You only get one good name

Letting fake chips slip into production can torch your reputation overnight. You end up buried under endless warranty claims nobody pays for and a legion of seriously pissed-off customers.

Not long ago the entire electronics world was choking on component shortages. Legit distributors quoted lead times stretching 6–12 months. Real choice was simple: shut down production for a year… or hunt for alternatives and roll the dice.

Case Study – The TC4428 Smoke Show

I urgently needed TC4428ACOA gate drivers from Microchip. Official channels? 36 weeks. I could hang a “closed” sign and wait like a saint. Asia had stock ready to ship basically tomorrow. Risk was insane, but you know the saying: no risk, no champagne. Ordered several hundred to keep the line alive while the real order crawled.

Parts arrived fast enough. Started production. Whole batch stuffed, time for power-up and testing.

Power → flash firmware.
Board 1 – good.
Board 2 – good.
Board 4 – smoke. Set aside.
Board 5 – good… then more smoke.

Checked assembly to avoid torching the rest pointlessly. Nothing found. Out of first 20 boards – 3–4 dead. Full stop on the batch until root cause is found. Since the suspect part came from a dodgy source, diagnosis was quick: counterfeits.

Already-populated boards needed every single counterfeit desoldered and replaced with genuine ones. Hand rework is nowhere near SMT line speed. Time to fix two chips ≈ time to build one new board – except material + labor cost doubled. No matter how you look at it – pure loss.

Cheap fakes almost always cost way more (rework, lost time, reputation damage, angry returns)

Why it’s so damn easy to get screwed

Classic scam: blacktopping. Sand off original markings, apply fresh black polymer coat, laser-etch fake Microchip logo and part number.

Why didn’t I catch it sooner? Parts came on tape reels that are fed straight into pick&place feeders. Dumping them out for manual inspection would mean bent pins, feeder jams, misalignment nightmare.

Risk Assessment – most common counterfeit tricks

Before we discuss the practical checks you can run in your own lab, here’s what you’re usually up against:

• Ghost parts – semiconductor dies of unknown origin, original markings removed, new fake ones applied
• Recycled / refurbished – pulled from scrap boards, cleaned, relabeled
• Empty packages – plastic body + leads only, zero silicon inside
• Cloned dies – reverse-engineered (or stolen) layouts fabbed in dirty labs: thinner layers, defects galore, bond wires that don’t match spec, poor metalization

Minimizing Risk – Practical Steps

Supplier Selection & Verification

Rule #1: Never buy anything you can’t realistically return.
Chinese borders are a one-way osmotic membrane – goods flow out easily, almost never come back. Sellers will do anything to avoid returns.

Helped one company once with a multi-thousand-dollar fake return fight. Shipped junk back… and they actually refunded us. Few months later the same box showed up on our doorstep again – seller wrote it off as cheaper than customs hassle. True story.

How to Order Without Shooting Yourself in the Foot

Quantity
Buy a buffer so you have enough for production + a few to destroy / test.

Packaging: always confirm packing method. Some just dump loose parts into bubble envelopes. After 2 weeks transit – pins look like modern art. No feeder handles kilo-bags of bent chips.

LPT: Keep a few verified genuine chips as your golden sample. Your forever reference for visual & dimensional checks.

Incoming Inspection – Visual Check

Grab samples immediately and hit them under stereo microscope. Compare among themselves first, then vs golden sample. Look at:

• Index marks (dimples / notches) – must be consistent across batch. Sanded fakes have shallow/small mixed with deep/large. Shape & size must match golden / datasheet?
  

  

  

  

• Bottom side – many packages have circular mold mark or logo recess. Present & correct?

  

• Pin geometry & finish – especially power parts. Look at lead trim & form. Fakes often have sloppy cuts or weird angles – obvious side-by-side.

• Logo geometry – crooks mangle fonts & proportions. Overlay real logo on screen and compare curves, spacing, thickness.

• Date code – must make sense. Don’t buy “new” 2026 parts with 2018 codes.

• Molding compound – naked eye or USB cam won’t show it. Stereo scope reveals different color, texture, finish vs genuine. Here’s the photo anyway:

  

• Package dimensions – micrometer on thickness. Sanded = thinner. Also check body width/length – different mold = different dims.

Blacktopping / Marking Permanency Test (Acetone)

Cheap, fast test that catches most sanded + remarketed fakes. Non-destructive for the tested sample. Costs pennies, can save thousands.

Put a drop of acetone on small area (side or corner), rub gently with cosmetic stick 10–20 seconds. Check if:

• markings smear or change color
• stick picks up dark/black residue
• surface turns matte or texture changes
• worst case – old original marking appears underneath

For tougher coatings: warm acetone bath (~50 °C) in beaker with magnetic stirrer.

Passing the test does NOT mean 100% genuine.
It’s just one filter – but a very good one.

Price is no protection

Think paying “market price” guarantees real parts? Wrong.

Another real case: company buyers scored a hard-to-get chip at catalog price. Showed me one piece – spotted fake with naked eye. Someone dumped counterfeits on the market at original price. Their supplier probably got burned too and accepted return… I wonder where those chips ended up? Probably in production at places with zero incoming inspection.

Aftermath

Always buy from trusted, authorized distributors.
Lead times suck, prices hurt – negotiate volume deals, be patient. At least you sleep at night.

Buying from unverified sources is playing Russian roulette. Don’t do it unless you have no choice – and if you must, NEVER leave it to purchasing alone. Buyers are salespeople, not engineers. Get an expert consultation to vet suspicious sources.

Build on bad components → ship bad products → destroy your reputation

I shared a few real stories and the simplest checks that actually work in a small lab. Do you already run similar routines? How effective are they in practice? Or is this whole counterfeit nightmare still new to you – and now you realize how critical it really is?

Drop a comment – maybe it’ll spark the next article.