001 | 11.9% |
| 10.0 | 1.010 | 10.8% |
| 11.0 | 1.016 | 10.0% |
| 12.0 | 1.024 | 8.9% |
| 14.0 | 1.041 | 6.7% |
The Wort Correction Factor (WCF)
What Is WCF?
Brewing wort is not a pure sucrose solution. It contains maltose, maltotriose, glucose, fructose, dextrins, amino acids, and other compounds — each of which refracts light slightly differently than sucrose. The Wort Correction Factor adjusts for this difference.
Corrected OBrix = Measured OBrix / WCF
Typical WCF Values
| Wort Type | Typical WCF | Notes |
|---|---|---|
| All-malt beer wort | 1.040 | Industry standard default |
| Wort with adjuncts (rice, corn) | 1.030–1.035 | Simpler sugar profile |
| Wort with sugar additions (10%+) | 1.020–1.035 | Closer to sucrose |
| Wine must (grape) | 1.000–1.010 | Nearly pure sucrose/fructose/glucose |
| Mead must (honey) | 1.000–1.020 | Depends on honey variety |
| Cider must (apple) | 1.000–1.015 | Mostly fructose/glucose |
How to Determine Your Personal WCF
- Take a pre-fermentation sample.
- Measure with both your hydrometer (at calibration temperature) and your refractometer.
- Convert the hydrometer SG to Brix: Brix ≈ (SG − 1) × 1000 / 4 (approximate) or use the full polynomial.
- Calculate: WCF = Refractometer Brix / Hydrometer Brix
- Repeat across 3–5 batches and average the results.
Example:
| Measurement | Reading | Brix Equivalent |
|---|---|---|
| Hydrometer SG | 1.052 | 12.87 °Bx |
| Refractometer | 13.4 °Bx | 13.4 °Bx |
| WCF | 13.4 / 12.87 | 1.041 |
A WCF of 1.04 is standard. If yours consistently comes out at 1.06, use that — it will significantly improve your correction accuracy.
Common Mistakes
Mistake 1: Forgetting to Correct Entirely
This produces wildly inaccurate FG and ABV values. If your refractometer says 7 °Bx but your beer tastes finished and dry, the problem is not a stuck fermentation — it is an uncorrected reading.
Mistake 2: Using the Wrong OBrix
The correction formula requires the original Brix reading, not just any pre-fermentation number. If you added sugar or fruit after your OBrix reading, the formula will be off. Always take OBrix as the last reading before pitching yeast.
Mistake 3: Not Applying WCF to OBrix
If your WCF is 1.04 and your measured OBrix was 13.0, the corrected OBrix is 13.0 / 1.04 = 12.5. Using 13.0 in the Terrill formula instead of 12.5 will shift your corrected FG by approximately 1–2 gravity points.
Mistake 4: Confusing Brix and Plato Post-Fermentation
While Brix and Plato are nearly identical in pre-fermentation wort, they diverge in fermented solutions. Most refractometers report in Brix. The correction formulas above use Brix. Do not mix scales.
Mistake 5: Temperature Extremes
Handheld refractometers with ATC (Automatic Temperature Compensation) typically work within 10–30 °C (50–86 °F). Outside this range, the ATC cannot fully compensate. Always take readings within this range. See our Gravity Temperature Correction Guide for more on temperature effects on measurement instruments.
Practical Workflow: Refractometer-Only Brewing
Many homebrewers prefer refractometers for brew day because they require only a few drops per reading. Here is the complete workflow:
| Step | Action | Reading Type |
|---|---|---|
| 1 | Pre-boil gravity check | Raw Brix (no correction needed) |
| 2 | Post-boil OG | Record as OBrix (apply WCF for calculations) |
| 3 | Fermentation monitoring (days 2–7) | Raw Brix + correction = tracking FG |
| 4 | Final gravity confirmation | Corrected Brix → FG |
| 5 | ABV calculation | Use corrected OG and FG in ABV formula |
For step 4, if your corrected refractometer FG and your expected FG (based on yeast attenuation) agree within ±2 gravity points, you can be confident fermentation is complete. If in doubt, take a hydrometer reading to confirm.
For the ABV calculation in step 5, use our Abv Calculator Complete Guide with the corrected gravity values.
When to Use a Hydrometer Instead
Despite its convenience, there are times the refractometer should yield to the hydrometer:
| Situation | Why Hydrometer Is Better |
|---|---|
| Final gravity for competition entries | Eliminates correction formula uncertainty |
| Very high gravity ferments (> 1.100) | Correction formulas less reliable at extremes |
| Validating your WCF | Need a reference instrument |
| First time using a refractometer | Build confidence by taking parallel readings |
| Troubleshooting a stuck fermentation | Confirm refractometer correction is not the issue |
Our Hydrometer Buying Guide covers which model is right for your needs and budget.
Digital Refractometers: Do They Solve This Problem?
Digital refractometers (like the Milwaukee MA871 or ATAGO PAL-1) offer advantages in display readability and repeatability, but they measure the same physical property — refractive index. They are subject to the exact same post-fermentation error. You still need to apply a correction formula.
The one exception is a digital density meter (such as the Anton Paar DMA 35), which measures density via oscillating U-tube technology rather than refractive index. These give true SG readings regardless of alcohol content, but cost 1,500+ EUR — firmly in professional brewery territory.
Summary
| Key Concept | Detail |
|---|---|
| Why raw Brix is wrong post-fermentation | Ethanol changes refractive index |
| Typical error magnitude | 10–30+ gravity points |
| Best correction formula | Terrill linear for most brewing |
| When to use cubic | High gravity (> 1.080) or extreme attenuation |
| WCF default | 1.04 for all-malt beer wort |
| Best practice | Determine personal WCF with parallel readings |
| When to use hydrometer instead | Competition, extreme gravities, troubleshooting |
Affiliate Disclosure: FermentationTools.net is reader-supported. When you purchase through links on this page, we may earn a small affiliate commission at no extra cost to you. This supports our ability to maintain free calculators and guides.
Methodology
The Terrill linear and cubic correction formulas are from Sean Terrill’s original research, published on his blog (seanterrill.com, 2011). Terrill developed these formulas by collecting paired hydrometer and refractometer readings across 57 fermentation data points spanning OG values from 1.033 to 1.093, then fitting polynomial regression models. His work has been independently validated by multiple homebrewing forums and the Brülosophy experimental brewery (brulosophy.com).
The Bonham correction formula was originally published by Louis Bonham in Brewing Techniques magazine (September/October 1997, Vol. 5, No. 5). The Wort Correction Factor concept is discussed by both Terrill and Bonham, with the standard value of 1.04 documented in the original New Brewing Lager Beer by Greg Noonan (Brewers Publications, 1996).
Refractive index values for water (1.3330), ethanol (1.3611), and sucrose solutions are from the CRC Handbook of Chemistry and Physics (102nd Edition, 2021). The ICUMSA (International Commission for Uniform Methods of Sugar Analysis) Methods Book, Method GS2/3-1 was referenced for the Brix-to-refractive-index standard conversion.
John Palmer’s How to Brew (4th Edition, Brewers Publications, 2017) and the BJCP 2021 Style Guidelines provided context for typical OG/FG ranges and acceptable measurement tolerances. Error magnitude estimates were generated by computing the Terrill correction across a range of common OG/FBrix pairs and comparing to the naive Brix-to-SG conversion.