Varying wood AV by physical qualities

[Vespers War]

Wood AV is a flat 0.2 per cm, but this overlooks (for the sake of simplicity) that woods have widely varying densities and hardnesses. One of the tests used on wood is the Janka test, which examines how many lb-ft of force it takes to embed a .444" steel ball halfway into a sample of wood. It varies from 380 lbf for Eastern White Pine to around 4-5,000 lbf for Australian Buloke. Buloke is also noted for being as hard as aluminum. I haven't worked this all the way through, but my first idea is that roughly every 500 lbf of Janka hardness might be 0.1 AV per cm, which would give the following examples:

0.1 - white pine, yellow poplar, chestnut
0.2 - red maple, black walnut, cherry
0.3 - hard maple, red pine, zebrawood
0.4 - osage orange, African rosewood, tigerwood
0.5 - live oak, southern chestnut, red mahogany
0.6 - bloodwood, ebony, ironwood

Part of this originates from reading recently about how the USS Constitution's layer of live oak in its hull is thought to be the reason the ship was so hard to penetrate. Near the waterline, it had roughly 12 inches each of white oak and live oak, in a 7-12-5 thickness from outside to inside. This would be AV 12 using the 0.2 AV per cm calculation. Using variable wood, it would be between AV 21 and 24 (white oak is right on the cusp between 0.2 and 0.3). That may not seem like a huge difference when looking at modern artillery, but black powder smoothbores tend to max out penetration at around 10-8-7-5 (it's hard for them to get over the 5 MJ threshold that allows Pen to go up to 28-25-21-13), so at medium range an AV 12 is often penetrated, while either of the higher AVs will see the shot bounce.

Feedback on how to improve the system or other systems for varying the quality of wood are highly welcome, since this is a very rough first draft of an idea.