I have read of many 2.5"-3" swaps with no significant gain, if any on some stroked 351s.
The camshaft timing has to be tied into this. With a camshaft that has less and later exhaust activity, as opposed to "vice versa," you may see some gains. With later exhaust timing, and adding larger exhaust to what already exist, it can decrease cylinder pressure, and there-by relieving you of some torque you once had.
HP = Torque x RPM/5252. What happens to the end product of horsepower, if your "plugged-in" torque is decreased (loss of cylinder pressure)?
This is why the camshaft has got to be tailored for the big exhaust.
The primary tubes can vary from 12"-36", or are you alluding to something else?
I do not see a 347 using anywhere near the flow capabilities of a 90mm TB.
A 90mm MAF will flow differently than a 90mm TB, because of obstructions and design.
Displacement * RPM * VE / (1728*2) = Volumetric flow of the engine in CFM
The 1728 converts from in^3 to ft^3, and the 2 comes from dividing RPM by 2 in order to get the actual number of intake strokes.
Make sure you type in 1.00 for VE instead of 100.
Take a 7,000 RPM 347: You are looking at 703 CFM. A 90mm TB flows almost 1400 (1369) CFM. Why do you need the extra 95% more flow? Keep in mind, this 347 is at 100% VE. Not to many of those around...
Now an engine does not see 28" all the time of course, it can be up and it can be down, but there are conversion factors for this, which will end up drawing even more air in. As the vacuum goes up, so with the CFM.
More air does equal more power, but only when you can use it.
Not trying to offend, just trying to see where you are coming from 347HO...