In ASME Section VIII, Division 1, rules are provided for calculating the thickness of 2:1 ellipsoidal heads in UG-32. UG-32(c) also states that “an acceptable approximation of a 2:1 ellipsoidal head is a torispherical head with a spherical radius of 0.9D and a knuckle radius of 0.17D”. However, calculating the thickness of a torispherical head with those “equivalent” dimensions results in a thicker head. This result is inherently inconsistent, which starts to bring into question the so-called equivalency.
Code Case 2260 further perpetuates this equivalency by providing alternative rules for calculating the thickness of torispherical heads, and then permitting the engineer to calculate 2:1 ellipsoidal heads implementing this 90-17 equivalency.
Additionally, the calculation methodology for a 2:1 ellipsoidal head in ASME Section VIII, Division 2 uses the torispherical head calculation methodologies and directly implements this 90-17 equivalency. However, this calculation method results, for the same allowable stress basis, in a completely different thickness from the above three methods.
This paper reviews the past 90+ years of work on this topic, and presents some theoretical treatment of the different head geometries. A review of the current Code rules is presented, with a comparison of results for several sizes. A survey of head fabricators is presented to show the actual geometries produced for use in ASME pressure vessels. Finally, conclusions regarding whether or not the 2:1 ellipsoidal head is in fact equivalent to the 90-17 torispherical head are presented, and recommendations for future revisions to both ASME Section VIII, Division1 and Division 2 are provided.