First reaction is that the horizontal burn tube is too long. Second, not enough rise in the stack to achieve barometric differential (chimney action), and third -- have no fear -- solution is simple: just add (6"?) by 2 ft. sections of stovepipe until you get that audible "rocket effect". Also, if new stack section(s) are insulated, depending on your application, it will be safer, and enable the chimney action to take effect much more quickly. My stove is also a boiler with 2 coils --ergo, high restriction, so I use a small draft inducer fan just to get things established. Further, we must remember the simplicity of the "rocket" concept. which is the shortness of the combustion area as it relates to the 90 degree upturn of the "stack", and the turbulence caused by the turn. I complicated the basic principle by using "on-hand" materials, and by preferring to use H2O as my mass heat storage medium. Forgive me for that, since I am an HVAC pro, enamored of water as a transferral method, since water lends itself to being heated by a great variety of means. Also, storage is easier, provided one has midrange plumbing skills. The draft problems in my system are mainly caused by the efficinecy of heat transferral to the water medium, thereby resulting in much lower "stack temperatures". The addition of a draft inducer creates much the same effect as it would in a "90% plus" efficiency furnace, in that it tends to create a vapor plume and condensate in the flue. These problems are not insurmountable -- all one has to do is anticipate the "low" spots, and provide passive drainage. Also, the draft inducer can be thermostaticallly controlled, so that it is not running 100% of the time. Actually, the higher your combustion temperature in the burn tube, the less likely you will have to handle condensate accumulation. My RMH boiler runs at temperatures where the flue temperature has yet to exceed 90 degrees Fahrenheit. I do get a little vapor plume at this temp, but not enough to have a condensate problem. This could probably change, depending on the moisture content of the fuel I am burning. In fact, most of the time after the combustion is established, the effluent is CLEAR. Oh, P.S.: Remember that controlling the burn is instrumental in the absolute maximum amount of heat obtainable from your chosen fuel, ergo, the maximum efficinecy. Therefore, the lower your final exhaust temperature. the greater the amount of heat you have "reclaimed".
Why not consider putting copper coils around the rocket riser, then run "PEX" piping underground to the pond into another copper coil which would be your heat exchanger for the pond water. At this scale, I would imagine that 3/4" tubing would be the minimum size required. If idle-use freezing is an issue, your could use just enough propylene glycol (RV antifreeze) to fill the tubing, and a small, atmospheric feed reservoir. You would also need a small circulator pump for the system. It is possible to eliminate the need for a pump, by upsizing the tubing on the return side of the system, thereby achieving "gravity" flow. The pump would be much more efficient, though, in terms of speed of heat transfer. :
I have built a "combo" RMH unit that will produce steam and hot water that runs 2 separate coils, 1 around the 4" (insulated)riser, and another looped through the secondary chamber. It works quite well, except it can be a little ticklish to get "kindled up". My stack temperature has yet to exceed 80 degrees F. I have overcome the initial draft problem by the addition of a draft inducer blower, installed in such a way that it can be shut down once the secondary temp is sufficient for sustained drafting. Basically, this works just like a modern 90+% efficiency gas furnace, except it is not necessary to leave the blower in operation during firing. The coils are valved in such a way that I can use them separately or in tandem for water & steam, steam, or just water.