Skipped stitches and seam failure after a needle change on a 4-thread overlock
A leggings line runs a 4-thread overlock (2 needles, 2 loopers) at 5,500 stitches/min on 220 gsm cotton/spandex single jersey. Last week the needle was changed from 14/90 to 11/75 because the buyer complained about needle holes at the waist seam. Since then: skipped stitches appear on roughly 3 in 10 pieces, mostly in the first 20 mm of the seam and on the gusset curve; the cut edge curls and the seam grins when pulled across; the needle threads sit loose on the face side. Settings: stitch length 3.0 mm, differential feed 1.0, needle thread 2/60 spun polyester, both loopers 2/120. A stretch test on 5 waist seams cracks at 11-13 kgf; the tech pack requires at least 15 kgf and 80% elongation before the seam cracks. The mechanic's first response was to tighten both needle threads two turns and raise the speed to 6,000 to pull the seam tight. Output is 620 pieces per shift against a target of 750.
1. Rank the probable causes of the skipped stitches and the low seam strength, and state which single change is most likely responsible. Justify each rank with evidence from the setup.
Read the worked answer
Rank 1 - wrong needle type and size (the needle change itself). A 220 gsm cotton/spandex knit needs a medium ballpoint (SES/SUK) needle. A sharp point severs knit loops, which explains the curling edge and, ironically, the needle holes the buyer complained about; an 11/75 needle carrying 2/60 thread in a 4-thread overlock is also light for the load, so it flexes and the needle loop is formed late or not at all. Skipped stitches are the direct result.
Why the skip pattern supports this: skips cluster at seam start (no fabric support under the needle, maximum deflection) and on the gusset curve (side load bends the needle). Thread-path faults usually skip randomly, not at these two places.
Rank 2 - speed raised to 6,000 spm. Higher speed increases needle deflection, vibration and needle heat, and shrinks the tolerance window for loop timing and take-up.
Rank 3 - differential feed 1.0. On a stretch knit the seam is extended while being sewn and springs back afterwards; that gives grin, edge curl and lost elongation even when every stitch is formed.
Rank 4 - two extra turns of needle tension. This masks the symptom, puckers the seam and reduces elongation; it cannot create a needle loop that was never formed.
The drop to 11-13 kgf is a consequence, not a separate fault: one skipped stitch breaks the interlock chain, so the seam fails at the weakest stitch rather than at the average.
2. Give the corrective sequence you would follow, in order, and state how you would verify each step with a measurable test before releasing the line.
Read the worked answer
Sequence
- Record the current settings (tensions, stitch length, differential, speed) on a settings sheet so the machine can be returned to a known state.
- Fit a 12/80 medium ballpoint needle as a compromise between needle holes and loop formation; check needle bar height, needle orientation per the manual, and inspect the needle plate hole and feed dog for burrs.
- Re-thread from scratch and reset all tensions to the standard recipe rather than leaving the two-turn correction in place; check the thread path and take-up spring.
- Set stitch length 2.5-3.0 mm and differential feed 1.2-1.3 so slightly more fabric is fed than the needle line takes; reduce presser foot pressure.
- Run trials at 4,500-5,000 spm before returning to full speed.
Verification - each step must produce a number, not an opinion
- Skips: sew 30 consecutive waist seams; target 0 skipped stitches (acceptance: 0 in 30).
- Strength: 5 seams on the tensile tester, crack at 15 kgf or more and elongation 80% or more.
- Appearance: inspect needle holes on a light box after pressing; seam grin measured as seam opening under a 5 kgf load.
- Stability: 3 home washes, seam growth not more than 5%.
- Only after all four pass should the machine go back to 5,500 spm, with the first 100 pieces 100% inspected.
3. The buyer still complains about needle holes. Explain what you would tell the buyer and the production manager about solving needle holes without going back to conditions that cause skipped stitches.
Read the worked answer
The two complaints are not opposite - they are the same physics. Needle holes come mainly from point shape and needle heat, not from needle size alone. A sharp point cuts filaments and the hole never closes; a ballpoint pushes filaments aside so the hole closes after washing and pressing. So the correct answer to the buyer is not a finer sharp needle but the right point on an adequate size.
What I would say to both:
- We caused the original complaint by using the wrong point, and we made it worse by fitting a needle too fine for 220 gsm fabric with 2/60 thread.
- If holes persist with a 12/80 ballpoint, check, in this order: needle plate hole and feed dog burrs, needle heat from running at 6,000 spm, excessive presser foot pressure, and stitch density. These mark or perforate the fabric without any needle being too large.
- We will submit a needle-hole panel: 10 pieces at 11/75 ballpoint and 10 at 12/80 ballpoint, pressed and washed once, photographed on the light box, with a signed golden sample before bulk.
To the production manager: tightening tension to hide skips is not an option - it puckers the seam and costs elongation, and if the seam fails at 11 kgf in the field the claim cost is far higher than a needle change. Approve the trial panel and the settings sheet, and let the numbers decide.