Sleep Calculator
Find the best times to wake up or go to sleep based on 90-minute sleep cycles. Wake up refreshed, not groggy. Free sleep cycle calculator — no signup.
Sleep Calculator Tool
Planning a nap right now?
This calculator provides general sleep timing guidance. Consult a doctor if you have persistent sleep difficulties.
How to Use the Sleep Calculator
Choose your planning mode
Select "Wake Up At…" if you have a fixed alarm time and want to know when to fall asleep. Select "Sleep At…" if you have a fixed bedtime and want to know the best time to set your alarm.
Enter your target time
Use the time picker to set your wake-up time or bedtime. Results recalculate instantly as you change the time.
Choose a green-highlighted time
The rows marked "Recommended" are 5 and 6 cycles, 7.5 and 9 hours of sleep. They sit inside the 7 to 9 hours that the US National Institute of Neurological Disorders and Stroke (NINDS) says most adults need. Four cycles, 6 hours, is marked "Under 7 hours".
Set your alarm and complete a full cycle
Set your alarm for the time you chose. The calculator counts whole 90-minute cycles so that the alarm falls near the end of one rather than the middle. Ninety minutes is an average, and your own cycles will run a little longer or shorter, so treat the times as a guide rather than a precise schedule.
Sleep Cycles — The Science Behind Waking Up Feeling Refreshed
What a sleep cycle is
Sleep is not one state. NINDS describes two kinds, REM (rapid eye movement) sleep and non-REM sleep, with non-REM divided into three stages. Stage 1 is the changeover from waking and usually lasts several minutes. Stage 2 is light sleep, when heartbeat and breathing slow and body temperature drops; you spend more of the night in stage 2 than in any other stage. Stage 3 is the deep sleep "you need to feel refreshed in the morning", and it comes in longer periods during the first half of the night. REM sleep, when most dreaming happens, first occurs about 90 minutes after falling asleep.
You go through non-REM and REM several times a night, and the REM periods grow longer and deeper later in the night. That is the idea behind this calculator. It counts back, or forward, in blocks of 90 minutes so that your alarm is more likely to land near the end of a cycle than in the deep sleep of stage 3. It also allows 14 minutes to fall asleep. That figure is this tool's assumption, not a measured value for you. If you usually take longer, move your bedtime earlier by the difference.
How much sleep adults need
NINDS says most adults need 7 to 9 hours of sleep a night, "even older people", while school-age children and teens need about 9.5 hours. It adds that there is no magic amount that works for everybody of the same age. The calculator's recommended rows, 7.5 and 9 hours, fall inside the adult range. Six hours falls below it and is labelled that way.
"Even relatively moderate sleep restriction can seriously impair waking neurobehavioral functions in healthy adults."
Van Dongen et al., Sleep 2003Why six hours is not enough
The clearest evidence comes from a laboratory study by Hans Van Dongen and colleagues, published in 2003. Forty-eight healthy adults aged 21 to 38 were allowed 4, 6 or 8 hours in bed a night for 14 days, or kept awake for three nights. Restricting sleep to 6 hours or less produced cognitive deficits equivalent to up to two nights of total sleep deprivation. The participants' own ratings of sleepiness barely changed after the first days, so they were largely unaware of how impaired they had become.
The cost is also economic. A 2016 RAND study of five countries estimated the loss from insufficient sleep at up to US$411 billion a year in the United States and up to US$138 billion in Japan. As a share of the economy, Japan's loss (1.86% to 2.92% of GDP) was the highest of the five. RAND did not study Singapore.
Catching up at the weekend
Sleeping in on Saturday and Sunday helps less than it feels. In a 2019 study in Current Biology, Christopher Depner and colleagues restricted young adults to 5 hours a night for a working week, then let one group sleep as much as they liked at the weekend. They slept about 1.1 hours more than at baseline, but when short nights resumed, their body clocks were delayed, evening eating and weight went up, and insulin sensitivity fell by roughly 9% to 27%. A regular sleep schedule works better than a short week followed by a long weekend.
What NINDS and the sleep-restriction studies found
NINDS says REM sleep first occurs about 90 minutes after falling asleep, the average cycle length this calculator uses.
You spend more of the night in stage 2 non-REM sleep than in any other stage, according to NINDS.
Deep stage 3 sleep comes in longer periods in the first half of the night, and REM periods grow longer later on.
NINDS says most adults need 7 to 9 hours of sleep a night, even older people.
School-age children and teens need about 9.5 hours a night, NINDS says.
In Van Dongen's 2003 study, 14 nights of 6 hours or less produced deficits equal to up to two nights without sleep.
The same participants were largely unaware of their growing deficits: their sleepiness ratings barely moved after the first days.
RAND estimated that insufficient sleep costs the US up to US$411 billion a year, and Japan up to US$138 billion.
In Depner's 2019 study, weekend lie-ins did not prevent the drop in insulin sensitivity when short nights resumed.
This calculator allows 14 minutes to fall asleep. That is its own assumption; adjust your bedtime if you take longer.
Frequently Asked Questions
-
About 90 minutes on average. NINDS says REM sleep first occurs about 90 minutes after falling asleep, and that you go through non-REM and REM several times a night, with REM periods growing longer later on. Real cycles vary from one to the next and from person to person, so the 90 minutes this calculator uses is an average, not a fixed clock.
-
There are several possible reasons. Being woken from deep stage 3 sleep, which NINDS says can be difficult to wake from, is one. Others include a late or irregular bedtime, caffeine or alcohol, or simply less sleep than the hours in bed suggest. Counting in 90-minute cycles gives your alarm a better chance of landing near the end of a cycle, but it cannot guarantee it, because real cycles vary in length.
-
Sleep inertia is the groggy, slowed feeling just after waking, before you are fully alert. It is usually stronger when you wake from deep sleep, which is why this calculator aims your alarm at the end of a 90-minute cycle rather than the middle. It passes as you stay awake.
-
NINDS says most adults need 7 to 9 hours of sleep. In 90-minute cycles, that is five cycles (7.5 hours) or six (9 hours), which are the rows this calculator marks "Recommended". Four cycles is 6 hours, which is below the range; in Van Dongen's 2003 study, two weeks at 6 hours a night produced cognitive deficits comparable to up to two nights without sleep. Three cycles, 4.5 hours, is not enough.
-
REM (rapid eye movement) sleep is the stage in which most dreaming happens. NINDS describes brain activity close to that of waking, faster and irregular breathing, and heart rate and blood pressure rising towards waking levels, while the arm and leg muscles are temporarily paralysed. REM periods grow longer and deeper later in the night, so cutting the end off a night's sleep cuts a larger share of REM than of other stages.
-
The calculator adds 14 minutes between lying down and the first cycle, so that it counts cycles from when you are actually asleep. That figure is the tool's own assumption rather than a measurement. If you know you usually take 30 minutes, move your bedtime about 16 minutes earlier than the one shown; if you fall asleep at once, you can go to bed a little later.
-
Only partly. In a 2019 study in Current Biology, Depner and colleagues had young adults sleep 5 hours a night for a working week, then let one group sleep as long as they liked at the weekend. They slept about 1.1 hours more than at baseline, but when short nights resumed their body clocks were delayed, they ate more after dinner and gained weight, and insulin sensitivity fell by roughly 9% to 27%. A regular schedule works better.
-
We have not found a Singapore figure we could check against its original study, so this page does not quote one. The standard worth measuring yourself against is the one NINDS gives for adults: 7 to 9 hours a night.
-
They do different jobs. A short nap is less likely to reach deep sleep, so it is easier to wake from. A nap of about 90 minutes spans roughly one full cycle, which also makes it easier to wake from than a nap that ends partway into deep sleep. Neither replaces a full night. A long nap late in the day can make it harder to fall asleep at bedtime.
-
For a 6:00 AM wake-up, the recommended bedtimes are: 8:46 PM (6 cycles, 9 hours — ideal for recovery), 10:16 PM (5 cycles, 7.5 hours — recommended), 11:46 PM (4 cycles, 6 hours — under 7 hours), or 1:16 AM (3 cycles, 4.5 hours — not enough). These times include the 14-minute fall-asleep buffer. Aim for the 10:16 PM bedtime for most weekday nights. You can enter 06:00 directly into the calculator above to verify these times.
Method & sources
How it computes
Counts whole 90-minute cycles back from a wake time or forward from a bedtime, allowing 14 minutes to fall asleep: bedtime = wake − n × 90 − 14 min, for n = 3 to 6. The 90 minutes follows NINDS (REM first occurs about 90 minutes after falling asleep); the 14 minutes is this tool's own assumption and the page says so. Five and six cycles (7.5 and 9 h) are marked Recommended against NINDS's 7-9 hours; four is marked Under 7 hours.
What this tool implements
- NINDS: most adults need 7-9 hours, even older people; teens about 9.5; stage 2 takes the largest share; deep sleep comes early; REM lengthens later in the night.
- Van Dongen 2003: 14 days at 6 h or less produced deficits equivalent to up to two nights of total sleep deprivation, with little change in rated sleepiness. The page said 'equivalent to 24 hours without sleep'.
- Removed: a pull-quote and a fact attributing 'S$9.1 billion' of lost productivity to Singapore to RAND. RAND RR1791 studied the US, UK, Japan, Germany and Canada only; it is now quoted for what it says (US up to $411bn, Japan up to $138bn). Also removed: Philips survey figures, 'Japan 6.1 h', a growth-hormone/athlete claim, a caffeine half-life and a '60-minute nap' rule. Depner 2019 is cited for weekend recovery sleep, and only for the metabolic finding it reports.
Sources
- National Institute of Neurological Disorders and Stroke. Brain Basics: Understanding Sleep. https://www.ninds.nih.gov/health-information/public-education/bra…
- Van Dongen HP, Maislin G, Mullington JM, Dinges DF. The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 2003;26(2):117-126. https://doi.org/10.1093/sleep/26.2.117
- Hafner M, Stepanek M, Taylor J, Troxel WM, Van Stolk C. Why sleep matters: the economic costs of insufficient sleep. RAND Corporation; 2016. https://doi.org/10.7249/RR1791
- Depner CM, Melanson EL, Eckel RH, et al. Ad libitum weekend recovery sleep fails to prevent metabolic dysregulation during a repeating pattern of insufficient sleep and weekend recovery sleep. Curr Biol 2019;29(6):957-967. https://doi.org/10.1016/j.cub.2019.01.069
Pick up where you left off
Stored only in this browser — never sent to our servers.