leedrong-
You weren't too specific in your application and the previous responses assume a Section VIII application in which case I mostly agree with their responses. If your design temperature is 1004°F and you are concerned about using allowable tensile stresses limited to 1000°F then you need a quick study in significant figures and engineering judgement: I'll just bet that the 1004°F (4 sig. figs) value is actually derived from a nice round 540°C (2 sig. figs). In other words, consider the design temperature 5.4*10^2 °C or 1.0*10^3 °F and you're fine.
However... you are talking about tubes at high temperatures, and that makes me think that you are looking at fired heater tubing such as the coils in a coker furnace. In this case, the design is not Section VIII, but API 530. I didn't find SA213-T1 in IID (I'm sure it's there; just didn't look enough) but if its a C-½Mo material then API 530 will allow you to use it up to 1100°F and 1325°F for limited durations.
External pressure on tubes is usually a joke. Run the numbers on a 5" tube 0.1" thick (L/Do>10, Do/t = 50 so A ~ 0.0005). Use Young's modulus (~20e6)instead of temperature to enter the curves in Fig. CS-2 and you'll find that the allowable external pressure (at 1000°F)is around 100 psi. You can check this result using VIII-1 UG-28(c) Step 7.
jt
