____ Thanks further-much for the follow-up info on the Mk2-carb's common flooding-issue, as I believe that's more likely what Dave probably has to deal with.
However before Dave jumps-forward with THAT probable-possibility, I hope he will first try lowering his throttle needle-valve back-down all the way, and lean-out his low-speed idle/pilot-jet circuit, (to find-out if doing that may then make his Duke run well enough).
Also, I've now seen that lower float-settings (than just "2mm") can be achieved, so if the two adjustments I've suggested don't achieve the desired leaned-out results, then next Dave should try lowering the float-level down to as much as about 6mm.
__ And then if Dave STILL has rich-running issues, he can THEN expect that HIS-own Mk2-carb is also suffering from the common flooding-issue that others have been dealing with.
____ Okay, what seems to be claimed here is that the (original/small-sized) float doesn't have enough lifting-pressure to keep the flow-valve's needle-tip properly seated against the Amal-Mk2's larger surface-area/opening in it's valve-seat, (which is supposed to control fuel-flow into the float-bowl/fuel-chamber, despite the increased pressure exerted by the fuel [due to the larger valve-seat area] ), WHENever the engine shakes the carb.Take a look at the float valve seat and compare it to a MK1. What you'll see is that the the MK2 has a bigger valve area than the Mk1. The problem is that as standard they use the same float. The result is that when you bounce the Mk2 around, and Commando's do a lot more shaking than most bikes, there's a much greater tendency for the Mk2 float valve to unseat and flood the carb.
__ So if indeed that's the actual case, then (besides the suggested updated float), clamping-down the inlet-end of the carb.body should cut-down on it's being shook (to a lesser degree, at least) by engine-vibration, and thus-then help alleviate the flooding-issue.
Enlightened-Cheers,
-Bob