Guide
Over-Carbonated, Under-Carbonated, or Just Wrong? Troubleshoot by Stage
A symptom-led carbonation troubleshooting guide for small-batch makers, from tank checks to packaging and stability.
A flat pour, a rush of foam, or packages that do not agree with one another can tempt a maker to change the carbonation setting first. Pause there. The same symptom can begin in the tank, on the way to the filler, or after the package is sealed. Find the stage before changing a process condition.
Name the stage before naming the fault
Carbon dioxide in a beverage responds to both temperature and the pressure of CO₂ above the liquid. A warmer product is less able to hold dissolved gas; a colder one holds more at the same pressure. That makes a temperature change an important clue, not a reason to reach for a universal pressure number.[1]
A tank gauge tells you about a process condition. It does not directly tell you the dissolved CO₂ in the product. Start with a representative sample and a suitable calibrated method for your product, then compare the result with the intended specification. Only then can you call the tank under- or over-carbonated with confidence.
Before an adjustment, make a small record of the batch:
- Where does the symptom first appear: tank sample, filler, early run, late run, or sealed package?
- What were the product temperature, tank condition, actual carbonation result, and time of the observation?
- Did the issue affect every package or only a pattern of packages?
- What changed since the last sound batch: transfer, product stability, stone or line service, filler setup, closure or seam work, or cleaning verification?
This short record turns an anxious guess into a useful comparison.
If the tank sample is flat or too lively
First, confirm the result with a representative, properly handled sample. If it is genuinely below target, check the temperature measurement, gas supply path, carbonation stone condition, contact time, and any mixing or installation factor in your own SOP. A pressure/temperature relationship describes equilibrium, but it does not prove that gas has transferred as expected through every stone, hose, and batch.
If the result is above target, do not treat a quick pressure release as a cure-all. Recheck the sample, the temperature reading, and the batch history first. A beverage that continues to ferment or otherwise changes after its initial check needs a product-stability review, not a blind reset of the carbonation process. Fermentation itself produces CO₂, so a live product can move the result after a tank has been judged ready.[1]
If the tank is right but the package is wrong
A sound tank result with weak packages points downstream. Compare the tank sample with packages from the same run using the same controlled sampling practice. Review package integrity, closure or seam performance, filling conditions, and whether product temperature changed between tank and package. A leak path or inconsistent seal can lose the result that the tank delivered.
Foam at the filler is also a location clue. It means gas is leaving solution somewhere in the transfer or fill path; it does not, by itself, diagnose the cause or prove oxygen pickup. Look for the first point of change: product warming, a restriction, an unstable fill condition, or a mismatch between tank and filler operation. Keep the tank condition steady while you isolate that downstream step.
If carbonation moves after packaging
When packages become progressively livelier during conditioning, review product stability and sanitation records alongside carbonation data. Residual fermentation or a process upset may be contributing gas after packaging. When packages become weak or empty, inspect the closure or seam process and look for a package-integrity pattern before altering the tank process.
Do not let one dramatic package define the whole batch. Compare samples by package time, line, closure or seam head, and storage history. A repeatable pattern is more useful than a single surprise pour.
If the run changes from first package to last
Packaging is not a static moment. As the liquid level falls, hydrostatic contribution changes, so headspace pressure and pressure at the tank outlet are not interchangeable. Record where the run began to drift, alongside tank level, product temperature, foam condition, and fill observations.[4]
That distinction matters when a system is maintaining headspace pressure. It may support consistent packaging conditions, yet it cannot replace package checks or a measured dissolved-CO₂ verification. Follow the installation and operating instructions for your equipment rather than borrowing a setting from another tank or maker.
Diagnose calmly, then choose the right tool
For a recurring process-control issue, automation may be worth assessing after the basic diagnosis is complete. FizzWizz uses product temperature and tank pressure to calculate and hold a carbonation process condition until packaging; PacWizz replaces gas in the tank headspace during packaging, with a foot-sensor configuration designed to account for hydrostatic pressure. Neither removes the need to verify the finished product with a suitable method.[3][4]
Treat any unexpected pressure behaviour as a safety event as well as a quality problem. Use pressure-rated equipment, intact relief protection, and the manufacturer-specific SOP. CO₂ is odourless and can accumulate in low areas or confined spaces, so use appropriate ventilation and monitoring practices; never improvise a release step or bypass a safety device.[2]
The reliable loop is simple: locate the stage, record the conditions, verify the product, change one controlled variable, and compare again. That is how a small-batch maker protects a lively pour without chasing every bubble.