Starting from scratch in the new department: Without external support, we can only pave the way for the process with problems
The newly established Liquid Crystal Module Department was like a newly born baby - there was no ready - made SOP (Standard Operating Procedure), no large - customer cases to refer to, and not even stable orders. Every day, we dealt with small domestic orders of just a few dozen pieces, with requirements so lenient that "as long as the screen can light up" was acceptable. We were like a group of people groping in the dark to find a light. Today, we solved the problem of misaligned IC mounting; tomorrow, we fixed the issue of curing bubbles. We had to write down the day's problems in a crumpled notebook before getting off work, and then gather in the meeting room on weekends to look through the notebooks. "Yesterday's misalignment was because the fixture wasn't fixed firmly. Next time, we should add a locating pin." "Today's bubbles were due to insufficient temperature. We need to raise the temperature of the curing furnace from 175℃ to 180℃." At that time, there was no "correct answer", only "try and see" - because all processes were built on "piles of problems", and all experiences were accumulated through "trial and error".
Receive modern orders: Hit the "quality iron wall" of world - class customers for the first time
The turning point came unexpectedly. The company said it would take an order from Hyundai Motor Company, and our small department, which had been established for only half a year, was supposed to take the lead. What does Hyundai represent? It is a world - class automotive parts supplier that demands "zero defects" in the quality of liquid crystal displays. The accuracy of IC mounting must be within ±0.05mm (previously, it was ±0.1mm for us). The curing pressure of ACF (anisotropic conductive film) needs to be precisely controlled at 3MPa ±0.5MPa (previously, a fluctuation of 1MPa was tolerated). Even the temperature and humidity of the production environment have to be maintained at 25℃ ±2℃ and 50%RH ±5% (previously, the temperature in the workshop could reach 30℃ in summer, and the humidity could drop to 30%RH in winter).
Some people say, "We have fully automatic COG machines (chip mounters) imported from Japan. What are we afraid of?" But only when you actually start working with them do you understand: The "diamond drill" is just a tool. Whether you can use it and use it to the "utmost" is the key to taking on the "porcelain work." For example, regarding the "visual positioning system" of the COG machine, we only used its basic functions before. As soon as Old Jin (the modern auditor) came, he pointed at the screen and said, "Your positioning accuracy is set too coarsely. You need to reduce the pixel error from ±2px to ±1px." - Just for this one adjustment, it took us three days and we tried the machine parameters back and forth 20 times.
The "devilish review" of the South Korean old man: Twist the "commonplace" into an "untouchable red line"
The auditor sent by the modernist school is 52-year-old Kim Seong-soo. He wears a navy blue suit and always holds a black hardcover notebook in his hand. He doesn't "reason" with you; he only looks at the "results":
- What we consider normal workstation cleaning (wiping once a day). He squatted on the ground and rubbed the workbench with a white glove. The glove got dusty, and he immediately crossed out qualified. If you wipe it every two hours, the dust will stick to the ACF, causing poor conductivity.
- We thought the IC mounting (±0.1mm) was qualified, but he took out the modern standard gauge and said, With a deviation of 0.05mm, it's a waste product – if the car screen freezes, it can cost lives.
- Even for the shelves where we store tools, he required that they be sorted by frequency of use: The most commonly used tweezers should be placed 15 cm to the right to reduce wasted movements.
The most headache - inducing thing is his "problem list". Every day before getting off work, he would list the day's problems on an A4 sheet. From equipment parameters to personnel operations, and from material storage to expiration management, no single detail could escape his notice. For instance, he found that the temperature display of our ACF refrigerator showed "5℃", but when actually measured with a thermometer, it was "8℃". "The storage temperature of ACF must be between 0 - 5℃. Exceeding 5℃ will cause premature resin curing. You're just wasting materials," he said.
"Implicit understanding in action" despite the language barrier: Replace communication with results and turn "not understanding" into "being able to do"
Communication is the biggest obstacle. Old Man Kim can only speak Korean, and my English is so poor that I have to think for two seconds before saying "Thank you". The sentences translated by the translation app often don't make any sense. Once, he pointed at the ACF volatiles in the microscope and pounded the table in desperation. I couldn't understand what he was shouting about. All I could do was rely on the engineer next to me to guess: "Is he saying that the mist-like substance will corrode the IC?"
Slowly, we developed a tacit understanding in actions: when he pointed at the microscope, I knew I had to adjust the ACF placement position; when he drew a line on the temperature and humidity curve in his notebook, I would go to check the air - conditioner settings; when he picked up a scrap product, I would look through the production records to find the cause. It was only then that I understood: in quality management, communication has never been about words, but result - oriented actions — no matter whether you understand what he says or not, as long as you can solve the problems he points out, it's right.
ACF's "Hazy Puzzle": Pushing "What's Taken for Granted" to "Precise Standards"
What Old Man Jin struggles with the most is the "foggy substance at the edge of the ACF". Let's first clarify what ACF is - it is the "nerve bridge" of the liquid crystal module: it sticks the IC (integrated circuit) and the LCD (liquid crystal display) together while enabling electrical conduction between the two. The temperature and pressure must be precisely controlled during the curing process, and even a slight deviation can cause problems.
Previously, when we inspected ACF, we only looked at three points: "Is the alignment accurate?", "Are there any air bubbles?", and "Is the curing sufficient?" As for the foggy stuff on the edges? "It's normal volatilization during curing. Don't worry about it." - This was our "accepted norm". But Old Man Jin said, "This isn't volatile matter. It's the incompletely cured resin in the ACF. Over time, it will corrode the electrodes of the IC."
To verify, I conducted a "half - pasting test": I found an IC. I pasted ACF all over one half of it and only one - third of the other half. After curing it according to the normal process, I observed it under a microscope - there were foggy substances on both sides. I thought I could "convince" him, but he took out a modern "Failure Analysis Report". On it was an IC that had been used for half a year, with its electrodes corroded and pitted. Next to it was written, "The foggy substances are epoxy resin monomers, which will gradually penetrate into the electrode plating, resulting in poor contact."
Finally, he set a rigid standard: the length of the ACF extending beyond the edge of the IC should not exceed 0.5mm.
The "cost" of this standard was immediate: In the first two weeks, we scrapped 327 products. The production line team leader came to me to complain, saying, "Previously, products with a deviation of more than 1 mm were still considered qualified, but now those with a deviation of 0.6 mm have to be discarded. The efficiency has dropped by 20%!" But Old Man Jin retorted with just one sentence, "Efficiency means 'doing things fast', while quality means 'doing things right'. It's better to do things 10% slower but get 100% right than to do things 10% faster but make a 5% mistake."
There's no other way. We had to "modify the equipment". We adjusted the accuracy of the "visual positioning system" of the COG machine from ±2px to ±1px and added an "ACF length sensor" to the fixture. It automatically measures the excess length for each piece pasted. After a week of debugging, we finally improved the positioning accuracy of the ACF from ±0.2mm to ±0.05mm, and the scrap rate was reduced to below 1%.
More importantly, this standard has directly reduced our "IC corrosion defect rate" from 1.2% to 0.03%. To this day, this "0.5mm" standard is still written on the first page of our "ACF Operation Instruction Manual" and has become an "untouchable red line".
From feeling the stones to taking the right path: The core of quality upgrade is reexamining the familiar
Half a month later, we finally passed the modern review. But more importantly, Old Man Jin gave us a lesson: the essence of quality management is "the subversion of the commonplace" - what you think is "normal" may be a "hidden danger"; what you think is a "problem-free" detail may be a "fatal flaw".
Later, when we received orders from Samsung, Panasonic and Delphi, we never panicked again:
- During the Samsung audit, when the auditors saw our "ACF Control Standard", they said, "Your standard is even more detailed than ours. Did Hyundai teach you this?"
- The auditor from Delphi went through our "Problem List" and said, "This is the'veritable archive' of quality management - each problem is a'milestone of growth'."
Looking back now, that half - month review was like a "secluded cultivation in martial arts training". Before, we were like those who only knew "flashy kung - fu" and could only "solve problems". After that, we had "profound internal strength" and learned to "prevent problems". The so - called "quality system" is nothing more than turning "problems" into "standards" and turning "what is taken for granted" into "the ultimate precision down to 0.1mm and 0.1℃".
Finally: There is no "end" but only a "starting point" for quality
Nowadays, our department is no longer a "small team". Our orders cover the world's top 10 automotive parts suppliers. However, I always remember what Old Man Jin said: "There is no end to improvement. Today's 'utmost' might become tomorrow's 'passing mark'."
For example, regarding the "foggy substance" in ACF, later we made an upgrade: We changed the "stepwise temperature increase" in the curing furnace to "linear temperature increase" to make the resin curing more complete; we added "humidity monitoring" to the ACF refrigerator to prevent moisture absorption. Now, the foggy substance is almost invisible, but we still retain the standard of "0.5mm" because it is the "starting point of quality".
Quality management has never been "done for the customers", but "done for oneself" — every time you are serious about the details, it is a way of "taking responsibility" for the products and "upgrading" yourself.