We fixed this before: why didn't the same fix work this time?
Summary
A past fix fails again when a condition it relied on has changed, or when it only paused the symptom without removing the cause. Check substrate, water hardness, concentration, temperature and equipment first. Compare the earlier and current case condition by condition, and reuse only the parts whose conditions still match.
Reliable, a maintenance and reliability publication, names treating symptoms rather than causes as the most frequent driver of recurring work orders (Reliable, accessed 2026).
Reusing a past solution means checking two things: the differences between the past and the current case, and which part of the past case can be transferred (Aamodt and Plaza, 1994).
Calcium and magnesium in hard water form insoluble soap scum when mixed with many metalworking fluids (OSHA Metalworking Fluids Best Practices Manual, accessed 2026).
Reliable suggests flagging any asset with three or more corrective work orders for the same failure mode within a rolling 12-month window for a root cause review (Reliable, accessed 2026).
This comparison method applies to process, fluid and maintenance problems where an earlier fix is on record or can be rebuilt. It does not replace a formal root cause analysis when no condition has changed, and it does not cover safety incidents, which follow your site's incident investigation procedure.
Why does a fix that worked before fail when the symptom is the same?
A fix that worked before fails on the same symptom when the symptom is shared but the cause is not, or when the cause is the same but a condition the fix relied on has moved. Symptoms are what people see: foam, rust spots, a bond that lets go, a filter that blinds early. Many causes produce each of them. Reliable names treating symptoms rather than causes as the most frequent driver of recurring work orders (Reliable, accessed 2026). A team that matched the new case to the old one by symptom alone has reused a conclusion without checking whether its premises still hold.
The case-based reasoning literature describes the same trap in general terms. Reusing a past solution means looking at the differences between the past case and the current one, and at which part of the past solution can be transferred (Aamodt and Plaza, 1994). When the differences are ignored, the old solution is copied as if the cases were identical, which works only when nothing relevant changed.
A second reason is that the earlier fix may never have removed the cause. OSHA's metalworking fluids manual warns that refilling without cleaning the delivery system exposes new fluid to "the same conditions that forced you to change the fluid in the first place" (OSHA Metalworking Fluids Best Practices Manual, accessed 2026). The earlier fix looked like a success because the symptom paused.
Which conditions does a past fix for the same problem depend on?
A past fix for a specialty chemical problem usually depends on five conditions: the substrate, the water, the product concentration, the temperature and the equipment. Each one can change without anyone treating it as a change.
Substrate
A cleaner, coating, adhesive or fluid that worked on one alloy, surface finish or part was proven on that surface. A new part supplier, a different heat treatment or a new machining step changes the surface the product meets.
Water hardness
Water-mixed fluids and cleaners depend on the water used to dilute them. OSHA's manual states that calcium and magnesium "will form insoluble soap scum when mixed with many MWFs" (OSHA Metalworking Fluids Best Practices Manual, accessed 2026). A new water source or a failed softener can shift hardness.
Concentration
OSHA's manual recommends checking actual concentration in machines frequently: low concentration may bring bacteria, fungus and corrosion problems, and high concentration foaming and heavy residues (OSHA Metalworking Fluids Best Practices Manual, accessed 2026).
Temperature
Cure, evaporation, reaction and microbial growth rates depend on temperature. A fix proven in a cool month may not hold in a hot one, or after a cooler is fouled or bypassed.
Equipment
A different pump, filter, nozzle, seal or machine changes delivery and contact. Reliable lists substituted parts, such as bearings that meet dimensional specs but lack the correct load rating, among the causes of recurring failures (Reliable, accessed 2026).
How can you tell what changed since the earlier fix of the problem?
You can tell what changed since the earlier fix by putting the earlier case and the current case side by side, condition by condition, and checking each condition with a record or a measurement rather than memory. Start from the earlier fix's record: what was done, on which line, with which product lot and under which readings. Then take the same readings on the current case. The table below decides which changed condition is most likely to stop the earlier fix from working on the current problem:
Check | What you see | Points to |
|---|---|---|
Substrate or part | New part supplier, alloy, finish or machining step since the earlier fix | The fix was proven on a surface that is no longer there; requalify on the current part |
Water | Hardness reading differs from the earlier record, or the water source or softener changed | Hard-water salts reacting with the product (OSHA Metalworking Fluids Best Practices Manual, accessed 2026) |
Concentration | Measured concentration sits outside the range recorded when the fix worked | Dosing, mixing or carry-out drift; the earlier dose no longer matches the system |
Temperature | Process or ambient temperature differs from the earlier case, or a cooler or heater changed | Rate-driven effects such as cure, evaporation or microbial growth |
Equipment | A pump, filter, seal, nozzle or machine was replaced with something not like-for-like | Delivery or contact conditions changed |
None of the above | All five conditions match the earlier record | The earlier fix probably treated a symptom; reopen the root cause analysis |
The row that changes the plan most is the last one: when all five conditions match the earlier record and the fix still fails, the earlier fix probably treated a symptom, and the case needs a root cause analysis rather than a third repeat. IEC 62740:2015 describes the basic principles of root cause analysis and specifies the steps that a process for it should include (IEC 62740:2015).
Which parts of the earlier fix still transfer to the new problem?
The parts of an earlier fix that still transfer to a new problem are the ones whose conditions match the current case; the rest has to be rebuilt or retested. Split the old fix into pieces before reusing it: the diagnosis, the method used to reach it, the action taken, the dose or setting, and the acceptance check. The case-based reasoning literature separates reusing a past solution from reusing the past method that constructed the solution (Aamodt and Plaza, 1994). On a site, the method often transfers when the solution does not: the sequence of checks that found tramp oil last time is still worth running even if the cause this time is water hardness.
Give each piece one of three verdicts. Transfers: the conditions it depends on match, so reuse it and confirm the result. Transfers with change: one condition moved, so adjust the setting and retest on a small scale before full use. Does not transfer: the piece depended on a condition that is gone, such as a part material or a water source. A past conclusion is evidence about the past case; it becomes a conclusion about the current case only after the comparison supports it.
Where the system is a process covered by US OSHA process safety management, changes to process chemicals, technology, equipment and procedures, other than replacements in kind, fall under written management of change procedures that address the technical basis before the change (OSHA 29 CFR 1910.119(l), 2026).
How do you record a fix so others know when it applies to the same problem?
Record a fix so others know when it applies by writing down the conditions it was proven under next to the action itself. A note that says "added defoamer, problem solved" tells the next person what was done but not when it works. Reliable recommends capturing four fields on every corrective work order: failure mode, failure cause, corrective action taken and recommended follow-up (Reliable, accessed 2026). For a chemical fix, add the conditions it depends on:
substrate or part, with its supplier
water source and hardness reading
product, lot and measured concentration
process temperature
equipment configuration, including any replaced parts
the acceptance check and the date the result was confirmed
Record failed attempts too. A fix that did not work, with its conditions, tells the next person which combination to avoid. In case-based reasoning, failures may be retained as separate failure cases, so a later failure can be recognized and corrected (Aamodt and Plaza, 1994). OSHA's manual likewise recommends tabulating fluid observations, laboratory analyses and additions so that relationships and trends can be spotted (OSHA Metalworking Fluids Best Practices Manual, accessed 2026). Finally, set a repeat trigger: Reliable suggests flagging any asset with three or more corrective work orders for the same failure mode within a rolling 12-month window (Reliable, accessed 2026).
When should you bring in a specialist when the same fix keeps failing?
Bring in a specialist when the same fix has failed twice, when the side-by-side comparison shows no changed condition, or when the fix would change a hazardous process. Reliable suggests that a second occurrence of a recurring work order should trigger a 5 whys analysis at minimum, and a third a formal investigation with engineering involvement (Reliable, accessed 2026). A specialist adds most when conditions interact, for example when water hardness and concentration drift together, or when the product supplier needs to test a retained sample of the current lot against the earlier one.
Prepare for that conversation with the side-by-side comparison, the records of both cases, the certificates of analysis for both lots and retained samples of the product and the in-use fluid. Handle samples, cleaners and biocides according to each product's safety data sheet (SDS) and your site's chemical safety rules. Lubinpla works a repeat problem like this from your own documents and site history: it fixes the similarity axes before searching past cases and judges what part of a past response transfers and where it stops, rather than reusing the past conclusion as the current one.
Frequently asked questions
Is a recurring symptom always the same problem coming back?
No. The same symptom can come from a different cause, and a cause that returns can arrive by a different route. Compare the conditions of the earlier and current case before calling it a recurrence, and treat a match on symptom alone as a hypothesis, not a finding.
Should you change several conditions at once to make the old fix work?
Not if you want to learn which condition mattered. Change one condition at a time where the process allows it, and record the result of each change, so the next case can be compared against a clear record.
What if the earlier fix was never written down?
Rebuild what you can from purchase records, work orders, lab reports and the people involved, and mark each rebuilt item as reconstructed. Then measure the current conditions in full, so the current case at least becomes a usable record.
Does a supplier's product change count as a changed condition?
Yes. A reformulated product, a new production site or a lot with different certificate values can behave differently under the same name. Compare the certificate of analysis and the safety data sheet revision date with the ones on file from the earlier fix.