Hi Buddy,
A quick background. When water boils against the tube wall it starts as a small bubble forming on some sort of irregularity on the wall surface, these are called nucleation sites. If you add more heat the bubble formation gets more rapid until you reach a point where there is a constant steam film residing on the wall, this is called DNB or Departure from Nucleate Boiling. DNB is harmful because while water is a good coolant, steam is not (water is denser, has a higher rate of heat transfer and absorbs a lot of BTU during the phase change to steam...all of which remove heat from the tube wall; steam cannot begin to remove heat as readily). If you can't remove enough heat, the temperature destroys the tube.
At supercritical pressures there are no steam bubbles, thus there is no insulating dry film forming on the tube wall. Harry also mentions parallel flow paths. In conventional boilers this can be an issue, if one tube forms a steam film it creates a rise in resistance causing the flow to move to the other parallel tubes. The other tubes remain cooler due to the increased flow and are less likely to form a steam film whereas the tube with restricted flow just continues to heat further til it fails.
All well and good, as we see Harry assumes that supercritical conditions preclude DNB and thus prevent tube failure. Well, they do preclude DNB, naturally. Where he concluded that this prevents tube failure is beyond me; I assume he didn't do enough research and tried to apply superheated steam conditions to supercritical operation and missed a step. The old literature talks about the difficulties encountered operating supercritical boilers, failures were common.
Simply because there is no steam film does NOT mean that the tube can't keep getting hotter. If add heat to the tube faster than you pump water in to remove the heat, it gets hotter. Tube burnout is not caused by DNB, it is caused by the temperature rise associated with DNB. If we can induce the same temperatures, we still get the burnout.
This is where things get tricky. The volume to surface area ratio goes up with diameter, double the diameter and the surface area goes up twice but the volume goes up four time. If there is twice as much surface but four times the water, there is a limit to how readily the tube wall can give up the heat to the water. Shrinking the diameter restrict flow, however. Putting tubes in parallel allows you to get more surface area for the same flow than you would get for a larger tube, but the flow is less than a single tube of the smaller diameter. Your margin of safety is the difference between the operating temperature of the tubes and their failure temperature.
Naturally, to get the peak performance possible, Harry stipulated a temperature very close to the top end of what the tube alloys could tolerate. He also put in parallel flow paths. There is almost no way to keep the temperature identical in each path, that would require exactly the same total heat, cleanliness of tube, identical tube length, water flow and so on. If the average temperature is OK but you have enough of a spread between the hottest and coldest, you might still get a failure in the hottest tube.
Just to make things better, nothing in the Cyclone literature seems to deal with an interactive control system to keep fuel and water flow perfectly proportioned nor to thermostatic control OF EACH PARALLEL PATH. Unless very nitpicking design is executed (none of which I see in the Cyclone patents) this design almost can't help burning out at the kinds of temperatures they envision.
AI
