ME 421 Machine Design · Fall 2025
Pressure Vessel
A class project to design, analyse, and build a small aluminum pressure vessel that safely contains 50–60 cc of liquid at 750 psig. Cylindrical body with a removable end cap, axial O-ring face seal, a six-bolt joint, and a single 1/4-inch NPT port for the Swagelok inlet/outlet.
Team of three (Nicole Kladakis, Pierre Gathy, Mehmet Guvenilir). Sealing, joint, fastener, and FEA work were shared across the team — no individual ownership of any one part.
What it is
A pocket-sized pressure vessel: a cylindrical 6061-T6 body with a thicker matching lid, sealed by an axial face-seal O-ring, clamped with six #10-32 socket-head screws on a bolt circle. It runs through one Swagelok 1/4-inch NPT port that doubles as both inlet and outlet. The whole thing fits a $100 parts budget, came in at $111 of bill-of-materials before subtracting the parts the lab provided, and is machinable from a single 4-inch aluminum rod.
How it works
The seal is an NBR 70A static face seal (McMaster 9262K534 — 70 mm ID, 1.5 mm cross-section) sitting in a 72 mm × 1.125 mm axial groove. That gives ~2.86% installed stretch, ~25% axial squeeze, and ~70% gland fill — all comfortably inside the manufacturer windows for a static face seal on a non- moving joint.
The bolted joint was sized end-to-end. We computed bolt and member stiffnesses for the steel #10-32 screws sandwiching aluminum, derived the joint constant C ≈ 0.308, then walked through the standard Shigley factor-of-safety set against the 4868 lbf separating force from the projected O-ring area: yielding FoS ≈ 1.17, overload FoS ≈ 2.45, and a joint-separation FoS of 3.27. Practical preload was set to 75% of proof load, giving ~70 lbf·in of tightening torque per bolt with K = 0.20 for zinc-plated fasteners.
On the FEA side, the von Mises peak hit at the bolt-hole contact (a known mesh-singularity artefact), but the representative away-from-edge probe came in at ~238 MPa against the 275 MPa yield of 6061-T6 — a ~1.2 material FoS, consistent with the 1.17 from the hand calc. Maximum lid displacement at the flange under full pressure was 0.028 mm, an order of magnitude smaller than the design 0.375 mm O-ring squeeze. So even at peak pressure the seal stays loaded.
Design changes from prototype
The first pass had a flange step between the body and the lid. We pulled it: the flange added machining steps and alignment risk without buying anything for the seal. The final design uses a constant outer diameter (matching the 4-inch lid stock), a thicker lid that uses up the excess material budget for stiffness, and longer 1-inch #10-32 screws to stay well above minimum thread engagement on the now-deeper joint.