High Frequency Welding for PVC and TPU Gloves

High Frequency Welding for PVC and TPU Gloves

A glove is the only common HF product where the part is closed in three dimensions and the electrode is not. You are pressing a flat bar onto a tube. That single mismatch explains most of the scrap in glove production: the seam looks bonded all the way round, the cuff feels solid in the hand, and then the glove fails an inflation test at one point on the thumb side. HF welding glove machine work is therefore less about generator power than about how the part is supported while the field is applied, and how the operator knows the difference between a sealed seam and a seam that merely looks sealed.

Why a Glove Is Harder Than a Bag

On a bag or a tarpaulin, the two plies lie flat and the electrode meets them over its full area at once. Pressure is uniform because the geometry is uniform. A glove breaks that assumption in three places.

Geometric Fact What It Does to the Weld
The cuff is a tube, typically 70–100 mm across A flat bar touches the crown of the tube only. The sides of the cuff sit below the electrode face, so pressure there is whatever the material's own stiffness allows — usually not enough.
Two plies meet at a curved overlap Where the shell doubles back on itself the stack is twice as thick. The same energy that seals the single-thickness zone over-heats the doubled zone.
The thumb crotch has a radius of a few millimetres Field concentrates at any sharp electrode edge. At the crotch, that concentration lands on the thinnest, most stretched part of the shell.
Most gloves are lined A jersey or fleece liner is a dielectric with its own loss factor. It absorbs energy, insulates the outer shell from the table, and can scorch on the inside while the outside looks perfect.

The practical consequence: on gloves you cannot fix a weak weld by adding time. Adding time fixes the crown of the cuff and burns the doubled overlap in the same press. The fix has to come from geometry — a mandrel, a radiused electrode, or splitting the cuff into two passes.

Which Glove Materials Actually Weld

"Glove" describes a shape, not a polymer, and the polymer decides whether the machine is the right tool at all. This is the first question to settle before any parameter is discussed.

Material Typical Product HF Behaviour
Flexible PVC, 0.3–0.6 mm Winter gloves, rain gloves, work gloves, waterproof over-gloves The best HF candidate of the group. Strong dielectric coupling, wide process window, forgiving of small pressure variation. The weak point is not welding — it is plasticizer.
PVC shell on jersey or fleece liner Insulated gloves, cold-store gloves Welds well, but the liner consumes energy and hides the seam. Scorch shows up on the inside of the glove weeks later, after washing.
TPU film, 0.15–0.5 mm Sports gloves, touchscreen-compatible shells, bladders Weldable but needs more energy and noticeably more pressure for the same bead. The window is narrow, and the failure mode is a surface seal that peels cleanly at low load.
PU-coated or PU-dipped fabric General handling gloves, assembly gloves Poor HF candidate. Where a sealed seam is mandatory, a coated tape or a different joining method is usually more reliable than chasing a process window.
Nitrile, latex, uncoated knit Disposable and general-purpose gloves Not HF materials. They are dipped or knitted as one piece and have no seam to weld.

The PVC-to-TPU difference is not a small correction factor. TPU needs longer dwell and higher pressure, and because it softens into a narrower band, the tolerance for a pressure drop at the edge of a curved cuff is much smaller. The underlying polymer mechanics are set out in how material properties affect the welding process.

The Three Failures That Get Past Inspection

In glove production the expensive defects are the ones that pass a visual check. A glove seam can look continuous, feel firm, and still leak — either immediately or after the first wash. These three account for most of them.

1. Cold weld at the cuff overlap. The doubled ply under the bar has more thermal mass and sits slightly higher, so it presses harder at the leading edge and lighter where the overlap steps down. The seam is continuous but the last 2–3 mm of the overlap never reaches melt temperature. It survives handling and fails in service. Tell: the seam peels from the overlap edge, not from the middle. Fix: rotate the part so the overlap is not at the crown of the bar, and reduce the step instead of increasing power.
2. Arc-through at the thumb crotch. The tight radius puts the electrode edge close to the material edge, and the field concentrates there. The result is a pinhole, one or two millimetres across, that no visual inspection will find. Tell: the glove inflates but drops pressure over a minute. Fix: radius every electrode corner to at least 1 mm, keep 3–5 mm of unwelded margin at the crotch, and use a sensitive anti-spark circuit rather than turning the power down.
3. Liner scorch behind a good-looking seam. The liner absorbs energy, the seam face looks correct, and the inside of the glove is brown and brittle. It fails after washing, or it fails as a hole when the glove is flexed. Tell: cut a finished glove open and look at the back of the seam. Fix: shorter dwell with a higher preheat, or a liner that is specified as HF-compatible.

Electrode and Fixture Design for a Curved Part

Most glove machine work is done with a narrow bar rather than a platen — a typical upper plate is in the region of 60 × 310 mm, which is deliberately just wider than a cuff seam and short enough to keep pressure even. That geometry is the reason glove welding is possible at all: a small electrode concentrates both the field and the force.

The same concentration has a cost. A 12 kW generator feeding an 186 cm² bar works out to roughly 65 W per cm² — several times the power density of a large platen press. Thin glove films therefore arc or puncture long before they under-weld, which is why machine selection for gloves is really a question of control resolution, not maximum power.

Part of the Glove Tooling Approach Common Mistake
Cuff, straight run Flat bar plus a mandrel or expanding insert that holds the tube round under the electrode Welding a free cuff with no internal support — the tube flattens and the underside never bonds
Cuff, full circumference Two or three indexed passes with a rotating fixture, each overlapping the previous by 20–40 mm One strike with a curved electrode — the ends of the arc receive less pressure than the crown
Thumb crotch Radiused electrode with generous edge margin; smaller electrode area, not more power Extending the bar to cover the crotch and the cuff in one press
Decorative emboss or logo Shallow engraved electrode, treated as a separate pass after the structural seam Embossing and sealing in the same press — the engraving changes pressure distribution across the whole seam

Edge distance follows the same rule as any HF seam: keep at least 3–5 mm of unwelded margin outside the bead. On a glove cuff that margin is often the first thing an operator gives up when chasing a leak, and it is exactly the change that produces arc-through instead of fixing the leak. The tooling principles are covered in more depth in designing electrodes for high frequency welding, and the consumable side — shoes, bars, anti-spark components — under high frequency accessories.

Two Stations, and Why Cooling Sets Your Output

A glove machine is usually built as a single-side double-head — two working positions facing one operator. One side presses and cools while the other is unloaded and reloaded. The throughput benefit is obvious; the less obvious point is that this arrangement is a cooling solution disguised as a speed feature.

On thin PVC the weld itself takes two to four seconds, and the part then has to stay under pressure while the bead sets. That cooling phase is commonly 40–60% of the total cycle, and it cannot be shortened by turning the power up. Shops that try to raise output by trimming cooling get seams that look fine on the table and delaminate during cuff trimming, because the bead is still soft when the trimmer pulls against it.

Load the second station during cooling, not after it. The two stations only pay for themselves if the operator's hands are moving in the "dead" seconds. If loading takes longer than cooling, the machine is waiting on the operator and a second head adds nothing.
Keep the two stations set identically. They share one generator but not one pressure regulator on every build. If the two sides drift apart, half your output is systematically weaker, and the drift is invisible until a customer complains.
Watch the electrode gap on a curved part. A gap in the region of 150 mm is generous for a flat panel and tight for a glove fitted over a mandrel. Check that the fixture plus the part still clears the bar at full open height.
Do not nest more than the table allows. Small tables — around 300 × 420 mm on many glove machines — are sized to one or two gloves. Trying to fit four across means the outer positions sit outside the pressure zone and weld weakly.

Testing a Glove Properly

A glove has an advantage over a tarpaulin: it is a closed volume, so it can be pressure-tested directly. Three tests cover the failure modes above, and none of them requires equipment beyond what a glove line already has.

Test Method What It Catches
Inflation hold Seal the cuff, inflate to a low pressure and hold for 60 seconds Pinholes at the crotch and any incomplete cuff circumference — the fastest test on the list
Cuff peel Cut 20 mm strips across the seam and pull at a steady rate Cold welds at the overlap. A good PVC weld tears the film before the seam opens; a weak one separates cleanly.
Wash cycle Two domestic wash cycles on a sample from each material lot Plasticizer-related seam failure and liner scorch, neither of which appears in a fresh-sample pull test
Cut-and-look Section the seam across its width and inspect the back face Liner scorch and a bead that has melted through the shell but not bonded to the ply beneath

The wash cycle is the one most shops skip, and it is the one that matches how a glove is actually used. A seam that survives a pull test on the day it was made can still open on the third wash if the plasticizer has migrated out of the weld zone. If you only run one extra test, run this one.

Plasticizer: The Slow Failure Nobody Plans For

Flexible PVC for gloves carries a much higher plasticizer loading than a tarpaulin coating, because the product has to feel soft. Plasticizer is not chemically locked into the polymer — it moves. It migrates towards the surface, into the liner, and out into whatever the glove touches, and the weld zone is where the loss matters most.

Two consequences follow. First, incoming material is not a constant: two lots of nominally identical PVC can need different energy because their plasticizer content or thermal history differs. Second, an over-welded seam is more vulnerable than a correctly welded one, because excessive heat drives plasticizer out of the bead and leaves a dry, brittle zone that cracks in flexing. This is why "weld it hotter to be safe" is bad advice on gloves specifically, and why the strength plateau on a glove seam is narrower than on heavier coated fabric. The general relationship is covered in what factors affect high frequency welding strength.

Specifying a Glove Machine: Five Questions

Glove machines look similar in photographs and differ in the details that decide whether the line runs. Ask these before comparing prices.

1. Does the pressure system suit the part, or only the catalogue? Pneumatic operation with a few tonnes of clamping force is normal for gloves, but ask whether the pressure is adjustable finely enough for a 0.3 mm film. Hydraulic conversion is worth it only when you need sustained force on thick stacks.
2. How sensitive is the anti-spark circuit? On gloves a false trigger costs one cycle; a missed trigger costs a leaking glove that ships. A high-sensitivity detection circuit is not an optional extra on thin film.
3. Can the recipe be stored per product? A PLC panel with stored recipes matters more on gloves than on flat work, because you change between cuff, crotch and emboss passes within a single glove.
4. What is the electrode changeover time? Glove styles change often. If swapping a bar and a fixture takes half a shift, the line will run long batches whether or not that suits the order book.
5. Is the cooling phase adjustable independently of the weld time? If one timer controls both, you cannot tune the bead without changing the seal. Separate control is the difference between a machine you can dial in and one you work around. The full range of configurations is set out under high frequency machines.

The same geometry problem appears on other three-dimensional coated products — a waterproof bag welding machine deals with corners and bound seams in much the same way, and the defect patterns overlap. Where a glove line is running high scrap, the general diagnostic sequence in reducing defects and downtime in high frequency welding is a reasonable starting point before changing tooling.

Frequently Asked Questions

Can a glove cuff be welded in a single press?
Only if the electrode follows the curvature and the cuff is supported internally. With a flat bar, a single press bonds the crown and leaves the underside weak. Two or three indexed passes around a rotating fixture are more reliable and usually faster overall than one rejected glove.
Why does my glove leak at the thumb crotch and nowhere else?
Field concentration at a tight radius, usually combined with too little unwelded margin. The electrode corner acts as a point source and punctures the film before the surrounding seam is fully bonded. Radius the corner, restore 3–5 mm of margin, and let the anti-spark circuit do its job instead of reducing power.
Do I need a different machine for TPU gloves?
Not necessarily a different machine, but a different recipe and often different tooling. TPU needs more energy and more pressure for the same bead, and it tolerates less pressure variation across a curved surface. The same machine can usually run both if pressure and cooling are independently adjustable.
Is a two-station machine really twice as productive?
Rarely twice, and never automatically. The gain is bounded by how much of the cooling time the operator can fill with loading. If loading and unloading already take longer than the cooling phase, a second station adds a head without adding output.
Can I weld a lined glove without scorching the liner?
Usually yes, by shortening the dwell and preheating rather than by raising the power. The liner absorbs energy, so a long slow cycle heats it as much as the shell. Where the liner is not specified for HF, no recipe will fully solve it and the liner has to change.
How do I know if my seam is over-welded?
Section it. An over-welded PVC seam shows a squeezed, translucent bead with a thin, dry zone at its edge and often a visible impression of the electrode outline on the outer face. It will pass a pull test today and crack at the bead edge after flexing.

Send Us a Sample Glove and a Reject

Chenghao Machinery builds high frequency welding machines for PVC and TPU gloves, with single-side double-head layouts, pneumatic pressure control and configurable electrodes for cuff, crotch and emboss passes. Tell us your material, thickness, liner and target output, and we will recommend the configuration and the tooling that suits the part.

Dongguan Shenghao Machinery Co., Ltd. (Chenghao Machinery) | sales11@chenghaointernational.com | +86 158 1765 5869
Dongguan, Guangdong, China