The System Never Worked Right From Day One: Decision Tree

Why this matters

The most expensive diagnostic mistake is chasing a failure on a system that never failed, because it was never right. You will test good parts, swap good parts, and still hand back a system that underperforms exactly as much as it always did. This tree keeps you from that trap by settling one thing first: did this system ever do its job? Establish the timeline before you touch a meter.

Start here: establish the timeline first

Before any test, ask the customer three questions and write the answers down.

1. When it was new, did it do the job you needed? Be specific: did it heat the whole house, hold pressure, keep up on the hot days?
2. When did the complaint start? A date, a season, or "always"?
3. Did anything change right before it started? A new appliance, a remodel, a storm, a prior repair?

Their answers, not your gut, pick the branch.

If it worked fine and then changed

You have a failure. Something was right and is now wrong, so a specific cause exists to be found.

• Anchor on the change. A new high-draw appliance, an added bathroom, a heavier vehicle, a remodel that added rooms, or a prior contractor's work is the first suspect.
• If nothing obvious changed, chase the normal failure list: a worn part, a fouled component, a leak, a bad connection, a control drift.
• Verify the repair restores the original performance. If the repair holds and the system is right again, you are done.

If it never worked right

You have a design or install fault, not a failure. Nothing broke. Stop hunting for a broken part.

• Confirm the individual components are actually healthy (a quick check, not a teardown). Healthy parts plus poor performance points away from failure.
• Ask what the system was supposed to deliver, and measure what it delivers. A real gap with no failed part is the signature.
• Sort the fault: undersized for the load, laid out or routed wrong, components mismatched, or a required piece never installed.
• Quote the honest correction, and be clear it is a change to the system, not a part swap.

If nobody knows: new owner, no history

When the customer bought the place recently or never paid attention, you cannot trust the timeline. Reconstruct the baseline instead.

• Read the equipment against the building: is the capacity plausible for the space, the fixture count, the load?
• Look for the peak-load tell. Run it hard. A system that keeps up in mild conditions but folds under full demand was almost certainly undersized, not failed.
• Check install quality: routing, clearances, connections, missing components. Sloppy or incomplete work that has been that way since install is a design/install fault regardless of history.

The trap: a slow decline that feels like "always"

A system can degrade so gradually that the customer swears it "always did that," when in fact a slow-building failure (creeping restriction, slow scale, a slowly loosening connection, gradual fouling) crept in. Do not take "always" at face value. Probe: was there truly no better period, even early on? If the parts test degraded rather than merely undersized, treat it as a slow failure you can repair, not a design flaw you cannot.

Recap

1. Get the timeline first: was it ever right, when did it start, what changed.
2. Worked-then-changed means failure. Find the change, fix the part, verify original performance.
3. Never-right means design or install. Confirm healthy parts, measure the gap, name the fault.
4. No history means reconstruct the baseline and run it under peak load.
5. Do not trust "always" over a tested slow-failure signature.

References

• Trade-standard practice for load calculation and equipment sizing
• Manufacturer application/design data for expected performance
• See related: Telling a Design Flaw Apart From a Failure; A Chronically Failing Part Points to a Design Problem (decision tree)