thermal expansion of a tube

[mielke]

when calculating thermal expansion for a tube with an outside temperature significantly higher than the tube inside temperature, what temperature is used? the average temp, the higher, or the lower?

 

[davefitz]

corus;

an earlier post suggested a 800F temperature difference between id and od. Such DT's across a wall thickness are rare , but possible in a radiant furnace of a large boiler firing oil and the waterwall tube heated one side by a flame at a radiant heat flux of about 125,000 btu/hr/ft2 and the back ,unheated side of the tube cooled by low temp liquid which is flowing at a rate less than required to avoid DNB. If immediate burnout does not occur , the hot OD facing the flame will want to expand but will not be able to due to restrained by the watewall membrane panel and the massive cold side buckstays- the result is yielding on the hot side of the tube and eventually a circumferential crack .

yielding occurs whenever the stress imposed by the DT and physical boundary conditions prevent relief of the stress. Such yielding occurs with a DT over 200F for most ferritic alloys, and certainly at lower DT's for austenitic steels with a lower yield stress and a higher thermal expansion coefficient.

But i repeat the heat transfer problem needs to be solved first - the example I give here is unusual in the extreme

 

[Christine74]

A general rule of thumb for fixed tubesheet heat exchangers is that they are normally limited to a maximum temperature difference of around 100°F, or maybe as much as 150°F, between the shell side and tube side fluids.

An 800°F difference is absurd.

-Christine

 

[racookpe1978]

I've got a fundamental problem with the assumptions that all the thermal stress will occur across the tube radially: He's defining a pipe heating up 800 degrees from the "cold end" to the "hot end" - The 800 degree delta T will occur only at the last few feet of the pipe as at the heat exchanger boundary since the cooling fluid will be at different temperatures from inlet to outlet.

But that leaves the thermal expansion from inlet to outlet longitudinally ..

 

[unclesyd]

Where is the shell going?

The worst case would be no water and everything would be 1000°F