I have some of these kicking around(4 feet long). I was told by a plumbing and heating suppier that a 4 foot section was good for 8700btu with 4gpm@180.F or 7000btu with2gpm@180.F. 700btu/foot seems awfull low.
I was thinking that a unit could be made that has sections 2 feet long. I would use a 1inch header top and bottom using the proper fittings. I have seen T's that are 1 inch straight thru with a 3/4 takeoff.
Nice polished copper after solder job would almost make it look like an antique radiator.
I have been researching heat systems for the new house we are planning, and 7000 BTU/HR sounds awfull high for a 4' unit. 4' of 3/4" copper pipe only has .7854 SQ/FT of surface area. The fins on the outside help transfer heat to the air, but it still has to transfer thru that .7854 SQ FT of pipe surface area to get to the press fit fins. The flow thru the pipe helps keep the fluid turbulated and the pipe heat saturated, but 7000 BTU/HR for 4' of radiator sounds overly optomistic. Here is one link I found that discusses this type heating system:
http://www.heatinghelp.com/heating_howcome4.cfmHere is a link to a manufactures brocure which has heat transfer tables in BTU/FT in 65F air at the bottom.
http://www.embassyind.com/pdfs/embassyind_system6.pdfYou will note both of these show between 500 and 2000 pounds per hour(1-4 GPM) fluid flow in their tables, which keeps the radiator heat saturated at the boiler temp along it's whole length. A listeroid providing thermosiphon flow rates will be unable to match these transfer rates, and will have a temp gradient/decrease from input to output, lowering the overall heat transfer. At those numbers, 16' in 65F air isn't enough for the full heat output of a 6/1. A fan on the coils would help, but I would not be overly optomistic of it's performance in 100F conditions...
Ron