PCB Microsection Analysis
Quality & Inspection Standards
PRODUCTS CENTER
Get Custom Quote
PRODUCTS CENTER
PCB Microsection Analysis
Microsection (cross‑sectioning) is the only way to see what really happens inside a PCB. We have seen boards that passed electrical test but failed microsection — voids hidden under pads, innerlayer separation only visible after thermal stress, and copper cracks that would have caused field failures within months. This guide walks through IPC‑TM‑650 Method 2.1.1, sample preparation pitfalls, and how to interpret real failure modes.
Table of Contents
1. Why Microsection – Beyond Electrical Test
Electrical test only confirms continuity and isolation. It cannot detect:
- Thin plating that will crack after thermal cycling
- Voids that reduce current carrying capacity by >30%
- Innerlayer separation (resin starvation or oxide failure)
- Microvia target pad separation (HDI killer)
- Backdrill stub residues that cause signal reflections at 25+ Gbps
Per IPC‑6012E §3.6.2.1, microsection is mandatory for qualification and periodic conformance testing. We have rejected production lots where electrical test passed but microsection revealed Class 3‑prohibited copper cracks — saving our customers from field returns.
Engineering Insight
A PCB that passes ICT (in‑circuit test) still has a 5‑15% probability of containing latent microvoids or innerlayer cracks, depending on the lamination process. Microsection is your only insurance against these hidden defects.
2. Sample Preparation per IPC‑TM‑650 2.1.1
Method 2.1.1 covers manual, semi‑automatic, and automatic microsection preparation. The critical steps are:
| Step | Key Requirement | Common Mistake |
|---|---|---|
| Sample extraction | Diamond saw or routing, minimal heat/mechanical stress | Shearing that delaminates edges |
| Mounting | Epoxy or acrylic, cure ≤93°C, vacuum to remove bubbles | Trapped air gaps that cause edge rounding |
| Alignment | Sample perpendicular to polishing plane | Tilted samples → measurement errors >0.025mm |
For high‑volume production, automatic systems with pin‑alignment fixtures (Method B) ensure consistent grinding to the exact hole center. We always request microsection coupons on the production panel to avoid destructive sampling of finished boards.
3. Grinding & Polishing – Where Most Errors Occur
IPC‑TM‑650 Table 5‑1 specifies a multi‑step sequence (manual method). The goal is to reach the exact center of the plated hole (±10% of diameter) without introducing artifacts.
| Grit (ANSI/FEPA) | Purpose | Typical Time (manual) |
|---|---|---|
| P60 – P120 | Rough planarization | Until flush with mount |
| P220 – P400 | Remove deep scratches | 2‑3 min per step |
| P800 – P1200 | Approach hole center | Until hole cross‑section >90% of diameter |
| 6µm → 1µm diamond | Rough & fine polish | 1‑2 min each (soft cloth) |
Critical: After each grit, rotate the sample 90° and grind until previous scratches are completely removed. If scratches from P400 remain after P1200, the microsection is invalid per Method 2.1.1 §5.4.
Pro Tip
Use intermediate ultrasonic cleaning between polishing steps. Epoxy‑based materials tend to trap abrasive particles in microvoids — a 1‑minute ultrasonic bath prevents false inclusions that could be misinterpreted as copper nodules.
4. Micro‑Etching to Reveal True Interfaces
Polished surfaces hide grain boundaries and true plating interfaces. IPC‑TM‑650 2.1.1 §5.5.2 specifies a micro‑etch solution:
Apply with a cotton swab for 3‑5 seconds, then rinse. Over‑etching (>10 seconds) can completely obliterate the copper‑electroless interface, making innerlayer separation assessment impossible. We have seen labs produce false‑positive innerlayer cracks simply due to excessive etching.
For tin‑lead finishes, use a specialized etch (e.g., 2% nitric acid in alcohol) to avoid galvanic effects. Always examine the sample in the as‑polished condition first — record any delamination or voids before micro‑etching (per §5.5.1).
5. Common Defects & How to Identify Them
The table below maps defect types to IPC‑TM‑650 observations and acceptance criteria.
| Defect | Microsection Appearance | Rejectable (Class 3) | Related Page |
|---|---|---|---|
| Copper Crack Type D | Through‑hole wall separation, visible at 100x | Yes (any length) | Copper Crack Guide |
| Plating Void | Missing copper in hole wall, often wedge‑shaped at innerlayers | Zero visible voids | Plating Void Guide |
| Innerlayer Separation | Gap between innerlayer copper and resin/oxide | Any separation | Innerlayer Separation |
| Microvia Target Pad Separation | Gap between microvia bottom and target pad | Any visible gap (Class 3) | Microvia Failure |
| Backdrill Stub Residue | Copper or epoxy left in backdrilled hole | Any residue exceeding specified stub length | Backdrill Stub |
Real‑world case: A 16‑layer telecom backplane passed A/B sample microsection but failed after thermal stress (IPC‑TM‑650 2.4.13.1). Re‑examination using reduced etch time (Method 2.1.1 note 6.2) revealed innerlayer separation that was smeared over during initial polishing — exactly the scenario that causes field failures after 6 months of thermal cycling.
6. Measurement & Acceptance Criteria
Per IPC‑TM‑650 2.1.1 §5.4, the observable hole length must be ≥90% of the drilled diameter (Figure 5‑2). Measurements are taken at 100‑200x magnification using a calibrated reticle.
Key IPC‑6012E Limits (derived from microsection):
- Minimum copper thickness (Class 3): 20µm average, 12µm minimum at any point
- Annular ring (Class 3): ≥50µm, no breakout
- Microvia target pad contact: fully captured with no separation
Always measure at the thinnest point of the wall. If measured thickness varies by >5µm around the circumference, check for drill wander or plating uniformity issues.
Microsection Quality Checklist
- ☑️ Sample perpendicular to mounting plane (visual check before potting)
- ☑️ No air bubbles between sample and mounting material (vacuum degassed)
- ☑️ Grinding steps rotated 90°, previous scratches fully removed (100x inspection)
- ☑️ Final polish with 1µm diamond on low‑nap cloth, ≤30 seconds to avoid edge rounding
- ☑️ Micro‑etch time ≤5 seconds, rinse immediately
- ☑️ Observed hole length ≥90% of drilled diameter
- ☑️ No copper smearing (re‑etch if needed per note 6.2)
- ☑️ Measure minimum wall thickness and annular ring at 200x
7. FAQ
Q: What is the minimum acceptable observable hole length in a microsection?
A: Per IPC‑TM‑650 2.1.1 §5.4, it must be ≥90% of the drilled hole diameter. For a 0.2mm drill, you need at least 0.18mm of plated wall visible. Less than that invalidates the measurement.
Q: Can microsection be performed on assembled boards?
A: Yes, but with caution. Components must be removed or ground away, and the solder joint may obscure innerlayer details. We recommend using dedicated coupon panels whenever possible to avoid destroying functional hardware.
Q: How many microsections are required for IPC‑6012 Class 3 qualification?
A: Per Table 4‑2, at least 3 coupons per panel, with at least 2 microsections per coupon (one for as‑polished, one for after thermal stress). Each microsection must include all plated hole types.
Q: What causes “pink ring” (oxide separation) in microsection?
A: Pink ring is a separation between the innerlayer copper and the black oxide treatment. It appears as a thin pink line after micro‑etching. Acceptable only if it does not reduce the remaining copper pad width below minimum annular ring requirements.
Continue Reading – Failure Analysis Series
Need a second opinion on your PCB microsections? Our engineers will review your cross‑sections against IPC‑TM‑650 and IPC‑6012 criteria.
Request Microsection Review →References: IPC‑TM‑650 Method 2.1.1 (2015), IPC‑6012E §3.6, §3.3; IPC‑A‑600J §3.3 (plated hole defects). IPC‑TM‑650 Official Method.
Get Custom Quote
PRODUCTS CENTER


