If you build corrosion-resistant lined valves, elbows, tees, or pipe spools, you have already run into FEP. Fluorinated ethylene propylene is one of the few fluoropolymers you can actually melt and push into a mold, and it keeps most of PTFE’s chemical resistance while doing it. That is the reason it ends up in chemical plants where the media would chew through a bare metal wall.
The catch is in the word “melt-processable.” It makes FEP sound friendly. It isn’t. The melt is thick, it hates shear, its workable temperature band is short, and the steel shell it lines moves about a tenth as much as the fluoropolymer when temperatures change. Miss any one of those and you get flow marks, flash, or a liner that splits a few days after it leaves the tool.
What follows is how FEP transfer molding actually runs on the floor: which grade to buy, how to treat the tool, where the temperature and pressure limits sit, and why cooling is the step that decides whether a liner lasts.
Why transfer molding beats the older routes
Lined parts used to start with isostatic molding of PTFE. Fine for simple shapes, miserable once you add branches, bends, or anything with geometry. Plunger (or “pot”) injection came next: load a heated pot, melt it, shove it into the cavity. The pot is the weak point. Fluoropolymers barely conduct heat, so a big charge insulates its own middle. The outside burns while the center is still solid. You get uneven melt every single time.
Transfer molding breaks the work into two passes:
- An extruder plasticates the resin nonstop and feeds a heated accumulator.
- The accumulator pumps the melt into a preheated mold, holds, then cools.
Splitting it fixes the uneven-melt headache, because plastication is continuous and the accumulator holds one temperature. It also scales: small fittings and big valve bodies run on the same machine, and you change temperature, pressure, and time instead of buying two presses. The tricky part is not either stage on its own. It is the handoff. You have to keep the melt and the accumulator at the same temperature and watch how long the resin sits before it gets injected.
Picking the FEP grade: melt flow versus crack resistance
FEP is graded by melt flow rate (MFR, at 372°C / 2.16 kg). Shops tend to sort it like this:
- High MFR flows easy and fills fast, but it flashes and it resists stress cracking poorly.
- Low MFR has the molecular weight and the crack resistance you want, but you need more pressure to fill the cavity.
On a lined part this is not a theoretical choice. The liner is trapped inside a metal shell, and when the assembly cools, FEP shrinks far more than the steel around it. FEP’s expansion coefficient runs about 90 to 110 × 10⁻⁶ /K; carbon steel sits near 12 × 10⁻⁶ /K. Roughly ten times the difference, which means the liner is stretched in two directions at once. Pick a resin that can’t take that and the part cracks within hours or days of coming out of the tool.
So the habit for lined work is to take a low-MFR grade, around MFR 1 (the “M3” class), and recover the lost flow with transfer pressure instead of dropping to a thinner grade. The major suppliers line up on this. One Daikin Neoflon grade and one Chemours Teflon FEP grade are both sold by their makers for exactly this: high-stress lined service, best crack resistance in the family. When a part sits under pressure, heat, and corrosive fluid for years, a low-MFR grade is the bottom line, not a preference.
Tool prep that lined parts really need
A lined mold is two parts: the shell side that shapes the metal body, and a core that forms the liner’s inner face. The core is what decides whether the part lets go cleanly and how the liner looks.
What actually matters:
- Core finish around Ra 0.40 μm or better, with a hard-chrome layer of 0.01 to 0.02 mm. The chrome does more than stop rust. It is what keeps FEP from grabbing the steel.
- Release agent is usually a silicone resin cut with toluene at about 1:9, then baked at 250 to 300°C for one to one and a half hours so it cures into a hard, slick film. This is a cure, not a coating you brush on.
- Before you build the tool, wipe the core with alcohol, let it dry, then lay on the release agent.
- Soak the fasteners in a 30% molybdenum-disulfide high-temperature lube, or the threads lock up in the heat.
- Once the tool is assembled, preheat it at 260 to 300°C for one to two hours and hold at temperature at least an hour.
That last step is the one shops drop when they are behind schedule, and it is a slow-motion failure. The metal shell dumps heat fast. The core, wrapped in FEP, holds it. Skip the even preheat and the temperature gap at injection becomes internal stress that surfaces as cracks long after the part looks perfect.
The temperature window is the tightest knob you have
FEP starts melting near 270°C (DSC puts it at 260 to 275°C) and degrades clearly above 420°C: color shifts, mechanicals drop. The mold cavity in transfer molding usually runs 300 to 330°C, and the published lining window for FEP sits around 332 to 349°C.
The extruder plastication zone climbs along the flow: roughly 270°C at the feed, 290°C at the accumulator, 300°C at the nozzle. The reason is dwell time. FEP sits in the barrel and accumulator longer than in straight extrusion, so each zone rides a bit above the textbook minimum while staying clear of the breakdown line. Old DuPont data showed FEP held at 360°C for 30 minutes already picks up about 10% melt flow. Heat drift wrecks the resin in ways you can’t see on the part.
The number I’d pin to the wall is critical shear rate. FEP’s is low: one grade reads about 4 sec⁻¹ at 343°C, while PFA runs 10 to 50 and ETFE hits 200 to 3000. Because it is that low, nozzle diameter and transfer rate have to be figured together with γ = 32q / (πD³). Cross the critical shear and you get melt fracture, a shark-skin liner face, and orientation you can’t undo in cooling.
Transfer pressure: fill it, don’t flash it
Pressure does one thing: beat nozzle resistance and shove the melt to fill the cavity. Too little and the part is short. Too much and you flash, plus you pile on internal stress.
Lined fitting data tells the story. At a mold temperature of 300°C, a DN100 part fills steady at about 7 to 7.5 MPa; drop to 6 MPa and flow marks show up. A DN50 part can run a touch lower, but 5.5 MPa is already the edge of “won’t fill.”
Supplier ranges are wide: fluorocarbon transfer pressure is often quoted 1000 to 4000 psi (about 7 to 28 MPa), with 1000 to 2000 psi (7 to 14 MPa) suggested, while fluoropolymer molding books put FEP nearer 15 to 25 MPa up by the shear limit. Local lining lines run toward the low end, and that tracks with nozzle design, wall thickness, and cavity shape. Don’t chase one magic number. Work back from “fills, no flash” for the part in front of you.
Holding pressure and cooling: where liners live or die
This is the deepest step, and the most skipped. Plenty of shops dial in temperature and pressure and still crack the liner. Almost always it is holding plus cooling.
After fill, holding pressure packs the melt and feeds more material, counters the swell from low-molecular fractions and volatiles, fuses melt that arrived at different times into one piece, and tighter packing means less shrink and better strength. But too much holding packs cold material under load, and when pressure lets go the part rebounds: stiff demold at best, cracking at worst.
Cooling is medium plus timing, and three setups show the gap:
- Water spray right after holding: the gate-side FEP freezes fast and seals the feed, but the core is still hot and keeps shrinking, so the liner tears.
- Natural cooling: the core, buried in FEP, can’t shed heat, the tool lingers hot, and the resin camps in its crystallization zone (265 to 275°C) until it goes brittle and cracks.
- Synchronized cooling: blow the core first because the shell already cools quick, then water once the gap closes.
Two things drive the difference. Thermal stress: the shell contracts little and fast, the liner contracts a lot and slow, so the liner has to be “held” by the shell as it shrinks instead of being pulled both ways. Crystallization stress: FEP crystallizes fastest in that 265 to 275°C band, and the longer it hangs there, the higher the crystallinity, the more it shrinks, the more it breaks. Cooling the core first shortens the time in that band.
The shop rule is air on the core, then water, so both sides move together. Newer lines push to stepped or mist cooling: a 60 to 80°C warm-water pre-cool sets the skin without shocking it, room-temperature water does the main crystallization, cold water finishes. FEP bellows and heat-shrink lines already run that gradient; lined parts can borrow it with a simple air-plus-water two-stage setup.
Stress cracking from two sides
Put the whole line together and liner cracking is three things stacked: constrained shrinkage, brittle crystallization, and orientation from fast fill.
Material side: take a low-MFR, high-molecular-weight grade, and get the flow back with transfer pressure and a slightly bigger nozzle instead of dropping to a thinner resin.
Process side: don’t fill fast (FEP’s low critical shear means speed raises orientation and invites melt fracture); nudge mold and melt temperature up and open the nozzle to lift the critical shear; keep holding moderate and scale hold time to wall thickness; cool air-on-core-then-water so the resin doesn’t sit in the crystallization zone.
One thing shops miss: the part design itself. Corner radius, wall transitions, and gate location have to respect FEP shrinkage and orientation from the start. You can’t fix a bad design with process tweaks, which is why fluoropolymer lining manuals spell out minimum corner radii and tell you to keep gates off high-stress areas.
Where the process is going
When domestic FEP transfer molding got started in the 1990s, the numbers came from trial: screw speed, accumulator temp, pressure, hold time, cooling, all built run by run. Those hand-built values still work as a baseline, but the control side has moved.
A few shifts matter:
- Isothermal transfer molding runs the whole transfer inside a hot-air oven so the mold doesn’t cool when it leaves the press. That matters on big molds that would otherwise drop below the critical temperature mid-shot.
- Zoned PLC temperature control with in-mold pressure sensors gives each extruder zone, accumulator, gate, and mold its own PID loop, and ties holding pressure to the real cavity curve instead of a clock.
- Mold coating care is getting real attention, including laser-shock composite processes that clean and strengthen FEP-coated tooling.
- Critical shear is now designed, not guessed. With γ = 32q / (πD³) you size nozzle, transfer rate, and melt temperature together against the shear curve instead of tuning each alone.
The steps you can’t skip
If you remember one thing, remember the short list that keeps getting re-proven:
- Don’t wobble between a higher- and lower-MFR grade for lined parts. A low-MFR (~1) grade is the floor for parts under long-term pressure, heat, and corrosion.
- Don’t cut the one-hour preheat hold, or you plant a temperature gap that bites later.
- Keep the extruder-accumulator-nozzle gradient near 270 / 290 / 300°C. Push the accumulator past 300°C and degradation risk climbs.
- Set transfer pressure by part size: about 7 to 7.5 MPa for DN100, 6 to 7 MPa for DN50. More is not better.
- Cool the core with air first, then water. Synchronized cooling is the heart of a liner that doesn’t crack.
FEP transfer molding is an old process in most shops, and that’s fair. But the material sets hard limits: low critical shear, a fussy crystallization window, and a tenfold expansion gap with metal. Drop any one of those from your thinking and you’re back to finding the numbers by hand.