You take a gravity reading. You apply the temperature correction. You do everything “right”. Then the number still looks wrong, or the reading drifts, or the hydrometer says one thing and your refractometer says another.
That is not you being sloppy. It is the measurement environment lying to you.
Most correction charts only fix one variable, temperature. Beer and wort introduce several other variables that can shift readings by multiple points. If you understand those variables, you make better brew day and packaging decisions, and you stop chasing phantom problems.
Fast takeaway
If the sample contains CO2, has temperature gradients, or is not fully mixed, the hydrometer can read “wrong” even after correction.
The fix is not a better chart. The fix is better sampling.
If you want the short version of best practice first, use this guide to hydrometer technique: how to use a hydrometer correctly.
1 What temperature correction charts assume (and why beer rarely matches it)
Hydrometers measure density by buoyancy. The deeper the hydrometer sinks, the lower the density of the liquid. Temperature correction charts exist because density changes with temperature, so a warm sample needs a correction to match the hydrometer’s calibration temperature.
The catch is that correction charts quietly assume the sample is:
- CO2-free (no dissolved gas coming out of solution)
- single temperature (no hot top, cool bottom)
- fully mixed (no sugar, alcohol, or temperature layering)
- still (no convection currents moving the hydrometer)
- bubble-free (no microbubbles clinging to the stem)
Wort and beer, especially during and after fermentation, break these assumptions constantly. That is why a “corrected” number can still be off.
2 Dissolved CO2: the biggest hidden error source
The number one reason gravity readings misbehave after fermentation is dissolved CO2. Beer coming out of active fermentation is saturated with CO2. Even “quiet” beer can hold a lot of dissolved gas, especially if it has been cold or under pressure.
2.1 CO2 lowers density slightly, but that is not the main issue
In a strict physics sense, dissolved CO2 changes bulk density a little. In practice, the main problem is what happens as soon as you take the sample and start measuring.
2.2 Microbubbles stick to the hydrometer and fake a low gravity
When CO2 comes out of solution, it forms microbubbles. Those bubbles love to cling to the hydrometer, especially on tiny scratches, dust, dried wort film, or even the etched scale lines. The bubbles increase buoyancy, meaning the hydrometer floats higher and reads a lower gravity than the true value.
This is why you can get two different readings from the same sample, at the same temperature, depending on whether you swirled it, poured it hard, or waited an extra minute.
2.3 Temperature makes CO2 problems worse
Warmer samples release CO2 faster. That means warmer samples are more likely to form fresh microbubbles during the measurement. So yes, you can apply a temperature correction and still be wrong because CO2 has changed the measurement environment while you were taking the reading.
Practical brewer rule
If the beer has been fermenting, or you see tiny bubbles on the stem, do not trust the first reading. Degas the sample, then measure.
3 Convection currents: why readings drift over 30 to 60 seconds
A surprising source of error is temperature gradients inside the test jar. If the sample is cooling or warming unevenly, you can get convection currents.
Here is the basic mechanism:
- Cooler liquid is denser and sinks
- warmer liquid rises
- the jar slowly circulates
- the hydrometer floats in moving fluid
If the hydrometer is slowly rising or falling while you stare at it, you are not “seeing things.” You are measuring in a moving system. Wait until the temperature stabilizes, then read.