Two rooms, one for vegetative growth and one for flowering, plus a cutting schedule that keeps the flowering room full at all times. Perpetual harvest changes the schedule, not the way you grow. You don’t get one huge harvest every eight to ten weeks. Instead, a few plants go into flower at a fixed interval, and a small harvest comes in every few days or every two weeks. This article gives the general arithmetic so you can size the pipeline for your own cycle times. It also checks a widely quoted example against that arithmetic and covers the running costs that matter more in a tropical climate than in the temperate climates most grow manuals were written for. It is written for licensed small-scale and craft growers in Thailand.
1. The model in brief
- Vegetative / cloning room: propagation equipment and LED lighting, usually on 18/6.
- Flowering room: LED on 12/12.
Older manuals specify fluorescent tubes on 24/24 for veg and HPS or metal halide for flower. Treat that as legacy equipment. It still works, but in a hot climate you pay for its extra heat twice: once to make it and again to remove it (see section 5).
How it works: each time a plant or a small group is harvested, rooted clones from the veg room take its place. The flowering room never drops below capacity. A single room that switches between veg and flower cannot do this, so both rooms have to run at the same time. Topics 1218 (Mother Plants & Cloning) and 1332 (Vegetative Stage) cover cutting technique, rooting and veg-stage care.
Light isolation is not optional. The veg room is lit 18 hours a day, so its lights are on for at least six of the flowering room’s twelve dark hours. Separate the rooms with blackout material, sealed doors or distance. Then stand inside the dark flowering room with the veg lights on and look for leaks. A leak during the dark period can delay or disrupt flowering, trigger re-vegging, and in sensitive cultivars induce intersex flowers. It is the most common way a two-room setup fails.
2. What staggering buys you
Full use of the flowering room. The flowering room has the highest-draw lighting. Once you know your rooting, veg and flowering times, it can run at or near full occupancy all the time. In a batch system it sits partly empty for long stretches.
Smoother labour and better drying. Harvest is the labour bottleneck: cutting, trimming, hanging, drying and curing. If you do a whole room at once, you usually do it badly. In a humid climate this matters more than it sounds. A few plants can be dried properly in a controlled space, but a whole room’s harvest usually cannot (climate control: topic 1046).
Mixed flowering times. Once plants are out of sync, they don’t need the same flowering time. An 8-week cultivar and a 14-week sativa can share the room.
Long-flowering sativas become practical. This matters in a region where landrace and sativa-dominant genetics are part of the heritage. In a batch system, a 14–16-week sativa ties up the whole room for about four months. In a staggered room, shorter cultivars keep finishing around it. A long flowering time then costs you floor space but doesn’t stop your harvests. If you don’t want a full pipeline, try a simpler version: flip half your plants, wait about a month, then flip the other half.
Less water per session. Plants at different stages rarely all need water on the same day, so each session needs only a modest volume. This matters if you use RO water. A small RO unit makes only a few litres per hour and cannot fill a large reservoir on demand, but it can keep up with staggered demand.
The skill this requires: you must be able to judge each plant’s water needs individually, by pot weight rather than by calendar. In a batch crop, watering everything on the same day mostly works. In a rotation it doesn’t. Build that skill before you start.
3. The general formula
This is plain arithmetic, so you can check it yourself.
Let:
- C = clones cut per day (a rate, not a batch size)
- T_veg = days from cutting to entering flower
- T_flower = days in flower
- T_total = T_veg + T_flower = days from cutting to harvest
Then at steady state:
| Population | Size |
|---|---|
| Plants in the vegetative room | C × T_veg |
| Plants in the flowering room | C × T_flower |
| Total plants in the system | C × T_total |
| Harvests per day | C |
You actually need two numbers:
Clones per day required = (flowering room capacity) ÷ (flowering days)
Vegetative room capacity required = clones per day × vegetative days
Worked example for a real room. Say your flowering room holds 24 plants, your flower cycle is 63 days (9 weeks), and it takes 28 days from cutting to flip:
- Clones per day = 24 ÷ 63 = 0.38, so about 3 clones every 8 days, or 1 every 2–3 days.
- Vegetative population = 0.38 × 28 = ~11 plants.
- Harvest rate = 0.38/day, so roughly one plant every 2–3 days, or about 11 plants a month.
Rounding matters. You cannot cut 0.38 of a clone. In practice you cut in small batches, say 3 every 8 days, and accept that the flowering room’s population swings by a plant or two. That is fine. Cutting 24 clones once every 63 days is not fine, because it turns your perpetual system back into a batch system.
Plan a safety margin: cut more clones than the formula says. Many grow manuals advise cutting roughly double what you need. Clones vary, and the fastest-rooting ones are usually the strongest flowering candidates. Once your rooting rate is reliable, a 20–30% over-cut is a reasonable planning figure. Cull the surplus at selection rather than carrying it through the pipeline.
Use your own numbers. Run one batch cycle first and measure your actual times: cutting to root, root to flip, flip to harvest. The formula is useless without them. Then start the cadence below your design rate and let the flowering room fill over one full cycle before you run at capacity.
The cohort version
If you move a group into flower every two weeks instead of cutting continuously, the same logic works in cohorts (groups that enter flower together):
Cohorts in the flowering room = flowering weeks ÷ interval in weeks, rounded up
A 9-week flower at a 2-week interval gives 9 ÷ 2 = 4.5, so plan space for 5 cohorts. A 16-week sativa at the same interval needs 8. Divide the flowering floor space by the number of cohorts to get the footprint per cohort. Long-flowering genetics take up proportionally more floor space, and that cost offsets their benefit.
Checking a published example
A commonly cited grow manual gives a worked two-room example:
- an 8-week flower
- two clones cut every four days
- one mature plant harvested every other day
- about 90 days from cutting to harvest
- a steady state of 30 flowering plants, 10 vegetative plants and 30 clones
Run it through the formula. Two clones every four days is 0.5 per day. In flower, 0.5 × 56 days = 28 plants, close to the quoted 30. Harvests come to 0.5 per day, one plant every other day, which also matches. The flowering side reconciles.
The pre-flower counts run high. From cutting to flip is 90 − 56 = 34 days, so 0.5 per day should give about 17 plants across clones and veg. The example lists 40. The likeliest explanation is the deliberate over-cut. If you take roughly double the cuttings you need (about one a day) and cull at selection, the clone and early-veg stages carry far more plants than ever reach flower. That accounts for much of the gap. So the example is broadly consistent, but its clone and veg counts include that surplus. Size your rooms from your own measured numbers rather than copying them.
4. Practical constraints
Plants must be movable. You will take out a plant every few days and reposition others, so each plant has to move on its own. Pots (soil or coco), drip and ebb-and-flow systems work well, and pots are the simplest. Aeroponics, NFT channels with fixed positions, and any system where roots tangle together make it hard to remove one plant without disturbing its neighbours.
Canopy height. Plants at different stages are at different heights. Stand shorter plants on upturned pots or crates and remove these risers as the plants grow. A set of risers in assorted heights is all the equipment you need. For canopy management in general, see topics 1161 and 1200.
Labels and records. Every plant needs its cut date and flip date on the pot and in your records. After two months you will not remember. On a licensed site, those records also have to match what the licence permits (see the regulatory note).
It’s a permanent commitment. The pipeline only works if it never stops. A missed cutting week shows up as a gap through the whole cutting-to-harvest time, roughly three months in the example above.
5. The costs nobody puts in the model
No crop break, so no pest reset. A perpetual room never empties. Any pest population that gets established has a continuous supply of host plants and can build up over months. In Southeast Asia, mite and mildew pressure is high all year. That makes this the strongest argument against the model, and you should weigh it honestly.
Mitigations:
- Quarantine everything that enters the room, including mother plants.
- Take cuttings from your own quarantined mothers rather than bringing in clones from outside. Outside clones are a common way powdery mildew, spider mites, aphids and thrips get in.
- Deep-clean the room in sections on a fixed schedule.
- Write down a plan for breaking the cycle if you have to.
A perpetual operation with no plan for emptying the room has no plan for a serious infestation. IPM is covered in topic 1301.
Continuous odour, so continuous filtration. A batch room has no odour between crops. A perpetual room always does. In Thailand, odour that reaches neighbours can lead to public-nuisance complaints and is a licence-compliance issue for the site. Plan carbon filtration and negative pressure from the start, and size them for continuous operation, not just the peak weeks of a single crop.
Continuous heat load. With no off period, the room never cools down. In a tropical climate, cooling then runs non-stop, which makes it the largest running cost after lighting. That is the main reason to use LED in a perpetual room here: you don’t have to pay to remove heat your lights never produce. The two-room electricity logic still holds. Your highest-draw fixtures run 12 hours a day over a full room, not over a half-empty one. To find the saving for your setup, calculate it from your own fixture wattages and hours.
Mother plants are a single point of failure. Every clone comes from a mother. A perpetual system uses cutting material continuously, so you need mothers kept in vegetative growth indefinitely. That is one more population to plan space for. Keep at least two of any genetics you cannot replace. If the mothers fail, the whole pipeline fails. Mother care is covered in topic 1218.
6. Living soil in a perpetual room
Soil preparation never stops. Instead of one big batch between crops, you need cured soil ready at every transplant interval. Recycled soil suits this, because root balls and spent soil come out of the room continuously. Here is how many bins you need with a 30-day cure and a 2-week rotation:
- A bin mixed on day 0 is ready on day 30 and gets used at the next transplant slot, day 42.
- So you need 3 bins, started two weeks apart, if each batch is mixed and closed within a day or two.
- If a bin takes the full two weeks to fill with spent soil before its cure starts, plan for 4.
- Two bins offset by two weeks only work if filling plus curing takes 28 days or less.
Top-dressing or core feeding happens at each transplant. That makes it a small routine job every two weeks rather than a big one every two months, the same smoothing benefit at the other end of the cycle.
7. Is it right for you?
It suits you if:
- you want to run long-flowering sativas alongside shorter cultivars
- drying and trimming capacity is your bottleneck
- you can only prepare limited amounts of water at a time
- you already have a separate veg space and reliable rooting
- your licence covers the plant count the pipeline holds
It does not suit you if:
- you can’t yet judge each plant’s water needs reliably
- you can’t run odour control continuously
- your pest pressure is high and your quarantine discipline is weak
- you need the room empty from time to time for any reason
The honest summary: this is a scheduling change with real benefits and one real cost: the room never gets a break. Everything in section 5 follows from that.
Regulatory note (Thailand)
In Thailand, cannabis flower is a controlled herb under the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis) B.E. 2568, issued 26 June 2025.
- Licence and GACP: commercial cultivation needs the relevant licence, and sold or exported flower must come from a GACP-certified cultivation site.
- Plant counts and records: a perpetual operation holds plants continuously by design. The plant counts your model produces, and the records behind them, must stay within what your licence permits. Check this before you build the second room.
- Sales: flower may be sold only through licensed channels and dispensed only against a prescription on form PT 33 from an authorised practitioner.
- Online: advertising or selling flower online is prohibited, and recreational use is not permitted.
- 2026 changes: two Ministerial Regulations B.E. 2569 tightened licensing this year.
- Draft Act: the Cannabis Control Act is still a draft. The Cabinet approved it on 22 September 2026 and sent it to the House.
Cultivation is illegal in most other Southeast Asian countries. Nothing here encourages growing anywhere it is not permitted.
For licensed growers running perpetual rooms in this region: what interval did you settle on, and what did you actually do about the no-crop-break problem? The efficiency arguments are just arithmetic. The pest question decides whether the model works, and grow manuals rarely address it. If you have run both a batch room and a perpetual room through a wet season, that comparison is worth more than the formula.
Related topics on the forum: 1218 (Mother Plants & Cloning), 1332 (Vegetative Stage), 1161 and 1200 (canopy management), 1046 (Climate Control), 1301 (IPM), 1064, and the Advanced Techniques category (244).
This article is technical horticultural information for readers operating legally in their jurisdiction. In Thailand, cannabis flower is a controlled herb under the MOPH Notification Re: Controlled Herbs (Cannabis) B.E. 2568, cultivation requires a licence, and the Cannabis Control Act is still a draft (Cabinet approved 22/09/2026, sent to the House). Cultivation is illegal in most other Southeast Asian countries. Nothing here is legal, financial or business advice.