racookpe1978
Nuclear
Vendor (a competitor) is quoting NEMA SM 23 limits on their new turbine nozzles to replace a unit that crashed a few weeks ago. As you would expect, those very restrictive turbine nozzle forces and moments are causing extreme cost and material increases in the replacement steam piping, pipe snubbers and restraints, and the new steel holding those new restraints and supports.
NEMA SM 23 (Mechanical Drive Turbine Specifications) is no longer maintained by NEMA; worse, NEMA 23 is not strictly relevant to steam turbine limitations for generator turbines. (NEMA SM 24-1991 is the right spec for generator turbines.) Reading both NEMA standards, it is obvious both SM 23 and SM 24 merely maintain the original 1950-1960 era limits of NEMA 21-22. (NEMA 23 Section 8.4.6 is word-by-word duplicated in NEMA 24. The sample problems are also copied from 23 to 24.))
Are there any other standards appropriate?
It appears that NEMA 23/24 are the most basic, most restrictive documents "in use" although neither is maintained by the NEMA, neither is theoretically nor practically appropriate, and neither offers a realistic evaluation of the actual force and moments on either individual nozzles nor the turbine casing as a whole.
Per NEMA 23/24, maximum allowed forces and moment on each individual nozzle are directly proportional to nozzle dia if less than 8 inches, and increase by 1/3 dia if dia is greater than 8 inches.
D_eq = (16 + Dia)/3
If Dia = 6, D_eq = 6.00
If Dia = 8, D_eq = 8.00
If Dia = 10, D_eq = 8.66
If Dia = 16, D_eq = 10.66
If Dia = 20, D_eq = 12.00 etc.
On each nozzle, maximum forces and moments are: 3*Fr + Mr < 500*D_eq (Section 8.4.6, Limit 1) for Fr in lbf and M in Ft-lbf.
Problem 1: The calculation uses inconsistent units (unless "3 ft" is assumed an arbitrary nozzle length for some reason).
Problem 2: The NEMA limit cannot be translated to metric units.
Problem 3. Incorrect calculation.
But even these assumptions are physically wrong: stiffness and strength of a nozzle is not linearly proportional Dia, much less Dia/3 just because the nozzle is larger than 8 inches. Turbine casing movement (and distortion is the most important restriction on a nozzle load on the casing) is not proportional to arbitrary nozzle face forces and moments calculated this way on real turbines. Even in the slide rule era, turbine distortion was known to be proportional to nozzle lever arm; nozzle position on the casing; nozzle diameter; nozzle wall thickness and nozzle pad reinforcement; turbine casing thickness, dia, length, and axial position.
Peng & Peng (Pipe Stress Design) shows measured turbine deflection strength is actually proportional to nozzle dia^2.
Even after all of these (flat-out errors) inconsistancies, the NEMA 23/24 Limits 2 totals all forces and moments from all nozzles together, then compares the total nozzle forces and moments to a single mythical Dc (proportional to the area of all nozzles added together.
Thus 2*Fc + Mc < 250*Dc (Section 8.4.6.2, Limit 2)
where Dc = sqrt((sum of all nozzle areas)/4*pi)
NEMA 23/24 Limit 3 is more arbitrary:
Fx < 50*Dc (Section 8.4.6.2, Limit 3)
Fy < 125*Dc
Fz < 100*Dc
Mx < 250*Dc
My < 125*Dc
Mz < 125*Dc
where Fx is the sum of all axial nozzle forces,
Fy is the sum of all vertical nozzle forces,
Fz is the sum of all horizontal nozzle forces,
and Mx, My, and Mz are the sum of all of the nozzle moments.
(Yes, units are inconsistent; and no, the requirement cannot be translated to metric units.)
/rant
Now that the industry has replaced slide rules with spreadsheets, and replaced 2-D spreadsheets with 3D finite element analysis, and now that we (collectively) must buy and sell internationally in international units, how does the turbine industry get rid of these demonstrably wrong 1950-1960 NEMA nozzle force and moment limits?
Or, at least update them to a rational criteria?
NEMA SM 23 (Mechanical Drive Turbine Specifications) is no longer maintained by NEMA; worse, NEMA 23 is not strictly relevant to steam turbine limitations for generator turbines. (NEMA SM 24-1991 is the right spec for generator turbines.) Reading both NEMA standards, it is obvious both SM 23 and SM 24 merely maintain the original 1950-1960 era limits of NEMA 21-22. (NEMA 23 Section 8.4.6 is word-by-word duplicated in NEMA 24. The sample problems are also copied from 23 to 24.))
Are there any other standards appropriate?
It appears that NEMA 23/24 are the most basic, most restrictive documents "in use" although neither is maintained by the NEMA, neither is theoretically nor practically appropriate, and neither offers a realistic evaluation of the actual force and moments on either individual nozzles nor the turbine casing as a whole.
Per NEMA 23/24, maximum allowed forces and moment on each individual nozzle are directly proportional to nozzle dia if less than 8 inches, and increase by 1/3 dia if dia is greater than 8 inches.
D_eq = (16 + Dia)/3
If Dia = 6, D_eq = 6.00
If Dia = 8, D_eq = 8.00
If Dia = 10, D_eq = 8.66
If Dia = 16, D_eq = 10.66
If Dia = 20, D_eq = 12.00 etc.
On each nozzle, maximum forces and moments are: 3*Fr + Mr < 500*D_eq (Section 8.4.6, Limit 1) for Fr in lbf and M in Ft-lbf.
Problem 1: The calculation uses inconsistent units (unless "3 ft" is assumed an arbitrary nozzle length for some reason).
Problem 2: The NEMA limit cannot be translated to metric units.
Problem 3. Incorrect calculation.
But even these assumptions are physically wrong: stiffness and strength of a nozzle is not linearly proportional Dia, much less Dia/3 just because the nozzle is larger than 8 inches. Turbine casing movement (and distortion is the most important restriction on a nozzle load on the casing) is not proportional to arbitrary nozzle face forces and moments calculated this way on real turbines. Even in the slide rule era, turbine distortion was known to be proportional to nozzle lever arm; nozzle position on the casing; nozzle diameter; nozzle wall thickness and nozzle pad reinforcement; turbine casing thickness, dia, length, and axial position.
Peng & Peng (Pipe Stress Design) shows measured turbine deflection strength is actually proportional to nozzle dia^2.
Even after all of these (flat-out errors) inconsistancies, the NEMA 23/24 Limits 2 totals all forces and moments from all nozzles together, then compares the total nozzle forces and moments to a single mythical Dc (proportional to the area of all nozzles added together.
Thus 2*Fc + Mc < 250*Dc (Section 8.4.6.2, Limit 2)
where Dc = sqrt((sum of all nozzle areas)/4*pi)
NEMA 23/24 Limit 3 is more arbitrary:
Fx < 50*Dc (Section 8.4.6.2, Limit 3)
Fy < 125*Dc
Fz < 100*Dc
Mx < 250*Dc
My < 125*Dc
Mz < 125*Dc
where Fx is the sum of all axial nozzle forces,
Fy is the sum of all vertical nozzle forces,
Fz is the sum of all horizontal nozzle forces,
and Mx, My, and Mz are the sum of all of the nozzle moments.
(Yes, units are inconsistent; and no, the requirement cannot be translated to metric units.)
/rant
Now that the industry has replaced slide rules with spreadsheets, and replaced 2-D spreadsheets with 3D finite element analysis, and now that we (collectively) must buy and sell internationally in international units, how does the turbine industry get rid of these demonstrably wrong 1950-1960 NEMA nozzle force and moment limits?
Or, at least update them to a rational criteria?
