Key takeaways
- Eight hours is a defensible default for almost any tank, and it is a starting point rather than an answer.
- Low tech tanks want 6 to 8 hours, medium light tanks 7 to 8, and injected high light tanks 8 to 10.
- Change the photoperiod in 30 minute steps and hold each setting for fourteen days before judging it.
- How fast film returns to cleaned front glass is the cheapest and most honest photoperiod feedback you have.
- Algae that survives a cut to six hours is a nutrient or carbon problem, not a lighting one.
Eight hours. If you want one number and no reading, put the timer on eight hours and walk away.
That answer is right often enough to be useful and wrong often enough to be worth ten more minutes. Tank depth, plant choice, and whether you inject carbon dioxide all pull it in different directions.
Below is the procedure I use on a tank I have never seen before. Pick a starting figure, hold it for fourteen days, then move it in half hour steps based on what the glass tells you.
Step 1: work out which of the three tanks you own
Photoperiod follows light intensity, not tank size. A shallow 40 litre (11 gallon) nano under a strong bar can be a high light tank, and a 200 litre (53 gallon) tank under a cheap strip can be a low light one.
Sort yourself by the PAR reaching the substrate, or if you have never measured, by what your plants are doing. Stretching stems mean too little. Green film on the glass every third day means too much.
If the category is genuinely unclear, assume the lower one. Running short is a slow disappointment. Running long is an algae bloom that takes six weeks to clear.
Step 2: set the starting photoperiod
These are starting figures, not final ones. Every one of them assumes the light comes on and goes off in one continuous block with no ambient daylight bleeding in.
| Tank type | Substrate PAR | Starting hours | Notes |
|---|---|---|---|
| Low tech, no injection | 15 to 30 | 6 to 8 | Anubias, Java fern, Cryptocoryne |
| Medium light, liquid carbon | 30 to 50 | 7 to 8 | Swords, Ludwigia, Vallisneria |
| High light, pressurised carbon dioxide | 50 to 80 | 8 to 10 | Carpets, Rotala, red stems |
| Fish only, no live plants | Irrelevant | 6 to 8 | Set for viewing hours, not growth |
The high light row is the one people get wrong. Ten hours at 70 PAR only works with gas running, a full fertiliser dose, and weekly water changes of 40 to 50 percent.
Set the block so it ends two to three hours before you go to bed. You get evening viewing time and the tank still gets a proper dark period.
Step 3: leave it alone for fourteen days
A planted tank responds to a photoperiod change over roughly two weeks, not two days. Changing the timer every few days means you never learn which change did what.
During those fourteen days keep everything else fixed. Same fertiliser dose, same water change schedule, same feeding, same gas rate if you inject.
Photograph the front glass on day one, day seven, and day fourteen from the same spot. Memory is a poor instrument for judging algae, and phone photos are honest.
Step 4: read the glass, then adjust
The front glass is the fastest reporting surface in the tank. How quickly a film returns after you scrape it is a usable signal.
- Glass stays clear for 10 to 14 days: photoperiod is fine, or slightly short
- Film returns in 7 to 10 days: normal, leave it
- Film returns in 3 to 5 days: cut 30 to 60 minutes
- Green dust within 48 hours: cut to 6 hours and check your nitrate
- Plants leggy and pale with clean glass: add 30 minutes, or raise intensity
Move in 30 minute steps and wait another fortnight between each. Cutting from ten hours to six in one go usually stalls the plants and the algae keeps going anyway.
Algae that survives a shorter day is not a light problem. Once you have cut to six hours and the growth continues, the cause sits in the water column, which is where the algae that traces back to photoperiod stops being the right diagnosis.
Splitting the day in two
A siesta schedule runs the light in two blocks with a gap in the middle, commonly 4 hours on, 3 to 4 hours off, then 4 hours on. Total light stays at eight hours.
The reasoning is that carbon dioxide levels recover during the dark gap in a tank without injection, giving plants a stronger second block. Results are mixed and the effect is easier to demonstrate in low tech tanks than in injected ones.
It also fits people who are out of the house all day. The full case for and against, including what it does to a gas schedule, is in splitting the photoperiod into two blocks.
Ramping instead of switching
Cheap fixtures snap from off to full output in a single step. Fish scatter, and on a tank with skittish species like rummynose tetras you will see them hit the hardscape.
A 15 to 30 minute ramp at each end solves it, and most controllable fixtures now include one. A Fluval Plant 3.0 or a Chihiros unit will do sunrise and sunset curves out of the box.
Ramp time counts toward the photoperiod at roughly half value in my experience, since output sits below useful levels for much of it. Set an eight hour block with 30 minute ramps and call it eight hours, not nine. There is more on curve shapes in how ramping and dimming change the effective day.
Tanks with no plants at all
A fish only tank has no photosynthetic reason to run any light, so the schedule is for your benefit and the fish’s routine. Six to eight hours is plenty.
Longer photoperiods on a bare tank still grow algae on the glass, on the heater, and inside the filter intake. There is nothing consuming the nutrients, so the algae takes all of it.
Nocturnal species change the calculation. Plecos, kuhli loaches, and most catfish will stay hidden through a ten hour day and come out at hour eleven, so a shorter block gives you more to watch.
Timers, and the ways they fail
A mechanical pin timer costs around $8 to $15 and drifts by several minutes a week. It survives power cuts by simply losing the time, which is annoying rather than dangerous.
Digital timers at $15 to $25 hold schedules better but reset to a default programme after a long outage, and I have seen tanks run 24 hours for a week because nobody checked. Smart plugs around $12 to $20 restore correctly but stop working when the wifi does.
After any power cut, check the timer before you check anything else. A fixture stuck on continuously will produce a serious algae bloom in about four days and a green water event in about ten.
One tank, one timer change
A 180 litre (48 gallon) tank came to me with hair algae across the driftwood and a hobbyist convinced the fixture was faulty. It was running a mid range LED at full output for eleven hours a day, no injection, dosing a comprehensive fertiliser twice weekly.
We cut to eight hours and dropped output to 70 percent. Nothing changed for twelve days, then the hair algae started going pale at the tips and the Cryptocoryne pushed three new leaves.
Six weeks later the tank was clean at eight hours flat. No new equipment, no chemicals, one timer change and a slider. That is the usual ending, and it is why I ask about hours before I ask about anything else.
What the hours cannot fix
Photoperiod is one of three inputs and it is the cheapest to change, which is why it gets blamed for everything. Intensity and spectrum still matter, and so does what you are trying to grow.
Plants that want 60 PAR will not thank you for twelve hours at 25 PAR. Total daily light integral is a real concept, but stretched hours at weak intensity mostly feed algae, and the species list in the light demand table for common species is the honest way to check whether your ambitions match your fixture.
Tropica publish a light requirement against each plant they sell at Tropica, which is a reasonable cross reference before buying. If the plant wants more than your tank delivers, no timer setting rescues it.
Start at eight hours if you cannot decide. Then let the glass tell you what to do next, because it always will. The tank in front of you outranks every rule of thumb, including this one, and the details of intensity behind these numbers are set out in the wider picture of PAR, Kelvin, and hours.
Questions readers ask about this
Yes, always. Fish and plants both need a dark period, and continuous light stresses fish and drives algae hard. Aim for at least twelve unbroken dark hours. A fixture left running around the clock after a power cut will produce a visible bloom within roughly four days.
For almost every home tank, yes. Twelve hours only works alongside pressurised carbon dioxide, daily fertiliser dosing, and weekly water changes near 50 percent. Without that support the surplus light goes to algae, since your plants hit a carbon ceiling several hours before the timer switches off.
It counts, and it is usually the hidden cause of stubborn algae. A tank within two metres of a window can receive several hundred lux for hours outside its scheduled block. Move the tank, fit a blind, or subtract an hour or two from the timer to compensate.
Six hours at around 50 percent output for the first four to six weeks. New tanks have no established biofilter and very little plant biomass, so nutrients sit unused in the water. Raise by 30 minutes a fortnight once plants show visible new growth.
Most do. Fish lack eyelids but still run daily rest cycles, and species such as tetras and rasboras visibly settle and lose colour intensity at night. Constant light disrupts that cycle and shows up as poor colour, reduced feeding, and increased aggression in the shoal.
