Data Recovery Case File · Portable Drives · The Heads Were Flying
Dropped While Running: Why That Answer Matters
Her enquiry answered, without being asked, the question that every dropped-drive page in this archive has to raise and most owners cannot resolve. A Samsung external hard drive: "it got dropped while connected to my laptop, and now when I plug it in it makes a short whirring noise and then stops and doesn't show up. I think it has a mechanical failure. The drive is 1TB and about two-thirds full. I'd like to try to get as much of the data from it as possible." Two things make this case worth its page. She has confirmed the drive was powered and in use at the moment of impact, which is the single most consequential fact after a fall. And her expectation — as much of the data as possible — is exactly the right one, for reasons this page explains.
| Media | Samsung 1TB portable external hard drive, approximately two-thirds full — dropped while connected and running; produces a short whirr then stops; not presented to the host |
| Reported situation | Impact sustained while powered and in use · brief spin attempt followed by silence on each connection · drive not appearing · owner seeking maximum partial recovery rather than a guarantee |
| Fault class | Impact damage sustained with heads flying — head assembly damage likely with possible surface contact; completeness governed by scoring extent |
| Equipment used | No power applied on arrival · laminar flow bench inspection of heads and platter surfaces first · matched donor head service · PC-3000 Express with Data Extractor prioritised head-mapped imaging around damaged zones · losses reported per file |
The decode: why powered matters, and what a short whirr means
What "powered at the time" changes: everything about the expected damage. A drive that is switched off has its heads parked — retracted off the data surfaces onto a ramp at the edge, held clear. In that state a fall transmits shock to the chassis and the heads have nowhere to strike, which is why powered-off drives frequently survive drops with no data damage at all. A drive that is running has its heads flying nanometres above platters turning at thousands of revolutions per minute. An impact at that moment can drive them into the surface. So her fall belongs to the more serious category, and knowing that in advance changes what the inspection expects to find rather than leaving it to be discovered.
What the short whirr then silence means: the drive attempts to start, gets some rotation, and stops. That is consistent with a damaged head assembly: the drive spins up, tries to unpark and calibrate its heads, cannot complete it, and shuts the attempt down rather than continuing. It is a drive protecting itself as much as failing — and it means the mechanism is not seized, which is a small piece of good news within a serious fault.
Why her expectation is the right one: she asks for as much of the data as possible rather than a guarantee, and that framing is correct for this fault class. Where heads have contacted the platters, the magnetic coating is physically removed in those zones, and no technique at any price restores scored tracks. But a head crash rarely destroys an entire surface — damage is usually confined to zones, and material outside them reads normally. So the honest deliverable is not a percentage promised in advance but a map: the drive serviced, imaged with the damaged regions deliberately deferred, and every unreadable area resolved to the specific files it affects. Two-thirds of a terabyte with a bounded damaged zone usually means most of it comes back and the losses are named.
Why she must not connect it again: the instruction that governs the outcome. Each attempt is a possibly-damaged head assembly moving across surfaces holding her data, and if any contact is occurring, it is extending the scored area every second. The number of power-ups after an impact is, on this fault, one of the largest determinants of what comes back — and it is the only variable still under her control.
On the bench
The drive was not powered on arrival — after an impact sustained while running, the surfaces are inspected before anything is switched on, because their condition sets the plan rather than following it. Under the laminar flow bench the head assembly was examined for the damage the fall had caused, and the platters for scoring, with the extent measured against the intact area. Matched donor heads were fitted, and imaging ran on the PC-3000 Express under Data Extractor: head-mapped and prioritised, the healthy expanse captured completely before any damaged zone was approached, all in one supervised session. Every unreadable region was resolved to the individual files it affected, and the recovered set was delivered with those losses named rather than summarised.
The outcome
The head assembly replaced, the readable surface imaged in one supervised session, and the result delivered as a per-file map of what survived. Free assessment, one fixed written figure including VAT, 50% of parts and labour upfront with the balance only on successful recovery. The decode, for anyone who dropped a drive: whether it was powered is the fact that matters most, because parked heads sit clear of the data surfaces and survive falls that flying heads do not — so say which it was, since it changes what the inspection expects; a short whirr then silence means the drive spun, failed to calibrate its heads and stopped, which at least rules out a seized mechanism; asking for as much as possible is the right expectation, because head damage is usually zonal rather than total; and every further connection extends whatever scoring is occurring.
Drive dropped while it was plugged in and working
Stop connecting it, and tell whoever looks at it that it was running at the time — that single fact is the most useful thing you can report. A powered-off drive has its heads parked clear of the platters and often survives a fall intact; a running drive has them flying nanometres above surfaces spinning at thousands of revolutions a minute, and an impact can drive them into the surface. So this belongs to the more serious category and the response should match: no second cable, no other computer, no trying again tomorrow. If any surface contact is occurring, every power-up widens it, and scored tracks cannot be recovered by anyone. Expect an honest deliverable rather than a percentage: head damage is usually confined to zones, so most of the drive typically reads normally and the losses should come back named file by file rather than estimated.
Don't switch it on again — call Glasgow Data Recovery on 0141 404 0294; surfaces inspected before power, heads replaced with matched donor parts, imaged around the damage with losses named per file.
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.