Extruding a Profile, and Blowing Film

A 3D printer nozzle is an extrusion die, and it swells like one. This is the cleanest demonstration available on a bench. Heat the hot end to printing temperature with the nozzle clear of the bed, and extrude slowly into air. Catch the strand, let it cool, and measure it with the micrometer. It is **larger than the nozzle**. A 0.4 mm nozzle typically gives a free strand of 0.45 to 0.55 mm, depending on polymer and temperature. Do it for PLA, PETG and ABS and tabulate the swell ratio for each. Then change the conditions and watch the ratio move: - **Extrude faster.** Swell increases, because the chains are stretched harder in the die and recoil more. - **Extrude hotter.** Swell decreases, because the chains have time to relax before they freeze. Now cancel it. Extrude into air and, with tweezers, pull the strand away faster than it emerges. Measure again: the strand is thinner than the nozzle. That is **draw-down**, and in a real line it is the haul-off doing exactly this at a controlled rate. Those two effects, in opposition, are the whole of profile extrusion control. Step 4 puts numbers on the combination.

Now make a shaped section rather than a round one. You need a small die: a plate with a slot or a cross cut through it, mounted where the strand leaves. The cheapest working version is a **die plate for a printer nozzle**: a thin brass or aluminium plate, drilled and filed with a 2 mm slot, held a millimetre below the nozzle in a clamp so the extrudate is forced through it. Crude, hot and fiddly, and it demonstrates the two things that matter. **The land.** The parallel section of the die, after the taper. A short land gives more swell and a rougher surface; a longer land lets the chains relax inside the die and gives a steadier, more accurate section. Make two plates, one with a 1 mm land and one with 4 mm, and compare the extrudate. **Uneven flow.** Extrude through a slot that is wider at one end. The melt comes out faster where the die is more open, so the profile bends towards the slow side as it leaves. Every asymmetric profile does this, and a real die is cut with deliberately uneven land lengths to balance the flow — which is why an extrusion die is hand-tuned on the machine and not finished at the CAD stage. Cool the extrudate as a real line does: run it through a shallow tray of water a few centimetres from the die. Note that cooling too fast freezes the outside while the inside is still shrinking, and the section bows. Molten polymer at 200–260 °C, a hot brass plate and water in close proximity. Face shield, gloves, a stable clamp, and never chase a falling hot strand with your hands.

Blown film is extrusion with the stretch turned up: melt leaves an annular die as a thin tube, air inflates it to several times its diameter while a nip pulls it upward many times faster than it left. It freezes mid-air, stretched both ways at once. You cannot blow film on a bench, but you can produce the *property* it has and prove it matters, which is the part worth knowing. Take a carrier bag or a piece of packaging film. Cut two strips, one along the roll direction and one across it — the roll direction is usually obvious from the fold lines or from any print. Tear both by hand. One tears cleanly and easily in a straight line; the other resists and wanders. Then pull both to destruction with the spring scale and record the force. That difference is **orientation**: the chains are lined up more in one direction than the other, so the film is strong along them and splits between them. A bag that tears beautifully down its length and not across it is a bag whose blow-up ratio and draw-down were not balanced. Now do the same test on a piece of **cast** film — cling film, or a sheet cut from the vacuum-forming rung's stock. It behaves much more evenly in both directions, and it is also weaker. That is the trade: orientation buys strength in the direction you orient, and takes it from the other one. Hold that result. The next rung takes it much further and makes a bottle out of it.

**Attribution, stated honestly.** No single patent is claimed for this rung. Screw extrusion of plastics grew out of the rubber industry across the nineteenth century and was thoroughly established by the 1930s, and the catalogue already holds the lead pipe extrusion rung as its metal-working ancestor. Dozens of filings cover particular dies, screws and die-swell compensations; none of them is *the* extrusion patent, and asserting one would be inventing a tidiness that history does not have. **What it does for economics.** Extrusion is the cheapest way to convert polymer into useful geometry, because the tool runs continuously and its cost is spread over length rather than over parts. That is why the cheapest plastic objects in the world — film, bag, pipe, profile — are extruded, and why a small moulded part can cost more than a metre of extruded section that weighs ten times as much. **Orientation is the idea to carry forward.** Everything downstream of the die in a film or fibre line is there to stretch the polymer while it freezes, because stretching lines the chains up and lined-up chains are far stronger. The next rung is that same principle, done deliberately in two directions, to make a bottle that holds five bar. **Honest limits.** Extrusion makes constant sections and nothing else — any variation along the length has to be added afterwards. Tolerances are much looser than moulding, and they wander with line speed, melt temperature and cooling. Die swell is not calculable to useful accuracy for a real compound, so dies are cut, run, measured and recut; the bench work in step 2 is a small honest version of exactly that loop. And the process is unforgiving of wet feedstock: hygroscopic polymers must be dried or the water flashes to steam in the die and the extrudate comes out foamed and streaked.