The Fault That Only One User Can Trigger
Why this matters
Several people use the same equipment. One of them, and only that one, makes it fail. Everyone else, including you, runs it fine. This is one of the most useful situations you can walk into, because the other users are a ready-made control group. The equipment is the same for all of them, so whatever is different sits with the one person who triggers it. Most techs waste the gift by testing the machine their own way, finding nothing, and calling it no fault found. The person who fails is not imagining it, and they are not lying. Find the one thing they do differently and you have the fault.
The peer contrast is your best tool
When only one operator among several triggers a fault, you already have half a controlled experiment done for you. Do not throw it away by only observing the person who fails. You learn the most from the delta between the failing user and a succeeding user doing the same task.
- Watch the failing user do the task their normal way, without coaching, the way they would if you were not there.
- Watch a succeeding user do the same task.
- Look for the single difference. It is usually one variable, not a whole different method.
The fault lives in that difference. Everything the two users do the same can be ruled out.
What makes one person different
The variable is almost always physical or habitual, not a mystery. A field key of what to look for:
- Force and pressure. A stronger person pushes a control, a handle, or a door harder and overtravels a switch, cracks a seat, or trips a limit that a lighter touch never reaches.
- Body size, reach, and handedness. Someone taller, shorter, or left-handed approaches a control from a different angle, leans on a panel, or blocks an airflow or sensor the others do not.
- Pace and rhythm. A fast operator cycles power, valves, or steps quicker than the equipment can reset, discharge, or re-pressurize between actions.
- Sequence habit. One person always does step B before step A, or turns on an accessory first, in a fixed order that exposes an interaction others avoid by luck.
- Sensory and reading habits. Someone misreads a label, cannot hear a warning tone, or does not see an indicator, so they push past a cue the others heed.
- A fixed routine. They always lean on the unit, rest a foot on a pedal, prop a lid, or hold a button longer than needed.
None of these make the person careless. They make them a different, consistent input the equipment was never tested against.
Draw it out, do not just ask
People cannot describe the habit that causes the fault, because to them it is automatic and invisible. Asking "what do you do differently" gets you a shrug. Instead:
- Have them show you, not tell you. "Run it the way you always do, and I will watch."
- Compare against a peer live if you can get both, so the difference shows itself rather than relying on either person's summary.
- Ask targeted questions about the delta you spot ("do you always push it that far," "how quickly do you go from this to that"), not general ones.
The outcome: repair, or coach the technique
Once you can reproduce the fault under that one person's input, you can finally diagnose from real readings. Then decide honestly which of two things you have. Either the equipment has a genuine weakness that a normal-range human input should not expose, in which case it is a repair or a more robust part, or the person is operating outside the equipment's designed input range, in which case showing them a different technique is the correct fix and not a cop-out, as long as their input really is outside the normal range and not just outside your gentle test. Do not decide which one before you have reproduced and watched the failure; a fragile design and a hard-handed user look identical until you see the numbers.
References
- Manufacturer documentation on intended operating inputs, forces, and duty limits
- Trade-standard practice for reproducing operator-dependent symptoms
- See related: It Only Fails When the Customer Operates It a Specific Way; Reproducing a Fault by Matching the User's Technique; The Fault Follows the Operator, Not the Equipment (decision tree)