The failure log: every hydroponic problem encountered in 6 months and what fixed it

Key takeaways

In 6 months of logged hydroponic failures across NFT, DWC, and kratky systems in a Bangkok apartment, reservoir temperature above 28 degrees Celsius was the single most common root cause -- present in 14 of 23 documented failure events either as a primary or contributing factor (Plant Mood grow log data, 2024).

Tip burn in hydroponic lettuce is caused by calcium transport failure at the leaf margin, not calcium deficiency in the reservoir -- raising EC to fix tip burn makes the problem worse in approximately 70 percent of cases because it increases osmotic load without improving calcium mobility.

pH drift above 7.2 in a DWC reservoir during Bangkok's hot season (March to May) accelerates to a rate that requires twice-daily correction rather than daily -- growers who check pH once per day lose crops to nutrient lockout that a sensor or a second daily check would have caught at the tipping point.

Root browning in the first 72 hours of a DWC cycle is almost always dissolved oxygen depletion from warm water, not pythium -- the fix is reservoir temperature reduction and aeration increase, not a hydrogen peroxide treatment that strips beneficial microorganisms and does not address the underlying cause.

Kratky method fails reliably above 30 degrees Celsius ambient in Bangkok because the static reservoir reaches temperatures that collapse dissolved oxygen faster than the plant can compensate -- this is not a user error, it is a physics problem that no passive system modification fixes.

TL;DR

Most hydroponic failures in a Bangkok apartment trace back to heat, and most of the standard fixes for those failures treat the symptom rather than the cause. Six months of logged failures across three system types produced a clear pattern: the growers who fix root rot with hydrogen peroxide, tip burn with calcium supplements, and germination failure with humidity adjustment are solving the wrong problem. The underlying cause in almost every case was reservoir temperature, dissolved oxygen, or osmotic stress from EC set too high for Bangkok conditions. The fix is almost always environmental, not chemical.

Month two, three system failures in eight days

The DWC lettuce crop started showing root browning on day 4. The NFT basil had tip burn on the third leaf pair by day 6. The kratky spinach had stalled entirely -- leaves present, no new growth, roots sitting in a reservoir that smelled faintly wrong. Three different crops, three different systems, eight days. I had been running the dashboard long enough to pull the sensor logs and look at what had been happening in the room during those eight days. Ambient temperature had averaged 33.4 degrees Celsius between noon and 6pm. Reservoir temperature in the DWC had peaked at 31 degrees Celsius twice. The kratky jar reservoir was uninsulated and sitting in direct afternoon light for approximately two hours per day.

The three failures looked different at tray level. The logs showed they were the same failure. Heat above a threshold was collapsing dissolved oxygen, slowing calcium transport, and in the static kratky system, creating a reservoir chemistry that the plant simply could not function in. I had been treating each symptom separately -- hydrogen peroxide for the roots, cal-mag top-up for the tip burn, nutrient refresh for the kratky. None of it worked because none of it was aimed at 33 degrees Celsius. That week changed how I read every failure that came after it.

What the research says about heat, oxygen, and hydroponic failure

Dissolved oxygen solubility data, standard aquatic chemistry tables

Water at 20 degrees Celsius holds approximately 9.1 mg per liter of dissolved oxygen at atmospheric pressure. At 30 degrees Celsius that figure drops to approximately 7.6 mg per liter -- a reduction of roughly 16 percent. At 35 degrees Celsius it falls to approximately 7.0 mg per liter. Root respiration demand does not decrease proportionally with temperature -- it increases. The result is that a hydroponic root zone in Bangkok conditions during hot season is operating under oxygen stress that has no equivalent in the temperate-climate systems that most grow guides are written for.

A Bangkok DWC grower running reservoir temperatures above 29 degrees Celsius is not running the same system as a European grower following the same guide at 20 degrees Celsius reservoir temperature. The oxygen budget is categorically different. Aeration that is adequate at 22 degrees Celsius is inadequate at 30 degrees Celsius for the same root mass. This is not a matter of degree -- it crosses a functional threshold.

Calcium mobility research in hydroponic lettuce, multiple university extension sources

Calcium moves through plant tissue via the transpiration stream -- it is pulled into the leaf margins by water movement driven by evaporation from the leaf surface. In high-humidity environments, transpiration rate drops because the vapor pressure differential between the leaf interior and the surrounding air is reduced. This is independent of calcium concentration in the reservoir. A plant in 85 percent relative humidity transpires less than the same plant at 60 percent relative humidity, and therefore delivers less calcium to the leaf margins regardless of EC or nutrient formulation.

Tip burn in Bangkok is a humidity and airflow problem before it is a calcium problem. A grower who responds to tip burn by increasing cal-mag concentration is adding calcium to a transport system that is already constrained by low transpiration. The calcium is present in the reservoir. The problem is that it cannot reach the leaf margin. Improving airflow across the canopy increases transpiration rate and calcium delivery without any change to nutrient formulation.

Pythium and Fusarium temperature range data, plant pathology literature

Pythium species most associated with hydroponic root rot -- including Pythium aphanidermatum, which is the primary pathogen in warm-climate systems -- have optimal growth temperatures between 32 and 36 degrees Celsius. Below 20 degrees Celsius, Pythium growth rate drops substantially. The implication is that a Bangkok reservoir at 30 to 32 degrees Celsius is not merely warm enough for Pythium -- it is near the optimal temperature range for the specific pathogen most likely to be present in a tropical hydroponic environment.

Treating Pythium with hydrogen peroxide after it is established addresses the existing infection but does not change the reservoir temperature that enabled it. In a Bangkok system that returns to 30 degrees Celsius within hours of treatment, Pythium reestablishes faster than beneficial organisms recover from the peroxide dose. Reservoir temperature reduction is the only sustainable Pythium management strategy in a tropical climate.

Six months of failures by the numbers

Failure events across NFT, DWC, and kratky systems in a Bangkok apartment grow operation over a 6-month logging period where reservoir temperature above 28 degrees Celsius was identified as a primary or contributing factor. The remaining 9 failures had distinct causes including pH management errors, light spectrum mismatch, and mechanical system failures (Plant Mood grow log data, 2024).

31 degrees Celsius

Peak DWC reservoir temperature recorded during the worst failure cluster in the logged period, corresponding to an ambient room temperature of 33.4 degrees Celsius averaged over a 6-hour afternoon window. At this reservoir temperature, dissolved oxygen availability and Pythium growth rate both cross thresholds that produce visible root damage within 48 to 72 hours (Plant Mood sensor log data, 2024).

Failure events that were initially misdiagnosed -- where the first corrective action taken addressed a symptom rather than the root cause and did not resolve the failure. In 8 of these 11 cases the misdiagnosis involved a chemical or nutrient intervention applied to a problem that was fundamentally environmental (Plant Mood grow log data, 2024).

Median time between the first logged indicator of a developing failure -- a pH reading outside the target range, a root color change, a growth rate drop -- and the point at which the crop was no longer recoverable, across all 23 documented failure events. In 17 of 23 cases the first indicator was present in the log data before any visible symptom was apparent at tray level (Plant Mood grow log data, 2024).

The 23 failures, grouped by root cause

The largest single failure category across the 6-month period. All 9 events occurred during periods when ambient temperature exceeded 32 degrees Celsius for more than 4 consecutive hours. Reservoir temperatures in DWC systems reached 29 to 31 degrees Celsius. Root browning appeared at the tips within 48 to 72 hours of the temperature exceedance. In 6 of 9 events the initial response was hydrogen peroxide treatment at 3 mL per liter, which produced temporary improvement followed by recurrence within 5 to 7 days. In the 3 events where reservoir temperature was reduced first -- using frozen water bottles to bring the reservoir to 24 to 25 degrees Celsius and adding a second air stone -- recovery was complete within 4 to 5 days without chemical intervention. The fix that worked was always the temperature reduction. The fix that did not work was always the hydrogen peroxide.

Heat-driven dissolved oxygen collapse -- 9 failure events

Any Bangkok DWC grower who has treated root rot more than once with hydrogen peroxide and seen it return is almost certainly managing a temperature problem with a chemical solution. The recurrence rate is the diagnostic. Pythium treated at source temperature does not recur at the same rate as Pythium treated with peroxide at optimal Pythium growth temperature.

All 5 events occurred in DWC systems during March to May. The pattern in the sensor logs was consistent: pH would be within range at a morning check, drift upward through the day as plants consumed nutrients and CO2 outgassing from the warm reservoir raised pH, and be measurably outside the effective nutrient uptake range by the following morning check. In 3 of 5 events, by the time visible symptoms appeared -- interveinal chlorosis on new growth, stunted leaf expansion -- the pH had been outside the target range for 36 to 48 hours. The fix was twice-daily pH checks during hot season and a phosphoric acid top-up at 6pm each day as a preventive adjustment rather than a reactive correction. Buffering capacity of the solution also dropped as temperatures rose and CO2 solubility decreased, meaning the same pH down dose that held pH stable in February was insufficient in April.

pH drift leading to nutrient lockout -- 5 failure events

A once-daily pH check protocol that is adequate during Bangkok's cooler months (November to February) will miss the drift window during hot season. The check frequency needs to be seasonal, not fixed.

All 4 tip burn events occurred during periods of sustained ambient humidity above 82 percent. Reservoir calcium levels were measured in all 4 cases and were within normal range. Initial response in 2 of 4 events was a cal-mag supplement increase, which produced no improvement over 4 days. In both of those events, adding a fan directed across the canopy at low speed resolved the tip burn within 3 days without any change to nutrient formulation. The 2 events where airflow was the first intervention resolved faster and without any chemical change. The sensor logs showed that ambient humidity dropped by 6 to 9 percent in the immediate canopy zone when the fan was running, which was sufficient to restore enough transpiration to deliver calcium to the leaf margins.

Tip burn from low transpiration in high humidity -- 4 failure events

Tip burn in Bangkok is a ventilation problem. The diagnosis test is simple: if tip burn is present and reservoir calcium is in range, add airflow before adding cal-mag. If airflow resolves it, the problem was never calcium deficiency.

All 3 kratky failures occurred in jars or containers with no insulation, positioned where afternoon light reached the reservoir. Reservoir temperatures reached 31 to 33 degrees Celsius. Plants stalled completely -- no visible new growth, roots present but inactive, no visible pathogen. In all 3 cases the system was not recoverable after 4 days at these temperatures. No fix was found that salvaged the crop. The only preventive measure that worked across subsequent cycles was reservoir insulation (wrapping jars in reflective foam) and repositioning to eliminate afternoon direct light exposure. Kratky as a method does not have a reactive fix for reservoir overheating -- the thermal mass of the static reservoir is too small to absorb Bangkok afternoon heat passively.

Kratky method failure from static reservoir heat -- 3 failure events

Kratky is not a viable warm-season system in a Bangkok apartment without active reservoir temperature management or full insulation. This is a structural limitation of the method, not a management failure. Any grow guide recommending kratky for tropical apartment growing without addressing reservoir temperature is incomplete.

Both events followed a period of high water uptake in hot conditions where the reservoir EC rose as plants consumed water faster than nutrients. The grower response in both cases was to top up the reservoir with fresh nutrient solution at the original EC target rather than plain water, which pushed the reservoir EC further above the target range. In one event reservoir EC reached 3.1 in a lettuce DWC system where the target was 1.4 to 1.6. Leaf margins showed burn within 48 hours. The fix was a 50 percent reservoir flush with plain pH-adjusted water, which brought EC back to range within 24 hours. Both events were preventable by topping up with plain water first when EC rises during hot periods rather than nutrient solution.

EC-related osmotic stress from overcorrection -- 2 failure events

When a Bangkok hydroponic reservoir is running hot and plants are drinking heavily, EC rises because water is being consumed faster than nutrients. The correction is plain water, not nutrient solution. This is the opposite of what a grower trained to top up with nutrient solution will do by default.

Plant Mood grow log and sensor data, January to June 2024

What the failure logs revealed that the grow guides did not

In 17 of 23 documented failure events, a measurable signal was present in the sensor or log data at least 24 hours before any visible symptom appeared at tray level. pH trending upward across three consecutive readings. EC rising faster than the top-up schedule accounts for. Reservoir temperature logging a new daily high for the third day running. These are leading indicators, not lagging symptoms. A grower who only checks the plant is always 24 to 48 hours behind the failure. A grower who reads the trend in the data has a recovery window. The single most impactful change in the second half of the logging period was learning to read the trend rather than the point reading -- not whether today's pH is in range, but whether it is higher than yesterday and the day before.

The first indicator is always in the data before it is visible in the plant

In 11 of the 23 failure events, the first corrective action was a chemical or nutrient intervention -- hydrogen peroxide, cal-mag, pH buffer, nutrient flush. In 8 of those 11 cases the intervention did not resolve the failure and added 3 to 5 days to the failure timeline before the environmental cause was identified and addressed. The pattern is consistent enough to function as a diagnostic rule: if a chemical intervention does not produce visible improvement within 48 hours, the root cause is environmental, not chemical. The corollary is that in a Bangkok grow operation, environmental should be the first hypothesis, not the last resort.

Chemical interventions applied to environmental problems extend the failure timeline

The 6-month log period covered Bangkok's transition from cool season (January to February) into hot season (March to May) and the beginning of rainy season (June). Failure rate was 1 to 2 events per month in January and February. It rose to 4 to 5 events per month in March and April. A grower who sets up their systems in November, runs flawlessly through February, and then encounters multiple simultaneous failures in March is not doing anything differently -- the environment changed. The failure cluster in March and April is not a sign that the grower made a mistake. It is a sign that the systems were designed for Bangkok's cool season temperatures and were not adapted before hot season arrived. Planning a system adjustment in February -- insulating reservoirs, adding aeration, reducing EC targets by 10 to 15 percent -- would prevent the majority of hot season failures before they occur.

Failure clustering by season is predictable and almost nobody plans for it

The median time between the first logged indicator and the point of no return was 72 hours across all 23 events. In the failure events where corrective action was taken within 24 hours of the first indicator, recovery rate was approximately 70 percent. In events where corrective action was taken after 48 hours, recovery rate dropped to under 30 percent. After 72 hours from the first indicator, no crop in the 6-month log period was recovered successfully. The implication for time allocation is direct: 20 minutes spent checking sensor trends each morning is worth more than two hours of corrective intervention after visible symptoms appear. The recovery window is narrow and closes predictably.

Recovery attempts on crops past the 72-hour window rarely succeed

The failure explanations that turned out to be wrong

I treated the root rot with hydrogen peroxide and it came back within a week.

Hydrogen peroxide kills the surface pathogen but does not change the reservoir temperature that created the conditions for Pythium growth. In a Bangkok system returning to 29 to 31 degrees Celsius within hours of treatment, Pythium re-establishes from any surviving spores faster than the root system recovers. The peroxide dose also strips whatever beneficial microbial community existed in the reservoir, removing a competitive layer that would otherwise suppress Pythium recolonization. The fix for recurring root rot is reservoir temperature reduction to below 24 degrees Celsius. Everything else is symptom management.

My pH keeps drifting up no matter how much pH down I add.

Warm reservoir water loses CO2 faster than cool water, and as dissolved CO2 decreases, pH rises. In a Bangkok DWC system at 30 degrees Celsius, the buffering capacity of the solution is lower than at 22 degrees Celsius, meaning the pH down dose that held pH stable in cooler conditions is insufficient. The reservoir is also being supplemented with water that has its own pH, and if that top-up water is not pH-adjusted first, each addition pushes the reservoir toward the water's natural pH. Sustainable pH management in Bangkok hot season requires reservoir temperature reduction, pH-adjusted top-up water, and twice-daily checks -- not larger pH down doses.

I added more cal-mag for the tip burn and it made no difference.

Tip burn in high humidity is a calcium transport failure, not a calcium availability failure. The reservoir contains adequate calcium -- the problem is that low transpiration in humid conditions means calcium is not being pulled to the leaf margins at the rate the new growth requires. Adding cal-mag increases the concentration of a nutrient the plant already cannot move efficiently. The fix is airflow across the canopy, which increases transpiration rate and calcium delivery. In all 4 tip burn events in the 6-month log, cal-mag supplementation produced no improvement and airflow addition resolved the symptom within 3 days.

My kratky system worked perfectly in December and completely failed in April.

Kratky works in Bangkok cool season because ambient temperatures keep reservoir temperatures in a range where dissolved oxygen is adequate and Pythium growth is slow. In April, the same system with the same setup has a reservoir temperature 8 to 10 degrees Celsius higher, dissolved oxygen below the threshold for healthy root function, and Pythium at near-optimal growth temperature. The system did not fail because of a management error -- it failed because the method is not thermally resilient in Bangkok hot season. The grower who succeeds with kratky in December and fails in April has not changed anything. The season changed the physics.

I refreshed the nutrient solution completely and the plants still did not recover.

A full nutrient refresh corrects EC and pH but does not address root zone damage that has already occurred. Roots that have experienced 48 to 72 hours of dissolved oxygen depletion or Pythium infection have structural damage to the root hairs responsible for nutrient uptake. A fresh reservoir presented to a damaged root system recovers slowly or not at all because the uptake mechanism is compromised regardless of solution quality. The refresh is the right action but it needs to be accompanied by root zone inspection and trimming of any visibly dead root material, reservoir temperature management to create recovery conditions, and an expectation that visible improvement will take 4 to 7 days if it comes at all.

The fix that makes the most Bangkok hydroponic failures worse

In a Bangkok tropical apartment, the majority of crop failures have environmental causes -- reservoir temperature, dissolved oxygen, humidity, airflow -- that chemical interventions cannot fix and often worsen. A hydrogen peroxide treatment applied to a warm reservoir does not cool the reservoir. A cal-mag boost applied in 85 percent humidity does not improve calcium transport. A nutrient flush followed by a strong fresh mix applied to heat-stressed roots increases osmotic load on a root system that is already struggling to function. Each of these interventions is the correct response to the symptom as described in a temperate-climate grow guide. Each of them is the wrong response to the actual cause in Bangkok conditions. The pattern across 11 misdiagnosed failures in the 6-month log was identical: chemical intervention applied to an environmental problem, no improvement within 48 hours, environmental cause identified and addressed 3 to 5 days later than it would have been if it had been the first hypothesis.

The diagnostic rule that emerged from the failure log is this: if a chemical or nutrient intervention does not produce a measurable improvement -- not visible recovery, just a measurable improvement in the trend -- within 48 hours, stop the chemical intervention and check the environmental variables first. Reservoir temperature, dissolved oxygen proxy (air stone function, aeration rate), ambient humidity, and airflow across the canopy. In Bangkok conditions, the environmental variable is the cause in more than half of all failure events. Treating the environment first and the chemistry second would have resolved 8 of the 11 misdiagnosed failures faster and with less crop loss.

When something goes wrong in a hydroponic system -- root browning, tip burn, slow growth, nutrient deficiency symptoms -- the instinct is to adjust the chemistry. Add a supplement, change the EC, treat with a solution, flush and start fresh with a stronger mix. This feels like active problem solving. The nutrient solution is the one variable the grower controls directly and visibly, so it becomes the default intervention regardless of what the actual cause is.

What the failure log changed about how the systems are managed

Any Bangkok hydroponic grower running DWC or NFT systems year-round

Set a seasonal system adjustment date in February, before Bangkok hot season arrives. Insulate reservoirs with reflective foam or wrap. Add a second air stone to every DWC reservoir that ran with one stone through cool season. Reduce EC targets by 10 to 15 percent across all systems. Move any kratky containers away from afternoon light exposure and wrap them. These adjustments take two to three hours and prevent the majority of the failure events that will otherwise cluster in March and April. The cost of not doing them is paid in crops, not preparation time.

A hot season failure rate comparable to cool season, rather than the 3 to 4 times higher failure rate that unmodified systems produce when ambient temperatures cross 32 degrees Celsius.

Growers who check pH and EC once per day

Add a second pH check at 6pm during Bangkok hot season (March through May). The afternoon check catches the pH drift that accumulates during the hottest part of the day and allows a small phosphoric acid adjustment before overnight -- which is cheaper and more effective than a large correction the following morning after another 12 hours of drift. The 6pm check takes four minutes. The crop loss it prevents in a single hot season event is worth more than the cumulative time cost of the checks across the entire season.

pH staying within the effective nutrient uptake range throughout the day rather than drifting outside it during afternoon hours and causing nutrient lockout that is not caught until morning symptoms appear.

Growers who have treated root rot with hydrogen peroxide more than once

Measure reservoir temperature at the point of root contact before the next treatment. If the temperature is above 26 degrees Celsius, address the temperature first -- frozen water bottles, reservoir insulation, repositioning away from heat sources -- and do not apply hydrogen peroxide until the reservoir is at or below 25 degrees Celsius. The peroxide treatment without temperature reduction is a two-week delay, not a fix. The temperature reduction alone, without peroxide, will resolve early-stage root browning in 4 to 5 days in most cases.

Root rot that does not recur within 7 days, because the treatment addressed the cause rather than the active infection at optimal reinfection temperature.

Growers diagnosing tip burn on hydroponic lettuce or basil

Before adjusting any nutrient parameter, measure ambient humidity at canopy level and check whether a fan is directing airflow across the leaf surface. If humidity is above 75 percent and there is no airflow across the canopy, run a fan at low speed across the plants for 48 hours before drawing any conclusion about calcium levels. If tip burn improves with airflow alone, the cause was transpiration rate, not calcium availability. Only if tip burn persists after 48 hours of improved airflow should EC or cal-mag be evaluated.

Correct diagnosis on the first intervention in the majority of Bangkok tip burn cases, avoiding 4 to 7 days of ineffective cal-mag supplementation before the actual cause is addressed.

Growers reading daily sensor logs or manual measurements

Read your measurements as trends, not points. A pH of 6.4 today is a normal reading. A pH of 6.4 today after 6.2 yesterday and 6.0 the day before is a trend that reaches 7.0 within 4 days at the same drift rate. An EC of 1.8 today is normal. An EC of 1.8 after 1.6 and 1.4 on the previous two days means the reservoir is concentrating and a plain water top-up is needed now, not when EC hits 2.2. The recovery window for every failure in the 6-month log was open in the trend data before it closed at tray level. Reading the trend takes the same time as reading the point. The outcome is categorically different.

Corrective actions taken 24 to 48 hours earlier than a point-reading protocol produces, which based on the 6-month failure log corresponds to a recovery rate improvement from under 30 percent to approximately 70 percent.

Checking reservoir temperature before diagnosing any other variable. Every experienced Bangkok hydroponic grower knows that heat causes problems. Almost none of them check the reservoir temperature as the first diagnostic step when something goes wrong -- they check pH, EC, root appearance, leaf symptoms. Reservoir temperature takes 10 seconds to measure with a basic aquarium thermometer and is the primary cause or a contributing cause in more than 60 percent of Bangkok hydroponic failures. It should be the first measurement taken when anything looks wrong, not the last one considered after the chemical interventions have not worked.