Obsolete Electronic Component Sourcing for Legacy Boards
Electronic components go out of production far faster than the equipment containing them goes out of service. A production line, a medical device or an aircraft system routinely stays in operation for 15 to 25 years, while the semiconductors inside it are typically supported by their manufacturer for seven to ten. The gap between those two numbers is the obsolescence problem, and it is the reason a board that works perfectly today can become unrepairable next year.
This page covers what we do about it: sourcing parts that are no longer manufactured, verifying they are genuine before they ship, and knowing when substitution is a better answer than sourcing.
How a part becomes obsolete
Obsolescence is not a single event. It arrives in stages, and the cost of responding rises at each one.
| Lifecycle stage | What it means | Urgency |
|---|---|---|
| Active | In production, normal distribution | None |
| Not recommended for new designs | Still produced, but flagged for replacement | Plan now |
| Last-time-buy announced | A final order window before discontinuation | High — this is the cheapest moment to act |
| End of life | Manufacturing has stopped | Broker market only |
| Obsolete / unavailable | Stock depleted across the open market | Redesign or recover from surplus |
The last-time-buy window is the single most important point on that timeline. It is the last moment at which the part can be bought at a known price, with full traceability, from an authorised source. Once it closes, every subsequent purchase carries counterfeit risk and price volatility. If your BOM contains parts flagged “not recommended for new designs”, that is the signal to act — not the eventual end-of-life notice.
How obsolete component sourcing actually works
Finding a discontinued part is not a database search. Distributor listings go stale, and the parts they claim to stock frequently are not there. Real sourcing works through relationships and inventory archaeology:
- Surplus and excess inventory. A large amount of obsolete stock sits inside organisations that bought more than they needed and have no active process for selling it — contract manufacturers, defence contractors, research institutions, and OEMs that have since ended a product line.
- Franchised and semi-franchised distributors with residual stock of parts they once carried in volume.
- Broker networks, which are the fastest route but also the highest risk, and require the testing discipline described below.
- Component recovery from surplus assemblies, where boards are de-soldered for their parts — legitimate in some contexts, and something we will only do when the source and condition can be documented.
Counterfeit risk is the real cost
Sourcing an obsolete part is easy. Sourcing one that is genuine is the actual service. The open market for discontinued semiconductors contains recycled, re-marked, out-of-specification and outright fabricated parts, and a counterfeit that passes a visual check can still fail in the field — which, in a medical device or an aircraft system, is not an acceptable outcome.
Risk control has three parts:
- Traceability. Where a documented chain of custody exists, we establish it. A part with a verifiable history is worth more than an identical part without one, and the price difference reflects that.
- Supplier assessment. Sources are vetted on reliability, documentation practice and history, not just on being the first to answer a quotation request.
- Independent testing. Where the application is safety-critical, components are tested before they are accepted, not after a problem appears.
How obsolete parts are tested
Testing is staged by risk. Not every part needs every test, but every part gets the first two.
| Test | What it detects | When it is applied |
|---|---|---|
| Visual inspection | Re-marking, resurfacing, damaged or mixed-date packaging | Always |
| X-ray inspection | Die and lead-frame mismatch against a known-good reference | Always for critical parts; sampling elsewhere |
| Electrical / functional test | Performance outside specification | Safety-critical and high-value parts |
| Decapsulation and die inspection | Die substitution — a different device inside a correctly marked package | Where traceability is weak and the application is critical |
| Solderability testing | Oxidised or re-tinned leads that will not wet properly | Parts with long storage histories |
The testing standard applied for counterfeit detection in aerospace and defence supply chains is AS6171; industrial and medical customers often specify equivalent in-house protocols. Whatever standard applies, the output should be a documented report, not a verbal assurance.
Last-time-buy and long-term storage
Where a customer knows a part is going end-of-life and the product will be supported for years, the cheapest long-term answer is often to buy the remaining requirement in advance. We can bulk-purchase end-of-life items and either ship immediately or hold them in controlled storage for release as needed.
Two conditions make this work: the storage has to control humidity and ESD to avoid degrading the parts, and the quantity decision has to account for the expected failure rate over the support period, not just the production requirement. Buying the production quantity alone leaves you unable to repair the units you already shipped.
When substitution is the better answer
Sourcing is not always the right call. If a part is genuinely unobtainable, or if the price of the remaining stock exceeds the cost of a small redesign, substitution wins.
Substitution ranges from trivial to substantial:
- Pin-compatible replacement — a drop-in part from the same family. Minimal work, but it must be verified against the original’s electrical specification, not just its footprint.
- Package-compatible alternative — requires a footprint review, and often a check on thermal and EMC behaviour.
- Functional equivalent with redesign — a modern part that achieves the same function but needs different support circuitry. This is a small engineering project and is priced as one.
Where no drop-in exists, we say so rather than shipping a substitution that will pass inspection and fail in service. This is the same principle we apply to schematic recovery: document the limit, do not paper over it.
How this connects to reverse engineering
Obsolescence and reverse engineering are usually the same project seen from two ends. When a board has to be reproduced because the original supplier is gone, the BOM is recovered through reverse engineering, and the parts on it are then sourced or substituted. Delivering a reproduced board with an unsourced BOM leaves the job half done, which is why we treat component sourcing as part of the same service rather than a handoff to somebody else.
Frequently asked questions
How long does it take to source an obsolete part?
Common discontinued parts are often located within days. Genuinely rare devices — limited-run automotive or military parts with no substitute — can take weeks, and sometimes cannot be found at all. We report feasibility before you commit rather than after a long search.
Can you guarantee a sourced part is genuine?
No one can guarantee that, and a supplier who claims to is worth less than one who shows their process. What we guarantee is documented traceability where it exists, staged testing against the risk level, and a written report of what was found.
Is it cheaper to substitute than to source?
Sometimes. Rare obsolete parts can be priced far above their original cost, and where that happens, a small redesign using a current part is often cheaper over the remaining product life. We quote both paths where both are viable.
What if the part cannot be found or substituted?
Then the remaining option is to recover the function at a higher level — reproducing the whole assembly or a functional block of it through reverse engineering, rather than replacing the individual device.
Do you handle RoHS conversion of old parts?
Yes. Where leaded parts must be replaced for compliance reasons, we can identify compliant equivalents and flag any process changes that follow from the different reflow profile.
If your BOM has a part you can no longer buy, send us the part number and the application and we will tell you whether it is sourceable, and what the realistic path is if it is not.
Related reading
- PCB reverse engineering — how a BOM is recovered from a physical board.
- Schematic recovery — recovering the circuit that tells you what each part does.
- Cost guide — how substitution decisions affect project price.