Design of Steel-to-Timber Dowel/Bolt Connections Under Combined Loading to Eurocode 5

Steel-to-timber dowel and bolt connections look simple on drawings, but these joints are where “clean” global actions become messy local demands, especially when the connection must carry moment, shear, and axial forces at the same time. The difficulty is that connection design sits at the intersection of group mechanics (how a fastener layout shares load under moment), shear plane bookkeeping (single vs double vs multiple shear planes), brittle timber behavior that can govern before a ductile Johansen mechanism is fully mobilized and compatibility.

 

At SPEC Toolbox, we develop a calculator for steel-to-timber dowel/bolt groups under combined actions. Our software team has since converted that workflow into a web-based application, and this blog explains the engineering logic behind it using the current Eurocode 5 framework and highlighting what is changing in the draft next generation (FprEN 1995-1-1:2025).

It focuses on steel-to-timber connections using dowel-type fasteners (bolts and dowels) under combined actions in the shear plane. The calculator supports:

• Actions: moment (M), shear (V), and axial force (N) acting together.
• Connection types: single shear, double shear, and multiple shear plane arrangements.
• Fasteners: bolts and dowels (with Eurocode-specific rope-effect limits and detailing requirements).

For a single dowel under a single shear force, the force path is straightforward: the fastener bears in timber (embedment), may yield in bending (plastic hinges), and if applicable, may develop some rope effect.

For a fastener group under moment, the question shifts from “what is the capacity of one fastener?” to:

• How does the group rotate and share load?
• Which fastener becomes critical?
• How does the simultaneous shear (horizontal/vertical) shift the critical fastener and change its force direction?
• How do multiple shear planes (built-up members, slotted-in plates) share resistance without violating compatibility?

This is a normal condition for portal frame, column shoes, diaphragm collectors, truss nodes, and retrofit interfaces. The key is not that there are three actions. The key is what that does to the force pattern inside the fastener group:

• Moment attracts demand to the fasteners farthest from the group centroid (outer fasteners often govern).
• Shear shifts the governing fastener (often a corner or edge fastener) and changes the force direction in the timber.

So the “design problem” becomes: Which fastener is most utilized, in which shear plane, in which direction, and does Eurocode allow you to combine the plane resistances the way you’re assuming?

Eurocode 5 frame: per fastener, per shear plane resistance

Eurocode 5 treats lateral resistance of dowel-type fasteners per shear plane. The characteristic resistance is obtained from the Johansen (European Yield Model) failure modes, combining timber embedment strength and fastener yield moment (EN 1995-1-1:2014, 8.2). For steel-to-timber joints, the standard provides dedicated expressions depending on steel plate thickness and plate position (EN 1995-1-1:2014, 8.2.3). For steel-to-timber joints, the characteristic lateral capacity per shear plane depends strongly on:

• steel plate configuration (thin vs thick, inner vs outer plate placement),
• timber embedment strength and fastener yield moment,
• whether the governing Johansen mechanism is embedment-controlled or hinge (ductile) controlled.

That “per shear plane” framing is not just bookkeeping; it becomes critical when you move from:

• single shear (one interface),
• to double shear (two interfaces),
• to multiple shear planes (three or more interfaces in built ups or multi-layer details).

With multiple shear planes, we need to differentiate between resistance per shear plane and per fastener; mixing these levels is one of the common sources of confusion in hand checks.

From joint actions to fastener forces

Eurocode 5 provides a resistance model. It does not prescribe a single universal method to distribute combined actions to individual fasteners for every real joint geometry. In practice, most timber engineers use a transparent rigid fastener-group interpretation for the moment component:

• The fastener group is assumed to rotate about a centre (often close to the centroid for symmetric layouts).
• Slip demand increases with distance from the rotation centre, so outer fasteners attract the largest moment-induced demand.

Direct shear is then combined with the moment-induced demand, typically by vector sum, to identify the governing fastener and its resultant force direction in the shear plane. The design then finds the resultant force on the critical fastener in the governing shear plane(s) and compares that demand to the per-shear-plane resistance, while also verifying brittle limits and spacing rules.

Multiple shear plane connections

Multiple shear plane connections tempt you to do a simple sum: “two planes, so twice the resistance.” Eurocode 5 is clear that this only works when the failure modes and deformation patterns are compatible. For multiple shear plane connections, Eurocode 5 instructs you to evaluate each shear plane as part of a series of threemember connections, and then it adds the critical limitation: You may only combine the resistances from individual shear planes if the governing failure modes are compatible.

The reason is Different Johansen modes mobilize at different deformation patterns (embedment-only vs hinge formation). If one plane is trying to develop a ductile hinge mechanism while another plane is governing in an embedment-controlled mechanism, the joint will not reliably mobilize the “sum of peaks” simultaneously. Load redistributes before you ever get there.

For steel-to-timber joints, the shear configuration controls both the available Johansen modes and the way resistance can be combined. Eurocode 5 explicitly requires failure-mode compatibility when combining resistances in multiple shear plane connections (EN 1995-1-1:2014, 8.1.3). Eurocode even flags specific failure mode families that must not be mixed when you are trying to sum plane resistances. The code identifies specific failure mode families that should not be mixed when summing plane resistances, because different modes mobilise at different deformation patterns and slip levels.

The draft FprEN 1995:2025 makes the same principle more explicit by trying summation to two conditions failure mode compatibility, and displacement compatibility of the members. It also becomes more prescriptive for connections with many planes, introducing thickness/geometry conditions intended to force compatible displacement behavior.

Rope effect: useful, but tightly controlled

In the current EN framework, rope effect is included in the lateral capacity formulation but is limited, and the allowed contribution depends on fastener type. In particular:

• Dowels: rope effect is not credited (effectively taken as zero for lateral capacity),
• Bolts: rope effect can be credited, but the code limits how much of it you can count relative to the Johansen part.

This aligns with how we should think in design: dowels are primarily shear devices; bolts can develop some axial restraint and clamp effects (depending on washers, plates, fit up, and gaps), but you still shouldn’t design a lateral shear connection assuming “full rope effect” unless detailing truly supports it.

The draft prEN 1995-1-1:2025 makes this topic more transparent by decomposing lateral resistance explicitly into dowel-effect plus rope-effect parts, with clearer conditions and limitation factors. Conceptually, it becomes: Lateral resistance per shear plane = dowel effect + rope effect, with clear conditions on when rope effect is allowed and how it is limited. That is an improvement from a transparency standpoint, and it will make it harder to over-credit rope effect inadvertently.

Brittle failure: where many real connections govern

A steel-to-timber dowel/bolt group can satisfy Johansen (ductile) resistance and still be governed by brittle behaviour in the timber around the fastener group. This is especially common when end distances are tight, spacing is tight, there are multiple rows, holes reduce net section, the load path introduces perpendicular-to-grain tension components, or you have multiple nearby connections interacting.

Current EN 1995-1-1 includes splitting checks for forces acting at an angle to grain (EN 1995-1-1:2014, 8.1.4). That verification is driven by the perpendicular-to-grain component of the transmitted force. This is one of those checks that can quietly govern a connection that “looks fine” in shear.

The draft prEN expands brittle verification into a structured design track for multiple-fastener connections, with explicit models for splitting, row shear, block shear, plug shear (partial penetration), and net tension failure, and with more explicit treatment of perpendicular-to-grain splitting and interaction of multiple nearby connections (FprEN 1995-1-1:2025, Sections 11.5 and 11.6).

Designing Steel-to-timber dowel and bolt connections under combined loading are not hard because Eurocode 5 is unclear. They’re hard because the design space is full of small decisions that change the mechanism: shear plane configuration, plate thickness, mode compatibility, rope effect limits, and brittle failure surfaces. The purpose of this blog is to make that logic transparent and consistent, grounded in EN 1995-1-1 as used today, while staying technically aware of where FprEN 1995-1-1:2025 is heading.

References

• CEN. EN 1995-1-1:2004+A1:2008. Eurocode 5: Design of timber structures – Part 1-1: General – Common rules and rules for buildings.
• CEN. FprEN / FprEN 1995-1-1:2025 (draft). Eurocode 5: Design of timber structures – Part 1-1. Draft provisions referenced in this post include Sections 11.2.3 (dowel effect, rope effect, and multi-shear-plane compatibility), 11.5 (brittle failure parallel to grain) and 11.6 (splitting and brittle failure perpendicular to grain).