A163 | Thread Tension and Time — What Sifu Chen Has Been Teaching for Thirty-Two Years

Sifu Chen started at this factory in 1994. She was twenty-three years old, had just finished a two-year technical training programme in Fujian, and knew enough about weaving to operate a standard shuttle loom without supervision. She did not know, she told us recently, anything worth knowing. What she knew then was the procedure. What she knows now, thirty-two years later, is the loom.

The difference between those two things is what this post is about.

What Procedure Teaches You

Procedure teaches you what to do when things are working correctly. You thread the warp through the heddles in the right sequence. You set the shuttle tension to the specification in the job sheet. You watch the selvedge and check the pattern repeat at the intervals the quality system requires. If everything is working correctly, this is sufficient. You will produce good ribbon.

The problem is that things are not always working correctly, and the procedure does not tell you what to do when they are not. A warp thread has been threaded with slightly more tension than its neighbours — not enough to trigger the automatic tension alarm, but enough to produce a selvedge that will start pulling inward after about forty minutes of running. A new shuttle has a slightly different weight distribution than the one it replaced, and the loom is throwing it with just enough inconsistency to produce a subtle pattern irregularity that appears every forty-third pick. The dye lot on the weft is within specification but at the cooler end of the acceptable range, and combined with a particular warp colour it will produce a finished ribbon that looks correct on the inspection table but reads differently under the buyer's showroom lighting.

None of these issues are in the procedure. The procedure is not wrong — it describes what to do when the inputs are as specified. But in a real factory, the inputs are almost never exactly as specified. They are within tolerance, which is not the same thing.

What Sifu Chen Teaches

Sifu Chen currently has three apprentices at various stages. The youngest has been with us for eight months. The most senior has been working alongside Sifu Chen for two and a half years and is approaching the point where she will be assessed for independent operation.

We asked Sifu Chen to describe what she tries to teach. She thought about this for a moment and said: "I try to teach them to read the loom." We asked what that meant. She said: "When you can read a loom, you know what it is about to do before it does it. You hear something change in the sound, or you see something change in how the warp is lying, and you already know what the problem will be in ten minutes. Then you fix it before it becomes a problem."

This is not something she can demonstrate in a classroom. It is learned by standing next to a running loom for a very long time, watching and listening, with someone next to you who can say: do you hear that? That is the sound the loom makes when the left-side shuttle spring is starting to go. You would replace it in three hours before the end of the shift, not tomorrow. The knowledge transfers through proximity and commentary, over months and years, and there is no shortcut.

The Selvedge Question

One of the things Sifu Chen teaches early is how to read the selvedge — the edge of the ribbon as it comes off the loom. A correct selvedge is consistent, lies flat, and does not pull. A selvedge that is beginning to pull inward tells you something about the warp tension on that side. A selvedge that is slightly loopy tells you something different, about the weft tension or the shuttle timing. To an untrained eye these are minor cosmetic variations. To someone who can read them, they are early warnings of problems that, if uncorrected, will produce defective ribbon twenty or thirty minutes later.

Sifu Chen's most experienced current apprentice can now read the selvedge reliably. The youngest cannot yet. This is not a reflection of ability — it is simply a reflection of how long each person has been watching. The youngest has eight months. Reading the selvedge, Sifu Chen says, comes after about eighteen months of consistent attention. It cannot be rushed.

A Conversation Worth Having

We sat with Sifu Chen for about an hour last month, talking about her work and what she has learned. At one point we asked her whether she thought the knowledge she carried could be lost — whether there was a version of the future where it simply was not there anymore. She did not dismiss the question. She said she had seen it happen at other factories she knew of: a senior weaver retires, the apprenticeship was not maintained, and within a few years the quality of a particular product line quietly declines. The managers know something has changed but cannot name it. The knowledge has not been lost in a single moment; it has simply not been passed on, and the gap only becomes visible when it matters.

She said this without particular drama. It was a statement of fact, not a warning. Then she said: "That is why I come in at six." She did not explain further. She did not need to. The jacquard section starts at six, and Sifu Chen has been there at the start of every shift for more than thirty years, reading the loom, teaching whoever is standing beside her to hear what she hears. That is the work. That is how it continues.

What the Factory Owes This Kind of Knowledge

We have thought about how to document what Sifu Chen knows. We have talked to her about it. She is willing, but she is also realistic. "You can write down that the loom sounds different when the spring is going," she said. "But writing it down does not help someone who has not heard the difference. They read the note and they think they understand. Then they hear the sound for the first time and they think: oh. That is different from what I expected." The gap between the written description and the actual experience is where the knowledge lives.

What the factory owes this kind of knowledge is not a documentation project — it is an apprenticeship programme that keeps the knowledge alive by ensuring that the transfer is always happening. Sifu Chen is fifty-five. She plans to work for another ten years. In that time, if the apprenticeship programme continues at its current pace, she will have transferred her knowledge to six or seven people. That is how the knowledge survives.