A partial joint penetration (PJP) groove weld is designed to deliver a specific weld strength without fusing through the full joint thickness. The throat dimension that controls that strength — the "effective weld size" in AWS D1.1 terminology — is not simply the depth of the groove. For groove angles below 60°, AWS D1.1:2025 applies a mandatory reduction known in practice as "z-loss." Getting this distinction right matters at three points: structural design, WPS preparation, and CWI inspection.
PJP groove welds and why groove depth isn't the whole story
When a structural engineer sizes a PJP groove weld, they compute a required effective throat — the minimum dimension through the weld throat that carries the design load. The design is based on this effective dimension, not on the physical depth of the groove preparation.
For a CJP groove weld, the effective throat equals the base metal thickness (full fusion). For a PJP groove weld, the effective throat is the depth of reliable fusion, which depends on the groove geometry. The deeper and wider the groove, the easier it is for the welder to fuse to the root. Below a critical groove angle, full root fusion becomes unreliable in production conditions, and the code accounts for that with a formal deduction.
The z-loss rule: how AWS D1.1:2025 treats narrow-groove PJPs
AWS D1.1:2025 establishes two cases for PJP groove weld effective weld size based on the included groove angle:
Groove angle 60° or greater: Effective weld size equals the depth of chamfer. No deduction. At 60° and above, the groove opening is wide enough for the welder to reliably reach and fuse the root of the joint, so the full groove depth counts toward weld strength.
Groove angle less than 60°: Effective weld size equals the depth of chamfer minus 1/8 in. (3 mm). This is the z-loss deduction. The narrower groove restricts electrode access and puddle manipulation at the root. Even a skilled welder cannot guarantee consistent root fusion at the tip of a sharp bevel, so the code deducts 3 mm from the claimed penetration.
The practical arithmetic: a 3/4 in. deep 45° single-bevel groove produces an effective weld size of 5/8 in. (3/4 minus 1/8). A 3/4 in. deep 60° double-bevel V-groove produces an effective weld size of 3/4 in. (no deduction). If the design requires 3/4 in. of effective throat, these two geometries are not interchangeable.
The z-loss deduction is a geometric correction, not a design safety factor. It's the code's acknowledgment that the "full depth" model breaks down for narrow-angle grooves in production. It does not change the allowable stress on the effective throat — it changes how the effective throat is calculated.
What this means for fabrication drawings and WPS preparation
A common coordination error between design and fabrication: the structural engineer specifies effective weld size on the design drawing (as is correct), but the fabrication shop preps the groove to match the effective weld size dimension without adding back the z-loss.
A design note reading "3/4 in. PJP" on a 45° bevel joint is interpreted by the shop as "cut to 3/4 in. depth." The actual effective weld size delivered is 5/8 in. — a 17% shortfall from the design requirement. This error is invisible after welding is complete because you can't see the groove depth in the finished weld.
The WPS for a PJP groove weld must specify both the groove angle and the groove depth — not just the effective weld size. The WPS is the production instruction; it must tell the welder and the CWI what to actually cut. A WPS that says "effective weld size: 3/4 in., groove angle: 45°" should also state "groove depth: 7/8 in." to make the z-loss math explicit and auditable.
When a prequalified PJP joint is used, AWS D1.1:2025 prequalified joint tables list both the groove preparation geometry and the resulting effective weld size, so the z-loss arithmetic is embedded in the table. The shop reads both values directly. Problems arise when a shop designs a custom groove angle that departs from the prequalified details, or when the detailer writes a joint note without checking whether z-loss applies.
For an overview of the prequalified joint detail framework and which groove configurations are available, see prequalified WPS joint design and Clause 5 geometries.
WPS coverage and essential variables for PJP joints
The WPS must qualify the joint type (PJP groove vs. fillet vs. CJP), the position, and the process for the specific production application. Essential variable changes that affect PJP groove weld coverage include:
Change in groove angle. Moving from a 60° V-groove (no z-loss) to a 45° bevel (z-loss applies) changes the effective weld size even if the groove depth stays the same. This is a change in joint geometry that must be evaluated against what the WPS and supporting PQR actually qualified. If the PQR was run at 60° and the production detail calls for 45°, the effective weld size delivered is different.
Change in base metal thickness. PJP groove weld qualification by test (when prequalified joints aren't used) has thickness qualification ranges that limit how much thinner or thicker the production base metal can be from the PQR test plate. The AWS D1.1:2025 qualification range for thickness applies to the base metal, not the groove depth.
Change in welding position. A PQR run in the flat position (1G) has limited coverage of other positions. Vertical (3G) and overhead (4G) PJP groove welds require separate qualification if the process is not prequalified for those positions. For the full table of position qualification limits, see welding position qualification limits in AWS D1.1.