Two Things Broke at Once: A Decision Tree
Why this matters
When two things fail together, your instinct says "bad luck, two repairs." Usually it is not luck. Most simultaneous failures share one root cause: one event or one upstream fault that took out both. Treating them as two separate problems means you fix the symptoms, miss the cause, and watch both come back. The discipline here is to assume they are connected until you prove otherwise, and to spend your first effort finding the single thing that could explain both.
Start here: assume one cause, not two
Two independent parts failing in the same hour is statistically rare. Far more often, one of these is true:
- A shared upstream caused both. One power event, one flood, one overheated component, one contaminated supply line damaged two things downstream.
- One failure caused the other. The first part failed, and its failure overstressed or starved the second. This is a cascade.
- They are not actually simultaneous. One failed a while ago and went unnoticed; the second failure just made the first one visible.
Your opening question to the customer: did both stop at the exact same moment, or did you only just notice both. That single answer often collapses the whole tree.
If both failed at the same instant: look upstream
Simultaneous death points at a shared cause. Find the one thing connected to both.
If both are electrical, look at the shared power path: a lost leg of voltage, a tripped breaker feeding both, a failed transformer that powers two circuits, a surge that hit everything on the panel. Two dead devices on the same circuit is one power problem, not two device problems.
If both involve flow (water, air, refrigerant), look at the shared line or source: a closed valve, a clogged common filter, a failed pump that feeds two paths, a contaminated supply. One restriction can starve two things downstream at once.
If both are mechanical and driven by the same source, look at the common driver: one belt, one motor, one shaft, one coupling. When the driver fails, everything it drives stops together.
The method is the same every time: draw the line backward from each failed part until the two lines meet. Where they meet is your prime suspect.
If one failed first and took out the other: trace the cascade
Cascades are the second big category. One fault creates the condition that kills the next part.
If a protective or supporting part failed and then a downstream part died, the protector was doing its job until it could not. Example patterns: a filter clogs, so a motor overheats; a safety fails, so a component runs past its limit; a lubrication path fails, so a bearing seizes.
If an electrical fault propagated, trace the energy: a shorted component can pull down a power supply and damage everything sharing it; a stuck contact can leave a device energized until it cooks; a failed cooling fan lets a whole assembly overheat.
To untangle a cascade, ask which failure could have caused the other, but not the reverse. That direction tells you which part is the root and which is the casualty. Fix the root, and verify the casualty was only a victim and not also a contributor.
If they only seemed simultaneous: separate them
Sometimes the timing is an illusion. If one part had been failing slowly (a slow leak, a degrading connection, a part running hot for weeks), it may have been broken long before the second, obvious failure drew attention. If a marginal condition existed for a while, a small trigger can reveal two pre-existing weaknesses at once.
Here you genuinely may have two repairs, but they are not coincidence; they are two parts that were both near the edge, exposed by one event. Diagnose each on its own evidence, then ask what common stress aged both (heat, vibration, contamination, age) so the customer understands why both went.
Decision summary: which pattern is it
| Clue | Likely pattern | Where to look first |
|---|---|---|
| Both died the same instant | Shared upstream cause | The common power, flow, or drive path |
| One failed, then the other | Cascade | The part that could cause the other, not the reverse |
| Both "just noticed," ages unclear | Two pre-existing weaknesses | Each on its own, plus the shared aging stress |
| Same circuit or same line | Shared upstream cause | The single supply they both depend on |
| Protector dead plus downstream dead | Cascade | The protector, as the root |
Confirm before you call it
Whatever pattern you land on, prove the link:
- Identify the single root that explains both, and verify it with a measurement, not a hunch.
- Confirm direction in a cascade: which is cause, which is casualty.
- Fix the root, then re-test both. If fixing one and curing the cause restores both, the link was real. If the second part stays dead, it has its own fault and needs its own diagnosis.
Two failures are a gift in disguise: they triangulate the root cause better than one failure ever could.
References
- Trade-standard root-cause-analysis and cascade-failure diagnostic practice
- NFPA 70 (National Electrical Code) for shared-circuit and protection principles
- Manufacturer service documentation for the specific equipment class
- See related: The Customer Already Replaced a Part: A Decision Tree; The Fault After a Power Blip: A Decision Tree