-
it's hard to do better than a round tube
Well yeah, that seems intuitively correct for sure.
But there'd certainly be some subtlety in a completely optimised shape, given you're joining two orthogonal tubes, and each end has clamps on it with chunky bits for bolts to thread into... But it probably only amounts to a handful of grams anyway.
The price of that Bastion stem! Sheesh
Kimmo-
there'd certainly be some subtlety in a completely optimised shape
There might be tiny gains at the ends which can be realised by additive manufacture which are impossible with forging plus a 5-axis closed loop CNC second operation, but as you say, even the most ardent weight weenie has other places to look before getting to $200/g savings on the stem, and sprinters looking for maximum stiffness can use 200g stems and still make the UCI minimumffor total machine weight.
gbj_tester
The Grammo seemed to offer a miraculous combination of high performance and light weight when introduced because it was hollow, and competing against solid stems. With metal stems, you tend to run out of stiffness before you run out of strength. Since torsional stiffness in the extension is where your trouble is, it's hard to do better than a round tube. Outside of weight-weenie contests, there's little to be gained by making a stem more expensive than a net-shape forged 7000-series aluminium alloy job, because those can already be done stiff and strong enough at just over 100g for a 100mm stem. Using either exotic alloys or additive manufacturing isn't going to get you 30g off that. The 3D printed stems currently available (Mythos, Nex-Gen, Bastion) show no benefit over much cheaper forged aluminium alloy models. At the high end, everything is heading in the direction of integrated bar-stem construction for the aeros.