-
Product Categories
-
Live Support Chat
Chip Failure Analysis Service
Failure analysis answers a specific question: why did this device stop working, and would it happen again? It is routinely required for supplier disputes, field-return investigations, incoming inspection of suspect batches and qualification of a new source or a reproduction part.
A useful analysis is sequenced. Electrical characterisation narrows the failure to a region before anything is opened, which means the destructive stage is targeted rather than exploratory. Opening a device first and looking around is how samples get consumed without an answer.
The analysis sequence
| Stage | Technique | What it establishes |
|---|---|---|
| 1. Non-destructive | External inspection, X-ray, XRF material analysis, acoustic microscopy | Package integrity, die attach voids, marking authenticity, internal structure |
| 2. Electrical | Curve tracing, pin-to-pin measurement, functional test, comparison against a known-good unit | Which pins or blocks are abnormal, and whether the failure is open, short or parametric |
| 3. Localisation | Emission microscopy (EMMI), OBIRCH, thermal imaging, liquid crystal | The physical location of the defect while the device is powered |
| 4. Physical | Decapsulation, layer-by-layer delayering, optical and SEM inspection | The defect itself: metal migration, oxide breakdown, ESD damage, voiding, contamination |
| 5. Conclusion | Correlation of electrical and physical evidence | Root cause plus an assessment of whether it is systematic or isolated |
Failure modes we identify
- Electrical overstress and ESD — characteristic melt filaments, junction damage and gate oxide rupture
- Electromigration and metal fatigue — voiding and hillock formation in power and ground routing
- Bond wire failure — heel cracking, intermetallic growth, wire sweep and lifted bonds
- Die attach and package defects — voids, delamination and thermal path degradation
- Corrosion and contamination — ionic residue, moisture ingress and dendritic growth
- Counterfeit indication — die substitution, re-marking, re-packaging and recycled parts
Where this matters commercially
- Supplier disputes: technical evidence that a failure is inherent to the part rather than caused by the application
- Incoming inspection: screening a suspect batch before it reaches production
- Field returns: determining whether a failure is isolated or the leading edge of a systematic problem
- Reproduction qualification: proving a clone or substitute part meets the reliability of the original
Frequently asked questions
How many samples do you need?
Ideally three: one to characterise electrically, one to open destructively, and one known-good reference for comparison. A single failed sample is workable but the conclusion is weaker because there is nothing to compare against.
Will I get the failed sample back?
The electrical characterisation sample and the reference sample are returned. The sample used for decapsulation is consumed by definition, and we return it in its opened state so the physical evidence remains available to you.
Can you determine whether a part is counterfeit?
Yes, and this is one of the more common reasons customers come to us. Marking inspection, XRF, X-ray and die comparison against a genuine reference will normally settle the question without destroying the part, which matters when the evidence may be needed for a claim.
How long does a failure analysis take?
A non-destructive screening runs two to three working days. A full electrical and physical analysis with localisation typically runs one to three weeks depending on how readily the defect can be isolated. Urgent returns can be prioritised — ask when you submit the request.
Do you provide the analysis for legal or insurance purposes?
We provide a technical report with the method, the observations and the conclusion, and we keep the raw imagery so that the findings can be independently reviewed. The report is written to be usable as evidence, but we are a technical laboratory rather than a legal advisor, and we do not present an opinion on liability.
Submit a sample for analysis
Tell us what failed, how it failed, what the field conditions were, and what question you need answered. We will propose the analysis sequence, state what each stage can and cannot establish, and quote it before starting. Work is accepted only where the requester can demonstrate ownership or written authorisation.