Handles That Peel at Three Kilos: A Joint Problem, Not a Glue Problem
A Divyang students' unit has an order for 25 shopping bags for a local shop. The body is stitched from old denim. Each handle is made from two used tyre-tube strips, 25 mm wide and 180 mm long, overlapped by 30 mm at each end and joined with a solvent-based rubber adhesive plus two aluminium rivets. The strips were wiped with kerosene, left for five minutes, spread with adhesive, left for two minutes and then pressed by hand. Ten bags went to the shop three days ago. Four handles have already peeled apart where the overlap begins, and a test on one handle gave way at 3.2 kg. The shop expects a bag to carry 8 kg. Six students make the bags; one has a weak grip in both hands and cannot squeeze a hand roller for long. A heat gun is available but the workshop has no fume extraction, and the only test equipment is a spring balance, a hook and a strong rail.
1. Diagnose why the handles peeled at 3.2 kg when the design load is 8 kg. Identify at least three process faults from the description.
Read the worked answer
The failure is at the overlap end, so it is a peel failure of a contaminated, badly bonded joint.
- Surface preparation is wrong. Tyre tube has talc, mould-release agent and road grime in the rubber skin. A kerosene wipe dissolves some of it and then leaves an oily film; kerosene is a solvent, not a cleaner, and it must be followed by abrading to fresh rubber and a clean dry wipe. The chalky surface mentioned is the release agent still sitting there.
- Open time is far too short. Contact adhesive needs to become tacky before joining, typically 10-15 minutes at this humidity, not two minutes. Joining while still wet traps solvent, gives a weak film and lets the joint creep under load.
- Pressure is too low. Hand pressing gives a fraction of the pressure a roller or a clamp gives, and pressure is what makes a contact adhesive wet the surface. With one student unable to squeeze a roller, the batch is inconsistent.
- Rivets at the wrong place. Two rivets at the very ends create a stress concentration and a peel crack that starts at the overlap edge; rivets need backing washers and should sit away from the edge, with the adhesive carrying the shear.
Why 3.2 kg: the joint is carrying peel rather than shear, so it fails at a fraction of the design load. The bag load must be turned into shear along the handle, and the overlap must be long enough (40 mm, not 30 mm) for the shear area.
2. Recommend the joining method and a step-by-step process that the workshop can actually run.
Read the worked answer
Method: abrade, prime, contact-adhesive plus stitched or riveted backup - belt and braces.
Steps
- Cut and template: strips 25 mm x 200 mm, overlap 40 mm, ends rounded with a 5 mm radius corner, because square corners start peels.
- Clean: wash with soap and water, dry fully, then abrade with 80-grit paper across the bond area until the surface is matt, and wipe with a clean dry cloth. No kerosene.
- Jig for alignment: a plywood board with two 25 mm slots and a 40 mm depth stop, so every handle has the same overlap without measuring.
- Adhesive: apply to both faces, wait 10-15 minutes until tacky and not wet, then join.
- Pressure without hand strength: a bench vice, G-clamps with ply cauls, or standing on a flat board for 60 seconds. This solves the weak-grip student's task - she can place and align, another student clamps.
- Mechanical backup: stitch with nylon thread at 5 mm pitch using a hand awl, or fit four rivets with washers, 8 mm in from the overlap edge and 10 mm apart. Stitching is better than rivets on thin rubber because it does not tear out.
- Cure: 24 hours flat, no load, before fitting to the bag.
- Heat gun: do not use it. Without extraction, heating rubber and adhesive fumes is an avoidable exposure; mechanical and adhesive joining is enough for an 8 kg bag.
3. Write an acceptance test the workshop can run with a spring balance, hook and rail, with pass or fail numbers.
Read the worked answer
Acceptance test for every handle
Setup: hang the bag from the rail by its own handle, load it with known weights or a bucket filled to a weighed amount, and pull with the spring balance where a cycle test is needed. Record each bag on a card.
Test 1: proof load (all 25 bags)
- Load to 16 kg, twice the working load, hang for 10 minutes, then inspect.
- Pass: no peeling at the overlap, no stitch tear, no rivet pull-through, no permanent stretch beyond 3 mm.
- Fail: any visible lift at the overlap edge.
Test 2: cyclic load (three samples per batch)
- 8 kg load, lift and lower the bag 50 times through 200 mm, about one lift per two seconds.
- Pass: no crack growth at the overlap ends; a crack of 2 mm or more is a fail.
Test 3: destructive check of one handle per batch
- Pull to failure with the spring balance on a scrap handle made in the same session.
- Pass: failure above 24 kg (three times working load). This tells you the process is still in control even when the proof test passes.
Judgement rules
- If one bag fails, quarantine that day's production and re-test five more; if two fail, stop and redo the process, because the cause is the method, not one handle.
- Record date, student, adhesive batch and test result, so a complaint from the shop can be traced to a session.