Blog — Data Recovery and Cybersecurity
Technical analyses, prevention tips and data recovery news from the experts at SOS Data Recovery, Swiss laboratory since 2006.
Flooded Hard Drive: Why You Must Not Let It Dry Out (and What to Do Instead)
A hard drive that has been in water is not doomed, provided you do not let it dry out as is. It is not the water that kills a hard drive, it is the corrosion that follows: when the water evaporates, the conductive salts and residues remain on the circuit board and keep eating away at the metal. The right steps fit into four lines: never plug the drive back in; remove its circuit board, rinse it with clean water and dry it quickly; never open the drive's own casing; and hand the whole thing over to a lab as fast as possible, because every day of corrosion counts.
What follows explains why these steps, and no others, with a detour through chemistry, two real-life stories and an insurance trap worth knowing about.
When the water rises, it does not pick and choose
In late June 2024, the historic flooding of the Rhône submerged entire neighbourhoods of Sierre and Chippis, in Valais: an industrial site under water, hundreds of homes affected, nearly a thousand people evacuated. In the flooded cellars and offices there were computers. Laptops, desktops, but also servers, NAS units, RAID systems, and magnetic backup tapes.
This is the classic flooding trap: backups are often stored right next to the machines they protect. On paper, the company is covered; in reality, the server and its backups soaked in the same water. And you end up having to recover the data directly from the damaged drives, the very ones you thought you would never have to open again.
Why water destroys: a little science
To understand the right steps, you have to understand what water actually does to electronics. Three mechanisms are at work, of increasing severity.
The short circuit. Water conducts electricity. Not chemically pure water, which is a decent insulator, but real water, charged with ions: mineral salts, limescale, pollutants. Salt water, saturated with sodium and chloride ions, conducts far better still than fresh water. If water bridges two points of a live circuit board, the current takes that shortcut: this is the short circuit. It can burn out the board itself, damage that is repairable in a lab. But it can also send current back to where it should never flow: electrical feedback reaches the read/write heads, components of extreme sensitivity, and destroys them. Hence rule number one: never plug a device that has been in water back in, "just to see".
Oxidation. A metal in the presence of water and oxygen oxidises: this is rust, and its equivalent on the copper, tin and nickel of circuits. Water accelerates this reaction, and salt water accelerates it further. This process is relatively slow, hours, days, but it does not stop on its own: as long as damp residues or salts remain on the board, corrosion progresses.
Electrolytic corrosion, the real killer. Electrolytic corrosion is the accelerated dissolution of a metal carrying an electric current in a conductive liquid: the ion-charged water acts like the electrolyte of a battery, and the metal of the traces is torn away ion by ion. When a still-powered board is bathed in water, the current flowing between two traces triggers electrolysis: on the anode side, the metal of the traces and solder joints is dissolved; these ions migrate into the liquid and redeposit elsewhere, sometimes forming metallic growths (dendrites) that create new short circuits. Unlike simple oxidation, this process is electrically forced: the more voltage and current there is, the faster, more sustained and more destructive the dissolution. A powered-off board corrodes over days; a powered board in salt water can be eaten away in hours.
The worst-case scenario is therefore not "my drive got soaked" but "my drive got soaked while it was plugged in", and the second worst thing is to plug it back in wet.
The cascade of damage
On a flooded hard drive, corrosion attacks in a predictable order.
The circuit board first: it is the exposed part, and it is the most common and most repairable damage. It is cleaned thoroughly or replaced, by transferring its ROM chip (which contains the adaptation data specific to each drive) to a healthy board.
The head connector next, and this is more insidious. The read/write heads are linked to the circuit board by a metal connector whose fine pins pass through the casing. These contacts corrode very quickly. A paradoxical result: a drive whose interior has stayed perfectly dry may still require a head swap in a clean room, solely because corrosion has destroyed the pins of that connection.
The inside of the drive lastly. A hard drive is not hermetically sealed, and a long stay in water eventually lets moisture penetrate, sometimes a few micro-droplets. The platters must then be cleaned before any attempt to start it up. After a very long immersion with significant internal damage, recovery becomes heavily compromised.
The motor, a lesser-known case: its mechanics can corrode and its lubricants can degrade. The solution is then a heavy one: to transfer the platters, in their exact alignment, to an identical donor drive whose motor works.
The right steps, one by one
| ✅ Do immediately | ❌ Never do |
|---|---|
| Unplug, never power it up again | Plug it back in "to see if it works" |
| Remove the circuit board (Torx screws, without touching the cover) | Place the drive on a radiator or let it dry out as is |
| Rinse the board with clean water to flush out salts and mud | Open the drive's cover |
| Dry fast: hair dryer or oil-free compressor | Use an oil compressor (it spits out oil) |
| Dry the head connector on the surface | Wait days before acting |
| Send to the lab as fast as possible | Let the cleanup company sort the media |
In concrete terms, in the hours following the incident, here is what we recommend.
- Do not plug anything back in. Neither the drive nor the computer that contains it.
- Remove the circuit board of the drive (a few Torx screws, without ever touching the cover of the casing).
- Rinse this board with clean water. Yes, with water: the aim is to flush out the salts, the mud and the conductive residues before they carry on their corrosion work. That is the whole paradox: clean water is here your ally against the traces of dirty water.
- Dry it immediately and completely: hair dryer, or oil-free compressor (oil compressors spit out oil droplets at the outlet, which would replace one contaminant with another). Also dry the head connector, on the surface.
- Never open the drive's cover. What happens inside can only be handled in a clean room.
- Send the drive to a lab as fast as possible. There, it will be handled in order: thorough cleaning of the circuit board, check of the board alone, test of the heads at a standstill, opening in a clean room to inspect the interior, and only then an attempt to bring it back into operation.
Why this half-measure of first aid rather than nothing? Because drying a drive as is, without rinsing, sets the salts and mud on the board: the water evaporates, the contaminants remain, and corrosion carries on its work in silence.
Mud, extinguishing powder, chemicals: the aggravated cases
Clean water is the most favourable scenario. Reality is often dirtier.
| Contamination | Severity | Lab treatment |
|---|---|---|
| Clean water (pipe, rain) | Moderate | Board cleaning, checks, clean room inspection |
| Flood water with mud | High | Complete and gradual cleaning of the entire drive before any opening (sometimes several hours per drive) |
| Salt water / sea water | High | Fast rinsing and treatment: maximum conductivity and corrosion |
| Extinguishing powder (fire) | Very high | Very fast rinses: extremely corrosive powder |
| Chemicals (acids, bases) | Critical | Specialised neutralisation and cleaning, uncertain prognosis |
The mud of floods requires a complete and methodical cleaning of the entire drive before any other operation, sometimes several hours of work for a single drive. The reason is simple: opening a still-soiled drive in a clean room would amount to contaminating its entire interior at the very moment of opening.
Extinguishing powder, when water accompanies a fire, is extremely corrosive: it demands very fast rinses, even before classic corrosion has begun.
Chemicals, lastly. We were recently commissioned for two hard drives sprayed with hydrochloric acid, neutralised on site, rightly, with a basic product, but whose residues were still attacking the metals. On the first drive, a complete cleaning of all substances and the repair of the affected parts made it possible to recover the data. On the second, the product had penetrated inside through the openings of the casing: despite the cleaning, a circuit board swap and several head-swap attempts, the drive's firmware was too badly damaged. Recovery failed. Had we been commissioned from the very first day, we could have guided the cleanup team on the steps to take, and those to avoid at all costs.
The time factor: every day counts
Acting fast repairs nothing, but it freezes a state. Every day that passes, corrosion advances, the salts work, moisture migrates. We regularly receive drives more than a week after the flood: on opening, the interior is still damp. That is never a good sign. A drive taken in at day 1 and the same drive taken in at day 10 are two very different cases.
Insurance often pays, but beware who touches your drives
Good news: water, flood and fire damage is typically covered by insurance, and data restoration can be part of the coverage. Contact your insurer as a matter of course.
But there is a trap, and we have seen it with our own eyes. After an incident, insurers commission restoration companies: cleaning, drying, clearing out. These companies are competent for buildings; they are not specialised in data recovery. In a recent case, one of them sorted the storage media itself and threw into the recycling centre everything it deemed unrecoverable, including backups on USB sticks, which would have been easily recoverable. We received the remaining hard drives more than a week and a half after the incident: major internal damage, recovery impossible. The only still-usable copies of that company's data had gone in the bin.
The lesson fits in one sentence: after an incident, bring a specialised lab into the loop from the start. It will not replace the cleanup team; it will tell them which media to set aside, in what condition, within what deadline, and will put in place the restoration procedure suited to each medium.
Frequently asked questions
What should you do immediately with a hard drive that has been in water?
Never plug it back in. Remove its circuit board, rinse it with clean water to flush out the salts and mud, dry it quickly (hair dryer or oil-free compressor), never open the drive's casing, and send it to a specialised lab as fast as possible.
Can you dry out a wet hard drive?
Not as is. As it dries, the water evaporates but the conductive salts and residues remain on the circuit board and keep corroding the metal. You must first rinse the board with clean water, then dry it quickly, and hand the drive over to a lab.
Why should you not plug back in a drive that has been in water?
Because ion-charged water conducts electricity: under power, it causes short circuits and electrolytic corrosion that dissolves the board's traces. Current feedback can also destroy the read/write heads. A powered board in salt water can be eaten away in a few hours.
Is a drive that has stayed under water for several days recoverable?
Sometimes, but the prognosis worsens every day: corrosion progresses and moisture eventually penetrates inside the casing, right down to the platters. We receive drives still damp inside more than a week after the incident. Only a lab analysis can settle the matter.
Does insurance cover data recovery after a flood?
Often, yes: water and flood damage is among the classically covered incidents, and data restoration can be taken on. Contact your insurer, but insist that a specialised lab be involved before the cleanup company sorts or throws away the storage media.
Is sea water worse than fresh water for a hard drive?
Yes, markedly. Salt water is saturated with sodium and chloride ions: it conducts electricity far better, worsens short circuits and strongly accelerates corrosion, including when powered off. It demands even faster rinsing and handling.