16
Mr. Hao's factory was called West City Servo, making Servo motors. The factory wasn't large, with about eighty people. They needed this kind of chip for their boards, four thousand pieces a month. Previously, they always sourced them from the south at 4.2 yuan per piece. Last year, prices in the south rose to 5 yuan, so they thought about looking locally. After searching for a year, they found us.
On September 18th at 10:00 AM, Mr. Hao signed the trial order on a piece of printed paper he brought himself—an A4 size, two pages, with a signature column left at the bottom of the second page. He raised his hand and felt in his jacket pocket for a pen. The pen barrel was wrapped in tape, three turns of it, blackened with grime. The nib paused on the paper, tearing it slightly. He didn't change pens and continued writing: two thousand pieces, 3.4 yuan, fifteen days. Once the goods arrived and passed inspection, they would talk about four thousand.
"Four thousand, is it decided once the trial order passes?" I asked.
"It's decided once it passes. After it passes, four thousand a month for three consecutive months." He reached out to put away the pen, snapping the cap on twice before it clicked tight. "If nothing goes wrong for three months, we'll talk about a year."
"For a year, how much?" I asked.
"For a year, the price is still 3.4 yuan." He pushed the paper over. "The price won't drop, but the volume will increase."
Actually, Mr. Hao didn't sign this trial order for nothing; he also wanted to see if the local area could do it. Later, when he snapped the pen cap twice and said to talk about a year if nothing went wrong for three months, I roughly understood that the trial order was an admission ticket, not an order. But the admission ticket was in hand. Anyway, those three months had to be taken one day at a time, so not a single one of those two thousand pieces could be wasted. Those chips, piece by piece, were all statements.
After signing, Mr. Hao didn't leave; he wanted to inspect the production lines. He looked at the six machines one by one. When he reached the No. 3 machine, he stopped, patted the casing with his hand, and said this machine's sound was wrong. What was wrong? The other machines hummed with a "weng," but this one had a slight hoarseness mixed into the hum. He pressed his ear close to the casing, listened for half a minute, raised his head, and said that hoarse sound came back every so often. Engineer Qin walked over, pressed his ear against it as well, listened for half a minute, straightened up, and said it was the feeder.
"What's wrong with the feeder?" I asked.
"The clearance is too big." He brought over a dial indicator, dismantled the feeder of the No. 3 machine, and measured it after taking it down. Once measured, he handed the gauge to me to look at. "Two point five. A new one is two point zero, and one that has run for a year is three point five. Two point five is the reading for something that has run for half a year."
"Didn't we replace three of these six?" I asked.
"We replaced three—Numbers 1, 2, and 4. Numbers 3, 5, and 6 are still the old ones." He put away the dial indicator. "When we replaced them last time, the clearance for No. 3 was 2.3, so it was still usable. After using it for three months, it wore down to 2.5."
"How much does 2.5 affect things?" I asked.
"Missed pickups." He raised his hand and flipped the feeder over, pointing to a groove inside. "When the clearance is too large, the tape strays off-center. If it deviates even a little, the nozzle misses the pick-up. Every missed pick-up means one less piece. Three thousand pieces a day means thirty fewer pieces."
"Change to a new one?" I asked.
"Change to an old one." Engineer Qin dragged a cardboard box out from the warehouse. Inside the box were three feeders, all old ones wrapped in plastic bags with numbers written on them: 2.0, 2.1, 2.1. A new machine replaced with old parts? The old parts were worn in. He took out the 2.0 one, placed it on the workbench, and said new parts were just rough blanks that had to run for three months before they ran smoothly. These three had run for half a year, worn smooth, and their clearance stabilized at 2.0, making them even more accurate than new parts.
Then why not use old parts exclusively? There were too few old parts. He installed the feeder onto the No. 3 machine. While installing, he aligned the grooves first, then tightened the screws after the grooves were properly aligned—four screws, tightened one by one until they were seventy percent tight, then shook it with his hand. He said the old parts were picked out from over a dozen machines; out of three chosen, the rest were worn out and scrapped. Even scrapped ones could still be used; they were installed on the slow lines.
"What about this 2.5 one that was taken off?" I asked.
"Keep it as a spare part." He handed the removed one to me. "If another machine breaks down someday, we'll use this to tide us over for half a day." He wrapped up the 2.5 feeder using two layers of plastic bags, and after finishing, wrote the date and clearance on the bag. Things like this couldn't be placed just anywhere; if scattered around, they couldn't be found, and if they couldn't be found, they'd have to be remeasured, which would take half an hour.
From September 19th to September 28th, they ran the two thousand pieces. The two thousand pieces weren't run all at once; they were divided into six batches of a little over three hundred each. They inspected after running and recorded after inspecting. Three of the six batches were run at night because the electricity was stable at night and the voltage didn't fluctuate. During the day, the factory area consumed a lot of electricity, causing low voltage. Low voltage meant the furnace temperature couldn't go up, and if it couldn't go up, it had to be burned for an extra ten minutes.
The first batch was 340, with 322 good ones; the second batch was 330, with 316 good ones; the third batch was 320, with 308 good ones; the fourth batch was 340, with 326 good ones; the fifth batch was 330, with 318 good ones; the sixth batch was 340, with 334 good ones. Adding the six batches together came to two thousand, with 1,924 good ones. Sun Qian filled the numbers into the sheet, pressed her hand against the grid after finishing, and said 96.2%, which was higher than expected. The expectation was 90%, so it was 1.2 percentage points higher. 1.2 percentage points meant twenty-four pieces.
"Where did these twenty-four pieces come from?" I asked.
"They came back because the No. 3 machine was replaced with an old part." She tapped her pen on the numbers for No. 3. "Before No. 3 was replaced, it missed pickups thirty times a day; after replacement, it missed pickups three times a day. Twenty-seven times means twenty-seven pieces."
Seventy-six pieces were scrapped—sixty-one due to thick solder on the pins and fifteen due to faint printing. Was the thick solder due to high temperature? Wu Dayong laid out the scrapped chips, looking at them one by one. After finishing, he pressed his hand against the area with thick solder and said the temperature wasn't high; there was just too much solder. Too much solder meant the scraper pressure was too low during printing. Who adjusted the scraper pressure? Sister Liu said from the side that she adjusted it.
On the day the No. 3 feeder was replaced, she also adjusted the scraper. When adjusting, her hand was a bit heavy, so the pressure became lower, resulting in excess solder. Adjusting it back—she said she adjusted it back, and the fourth batch had fewer. She reached out to collect the scrapped chips and said the fourth batch had eight scraps, the fifth batch six, and the sixth batch four—fewer each day.
"Fewer each day, what does that indicate?" I asked.
"It indicates that the hands are learning." Engineer Qin placed his hands on the workbench. "Once the hands learn, the machine stabilizes; once the machine stabilizes, people make fewer mistakes."
Actually, this 96.2% wasn't calculated; it was won back by replacing the old part on the No. 3 machine. Later, when Sun Qian filled the numbers into the sheet and said it was 1.2 percentage points higher, I roughly knew that only when the machine was stable could people be stable. However, only when people were stable could their hands learn quickly. Anyway, there were fewer and fewer each day, which was why we dared to take on those six thousand pieces. Those scrapped chips were fewer and fewer piece by piece.
On September 29th, Mr. Hao came to inspect the goods. He was still wearing that blue gown and holding that same tape-wrapped pen. During the inspection, he looked at them piece by piece, taking half an hour. After finishing, he put the chips back into the box and said 1,924. I said 96.2%. He said 96.2% was higher than what his own factory could do. Later, he closed the box, secured the lid, pressed his hand against the box cover, and said their factory also made this, with their best year being 94%. Even at 94%, it was still done by veteran masters.
"Are the veteran masters still doing it now?" I asked.
"Yes, they're in their sixties, and their eyes aren't good anymore." He took out his pen, wrote in his notebook, tore off the paper after finishing, and handed it to me. "Starting on the first of next month, four thousand a month. Price: 3.4 yuan. Payment term: thirty days upon arrival of goods."
"Thirty days, can it be squeezed down to fifteen days?" I asked.
"Thirty days is the fastest I can give." He put away his pen. "Our factory's accounts are also thirty days. If I give you fifteen days, I'll have to shell out the money myself."
"Shelling it out yourself, can't afford to?" I asked.
"I can afford it, I just won't do it." He tucked the notebook under his arm. "Once you shell out for it, you'll have to do it from then on. Once you get used to it, you stop taking it seriously."
He walked to the door, turned back, and said four thousand pieces, calculated from the first of next month; the first batch would be delivered on the 15th. He walked to the door again, stopped, and said wait a minute. He took the notebook out from his bag again, flipped to the page he had just written on, crossed out "four thousand," and wrote two characters next to it: "six thousand." I asked, six thousand? He said six thousand. He turned the notebook around to show me, saying our rate was 96.2%, while the best year for their own was 94%. The extra two thousand pieces wouldn't be given to another company; he was giving them to us.
"Six thousand, the six machines can't run that many." I said.
"If they can't run them, we'll slowly increase it." He put away his pen, snapped the cap on twice until it clicked tight. "Until you can't run them anymore."
After Mr. Hao left, Engineer Qin pinned that trial order to the wall, right beside the comparison chart. While pinning it, he pressed his hand against the paper, flattening it before hammering the pin in. After finishing, he stepped back to look at it, and then put the hammer back in the drawer. Sun Qian wrote the production schedule for the six thousand pieces in her notebook. After finishing, she calculated it three times; all three times came out to six thousand a month, two hundred a day. Two hundred a day distributed across six machines meant thirty-three per machine. She said thirty-three per machine was eleven more than before. Eleven wasn't many, but it was steady.
