Data Recovery Case File · Desktop Externals & Aging Drives · Suspect the Supply
Two at Once Points at What They Shared
His enquiry reports a machine and two casualties. A ten-year-old desktop that died, where "an initial investigation indicated that two of the internal drives — the system drive and a storage drive — were faulty. Would you be able to test those drives and see if it is possible to recover any data from them? The drives have been removed." Two drives failing simultaneously in one machine is rarely two independent events — and there is a good chance one or both are less damaged than the investigation suggested, for a reason worth understanding.
| Media | Two internal drives removed from a desktop of approximately ten years — both reported faulty following the machine's failure; assessment sought independently |
| Reported situation | Desktop machine of approximately ten years ceasing to function · initial investigation reporting both an internal system drive and a storage drive as faulty · both drives removed from the machine · independent testing and recovery sought |
| Fault class | Concurrent failure of two drives in one host — shared supply implicated; original diagnosis performed within the failing machine and therefore unreliable |
| Equipment used | Each drive assessed independently on known-good equipment · original diagnosis treated as unverified · current draw measured per drive on a controlled bench supply · board damage assessed before mechanical conclusions · imaging performed per drive on restored operation |
The decode: the shared component, and the unreliable diagnosis
Why one cause is likelier than two: the chance of a particular drive failing in a given month is low. The chance of two failing in the same machine at the same moment is far lower than the chance of something they had in common failing and taking both with it. The question is what they shared, and in a desktop the answer is almost always the power supply.
Why a ten-year-old power supply is the leading suspect: supplies are among the commonest things to fail in an ageing desktop, and they do not always fail cleanly. A supply that degrades can deliver voltage outside specification — which reaches every connected device simultaneously, and can damage drive boards while leaving mechanisms perfectly intact. A machine that "died" and took two drives with it is the classic presentation.
Why that would be relatively good news: board damage leaves the platters untouched, because an electrical event cannot reach magnetic patterns on a surface. Two drives with damaged boards are two component-level repairs and two intact sets of data — a materially better position than two mechanically failed drives.
Now the second point, and it may matter more: the initial investigation was almost certainly performed in the failing machine. A drive tested on a bad power supply behaves badly regardless of its own condition — it may not spin up, may not be detected, or may report errors that reflect the supply rather than the drive. A perfectly healthy drive can be diagnosed as faulty by a machine that is itself the fault, and that is not a criticism of whoever tested it; it is a limitation of testing anything inside a broken system.
So both drives are assessed independently: on known-good equipment, with the original finding treated as unverified. It is entirely possible that one is fine, that both are fine, or that both have board damage requiring repair — and those are very different outcomes that cannot be distinguished from inside the machine.
What must not happen: neither drive should go back into that computer, and the power supply should not be trusted with anything until it is replaced or tested. Connecting a third drive to establish whether the machine works is how a third drive gets damaged.
On the bench
Each drive was assessed independently on known-good equipment and the original diagnosis treated as unverified — a drive tested on a degraded supply behaving badly regardless of its own condition, so a healthy drive can be reported faulty by a machine that is itself the fault. Current draw was measured per drive on a controlled bench supply, and board damage assessed before any mechanical conclusions, since an out-of-specification supply damages boards while leaving mechanisms intact. Imaging was performed per drive on restored operation.
The outcome
Each drive assessed independently with the original finding treated as unverified, draw measured per drive and board damage assessed before mechanical conclusions. 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: two drives failing together in one machine points at what they shared, and in a ten-year-old desktop that is the power supply. A degraded supply damages boards while leaving platters untouched. And a diagnosis made inside the failing machine cannot be trusted, because a healthy drive on a bad supply looks faulty.
Old desktop that died with two drives reported faulty
Don't put either drive back in that machine, and don't connect a third to see whether it works — that's how a third drive gets damaged. Two drives failing at the same moment in one computer is far less likely than something they shared failing and taking both with it, and in an ageing desktop that's almost always the power supply. Supplies don't always fail cleanly: a degrading one can deliver voltage outside specification to everything connected, damaging drive boards while leaving the mechanisms and your data perfectly intact. That's a better outcome than it sounds. It also means the original diagnosis is unreliable, because a healthy drive tested on a bad supply behaves badly regardless.
Have them tested elsewhere — call Guildford Data Recovery on 01483 901310; each assessed independently on known-good equipment, draw measured per drive, board damage assessed before mechanical conclusions.
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.