One lab in Ohio put lettuce into different hydroponic systems three times and got a different winner each time. In the summer of 2023, NFT out-yielded DWC by 22.8%. A year later the same researchers repeated the trial in the fall, and DWC came out ahead on yield. When they swapped mineral salts for an organic fertilizer, both water-based systems lost to an ordinary container filled with growing medium.
That’s bad news for anyone hunting for “the best system”: yield isn’t built into the hardware. Season, solution temperature, and what you feed decide it. So this NFT vs DWC comparison, with Kratky and drip added, looks at what the system actually controls: where the roots live, what happens to them when the power goes out, how much solution each plant needs, and which crops the system can carry.
NFT, DWC, Kratky, and drip: where the roots live
The four systems differ in one thing: how much water and how much air the roots have at any given minute. That shapes yield, how forgiving the system is, and what happens when the pump stops.
- NFT (Nutrient Film Technique). A thin film of solution runs down a sloped channel. The lower part of the root mat sits in it, the upper part breathes air. There is no water reserve: all the solution is in the tank, and the channel holds only what is flowing right now.
- DWC (Deep Water Culture). Roots hang fully submerged in a tank of solution, and an air pump supplies oxygen through an air stone. Plenty of water, and only as much oxygen as has dissolved.
- The Kratky method. The same deep water, but no air pump and no topping up. The level drops, a moist air gap grows above the solution, and part of the root system moves up to live there.
- Drip irrigation. Roots grow in a substrate: coco coir, rockwool, perlite. An emitter delivers solution in doses on a timer, and the substrate holds moisture between irrigations.
A quick crop-and-scale matrix for these four systems already lives in our wiki article on choosing a hydroponic system. Here we go further: into trial data, and into what changes when rolling blackouts are part of the picture, as they are in Ukraine.
NFT vs DWC: lettuce yield depends on the season
The fairest NFT vs DWC comparison we found comes from Teng Yang, Uttara Samarakoon, and James Altland at Ohio State University. They set up both systems in the same climate-controlled greenhouse, four replicates per system, nine Butterhead lettuce plants in each. Same solution, same light; the only difference was the design.
Summer: NFT ahead
The trial ran through July and August, and the results appeared in Acta Horticulturae in 2023. In NFT, leaf area was 13.0% higher, fresh yield 22.8% higher, and dry yield 27.7% higher. Photosynthesis in the NFT lettuce stayed higher throughout the trial. NFT also used 9.6% more water.
There’s a discrepancy worth pointing out. The conclusion of the same abstract talks about a “30% yield increase,” while the results report 22.8% for fresh weight and 27.7% for dry. We go with 22.8%: if you sell lettuce by weight, fresh yield is what counts.
One more detail from that trial. NFT lettuce had 5.2% less chlorophyll and a full 41.0% fewer carotenoids than DWC lettuce. More grams doesn’t mean a better leaf.
Fall: DWC pulls ahead
In 2024 the same three authors published a two-season follow-up in Scientia Horticulturae: fall (October and November) and summer again. In the fall, DWC delivered better photosynthesis, faster growth, and higher fresh yield. The authors point to the water: in the large DWC tank, solution temperature swung less than in the NFT channels.
In summer, NFT again yielded more, but paid for it with tipburn. Tipburn showed up only in summer, and it was worse in NFT. The cause is calcium: NFT lettuce had lower calcium and magnesium in young leaves than DWC lettuce in both seasons. In the fall, DWC also won on quality: 9.4% more vitamin C, 34.6% more carotenoids, 40.6% more phenolic compounds.
The abstract doesn’t give an exact figure for the fall yield, and the publisher didn’t let us reach the full text. So we report only the direction: DWC won in the fall.
What this means in practice
If you grow lettuce in a greenhouse that peaks in summer, NFT will give you more pounds, but expect tipburn and keep an eye on calcium. If your main season runs from fall through spring, DWC with its steadier temperature will yield more and give you a better leaf. Picking a system “for yield” without tying it to your season makes no sense.
Hydroponic systems comparison table
| NFT | DWC | Kratky | Drip | |
|---|---|---|---|---|
| Where the roots are | Solution film plus air | Fully in solution | Solution plus a moist air gap | Substrate |
| What runs on power | Pump, nonstop | Air pump, nonstop | Nothing | Pump on a timer |
| Water reserve at the roots | Almost none | The whole tank | The whole tank for the cycle | Water held in the substrate |
| Crops | Lettuce, greens, spinach, arugula, basil | Lettuce, greens, single large plants | Short-cycle leafy greens | Tomato, pepper, cucumber, strawberry, anything |
| Scale | A few channels up to commercial | Up to a few dozen plants per tank | Windowsill, balcony, small tank | From 10 plants to 2.5 acres and up |
| Main risk | Flow stops | Low oxygen in warm water | Topping up drowns the roots | Clogged emitters, salt buildup in the substrate |
Hydroponics during a power outage: which system survives
In Ukraine this question isn’t theoretical. In a research greenhouse the power stays on; here it doesn’t, and the gap between systems is bigger than any yield difference.
NFT: the most vulnerable
When the pump stops, the film drains out of the channel, and the roots are left in air with no water reserve. Oregon State University Extension puts it plainly in a 2025 guide: power loss or pump failure can lead to immediate plant wilting. The leaves can indeed start to wilt almost right away, while the roots themselves, in our farm’s experience, dry out in the channel in 2-4 hours, faster the warmer it is. Scheduled outages here often run longer, so don’t leave NFT without backup power. Our wiki article on the NFT system covers film mechanics, slope, and flow.
DWC: water stays, oxygen runs down
Without the air pump the roots don’t dry out, since they sit in the tank. But the roots themselves use up the oxygen in the solution, and none comes in to replace it. How fast depends on temperature: warm water holds less oxygen to begin with.
| Water temperature | Maximum dissolved oxygen, mg/L |
|---|---|
| 59 °F (15 °C) | 10.1 |
| 68 °F (20 °C) | 9.1 |
| 77 °F (25 °C) | 8.3 |
| 86 °F (30 °C) | 7.6 |
These are saturation limits for fresh water at sea level; in a salt solution they run slightly lower. That makes a summer outage more dangerous for DWC than a winter one. More on aeration and temperature in our wiki article on the DWC system. Spinach is especially sensitive to warm solution: root rot hits it at 70 °F (21 °C), as we explain in our guide to hydroponic spinach.
Kratky: doesn’t notice the outage
The Kratky method has no component that runs on power. For a plant on a windowsill, the blackout schedule changes nothing, and in our conditions that’s the strongest argument for the method.
Drip: the substrate is the buffer
When the power goes out, the timer simply skips an irrigation. The plant drinks the water already in the substrate. The more substrate per plant and the cooler it is, the longer it holds out. The riskiest case is a small volume of substrate in hot weather: a rockwool cube under a big tomato dries out fast.
How to size backup power
NFT and DWC need backup power for the pump or air pump. Check the wattage on the housing and multiply it by the longest outage on your schedule in hours. A 20 W pump for four hours needs 80 Wh, plus a margin for inverter losses. For NFT a backup is mandatory; for DWC it’s strongly advisable in summer.
Kratky method: how much solution per plant
Bernard Kratky of the University of Hawaii described his method in Acta Horticulturae in 2009. All the solution goes in before planting. After that, nothing is added until harvest: no water, no fertilizer, no electricity.
The numbers from his paper:
- the tank is filled with 4-8 liters (1-2 gallons) of solution per plant;
- lettuce uses 3-6 liters over the cycle;
- the target head is 150-250 grams (5-9 oz), and leaf varieties are ready 6-7 weeks from seeding;
- a kilogram of lettuce takes under 20 liters of water, with a record of 11 liters.
That shows where the method stops working. All the solution for the cycle has to go in up front, and for lettuce that’s a few liters. Tomatoes and cucumbers live for months and drink far more, so a tank for every plant becomes unrealistic. Kratky wins where the cycle is short and the plant is small.
The method’s main trap is topping up. Following the Japanese researcher Imai, Kratky calls the roots above the solution oxygen roots: they breathe air. Raise the level and those roots end up underwater, and the plant suffocates. Our wiki article on the Kratky method explains why, and how to size the volume properly. The other failure modes, from salts in the water to rain, are covered in why Kratky setups fail.
Drip irrigation on substrate: for tomatoes, peppers, and organic feeding
Large long-cycle crops in commercial greenhouses grow on drip. Tomatoes, peppers, and cucumbers have heavy root systems that clog an NFT channel within weeks, and in DWC each plant needs its own big tank. Drip scales on the same logic from a dozen plants to 2.5 acres and beyond, and the substrate smooths out sharp swings in EC and pH. For a worked example with a stage-by-stage nutrient recipe and a pruning trial, see our guide to hydroponic peppers in Dutch buckets.
The price of that flexibility is complexity. Pump, timer, inline filter, irrigation frequency, runoff EC monitoring: every piece can fail. Our wiki article on drip irrigation covers irrigation schedules and salt buildup in the substrate.
Substrate also wins on organic feeding. In 2024 the same Ohio group, this time with Milon Chowdhury, grew lettuce on a liquid organic fertilizer in four systems: NFT, DWC, Dutch buckets, and regular containers with substrate. The containers beat NFT by 29-60% and DWC by 15-44% on head width, leaf number, leaf area, and fresh and dry weight. Roots in the water-based systems were 21-94% smaller. NFT on organic fertilizer grew the least.
If you plan to use organic fertilizer in hydroponics, go with substrate. Organic matter in the solution has to be broken down by microorganisms, and substrate gives them somewhere to live. How that works, and why none of these systems can get organic certification, is covered in our article on hydroponics vs aquaponics vs bioponics.
Which hydroponic system to choose: six situations
- Lettuce and greens on a windowsill, no equipment. Kratky. Zero electricity, zero dependence on the outage schedule.
- First try, a few plants. DWC. Simple mechanics, visible roots, and EC and pH stay stable in a large volume.
- Leafy greens for sale, main season fall to spring. DWC or floating rafts: steady solution temperature, higher fall yield, better leaf.
- Leafy greens for sale, many channels, summer peak. NFT, but with backup power for the pump and calcium monitoring against tipburn.
- Tomato, pepper, cucumber, strawberry. Drip on substrate at any scale. How many berries to expect from one plant at home is covered in our piece on hydroponic strawberries.
- Liquid organic fertilizer. Substrate with drip, or Dutch buckets. Not NFT.
Whichever system you pick, don’t pour solution straight into new hardware. Lubricant and dust left from manufacturing will go right to the roots. The flushing and disinfection sequence is in our wiki article on the first run of a hydroponic system.
Bottom line
- Yield isn’t a property of the system. In one lab, NFT beat DWC in summer (+22.8% fresh weight), lost in the fall, and on organic fertilizer both fell behind substrate.
- In summer NFT gives more pounds, but less calcium in the leaf and worse tipburn.
- In a power outage, NFT without backup is the most vulnerable, DWC holds as long as the tank has oxygen, drip lives on the substrate’s reserve, and Kratky doesn’t notice.
- Kratky needs 4-8 liters of solution per lettuce plant and doesn’t suit long-cycle crops.
- For tomatoes, peppers, and cucumbers, the standard is drip on substrate.
Don’t start with “which system is best.” Start with “when is my season, what am I growing, and how many hours a day am I without power.” The answers usually leave one system standing.
