FINITE ELEENT EXAMPLE

So for this one, I was analyzing a steel-portal-frame warehouse — the Mavoko warehouse job — in Robot Structural Analysis. I built out the full frame: columns, rafters, purlins, girts, bracing, the works, and ran it through wind loading at 30 m/s in the X-direction, using a proper simulation rather than just slapping on a flat pressure value. The first thing I always check is how the wind actually hits the building, because it's never uniform. You can see that in the pressure map — the roof and the leeward side are mostly in suction (pulling outward, shown in the blue/purple), while the windward wall gets pushed in (the red zones). That distinction matters a lot in practice, because roof cladding and purlin fixings usually fail from uplift, not from wind pushing down. So getting that map right upfront tells me where the risk actually is. From there I looked at how much the frame moves — the deflected shape. I exaggerate it visually so it's easy to see, but really I'm checking the actual drift number against code limits, because if the frame sways too much under wind, you start cracking cladding or people notice the building moving, even if it's structurally fine. Then I pulled the bending moment diagram, specifically at the knee — that's the joint where the column meets the rafter. That's basically always the hot spot in a portal frame. Here you can see the moment flips from about +65 kNm to -25 kNm right through that joint, and that sign change is the tell — it means the knee connection is carrying the worst combination of forces in the whole frame. So that's exactly where I'd focus the connection design — bolt count, end plate thickness, whether it needs stiffeners. Last, I checked the axial force running down the column — it builds up to about 260 kN compression near the base. That number goes straight into the column capacity check, where I'm verifying it can handle that compression combined with the bending moment without buckling, per whichever code we're using — Eurocode or AS4100. So basically, this one wind case walks through the full logic: figure out how the wind loads the building, check the building doesn't move too much, find where the internal forces peak, then verify the members and connections can actually take it. That's the same process I'd run on your structure — just with your geometry and your loads instead.