How to get a cleaner burn

Did my first fire of a RMH. I build glassblowing furnaces so I have a background using refractories. I used 2300 firebricks for most of it except where the raw wood goes in. My thinking was more insulation between the fire and stack the hotter it will be inside = cleaner burn. I’ve only ever seen one run on YouTube so I’m not sure how much smoke to expect during lighting and shifting of material but there was a lot. At my cleanest burn there was still visable smoke, much less but still there. Is that normal? Should it be 100% clear?

Hi Rajin;
It looks like you have your bricks "dry" stacked? no mortar yet?

If so then that would be one issue  to solve.
No matter how flat they are they will still leak air.
It needs to be completely sealed so the only incoming air is thru the feed tube.
I think if you seal it up you will be happier with the results!

Is that an L tube or a partially dismantled J tube Rajin?

You should only be seeing a small amount of smoke at the startup and then occasionally during the burn when the wood self adjusts.
I'm assuming your wood is dry and your not blocking too much of the incoming air with wood or ash.

Most days for me, I see quite a bit of steam coming out my stack that can be confused for smoke, but it is pure white and doesn't linger a long time in the air the way smoke does. It also doesn't have a smoky smell to it, but rather a pleasant mild woodsy smell.

The insulating firebricks and hard splits at the wood feed are both excellent choices; however, the dimensions don't look optimal. A 6" diameter or 6" x 6" square system is about the smallest that is easy to get to work reliably. You appear to have about a 4 1/2" x 6" riser and a 4 1/2" x 4 1/2" "burn tunnel"/port between feed and riser.

What kind of "rocket" did you see on youtube? There are tons of ill-informed experiments there. As you appear to be planning for a barrel around the riser, I assume you want to build a rocket mass heater. There are established proportions for the parts of a J-tube combustion core which your interpretation does not appear to follow. A close to square cross section (or round or octagonal for the riser)works best, and the right-angle transitions from feed to burn tunnel to riser generate beneficial turbulence for fuel/air mixing. The cross section should stay close to the same throughout the system except at a couple of critical places where more flow space is needed. There are three parts to a J-tube, the basic type of core, the vertical feed tube, the horizontal burn tunnel, and the vertical heat riser. Your feed should be as tall as the wood you plan to use, so usually 16", and the proportion of parts can be around 1:2:3, 1:2:4, or 1:1.5:3, all measured along the outside faces of the flow path. So a 16" high feed, a 24-32" burn tunnel, and a 48-64" high riser would be good. Your mockup looks like it would be 9"/16"/41" if there was a front on the feed to make it a J-tube configuration, which would be reasonable for short fuel.

I only see signs of fire in the base of the riser and port, which would not make it a proper rocket in function. The changes in direction give turbulence for mixing as well as the flow path length for full combustion to occur. If the fuel-air mixture reaches the top of the riser before combustion is complete, you will get smoke.