Tin whiskers on PCBs - causes of formation, experiences

megao 3144 34

TL;DR

A PCB assembly with two 0.5-mm-pitch ICs developed tin whiskers on the input leads, causing unexpected malfunctions after eight months of lead assembly.
Microscope inspection traced the failure to whiskers bridging the semiconductor actuator input pins; re-soldering the leads cleared the fault.
The pitch was 0.5 mm, with about 0.2 mm between leads, and the whisker width was estimated at 30-40 µm.
One whisker had already shorted a lead, another was growing, and the device worked properly again after re-soldering.
Possible causes considered include excess solder paste, poorly mixed paste, printer contamination from unleaded paste, and mechanical stress, but none were verified.

Treść została przetłumaczona

Zobacz oryginalną wersję tematu

Report a violation of the law

Reply Cool? Ranking DIY | New topic

Notify about new articles

📢 Listen (AI):

#31 21753838 17 Nov 2025 03:42

megao megao

Level 25

» | Topic author Helpful post? (+1)

Post #31
21753838 17 Nov 2025 03:42

While looking for information, I came across the monthly magazine SMT Magazine - Engineering Solutions for PCB Manufacturing .
In the February 2015 issue there are several articles treating tin whiskers.
A typical trade publication, rather for production technologists.
They still have a monthly PCB, Design section.
I am posting this as a curiosity.
ADVERTISEMENT
#32 21754172 17 Nov 2025 13:45

Homo_toxicus Homo_toxicus

Level 27

» | Helpful post? (0)

Post #32
21754172 17 Nov 2025 13:45

I had no idea about this phenomenon. But the occurrence of "tin whiskers" would explain why a non-working computer started to work after the motherboard and components had been "rinsed out" and vacuumed. I have never given it much thought, but more than once I have experienced situations in which equipment disassembled to the last screw, blown out with a compressor or "dusted off", then put back together on a table, started working and after putting everything back together again, worked as if nothing had happened. Even with a microscope, I would not have been able to find such "whiskers" among the dust that often lingers in enclosures.
ADVERTISEMENT
#33 21754210 17 Nov 2025 14:35

cranky cranky

Level 30

» | Helpful post? (0)

Post #33
21754210 17 Nov 2025 14:35

Homo_toxicus wrote:
equipment disassembled to the last screw, blown out with a compressor or "vacuumed" and then put together on a table, started to work

Well, there are various reasons for this. Disc slots, even gold-plated ones, can become so oxidised/fogged up that sometimes a little spirit can do wonders.
ADVERTISEMENT
#34 21754217 17 Nov 2025 14:43

Homo_toxicus Homo_toxicus

Level 27

» | Helpful post? (0)

Post #34
21754217 17 Nov 2025 14:43

cranky wrote:
....No, there are various reasons for this. Disc slots, even gold-plated ones, can become so oxidised/fogged up that sometimes flushing them with spirit works wonders.

I completely agree. Such a total disassembly is an opportunity to wash out card or memory connectors. But I'm inclined to think that the board didn't boot, because even a bare board with only the processor and memories didn't boot and the speaker (bios) was silent. Only the fans were spinning.
#35 21754249 17 Nov 2025 15:28

aadeer aadeer

Level 17

» | Helpful post? (0)

Post #35
21754249 17 Nov 2025 15:28

Homo_toxicus wrote:
I had no idea about this phenomenon

I was treating it as a production/assembly error, the NASA website was viewed with care. I think I've seen it in older equipment too, but at the time I suspected it was more like filings and dust.
Create an account, log in here. You will receive points by participating in discussions.
Join this discussion.

Install Elektroda application

Reply Cool? Ranking DIY | New topic

Notify about new articles

📢 Listen (AI):

Report a violation of the law

Home page
/
Forum
/
Articles
/
Tin whiskers on PCBs - causes of formation, experiences

FAQ

TL;DR: In 0.5 mm‑pitch ICs with ~0.2 mm clearance, tin whiskers caused test failures after 8 months; “Tin whiskers are spontaneously growing, conductive tin crystals.” [Elektroda, ElektrodaBot, post #21749854] Why it matters: This FAQ helps hardware designers and repair techs diagnose, prevent, and fix tin‑whisker shorts on modern fine‑pitch PCBs.

Quick Facts

High‑risk geometry: 0.5–0.4 mm pitch with ~0.2 mm clearance between leads. [Elektroda, ElektrodaBot, post #21749854]
Whiskers can grow without moisture or bias and even in vacuum. [NASA NEPP Tin Whisker]
Primary mitigations: Ni barrier or NiPdAu finishes, SnPb, matte Sn vs bright Sn, conformal coating. [Elektroda, ElektrodaBot, post #21749854]
Typical accelerated evaluation: 30–60°C storage, 60–90% RH, plus thermal cycling and microscopy. [Elektroda, ElektrodaBot, post #21749854]
Acceptance/testing references often cited: JEDEC JESD22‑A121 and JESD201. [Elektroda, ElektrodaBot, post #21749854]

What are tin whiskers, in plain terms?

Tin whiskers are needle‑like, conductive crystals that spontaneously grow from electroplated tin, especially pure tin finishes. They can bridge adjacent leads, creating intermittent or permanent shorts, and in high‑energy circuits may trigger arcing. “Tin whiskers are spontaneously growing, conductive tin crystals.” [Elektroda, ElektrodaBot, post #21749854]

Tin whiskers vs dendrites — what’s the difference?

Whiskers are solid‑phase tin filaments that can grow without voltage or moisture. Dendrites are tree‑like deposits from ionic migration, needing moisture and an electric field. Treat them differently: whisker risk is about metallurgy and stress; dendrite risk is about residues and leakage paths. [Elektroda, ElektrodaBot, post #21749854]

What actually causes tin whiskers to form?

Compressive stress in tin layers drives growth. Sources include Cu‑Sn intermetallic growth at the Cu/Sn interface, internal plating stress (bright tin), thermal cycling, and mechanical abrasion. Limiting those stresses and adding diffusion barriers reduces risk. [Elektroda, ElektrodaBot, post #21749854]

How fast or how far can whiskers grow?

Lengths vary from micrometers to millimeters over months to years. Importantly, growth does not require applied bias or humidity and has been observed in vacuum, which complicates field prediction. [NASA NEPP Tin Whisker]

My board failed after 8 months—could whiskers be the reason?

Yes. A real case in this thread showed shorts on two 0.5 mm‑pitch ICs after ~8 months of storage/use. Under microscopy, 30–40 µm whiskers bridged adjacent leads; re‑soldering cleared the fault. [Elektroda, megao, post #21749760]

How do I check for tin whiskers on fine‑pitch parts quickly?

Use a stereomicroscope and inspect above the solder meniscus on leads, between pins, and near tin‑finished shields. Three‑step check: 1. Power off and clean the area. 2. Inspect at 20–40× for filaments or nodules. 3. If found, remove and re‑solder the leads, then re‑inspect. [Elektroda, ElektrodaBot, post #21749854]

What’s the best prevention for new designs and builds?

Avoid pure tin on component leads. Prefer Ni/Sn (matte), NiPdAu, or SnPb where permissible. Choose ENIG/ENEPIG over immersion tin for PCB finish. Minimize mechanical stress, avoid abrasions, and add conformal coating in critical fine‑pitch areas. [Elektroda, ElektrodaBot, post #21749854]

Does reheating the PCB permanently solve whiskers?

No. Heating can melt existing whiskers, but it does not remove the stress sources that drive new growth. “Simply re‑baking removes whiskers only temporarily—they can grow back.” Use metallurgical fixes (barriers, SnPb, retinning). [Elektroda, ElektrodaBot, post #21749854]

Do conformal coatings help, and which ones?

Coatings are a barrier, not a cure. Parylene C provides a strong mechanical barrier; thicker polyurethane or silicone helps. Thin acrylics can be pierced. Target ≥50 µm over fine‑pitch ICs and high‑consequence nets. [Elektroda, ElektrodaBot, post #21749854]

Are lead‑free solders the root cause?

Lead significantly suppresses tin whiskers, so pure‑tin finishes and SAC joints carry more risk. That said, whiskers often originate in the component lead finish above the solder meniscus, not the bulk joint. [Elektroda, ElektrodaBot, post #21749854]

Could the component lead finish be the source of growth?

Yes. Whiskers frequently emerge from tinned component leads above the meniscus. Mitigate by specifying NiPdAu or using hot solder dip (SnPb) retinning services for critical parts. “Cleaning or Coating?” explains tradeoffs. [Tin Whisker Mitigation Strategies: Cleaning or Coating?]

Do ionic contaminants or flux residues matter?

A cited study shows ionic contamination on leads can correlate with whisker initiation. Clean supply chains, verify cleanliness, and avoid mixing paste chemistries across jobs to limit residues that complicate risk. [Elektroda, megao, post #21752728]

Which PCB finishes reduce risk the most?

Prefer ENIG or ENEPIG for reliability. Avoid immersion tin on high‑reliability products. If you must use tin, pick matte tin over bright tin and include a Ni barrier where possible. [Elektroda, ElektrodaBot, post #21749854]

What standards or tests should I reference?

Use JEDEC JESD22‑A121 for test methods and JESD201 for acceptance. For mission‑critical builds, consult GEIA‑STD‑0005‑2 and NASA‑STD‑8739.21 mitigation guidance and run accelerated storage/thermal cycling with microscopy. [Elektroda, ElektrodaBot, post #21749854]

What failures should I expect in the field?

Expect intermittent resets, data errors, or permanent shorts. In HV or energy systems, a whisker can vaporize into metal plasma and sustain an arc—an edge case with severe consequences. [Elektroda, ElektrodaBot, post #21749854]

What is Hot Solder Dip (HSD), and when should I use it?

HSD is a controlled retinning process that replaces pure‑tin lead finishes with SnPb to suppress whiskers. Use it for critical parts lacking Ni barriers or when alternate finishes are unavailable. [Elektroda, ElektrodaBot, post #21749854]