High alloy steel tubes used in superheaters and reheaters, unlike carbon steel, fail by creep rupture. Creep refers to the permanent deformation of tubes that are operated at high temperatures. Carbon steel tubes operate in the elastic range where allowable stresses are based on yield stresses, whereas alloy tubes operate in the creep-rupture range, where allowable stresses are based on rupture strength. The life of superheater tubes is an important datum that helps plant engineers plan tube replacements or schedule maintenance work. When a new superheater tube is placed in service, it starts forming a layer of oxide scale on the inside. This layer gradually increases in thickness and also increases the tube wall temperature. Therefore, to predict the life of the tubes, information on the corrosion or the formation of the oxide layer is necessary. The corrosion of oxide formation also reduces the actual thickness of the tubes and increases the stresses in the tubes over time even if the pressure and temperature are the same. The data on oxide formation were once obtained by cutting tube samples and examining them but are now obtained through nondestructive methods. There are also methods to relate the oxide layer thickness with tube mean wall temperatures over a period of time.
Creep data are available for different materials in the form of the Larson Miller parameter, LMP. This relates the rupture stress value to temperature T and the remaining lifetime t, in hours.
LMP = (T + 460)(20 + log t)
Every tube in operation has an LMP value that increases with time. LMP can be related to stress values and the relationship then used to predict remaining life. However, there are charts that give what is called the minimum and the average rupture stress versus LMP, and one can compute different life times with the different values. Also, it can be seen that even a few degrees difference, say 10°F in metal temperatures, can change the lifetime by a large amount, which shows how complex and difficult it is to interpret the results. Figure 3.22 Shows the relationships between LMP and minimum rupture stress values for T11 and T22 materials.
35 — |
|
30 — |
|
25 — |
|
20 — |
Cl O CD O |
15 ■ |
1» W V |
10 — |
M |
5 — |
|
N — |
32 33 34 35 36 37 38 30 40
LMP parameter CT+460X2a+togt)/10GD
Figure 3.22 Larson—Miller parameters for T11 and T22 materials.
Assume that a superheater of T11 material operates at 1000°F and at a hoop stress of 6000psi. What is the predicted time to failure? From Fig. 3.22, the LMP at 6000 is 36,800.
Solution: From the above equation, we can see that
36, 800 = (1460)(20 + log t), or t = 160,500 h
If a tube had operated at this temperature for 50,000 h, its life consumed would be 50,000/160,500 = 0.31, or 0.69 of its life would remain. If after this period of
50,0 h, it operated at, say, 1020°F and at the same stress level, then
36,800 = (1480)(20 + log t), or t = 73,250 h
And the number of operating hours at this temperature would be 0.69 x 73,250 = 50,728 h.
One can see from the above how sensitive these numbers are to temperatures and stress values. Hence we have to interpret the results with caution backed up by operational experience. Simplistic approaches to replacement of tube bundles are not recommended. It should also be noted that if the average rupture stress is used instead of the minimum value, the lifetime would be much higher, casting more uncertainty in these calculations.