Data Recovery Case File · Desktop Externals & Aging Drives · The Ambiguous Sound
A Hum Is the Only Noise That Might Mean Nothing
Her enquiry reports a sound and correctly declines to interpret it. A 4TB drive holding almost two years of work: "when plugged in it's making a bit of a humming noise, which we figure isn't good." Every other noise in this archive is a symptom. A hum is the exception — because a healthy drive at full speed hums, and telling that apart from a motor straining against something it cannot turn is a matter of what the sound does rather than what it is.
| Media | 4TB external hard drive — producing a sustained low-frequency sound on connection; character not yet distinguished; approximately two years of business content held |
| Reported situation | Drive failing during the working day · sustained humming reported on connection · sound character not established · approximately two years of work held · contents required |
| Fault class | Undetermined pending acoustic characterisation — normal rotation, motor stall and bearing fault all consistent with the reported description |
| Equipment used | Acoustic character established before any conclusion · rotation confirmed by touch rather than by sound · current draw measured on a controlled bench supply · laminar flow bench inspection before any spin attempt where a stall was indicated |
The decode: three hums, and how to tell them apart
The first — a drive working normally. Platters at several thousand revolutions a minute produce a steady low tone, and a desktop external with a fan produces another. That sound is present on every healthy drive, and people frequently notice it for the first time when they are already worried about something else.
The second — a motor straining. Where heads have come to rest on a surface and adhered, the motor applies torque against a stack that will not turn. That also produces a low sustained sound, and it is one of the more serious faults there is. The distinguishing feature is what happens next: a straining motor does not settle. It labours, and on most designs it gives up after a few seconds and either stops or tries again.
The third — a failing bearing. A spindle bearing that has worn produces a rougher, less even tone, often with a whine or a rasp riding on top, and the character changes as the drive warms.
So the four questions that separate them, and they cost nothing. Does it settle? A healthy drive reaches a steady tone within a couple of seconds and holds it; a straining motor does not. Does it stop? A drive that gives up after a few seconds was not turning. Can you feel rotation? A hand on the case detects the faint vibration of platters at speed — that is the most reliable test available and it requires nothing but touching it. And is it detected at all? A drive that hums normally and is not recognised has a completely different fault from one that never spins.
Why that distinction decides the urgency: if the drive is turning normally, the hum is a red herring and the real fault is elsewhere — electronics, filesystem, or the enclosure. If it is not turning, the mechanism has stopped and every further attempt applies full torque to potentially adhered heads. The same reported sound points to a routine case or an urgent one.
Why she was right to mention it: because it is describable, and describing it precisely is worth more than most diagnostic steps a person can take at home.
What not to do meanwhile: not repeatedly power it to listen more carefully. One attempt, with a hand on the case, answers all four questions.
On the bench
Acoustic character was established before any conclusion — a healthy drive at speed producing a steady low tone, a straining motor producing a similar sound that does not settle and typically ceases within seconds, and a worn bearing producing a rougher tone that changes with warmth. Rotation was confirmed by touch rather than by sound, platter vibration being detectable through the case and the most reliable available test. Current draw was measured on a controlled bench supply, and inspection ran under the laminar flow bench before any spin attempt where a stall was indicated.
The outcome
The sound characterised before any conclusion, rotation confirmed by touch and draw measured on a controlled supply. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode: a hum is the one noise that might be entirely normal, because platters at speed produce exactly that. Four questions separate them — does it settle, does it stop, can you feel rotation through the case, and is the drive detected at all.
Drive making a humming noise
Put a hand on the case and feel for vibration — that's the most reliable test available and it needs nothing but touching it. A hum is the one sound that might mean nothing at all, because platters spinning at several thousand revolutions a minute produce exactly that, and people often notice it for the first time when they're already worried. Three things sound similar: a healthy drive at speed, a motor straining against heads that have stuck to the platter, and a worn bearing. Ask whether it settles into a steady tone within a couple of seconds, whether it gives up after a few seconds, whether you can feel rotation, and whether the drive is detected at all. Do that once, not repeatedly.
Feel for rotation, then call Guildford Data Recovery on 01483 901310; acoustic character established before any conclusion, draw measured on a controlled supply, inspected before any spin attempt where indicated.
Request a quote online →
Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.