PTH Reliability Testing
Quality & Inspection Standards
PRODUCTS CENTER
Get Custom Quote
PRODUCTS CENTER
PTH Reliability Testing – IPC Class 2/3 Acceptance, Thermal Stress & Crack Limits
Plated through hole (PTH) reliability is the single biggest predictor of PCB field life in thermally stressed environments. We have seen telecom backplanes that passed electrical test but failed after 500 thermal cycles – not because of component solder joints, but because PTH barrels cracked from Z‑axis expansion. The mechanism is predictable: when the board heats above Tg, the dielectric expands, pulling the copper barrel apart. Per IPC‑6012E §3.6.2.1, Type D through‑hole cracks are not allowed for Class 3 products. This guide covers PTH thermal stress testing (IPC‑TM‑650 2.6.8), thermal shock (2.6.7.2), microsection evaluation, and acceptance criteria for Class 2 vs Class 3.
Table of Contents
1. Why PTH Reliability Matters – Field Failure Data
In our failure analysis lab, PTH cracks account for over 40% of field failures in thermally cycled products (telecom, server, automotive). The pattern is consistent:
- Infant mortality (first 6 months): Usually manufacturing defects – plating voids, wedge voids, thin copper.
- Mid‑life failures (12‑36 months): Typically fatigue cracks from thermal cycling – Z‑axis expansion pulling the barrel.
- End‑of‑life (after 5+ years): Ductile fatigue, CAF, or dielectric breakdown.
One major server OEM found that 68% of their field returns after 2 years were due to PTH barrel cracks – specifically Type D cracks that had propagated through the entire wall. The root cause was a low‑Tg laminate (140°C) combined with aggressive thermal cycling. Switching to high‑Tg (170°C) material with lower Z‑axis CTE reduced returns by 90%.
The physics: Each thermal cycle induces plastic strain in the copper barrel. After enough cycles (typically 200‑1000, depending on ΔT and CTE), the strain accumulates, and a crack initiates at the innerlayer interface where stress is highest. Once initiated, the crack propagates with each subsequent cycle until the electrical path opens.
2. Test Methods: Thermal Stress (2.6.8) vs Thermal Shock (2.6.7.2)
Two primary IPC methods evaluate PTH reliability. They serve different purposes:
| Method | Description | What It Reveals | Typical Cycles |
|---|---|---|---|
| 2.6.8 Thermal Stress | Solder float at 288°C for 10±1 seconds | Gross defects: brittle cracks, delamination, wedge voids | 1 cycle |
| 2.6.7.2 Thermal Shock | Air‑to‑air cycling, e.g., -40°C to +125°C, 15‑30 min dwell | Fatigue failures: ductile cracks, microvia corner cracks, CAF | 100‑1000 cycles |
For qualification, both methods are often required: thermal stress catches brittle manufacturing defects; thermal shock simulates field life. Per IPC‑6012E Table 4‑2, Class 3 requires thermal stress on every qualification panel. Thermal shock is required only when specified by the customer (e.g., aerospace, automotive).
In our lab, we run a hybrid test for high‑reliability programs: 500 thermal shock cycles (-40°C to +125°C) followed by 288°C solder float. This order reveals fatigue cracks (thermal shock) then tests remaining integrity (solder float). Boards that survive both have demonstrated excellent PTH reliability.
3. PTH Failure Modes – Type D, Corner Cracks & Barrel Fatigue
Per IPC‑A‑600J §3.3.9, PTH cracks are classified by location and extent. The most critical for reliability:
- Type D (through‑hole wall separation): Crack extends completely through the copper wall. Class 3: zero tolerance. Class 2: accept only if remaining copper thickness ≥18µm at the crack location.
- Type B (innerlayer interface crack): Crack initiates at the innerlayer pad edge and propagates into the barrel. Often a precursor to Type D. Class 3: reject if it reduces annular ring below 50µm.
- Corner cracks (microvia style, but also in PTH at pad/barrel junction): High stress concentration at the 90° corner. Often seen in thick boards or high aspect ratio holes.
- Plating voids (wedge voids): While not a crack, voids reduce current carrying capacity and create stress risers. Class 3: zero wedge voids. Class 2: accept if remaining Cu ≥18µm.
A 16‑layer board we analyzed had PTH cracks after 350 thermal cycles. Microsection at 200x revealed Type B cracks at 4 innerlayers. The supplier had changed oxide chemistry without requalification. The new oxide produced a weaker bond, and the Z‑axis expansion sheared the interface. Rejecting the lot saved $1.2M in potential field failures.
Engineering Insight
The most dangerous cracks are those that initiate at an innerlayer but are not yet through the wall (Type B). They do not cause immediate opens, so electrical test passes. But after months of field thermal cycling, they propagate to Type D and fail. Microsection after thermal shock is the only way to detect them.
4. Acceptance Criteria: Class 2 vs Class 3
IPC‑6012E and IPC‑A‑600J define the following limits for PTH after thermal testing:
| Defect | Class 2 Acceptance | Class 3 Acceptance | IPC Reference |
|---|---|---|---|
| Type D (through‑wall crack) | Accept if remaining Cu ≥18µm, isolated | Zero tolerance | IPC‑6012E §3.6.2.1 |
| Type B (innerlayer crack) | Accept if annular ring ≥25µm after crack | Annular ring must remain ≥50µm | IPC‑A‑600J §3.3.9 |
| Wedge void | Accept if remaining Cu ≥18µm | Zero wedge voids | IPC‑6012E §3.6.2.2 |
| Pinhole void (barrel) | ≤3 per hole, total area ≤5% of wall | Accept only if thickness ≥20µm at void | IPC‑A‑600J §3.3.8 |
| Corner crack (pad/barrel) | Accept ≤10% of circumference | Accept only if isolated and ≤5% of circumference | AABUS (common OEM spec) |
Table 1: PTH acceptance after thermal stress/shock. Class 3 imposes much stricter limits, especially on Type D cracks and wedge voids.
Important: IPC does not specify a "maximum allowed number of cracked vias" – instead, any crack that violates the above limits is a reject. For Class 3, a single Type D crack in any microsection coupon rejects the entire lot.
5. Sampling Plans – How Many Vias to Section?
Per IPC‑6012E Table 4‑2, minimum sampling for qualification:
- Class 2: 1 coupon per panel, 2 microsections per coupon (one as‑polished, one after thermal stress). Each microsection must include at least 3 PTHs of each type.
- Class 3: 3 coupons per panel, 2 microsections per coupon. Minimum 5 PTHs per coupon.
However, to achieve statistical confidence, we recommend:
- Prototype qualification: 10 coupons (5 as‑polished, 5 after thermal stress). This gives 95% confidence to detect a 5% defect rate.
- Production lot acceptance (Class 3): 5 coupons per panel, 5 PTHs per coupon = 25 PTHs inspected per panel.
- Aerospace / military: 100% of PTHs on the test coupon – typically a daisy‑chain pattern with 50‑100 vias.
In one program, the supplier followed only the IPC minimum (3 coupons, 2 PTHs per coupon). They missed a 12% defect rate (wedge voids). Our incoming audit with 10 coupons caught the issue, and the lot was rejected. The cost of the extra microsections was $600 – the cost of a field failure would have been $200,000.
PTH Reliability Test Checklist (IPC‑Anchored)
- Define test methods in PO. Specify both thermal stress (2.6.8) and thermal shock (2.6.7.2) if needed.
- Include daisy‑chain coupons. For thermal shock, monitor resistance continuously – a 10% increase is a failure.
- Pre‑bake samples at 125°C for 4‑6 hours. Moisture reduces T260 and increases crack risk.
- For thermal stress, control solder temperature to 288±5.5°C. Measure 25mm below the surface.
- For thermal shock, specify cycles and temperatures. Example: "500 cycles, -40°C to +125°C, 20 min dwell."
- After testing, microsection at 100‑200x. Examine for Type D, Type B, wedge voids, corner cracks.
- Document all cracks with 200x micrographs. Include scale bar and classification. Save for supplier corrective action.
6. FAQ
Q: What is the difference between IPC‑TM‑650 2.6.8 and 2.4.13.1?
A: 2.4.13.1 is "Thermal Stress, Laminates" – it applies to bare laminate samples (often etched). 2.6.8 is "Thermal Stress, Plated Through Holes" – it is performed on finished boards with PTHs and uses a shorter float time (10s vs 10s but with different handling). For PTH evaluation, use 2.6.8.
Q: How many thermal cycles are needed to simulate 10 years of field life?
A: Rule of thumb: 500 cycles of -40°C to +125°C approximates 10 years of daily power cycling for controlled environments. For automotive under‑hood, 1000 cycles is common. Validate with Arrhenius acceleration – there is no single IPC number.
Q: Can a PTH with a Type B crack be used in Class 3?
A: Yes, only if the remaining annular ring after the crack is still ≥50µm and the crack does not extend into the barrel. However, most Class 3 programs reject any Type B crack as a process warning. We recommend root cause analysis before accepting.
Q: What aspect ratio is considered high‑risk for PTH cracking?
A: Aspect ratio >10:1 (thickness / drill diameter) significantly increases plating difficulty and crack risk. For Class 3, we limit aspect ratio to 8:1 for PTHs. For stacked microvias, the limit is 1:1 per IPC‑6018.
Continue Reading – Thermal Reliability Series
Concerned about PTH reliability in your high‑cycle application? Our engineers will review your stackup, recommend test plans, and help interpret microsection results per IPC‑6012 and IPC‑TM‑650.
Request PTH Reliability Review →References: IPC‑TM‑650 2.6.8, 2.6.7.2; IPC‑6012E §3.6.2.1, §3.6.2.2, Table 4‑2; IPC‑A‑600J §3.3.8, §3.3.9; IPC‑6018 §4.4.3. IPC‑TM‑650 Official Method.
Get Custom Quote
PRODUCTS CENTER


