When an OEM discontinues a spare part, the machine does not stop caring. This is the method we use to find a safe replacement without guessing.
1. Confirm the part is really obsolete
Check the OEM’s successor list and the marketplace before assuming the end. Many “obsolete” parts have a direct successor with a new number.
2. Capture the full specification
Record the exact reference, revision, and — critically — the function and interfaces: dimensions, electrical ratings, materials, and how it connects to the rest of the system. A part number alone is rarely enough.
3. Search the three routes
- Direct/successor — same or upgraded OEM part, lowest risk.
- Cross-reference equivalent — another maker’s part meeting the specification.
- Surplus / refurbished — tested used stock, acceptable for non-safety-critical duty.
4. Verify, don’t assume
Compare datasheets field by field, and where the application is critical, test the candidate before install. Document what was verified and the residual risk.
5. Plan the long-term fix
If the part will fail again, decide whether to re-engineer, hold a strategic stock, or qualify a permanent alternate. The cheapest immediate fix is not always the cheapest over the machine’s life.
The four replacement routes, compared honestly
Once a part is genuinely obsolete, there are four routes. Most buyers try them in the wrong order, usually starting with the most expensive and slowest.
| Route | Typical lead time | Cost profile | Main risk | Use when |
|---|---|---|---|---|
| New-old-stock from distributors or other users | Days to 3 weeks | Moderate to high per unit | Storage history and re-marking | A known quantity of unused stock is known to exist |
| Original maker’s repair or exchange programme | 2–8 weeks | Lower than a new unit | Maker has closed the programme without notice | The maker still supports the product line |
| Reverse engineering a replacement | 4–12 weeks including tooling | High first cost, low unit cost | Material and heat treatment guesses | Simple geometric part, no certification attached, repeated demand |
| Redesign or upgrade to a current equivalent | 6 weeks to 12 months | Highest, but permanent | Project cost and production interruption | The equipment will run for years and the old design is a dead end |
The usual correct order is: check the maker’s successor first, then new-old-stock, then repair, and only then consider reverse engineering or redesign. Reverse engineering a part that a distributor still holds at a fair price is wasted effort, however satisfying it feels.
Getting the specification right the first time
Step 2 of the process is where most of the cost is decided. A specification captured badly means every subsequent route has to be re-run. The minimum set worth recording for each obsolete item:
- Identity — original maker, part number, machine model, position in the bill of materials.
- Function — what the part does, in one sentence, in mechanical or electrical terms. “Holds the shaft in position under 4 kN axial load” is a function; “bearing” is not.
- Critical dimensions — the dimensions that define whether it fits, with tolerances, measured on a reference where possible.
- Material and treatment — including surface finish or coating, which is the most commonly lost detail and a frequent cause of premature failure.
- Operating conditions — temperature, pressure, medium, duty cycle, lubrication.
- What went wrong — the failure mode. It changes the replacement decision: a part that failed from wear may be replaced like for like, while one that failed from a design weakness should not be.
What to record so the next event is cheaper
Obsolescence is not a one-off. The same machine will lose more parts, and the same part may exist on several machines. Two habits pay for themselves:
- Maintain a live obsolescence register. One line per part: identity, status (supported / limited / obsolete), remaining stock on the shelf, the replacement decision taken, and the date it was reviewed. Review it annually against the makers’ lifecycle announcements.
- Feed the failure back into the stock decision. If a part failed twice in three years and is now obsolete, the answer is to buy a lifetime quantity now, not to buy one unit and hope. Sizing that quantity is a calculation — see our guide to last-time buy planning.
Questions we are asked about obsolete parts
How do I find out whether a part is actually obsolete? Ask the maker’s technical support in writing and keep the reply. “No longer available” from a distributor is not the same as discontinued by the maker — the maker may still support repair or hold stock that distributors cannot see.
Is reverse engineering ever the fastest route? Rarely. Sampling, drawing approval and first-article inspection take weeks even when the workshop is fast. It is a route for repeated demand, not for one machine that is down.
What if the maker has gone out of business? Then the specification package becomes the whole asset. Photograph and measure everything while the equipment is still running; that evidence is what lets any supplier, anywhere, make or find a replacement years later.
Can a higher-specification part replace a discontinued one? For passive mechanical and electrical parts, usually yes — provided the interfaces are identical. For anything with a control or protective function, compare the response characteristic rather than the headline rating.
References we have recently quoted in this category
Need a specific reference? Send the exact number to [email protected] with quantity and target date — we reply with sourcing options, equivalent/replacement candidates, condition and lead time.
Related reading
- Obsolete PLC, Sensor & Motor Parts: A Cross-Reference Workflow
- MRO Spare Parts Management: A Practical Guide