The Scenario
Consider a civil engineer working as a consultant in Kano who has spent three years building a documentation library for her practice — AutoCAD drawings, structural calculation spreadsheets, project photographs, client correspondence, and the accumulated institutional knowledge of a growing consulting practice. All of it on a single desktop PC in a home office. None of it backed up anywhere else.
She knows she should back up. She means to back up. There is a portable hard drive next to the machine, used occasionally, when remembered. Then a surge arrives overnight: a dead PC, a burning smell in the room, and a blackened patch on the power strip where it entered.
The right first move, when that call comes, is counterintuitive: bring the machine in before touching anything. Do not try to turn it back on.
The Challenge
A power surge damage assessment is a triage exercise. When high-voltage power enters a PC through an unprotected outlet, it typically travels through the power supply first — the PSU takes the brunt and often dies. But depending on the surge magnitude and path, it can also damage the motherboard, storage devices, and GPU. The hard drive — where the data lives — may be intact, partially damaged, or completely destroyed, depending on exactly where the surge energy dissipated.
Disassembly and assessment come first. A typical pattern in a case like this: the power supply completely dead with visibly burnt internal components; the motherboard damaged on the VRM circuit near the power input, needing replacement; the GPU physically intact; and the HDD — the critical component — showing no visible damage and no burn marks on the PCB.
The question is whether the platters and read heads inside the drive are intact. Connected as a secondary device to a protected diagnostic machine, a surviving drive spins up, is recognised, and shows an intact partition table and readable folder structure.
The Recovery Protocol
A full image copy to a new drive comes before any attempt to access individual files. This is standard data recovery protocol and it is not optional: reading files directly from a potentially damaged drive risks further damage if the drive has bad sectors. Imaging the entire drive first creates a safe working copy.
Imaging a drive of this size takes several hours. Sector read errors clustered in the Windows system files rather than the data partition are the good outcome — the operating system is replaceable, the project archive is not. Once the image completes, it gets mounted and the project folders checked.
The Rebuild
Recovered data moves to a new drive immediately, and a copy onto a USB drive means the owner has her files back within days rather than waiting for a full machine rebuild. That sequencing matters more than it sounds — the data is the emergency, the machine is not.
The rebuild itself is a decision point. Rather than simply replacing failed components with equivalent parts, a full modern rebuild is usually the better use of the opportunity — a five-year-old machine on a motherboard the manufacturer no longer supports is not worth repairing to its original specification.
Rebuilt workstation — indicative ₦2.8 million:
- CPU: Intel Core i7-13700 — substantial upgrade from the original i5-8500
- RAM: 32GB DDR5
- GPU: NVIDIA RTX 3060
- Storage: 1TB NVMe (OS) + 4TB HDD (project files, mounted in rubber vibration-dampened tray)
- UPS: APC Smart-UPS 1500VA — pure sine wave, AVR, network-connected for shutdown management
- Surge protection: Legrand surge arrester installed at the wall socket, in addition to the UPS
Backblaze Personal Backup belongs in the rebuild — project data backing up to cloud storage nightly, automatically, without anyone needing to remember. That produces three copies of every project file: local NVMe, local HDD, and cloud.
What Changes
The rebuilt machine is faster than the original by a wide margin, because five years of CPU and storage progress separate them — AutoCAD loading from NVMe rather than a mechanical drive is the most visible part of that.
The more important change is that the backup infrastructure now runs without the owner's active involvement. A portable drive used "when remembered" is not a backup strategy; a nightly automated cloud backup is. Checking a Backblaze dashboard once a week is like checking the front door is locked — you just need to know it's there.
And the UPS with a wall-mounted surge arrester means Kano's variable power environment stops being a threat. The event that destroyed the original machine cannot reach this one the same way.
Key Takeaway
Data recovery from surge damage is possible — but it is not guaranteed, and the outcome depends heavily on how quickly you stop trying to use the damaged machine. Every attempt to boot a drive with potential platter damage risks further physical damage. The correct response to a surge event is: power off everything, do not attempt to turn the machine back on, and get it to a professional immediately. Speed matters. Panic does not.
The more important lesson is the one before the crisis: there is no excuse for a single-copy backup strategy in 2026. Cloud backup services cost less per month than a cup of coffee. Losing three years of work to a power surge is a preventable tragedy, and the prevention costs almost nothing.
Worried about your data protection or power resilience? Talk to our team before the problem happens. We design backup and power protection into every build.