Work out a starting bubbles-per-second figure for your planted tank and how long your CO2 cylinder will last at your photoperiod — and see why published cylinder-life figures disagree with each other. Everything runs in your browser; nothing is saved.
How it works
The calculator gives you a starting point, not a prescription. Bubbles per second is only a rough guide to how much CO2 your tank gets, because a "bubble" is not a standard size — counters make different-sized bubbles — so the drop checker is the authority.
The Dennerle starting point. Bubbles per second = tank volume (L) × 1.3 ÷ 100. A 100 L tank starts at 1.3 bps. Alongside it, the calculator shows the conservative start: 1 bps per 15–20 US gallons — 1.32–1.76 bps for that same 100 L tank. Begin at the conservative end.
Where the grams come from. A bubble counter passes roughly 25 g of CO2 per day at 1 bps running continuously. That figure is not a guess: 0.16 mL bubbles at a CO2 density of 1.84 g/L work out to 0.16 × 1.84 × 86,400 ÷ 1,000 = 25.4 g/day at 1 bps, so this tool uses 25 g/day. At your photoperiod the daily use is 25 × bps × hours ÷ 24, and cylinder life in days is cylinder grams ÷ daily grams.
Why sources disagree. A 5 lb (2,270 g) cylinder at 1 bps lasts about 91 days running 24 hours a day, but about 272 days at 8 hours a day on a solenoid. Both figures are correct — they just assume different hours. Any "months per bottle" claim is meaningless without the hours, so this calculator shows both.
The ×13 trap. Some calculators publish the same formula as ×13 — 13 bps for a 100 L tank. This tool starts at 1.3, because at 25 g/day per bps, 13 bps would drain a 5 lb cylinder in about a week (2,270 ÷ (13 × 25) = 7.0 days) and sits far outside the 1–2 bps range. If a calculator tells you 13 bps, check whether it means 1.3.
Surface agitation. A rippling surface loses roughly twice the CO2 of a gentle surface, so a tank with heavy surface movement needs more gas than the starting figure. This calculator does not take agitation as an input — judge the real need by the drop checker, aiming for 20–30 ppm CO2, green at 4 dKH.
Worked example
100 L tank, CO2 on 8 hours a day, 5 lb (2,270 g) cylinder, measured rate 1 bps.
Volume 100 L (26.42 US gal), photoperiod 8 h/day, cylinder 2,270 g, measured rate 1 bps.
Dennerle bps = volume (L) × 1.3 ÷ 100. Daily use = 25 g/day × bps × hours ÷ 24. Days = cylinder grams ÷ daily use.
Dennerle: 100 × 1.3 ÷ 100 = 1.3 bps. Conservative: 26.42 ÷ 20 = 1.32 to 26.42 ÷ 15 = 1.76 bps. Daily use at 8 h: 25 × 1 × 8 ÷ 24 = 8.33 g/day. Days: 2,270 ÷ 8.33 = 272.4, about 272 days. Running all day: 2,270 ÷ 25 = 90.8, about 91 days.
Start at 1.3 bps (conservative 1.32–1.76 bps). The 5 lb cylinder lasts about 272 days at 8 h/day — versus about 91 days running all day.
Without a measured rate the calculator uses the Dennerle figure itself: a 200 L tank starts at 2.6 bps, and the same 5 lb cylinder lasts about 105 days at 8 h/day.
How to read your results
The headline is the Dennerle starting rate in bubbles per second, with your tank volume, photoperiod and cylinder underneath. The stat grid shows the four numbers that decide the setup: the starting rate, the conservative range to begin in, the cylinder life at your photoperiod, and the CO2 use per day.
The figures table then splits the use two ways — running 24 hours a day and at your photoperiod — so the solenoid saving is visible, not assumed. If you entered a measured bubble rate, the cylinder rows use that rate; otherwise they use the Dennerle figure.
What to do next: start at the conservative end of the range, watch the drop checker over a few days, and only then step up toward the Dennerle figure. Measure at lights-on — CO2 should be at target when the lights come on. Needle valves are sensitive: adjust a fraction of a turn and wait a day before judging. If fish gasp at the surface, turn the CO2 down and add surface movement at once.
Common mistakes
- Copying a forum bubbles-per-second figure for a different tank size. bps scales with volume — a rate for a 200 L tank is double what a 100 L tank needs.
- Treating bubbles per second as exact. Bubble size varies by counter, so bps is a starting point and the drop checker decides.
- Judging the rate on day one. Adjust a fraction of a turn, wait a day, and watch the drop checker — it lags the tank by 1–2 hours.
- Forgetting the photoperiod in "months per bottle". The same cylinder lasts about 91 days running all day and about 272 days at 8 hours a day — the hours are the whole difference.
- Running CO2 24 hours a day without a solenoid and wondering why the bottle empties in a third of the time.
- Ignoring surface agitation. A rippling surface loses roughly twice the CO2 of a gentle one, so the starting figure underestimates a tank with heavy surface movement.
- Trusting the number over the fish. Gasping, lethargy or fish hanging at the top means turn the CO2 down now, however good the bubbles-per-second figure looks.
Starting rate and cylinder life by tank size
At the Dennerle rate, 8 hours a day, with a 5 lb (2,270 g) cylinder and ~25 g/day at 1 bps. Bigger tanks need proportionally more gas, so the cylinder empties proportionally faster.
| Tank size | Dennerle starting rate | CO2 use at 8 h/day | 5 lb cylinder life at 8 h/day |
|---|---|---|---|
| 10 US gal | 0.49 bps | 4.1 g/day | 554 days |
| 20 US gal | 0.98 bps | 8.2 g/day | 277 days |
| 40 US gal | 1.97 bps | 16.4 g/day | 138 days |
| 75 US gal | 3.69 bps | 30.8 g/day | 74 days |
| 125 US gal | 6.15 bps | 51.3 g/day | 44 days |
These are starting points at the Dennerle rate — start at the conservative end (1 bps per 15–20 gal) and let the drop checker set the final rate. A 125 US gal tank at 6.15 bps empties a 5 lb cylinder in about six weeks; that is the rate doing its job, not a leak.
FAQ
Why do two calculators give different bps for my tank?
Mostly because of the ×13 trap above — one of them publishes the Dennerle formula as ×13 instead of ×1.3, giving ten times the rate. The other common cause is a different starting rule: some calculators use the conservative 1 bps per 15–20 gallons instead of Dennerle. This calculator shows both, so you can see which one a figure came from.
What if I run CO2 24 hours a day?
The cylinder lasts about a third as long as at 8 hours a day — a 5 lb cylinder at 1 bps lasts about 91 days instead of about 272. Plants do not use CO2 in the dark, so the extra gas only pushes the overnight pH down. A solenoid on the light timer is the usual fix.
Do I need a solenoid?
Not to start, but it almost triples cylinder life at an 8-hour photoperiod versus running all day, and it keeps the overnight pH steadier. If you run CO2 without one, the "running 24 h/day" row in the results is the figure that applies to you.
My drop checker is still blue at the suggested rate — now what?
Blue means under ~15 ppm, so CO2 is low. Step the rate up a little — a fraction of a needle-valve turn — wait a day, and check again. Before going far, check for heavy surface agitation or a leak: a rippling surface loses roughly twice the CO2 of a gentle one.
Does bubble size matter?
Yes — it is the whole uncertainty in the grams figure. The ~25 g/day at 1 bps assumes 0.16 mL bubbles; real bubble-counter bubbles run roughly 0.05–0.16 mL, so a counter making small bubbles delivers less gas than the same bps on a counter making big ones. That is why bps is a starting point and the drop checker decides.
Can I use this for DIY CO2?
Not directly — DIY output is set by the recipe, not by a needle valve, so bubbles per second is an output you measure rather than a rate you set. Use the Citric Acid CO2 Calculator for the recipe and how long a batch lasts, then check the ppm with the CO2 Calculator.
What about a tank with very strong surface agitation?
Expect to need more gas than the starting figure — a rippling surface loses roughly twice the CO2 of a gentle surface. This calculator does not take agitation as an input, so treat the starting rate as a floor and raise it slowly while the drop checker stays blue. Do not remove surface movement to save gas: the fish need the oxygen it brings.
Sources
No URLs are on file for the sources below — they are listed by name, as cited in the research notes and the tool's own notes.
- Haynes (ed.), CRC Handbook of Chemistry and Physics — CO2 gas density about 1.84 g/L at 20 °C, 1 atm (Textbook).
- Barr, T., Barr Report / planted tank forums — CO2 loss at the surface; 30 ppm target (Hobby practice).
- Dennerle starting point, via MicroQit — bubbles per second = tank volume (L) × 1.3 ÷ 100 (Manufacturer).
- js11.eu.org CO2 calculator — 1 bps is about 25.2 g/day (Hobby calculator).
- AquaCalc — about 90 days for a 5 lb cylinder at 1 bps running continuously (Hobby calculator).
- Buce Plant — 6–9 months at 1–2 bps, 8 h/day (Retailer guide).