Brilliant design idea for the chimney extension! The idea is to make sure that the air flow is at least what it would be with just the factory chimney. To do this you have to look at a lot of factors, most of which I cannot calculate or measure; turbulent or laminar flow, friction, and so on. What I did is calculate what is called the hydrostatic diameter. This measurement relates the effective area of the opening of different shaped pipes. The formula is HD = 4 * area/p where HD the hydrostatic diameter is 4 times the area of the opening divided by the parameter of the pipe. This works for circular and rectangular pipes. You will notice that as the rectangular pipe becomes flatter and moves away from a square the effective diameter gets smaller. That means that a 2x6 inch opening is way better than a 1x12 opening. I did some basic friction calculations and found that a rectangular opening in a 2:1 ratio was no worse than a circular opening. You can calculate the hydrostatic diameter of your extension and compare it to the diameter of the factory chimney. Also, the shorter the extension the lower the friction.

You did a great job with your grill! Thanks for posting. I would not have thought of what you did with the chimney extension.

I'm currently trying to measure the effect that such a chimney extension makes in the temperature distribution. What I've found is that with the chimney extension the dome temperature is 15F hotter and the grill surface is 20F warmer. I don't think this is due to a constriction because that would have caused the fire to go out and when I remove the extension the temperatures dropped. I think I was getting good flow. This is the test rig I'm using. The final version will be made of smooth metal - too much friction and turbulent flow. I think the measurements I need are midway on the grill and at the far end (I pile the coals at one end for indirect heating).

I like your solution because it takes less room. My test rig sticks 4 inches into grill area. I could hide in in the corner but that makes it longer and I think I want the flow centered to help keep the temperature distribution to a minimum.