Shear Tab
AISC 360-22 · J3 + J4 ← Engineering
LRFD
ASD

A shear tab, screened one limit state at a time

A single-plate shear connection fails in one of a small number of ways, and the useful question at the bench is which one comes first. This page runs the four that live in the plate and its bolts — the bolt group (shear, bearing and tearout, computed bolt by bolt), plate shear yielding, plate net shear rupture, and block shear — reports the governing one, and draws the plate to scale so the geometry on screen can be checked against the geometry on the drawing.

It is a screening tool, not a connection design. The list of what it does not check is further down and it is not short. The most important omission: eccentricity. Read that section before you use a number from here for anything.

The bolt group, bolt by bolt — and why that is not the same as “n times one bolt”

AISC 360-22 Section J3.7 carries a user note that decides how this is done: “The effective strength of an individual fastener may be taken as the lesser of the fastener shear strength per Section J3.7 or the bearing or tearout strength at the bolt hole per Section J3.11. The strength of the bolt group is taken as the sum of the effective strengths of the individual fasteners.”

So each bolt gets its own answer. On a shear tab with a vertical reaction, the bolt nearest the loaded edge tears toward that edge over a clear distance of L_ev − h/2, while every other bolt tears toward the hole below it over s − h. A short edge distance therefore weakens one bolt, not all of them — which is why this page prints the weak bolt separately instead of multiplying a single number by n.

Per boltEquationWhere
Shear (single shear plane)R_n = F_nv A_bJ3.7
Bearing, deformation a considerationR_n = 2.4 d t F_uJ3.11
Tearout, deformation a considerationR_n = 1.2 L_c t F_uJ3.11
Effective strengththe least of the threeJ3.7 user note

φ = 0.75 and Ω = 2.00 apply to all three. F_nv comes from Table J3.2, which is the same set of values as RCSC Table 5.1: 54 ksi for a Group 120 bolt with threads in the shear plane, 68 with them excluded; 68 and 84 for Group 150; 65 and 81 for Group 144.

Two different hole sizes, and using the wrong one is a real error

Bearing and tearout use the nominal hole from Table J3.3. Net areas — shear rupture and block shear — use the nominal hole plus 1/16 in., because Section B4.3b says “the width of a bolt hole shall be taken as 1/16 in. greater than the nominal dimension of the hole.” That sixteenth is a net-area allowance; it does not belong in the clear distance L_c. This page prints both numbers on the first two lines of the tape so you can see which went where.

The nominal holes themselves are the current values: d + 1/16 for bolts 7/8 in. and smaller, d + 1/8 for bolts 1 in. and larger. The clearance for the larger bolts was increased by the 2020 RCSC Specification (Table 3.1, and its Commentary to Section 3.3.1, which states the clearances “were increased … for bolts 1-in. diameter and greater, and clearances for smaller bolts remained unchanged”), and carried into AISC 360-22 and the 16th Edition Manual. A chart printed before that shows 1 1/16" for a 1" bolt. If your reference disagrees with this page on a 1 in. bolt, that is almost certainly why — and it is a conversation with your engineer, not something to quietly reconcile.

The plate

Limit stateEquationφΩWhere
Shear yielding, gross sectionR_n = 0.60 F_y A_gv, A_gv = L t1.001.50J4.2, Eq. J4-3
Shear rupture, net sectionR_n = 0.60 F_u A_nv, A_nv = t (L − n d_h)0.752.00J4.2, Eq. J4-4
Block shearR_n = 0.60 F_u A_nv + U_bs F_u A_nt ≤ 0.60 F_y A_gv + U_bs F_u A_nt0.752.00J4.3, Eq. J4-5

Shear yielding is the one limit state on this page with φ = 1.00 — not a typo, and worth knowing, because it is the reason a plate that looks marginal on rupture often is not. The 0.60 in all three is the von Mises shear-yield ratio 1/√3 ≈ 0.577, rounded up.

Plate depth is derived rather than typed: L = (n − 1)s + 2 L_ev, a bolt row centred in the plate. Two numbers that must agree cannot be entered separately here, so they cannot disagree. If your plate is deeper on one end than the other, enter the smaller edge distance — that is the one that governs tearout and block shear.

Eccentricity — the thing this page does not do, stated plainly

A shear tab carries its reaction on the bolt line but delivers it at the support, and the distance between those two lines puts moment into the bolt group and the plate. Whether that moment can be ignored depends on the configuration, and the rules for it live in the AISC Manual, not in the Specification. Everything on this page is computed at zero eccentricity, so every number here is an upper bound. On a conventional tab with a short distance to the support the real answer is close; on an extended tab it can be a long way below. Treat a number from this page as “the connection is at best this strong,” and take the eccentricity to whoever is stamping the drawing.

What else it does not check

Checking it yourself

A 3/8" A36 plate, three 3/4" A325 bolts with threads included, standard holes, s = 3", L_ev = 1 1/4", L_eh = 1 1/2", so the plate is 8 1/2" deep. Bolt shear is 23.9 kips a bolt, but the bottom bolt tears out at 22.0 kips over a clear distance of 0.84", so the group is 22.0 + 2 × 23.9 = 69.7 kips. Plate shear yielding 68.9, shear rupture 76.7, block shear 81.8. The bolt group governs: φR_n = 52.3 kips (LRFD) or 34.9 kips (ASD). If this page says otherwise, do not use it — tell us.

Where these numbers were checked

Every value was confirmed in two independent published sources before a line of this page was written, with the primary document breaking any tie. The hole dimensions and the bearing, tearout and bolt-shear provisions were read out of the 2020 RCSC Specification itself — Table 3.1, Table 5.1, Sections 5.1 and 5.3 — and cross-checked against AISC's own Modern Steel Construction primer on bolted connection design written to the 16th Edition. The J4.2 shear yielding and rupture equations and their φ and Ω were confirmed in two independent connection-design references. Nothing here is recalled from memory, and where sources disagreed the disagreement is written into the page above rather than quietly resolved.