Take polyamide 6 and replace part of its flexible aliphatic backbone with rigid aromatic rings, and you get a semi-aromatic polyamide: polyphthalamide. The rings do not rotate, they stack, and they raise the melting point and the stiffness far above what an ordinary nylon manages.
That is the whole design intent. PPA exists because engine bays get hot, and because the polyamides that serve everywhere else in a vehicle give up before the components under the bonnet do.
What the aromatic content changes
Temperature capability, dramatically. Stored heat deflection runs from 140 °C right up to 250 °C — the widest band on this site, because it depends heavily on the fibre loading — against 60 to 100 °C for unfilled PA6. Tensile lands at 90 to 150 MPa.
Water uptake, downward. The aromatic sections carry no amide groups, so there are fewer sites per unit length for water to hydrogen-bond to. PPA still absorbs moisture, and it still needs drying, but a PPA part holds its dimensions in service far better than a PA6 one.
Stiffness, upward and permanently. Aliphatic nylons soften noticeably as they take up humidity, because water plasticises them. PPA's rigid segments do not soften in the same way, which is exactly why it is specified for parts that must stay tight — connector bodies, sensor housings, thermostat and pump components.
It is essentially always filled, and that is not incidental
Unfilled polyphthalamide is close to unprintable on any machine short of an industrial one: its shrinkage is high, its melt is stiff, and it needs chamber temperatures that hobby enclosures do not reach. What is sold as PPA filament is nearly always carbon- or glass-filled, and the reinforcement is doing the same job it does in PA6-CF — restraining contraction so the part stays on the plate.
That is why the stored shrinkage range of 0.3% to 1.0% is so wide. It is not measurement uncertainty; it is the difference between fill fractions, and it means you cannot compensate dimensionally from a published figure. Print a coupon, measure it, and work from that.
It also means the abrasion rules apply in full: hardened nozzle, expect wear on drive gears and the filament path, and use 0.6 mm rather than 0.4 mm.
The drying requirement is equipment, not diligence
100 °C for 8 to 16 hours. That is above what a food dehydrator reaches and above what most filament dryers reach, so printing PPA properly means owning a laboratory oven or a high-temperature dryer. There is no version of this where you dry it at 70 °C for longer and get the same result — a polyamide holds water tightly enough that the temperature, not the time, is what liberates it.
Nozzle 290 to 320 °C, bed 100 to 130 °C, and an actively heated chamber. That combination excludes almost every desktop machine, and it excludes some machines that manage PA6-CF comfortably.
Costing a bracket, failure rate first
Start from the failure rate, because on this material it is the assumption that decides whether a quote is honest. Set it at 18% for an under-bonnet bracket — 140 g, 9 hours, heated chamber drawing 500 W, a $130 spool, $1.20 an hour of machine time — and the allowance alone comes to $6.54. Filament adds $18.20, machine time $10.80, and electricity $0.79 for 4.5 kWh at the US average rate. $36.33 all told.
The failure allowance is worth dwelling on. At 18% it is nearly a fifth of the bill, and that rate is not pessimism — a material that shrinks unpredictably, needs 320 °C and depends on chamber temperature for its layer bonding does fail. Model your own rate honestly in the print cost calculator; a quote built on a 5% assumption for this material is a quote that loses money.
What PPA is genuinely bad at
- Every consumer printer without a heated chamber, and most with one.
- Price, at $90 to $180 per kilogram before the dryer and the nozzles.
- Predictable dimensions, for the fill-fraction reason above.
- Impact. Fibre-filled semi-aromatic polyamide is stiff and comparatively brittle; it is not the material for a part that gets dropped.
- Surface finish, which is matte, fibrous and functional-looking.
- Being bought before it is needed. The heat capability is the only reason to accept everything else, and if the part never exceeds 120 °C, a fibre-filled PA6 will do the job on a machine you already own.
- Being compared between brands. PPA is a family of semi-aromatic copolymers rather than a single polymer, and different suppliers build theirs on different diacids with different fill fractions. Two spools both marked PPA can differ more than PA6 differs from PA12, so a profile that worked on one is a starting point rather than a setting.
Under the bonnet, and in tooling
Under-bonnet automotive components — sensor mounts, connector housings, brackets near the exhaust side. Injection-mould tooling inserts and jigs used at temperature. Parts that see continuous heat rather than occasional peaks, where a material's heat deflection matters less than its ability to hold properties for years at temperature.
Between PPA and the aromatic polymers above it, the boundary is roughly where a heated chamber stops being enough. If the part must work above about 150 °C continuously, PEEK is the next step and it is an order of magnitude more expensive.