The Substituted Part Also Fails (Decision Tree)

Why this matters

There are three outcomes you plan for when you run a substitution test: the fault clears, the fault persists, or the result is ambiguous. There is a fourth outcome that catches people off guard because it does not fit that framework at all: the known-good part you just installed also fails, sometimes immediately, sometimes within minutes of energizing it. This is not the same as "the fault persisted." It means the component you swapped in, one you personally confirmed was working, is now also bad. That points somewhere specific, and it is usually more serious than the original complaint. Before you reach for a second substitute, stop and think about what just happened.

Safety first: stop and de-energize

If the substituted part failed with any sign of heat, smoke, arcing, a burning smell, or an audible pop or bang, or if a breaker, fuse, or protective device tripped, de-energize the circuit or system immediately and do not re-energize it until you understand why. Do not attempt a third part, known-good or otherwise, into a circuit that just destroyed a confirmed-working component. See the article on responding to a burning smell or active hazard if any of these signs are present.

If the part simply stopped functioning with no hazard signs (no heat, no smoke, no smell, no tripped protection), you can continue diagnosing, but treat the finding with the same seriousness below.

What this result actually tells you

A known-good part failing when installed in the suspect system is strong evidence that something upstream of that component is actively destroying parts, not that the component itself was ever the real problem. Common upstream culprits, by pattern:

• Excess supply feeding the component: voltage, pressure, or flow above what the component is rated for, coming from something else in the system, not the component itself.
• A short or a path to ground downstream of the component that overloads anything placed in that position, regardless of how healthy the part is.
• A seized or blocked load that the component is trying to drive or serve, causing it to work far outside its rated condition until it fails.
• A control or logic fault that is commanding the component into a condition it was never designed to sustain (holding a valve or contact energized far longer than intended, cycling something far more rapidly than its duty cycle allows).

Check the supply and the load before you touch another part

If you have not yet measured what is actually feeding this component (the voltage, pressure, or flow at the point of connection, with nothing installed or with a load resistor/test fixture in place if safe to do so), do that now, before installing anything else of value. You are looking for a supply condition outside the component's rated range.

If the supply measures within normal range, the upstream cause is more likely downstream of the component instead: check for a short, a blockage, or a seized load on the output side that the component is trying to drive.

If you find a supply or downstream condition outside normal range, that is your actual root cause. Fix that first. Installing a third part into an unaddressed upstream fault repeats the same failure a third time.

Once the upstream cause is addressed

1. Verify the upstream condition is now within normal range before installing anything further. Do not proceed on the assumption that a fix "probably" worked; measure it.
2. Install a replacement for the originally-suspect component, now that the condition that killed your known-good test part is corrected.
3. Test under the real triggering condition, not just a static power-up, and watch specifically for any recurrence of the failure signature you just saw.
4. Restore or replace the part you sacrificed during the test, per whatever arrangement applies (a borrowed part that failed on a donor's behalf typically needs to be replaced for that customer, not just noted as a loss).

If you cannot find an upstream cause

If a thorough check of supply and load conditions turns up nothing outside normal range, and the part still failed, consider two remaining possibilities: the substitute part was not actually as known-good as you believed (verify its source and testing history again, honestly), or the failure is genuinely intermittent and tied to a condition you have not yet reproduced (a specific load spike, a specific external event). Do not install a third part without addressing one of these; repeating the same swap a third time with the same unresolved unknown just burns another part.

Decision recap

1. Any hazard sign on the failed substitute means de-energize and stop before doing anything else.
2. A clean failure (no hazard signs) is still a serious signal: something upstream is destroying parts, not confirming the original component was ever really the issue.
3. Measure the actual supply and load conditions before installing anything further.
4. Fix the upstream cause you find, verify it is corrected, then install the real replacement and test under real conditions.
5. If no upstream cause is found, question the substitute's true condition or suspect an intermittent trigger before trying a third part.

References

• OSHA lockout/tagout guidance for de-energizing equipment before further work
• Manufacturer documentation on component rated limits and protection requirements
• See related: The Known-Good Swap Test Method
• See related: Correlating a Fault with a Recent Power Event