PREPRINT-ICPWS XV Berlin, September 8-11, 2008
Design Factors and Water Chemistry Practices - Supercritical Power Cycles Frank Gabrielli Alstom – Windsor, USA,
[email protected] and
Horst Schwevers Alstom – Mannheim, Germany,
[email protected] The paper provides an overview of current supercritical plants with emphasis on boiler design characteristics, materials of construction for both boilers and turbines and water chemistry requirements. The paper also discusses material requirements for the ultra-supercritical pressure cycle with pressures and temperatures approaching 375 bars and 760C. In this respect, it highlights water chemistry and material issues/requirements that still need to be researched and evaluated.
and designing for sufficiently high mass velocity that ensures cooling of the furnace tubes. Drum units maintain the proper mass flows by the use of either natural or forced circulation. These boilers are designed to generate steam in the furnace walls under nucleate boiling conditions. Nucleate boiling is characterized by formation and release of steam bubbles at the surface-liquid interface with the water continues wetting the inner surface of the tube. The heat transfer coefficients in the nucleate boiling regime are high and a temperature gradient between the metal tube and the fluid inside the tube is relatively small. In a once-through boiler, waterwall mass flow changes in direct proportion to steam flow. In the supercritical pressure region, the fluid inside the tubes is heated and the heat is directly converted into a higher temperature. In the subcritical once-through pressure region the process of heat transfer is more complicated and the process involves a change in phase from liquid to steam as well as superheating. In most practical situations, a fluid at a temperature below its boiling point at the system pressure enters a furnace tube in which it is heated so that progressive vaporization and slight superheating occurs. The process of heat transfer during vaporization depends on many variables. As the quality of the steam-liquid mixture increases, various two-phase flow patterns are encountered.
Introduction Optimal economy demands high operational flexibility from power plants, which in turn, requires that the plants are suitable for a variable load program and two-shift operation (i.e. cycling operation). In order to satisfy these requirements, excellent dynamic behavior and high load gradients are absolutely essential. Current supercritical designs have greater operational flexibility and are better suited for cycling duty than comparably sized drum units. However, the requirement for daily cycling and/or two- shift operation can create undesirable thermal stresses in the steam turbine and boiler pressure parts components. These concerns can be minimized by adopting sliding pressure operation that significantly minimize temperature differences in the boiler and steam turbine systems. The emphasis of designing more flexible units as well as extending component life has led to the wide use of sliding pressure operation. Basic Design Considerations Furnace walls are formed by finned or fusion welded tubes that form a continuous watercooled envelope. The biggest concern with any sliding pressure, supercritical design is created by the requirement for once-through operation
1
The designers must be concerned with two critical conditions: Departure from Nucleate Boiling (DNB) and Dryout (DO). DNB and DO are characterized by formation of a flow of steam which covers the inner surface of a tube, a sharp decrease in heat transfer coefficient, and a consequent high metal temperature rise. DNB is of major concern at operating pressures of 2900 psig (204 bar) and higher since, at these pressures, it is possible for DNB to occur even in sub-cooled and low quality regions of the furnace where heat fluxes are relatively high. While DNB may also occur at lower pressures, of greater concern at these pressures is DO which is unavoidable as long as a boiler operates in once-through mode. The requirement for once-through operation brings about a second design challenge; namely, avoiding potentially damaging stresses resulting from temperature differences at the furnace wall outlet. A drum unit always operates with saturation temperature in the waterwalls and all
circuits are at the same fluid temperature. With a once- through design, the steam outlet of the furnace walls is slightly superheated. Therefore, tube circuits can be at different temperatures due to the variation in heat absorption patterns around the furnace perimeter. These temperature differences must be maintained within acceptable limits. Design strategies to deal with the above concerns are available and are described below. Spiral Wall Design The spiral wall design as it stands today has over thirty years of experience behind it and can be applied to all unit sizes, pressures, and fuels. The basic concept of the spiral wall is to increase the mass flow per tube by reducing the number of tubes required to envelop the furnace wall without increasing the spacing between the tubes. Figure 1 illustrates this concept.
Figure 1. Spiral Wall Design
2
Figure 2. Evaporator Wall Design
furnace region. This transition to a vertical wall is accomplished in a zone where heat fluxes are relatively low and the requirements on tube cooling are not as high as in the lower furnace zone. The transition requires the use of an intermediate header or bifurcated/trifurcated fittings. Due to the fact that the furnace wall tubes are at an angle, the tubes are not self-supporting. The supporting load must be transmitted to the vertical tubes forming the upper section of the furnace enclosure.
Figure 2 shows a comparison between the inclined tube arrangement and the vertical tube arrangement. The spiral furnace wall system employs smooth bore tubes which require high mass velocity to provide acceptable tube cooling. The high mass velocity produces high film conductance that ensures low metal temperatures and thus economical selection of tube materials. Fewer and longer tubes combined with higher mass flow rate, however, produce a higher pressure drop in the furnace walls. For a given furnace size and fuel, the expected heat flux rate determines the mass velocity rate required to ensure the cooling of the furnace wall tubes. With a spiral wall design, the number and size of the tubes are selected to provide sufficient cooling over the entire load range. Additionally, by spiraling around the furnace, every tube is part of all four walls, which means that the difference in length between the furnace tubes is minimized and that the heat pickup by individual tubes is approximately the same. This makes the spiral wall system less sensitive to changes in the heat absorption profile in the furnace. Practical design considerations require a vertical waterwall configuration in the upper
Vertical Wall Rifled Tubing As an alternative to the spiral wall designs, Alstom Power has developed a design that uses conventional vertical tube walls for ease of construction and maintenance. The design is a derivative of the Combined Circulation® unit which we have continually improved since its introduction in 1961. Rifled tubing is used in the furnace walls. A vertical wall configuration typically employs either 1-1/4” (31.8mm) or 1-1/8” (28.6mm) O.D. internally rifled tubes (Figure 3).
3
Figure 3. Rifled Tube Rifled tubing offers significant advantages over smooth tubing in the evaporation range of sub-critical pressure boiler operation. The use of rifled tubing permits much lower mass flows with the same margin of protection against DNB and overheating. Extensive testing at the Alstom Power Plant Laboratories has characterized the heat transfer and flow behavior of rifled tubing for vertical tube furnace applications. Rifling promotes turbulence and aids in wetting the inside tube surface, consequently, increasing the nucleate boiling quality for a given heat flux, mass velocity, and pressure. Similar to the spiral-wound design, the waterwall panels can be formed by either fin or fusion welded tubes. Typical average mass velocity per tube is much smaller than for the spiral arrangement and still retain more than adequate margin of safety due to the combined effect of a smaller tube diameter and rifling.
All once-through supercritical units must be designed to minimize temperature differences in the furnace walls. Waterwall outlet temperature deviations are minimized in the same manner as on Combined Circulation® designs. This is accomplished by installing individual tube orifices that distribute flow in relation to the heat absorbed by each circuit. Detailed waterwall analysis based on operating experience and design practices for Combined Circulation units show that satisfactory temperature differentials throughout the entire operating load range can be achieved with sliding pressure operation. The temperature difference between the maximum and minimum value is well within acceptable range. In addition, there are orifices installed in a supply sphere (figure 4) to insure that the prescribed amount of flow is distributed to each wall.
4
Figure 4. Furnace Waterside Arrangement
Steam Components
system. This means that during low load and start-up, the steam generators are operated in a sub-critical pressure range. Therefore, to facilitate a satisfactory boiler service, a low load start-up system is provided. Selection of a minimum once-through flow/load depends on such factors as a mode of operation, circuit stability, and tube materials. For boilers that are primarily base loaded, the once-through minimum load should be selected as high as possible. This results in the lowest pressure drop in the waterwalls at a full-load condition. Steam generators that are required to cycle must be designed for a lower once-through minimum load so that once-through flow operation is extended to the lowest load practical. Commercial experience with a minimum once-through flow down to 35% to 40% load has proven to be successful. Lower once-through loads are also feasible. The start up system includes a water separator located between the waterwalls and the primary superheater, a water storage tank and a drain water discharge system with heat recovery capabilities. The water separator consists of one or more vertical vessels with tangential inlets (Figure 5).
Two principal arrangements of heating surfaces are utilized in Alstom for reheat steam generators. These are the pendant panel surface (two-pass) and the horizontal surface (tower) designs, both of which are used in sub-critical and supercritical applications. While the basic concepts of surface arrangements are similar, there are differences in the location of various sections in the furnace. The decision to use a pendant panel design versus a horizontal design is not dependent on the cycle choice (i.e. subcritical vs. supercritical). Each of these configurations has its advantages and allows for customer preference as a factor in the final arrangement of the heating surfaces. Start-up Systems General Design Considerations Today’s supercritical power plants are designed to follow a rigorous load program that often includes two shift or cycling operations. To effectively accommodate this operating requirement, a steam generator must be capable of sliding pressure operation in the whole
5
Figure 5. Steam-water separator system and water storage tank
Steam outlets are located in the upper part and the drain is discharged through the lower part. The water separator is in “wet condition” when it operates in flow recirculation mode and it is “dry” when the flow is once-through. The drain discharge systems with heat recovery capabilities can be of two types, indirect heat recovery and direct heat recovery There are also two types of direct heat recovery systems that are available. The first one is a system with a low load recirculation pump; the second one is a system that includes a drain return line via a heat exchanger into the deaerator/feedwater storage tank. The suitability of each system depends on the economic evaluation associated with operational requirements of a steam generator.
as waterwalls and finishing superheat/reheat sections are briefly discussed. Materials for Water walls Furnace tubes are subject to the highest heat flux in the furnace. Typical tubing alloys applied in the commercial designs are: 1.25 Cr-0.5Mo (T-12) material, 2.25 Cr – 1.0 Mo (T22) material and a European developed material 15Mo3. These alloys have excellent mechanical properties suitable for easy fabrication of the waterwall panels. Design limit of T12 was studied for a cycle with the superheater outlet steam conditions up to 4060 psig (280 bar) and 1112°F (600° C). For non-corrosive coals and based on the furnace outlet gas temperature of 2280°F (1250°C), T12 can be applied in the waterwalls of tower designs up to the waterwall outlet temperature of about 880°F (470°C). T12 is used for lower temperatures in panel designs. For higher steam conditions and higher water wall outlet temperatures respectively, different materials are required. For example, T22 has slightly higher creep rupture strength and higher oxidation limit and is frequently used in place of T12. Its application, however, doesn’t enable any significant increase in cycle parameters. In recent years, new ferritic alloys became commercially available that could be applied to water wall construction. They offer a substantial improvement in the creep strength and can be used instead of conventional alloys or enable higher steam parameters cycle. These materials
Materials and State-of-the-art Steam Parameters Except for the waterwalls, where tubing made of low Cr alloy is used instead of carbon steel, materials applied in supercritical boiler designs are similar to materials selected for the drum type steam generators. The creep rupture strength and oxidation limit often establish the temperature limits of a material. For a given temperature, the creep rupture strength decreases over time. As the steam parameters increase, available design margin of many conventional alloys decreases and, at some temperature level, their application becomes impractical. The use of alloys for critical pressure part components such
6
are 2.25Cr-1.6W-V (T23) alloy (ASME code approved) developed by Sumitomo Metal Industries and a 7CrMoVTiB1010 (T24) alloy developed by Vallourec & Mannesmann Tubes.
strength and greater corrosion resistance than the best boiler materials in use today. The most attractive alternative appears to be nickel-base alloys for the very highest temperature components. The creep strength is sufficient to allow operation with steam temperatures of close to 1300°F (705°C).
Materials for Superheater and Reheater Tubes When selecting superheater and reheater tube materials, the creep strength of the selected alloys must be high enough to provide adequate margin of safety in the operating pressure and temperature range. In addition to the requirement for high strength, the material’s corrosion resistance, both on the flue gas side and on the steam side, must be considered. Oxides will always form on the inside and outside surface of a tube. On some older boiler designs, exfoliating metal-oxide scale from the internal surfaces of superheater and reheater tubes, headers, and piping have been a principal source of solid particle erosion in steam turbine and valves. The erosion most frequently occurred in the intermediate pressure stages and seldom in the high pressure stages of steam turbines and valves. The exfoliated oxides are mostly ferritic types. Low chrome ferritic alloys are predominant materials throughout the reheater system and are more prone to exfoliation than austenitic alloys. Alstom conservative design practice generally limits application of these ferritic alloys to relatively low temperatures. In the past ten years new higher Cr ferritic containing 9-12% Cr alloys became commercially available. Ferritic materials allow for more economical design and have the advantage of avoiding the problems of dissimilar metal welds and the large coefficient of expansion of austenitic steels that are typically used for higher temperature tubes. These high chrome ferritic alloys are continuously being improved for high strength and higher resistance to oxidation. The addition of Tungsten (W) and other carbide and nitride forming elements have led to considerable improvements in the high temperature strength of the 9-12Cr steels. However, even for these higher strength alloys Alstom design practice is to limit their application to tube outside surface temperature less than 1200ºF (650ºC). For higher temperatures, austenitic alloys are used. A new generation of ferritic materials is being developed which will give enhanced temperature capabilities relative to those now available. Very high operating steam temperatures, considered for higher efficiency cycles, will require the application of materials with greater creep
Materials for Turbine Casing and Blades The HP turbine for live steam temperatures up to 1100°F (594°C) is state of the art. For higher temperatures the design remains the same, only the materials for valve casings and valve internals, the diffuser for steam feeding from the inlet valve to the inlet spiral and the inner casing itself in the area of increase temperatures are replaced. Traditionally the Alstom rotor is a welded construction. The thermally heavily loaded rotor disk will be adapted by inserting of a disk of 10 % chromium steel. For the first turbine blade level an austenitic, for the subsequent level a martensitic chromium steel is used. Same applies for the IP turbine, only the first blade rows are made from nickel base alloy. The LP turbine final blades suffer from higher centrifugal forces. Titanium material enables to realize longer blades with slender and aerodynamically favorable profiles. So they allow with the same vacuum bigger steam turbines with compact dimension. For titanium final blades there are numerous references in Japan. Ultra-Supercritical Cycle The fundamental need for improved cycle efficiency capable of variable pressure operation will require increase in steam temperatures and pressures. Available materials make cycles with steam conditions of 4350 psig/1110ºF/1150ºF (300 bar/600ºC/620ºC) feasible in today’s market. A number of material development programs in the US and Europe will enable primary steam temperatures higher than 1112ºF (600ºC) in not so distant future. For example, Ni-based alloys are being developed and tested for higher steam conditions. The European research project Thermie is aiming at the development of a cycle with steam conditions at 5440 psig/1290ºF/1330ºF (375 bar/700ºC/720ºC). The boiler water walls will need to be constructed of tubes made of higherstrength, corrosion resistant martensitic steels. The high-pressure outlet headers, piping, and the final stage of the superheater tubes will need to
7
be fabricated of Ni-based alloys. Predicted plant efficiency is approximately 50% based on lower heating value.
this process, the microstructure of the metal does not change and the tubing retains its ductility. 2. Hydrogen damage — This type of corrosion damage usually occurs beneath a relatively dense deposit. Although some wastage occurs, the tube normally fails by thick-edge fracture before the wall thickness is reduced to the point where stress rupture would occur. Hydrogen, produced in the corrosion reaction, diffuses through the underlying metal, producing decarburization and intergranular microfissuring of the structure. Brittle fracture occurs along the partially separated boundaries, and in many cases, an entire section is blown out from the affected tube. Ductile attack is more probable when the boiler water contains highly soluble alkaline chemicals such as sodium hydroxide. Hydrogen damage, on the other hand, is more apt to occur when a low pH boiler-water environment is produced as a result of condenser leakage (if bypasses or leaks through the polisher) or some other type of system contamination. It is recognized that the successful operation of a utility once-through boiler unit with respect to water technology is coupled to the proper operation of the condensate demineralizer system.
Cycle Corrosion The utility boiler-turbine cycle is subject to two basic types of corrosion damage. One type of corrosion is the general attack on ferritic materials during periods of adverse feedwater conditions such as high oxygen concentration in combination with dissolved contaminants, low pH, etc. This corrosion may take place throughout the system. The second type of corrosion is caused by acid or alkaline contamination of the boiler water in the presence of deposited products, and occurs during system operation. The effects of oxygen corrosion are two-fold. First, the resultant pitting can completely penetrate tube walls. Failures of this type have predominantly occurred in economizers during operation, and the superheater and reheater sections of the boiler under idle conditions. However, all carbon steel components of the cycle are potentially susceptible to this type of failure. Second, the metallic corrosion products or oxides resulting from oxygen corrosion will be transported throughout the cycle by the working fluid which can then foul heat transfer surfaces and flow sensitive components such as orifices, turbine passages, etc. Deviations from recommended chemistry limits that result in either depressed or elevated pH values, promote failures of boiler tubing. These types of attack are accelerated by the presence of internal metal oxide deposits that permit soluble contaminants to concentrate during the process of steam generation or nucleate boiling. Supercritical units that do not operate on a sliding pressure mode are generally not susceptible to this corrosion mechanism since high contaminant concentrations are not produced in a single- phase fluid. Also the presence of condensate polishers in the feedwater system prevents large amounts of soluble contaminants from entering the boiler. Although there are many variations from a descriptive basis, the majority of these types of failures can be classified into one of the following two categories: 1. Caustic corrosion — This type of damage is normally characterized by irregular wastage of the tube metal beneath a porous deposit. It progresses to failure when the tube wall thins to a point where stress rupture occurs locally. In
Water Treatment Practices The main objectives of water chemistry control are to insure the long-term integrity of the materials of construction and the successful operation of the boiler-turbine power cycle. The particular types of chemical treatment may vary depending on many factors such as the variety of materials, operating conditions, system design, etc. Chemistry control in once-through steam generating units is based on the following features: a. No phase separating devices (i.e. steam drum) in the system b. Feedwater, boiler water, and steam are the same fluid stream c. Sliding pressure operation includes fluid conditions typical of both supercritical and subcritical operation. The above design considerations require that the concentrations of feedwater contaminants be kept to a minimum and be within allowable turbine steam purity limits (as specified by the steam turbine supplier) as the solubility of
8
contaminants increases with higher steam parameters. Corrosion products transported to the boiler (or superheater/reheater/turbine) from the condensate and feedwater system must be kept at low enough concentrations to minimize fouling of tube and turbine surfaces, and thus the potential for damage and/or efficiency losses. For once-through systems, feedwater conditioning to minimize general corrosion and the production of iron oxide can be accomplished with either all-volatile treatment (AVT) or oxygenated treatment (OT). Due to the greater concern for copper transport at supercritical pressures and its impact on turbine performance, feedwater systems in these once-through units consist primarily of ferritic alloys and do not contain copper alloys downstream of the condensate polishers. Controlling feedwater contaminants to a minimum is critical in once-through units, as there is no mechanism for their removal once in the feedwater (downstream of the condensate polishing system) nor can their aggressive behavior be arrested by the typical feedwater chemical treatments (OT or AVT). Contaminant ingress (from condenser inleakage, makeup water, etc.) is generally controlled by a condensate polishing system. The most frequent contributor to boiler waterside corrosion, fouling, and failures has been the accumulation of metal oxide deposits. These deposits form principally on heat transfer surfaces but can also foul control orifices that can then cause overheating of water wall tubes A reduction in the amount of debris and metal oxide deposition within the boiler can be successfully accomplished throughout its life cycle by: a. Good storage and on-site erection conditions. b. Minimum metal oxide concentrations in the boiler feedwater during startup operations as well as at load conditions. c. Adherence to operational water chemistry guidelines. d. Adherence to optimum lay-up procedures that prevent or reduce standby corrosion. e. Periodic chemical cleanings. Materials of construction most commonly used in the condensate and feedwater systems of once-through utility cycles are ferritic alloys, stainless steels, and titanium. Copper alloys are not used in feedwater heaters since copper oxides can dissolve in the supercritical fluid resulting in turbine fouling. Carbon and stainless steel tubes are therefore utilized in feedwater heaters and
stainless steels and titanium are used for condenser tubes. Water Treatment Regimes The Alstom once-through cycle chemistry guidelines are presented in the appendix (all listed pH refer to a sample temperature of 25°C). More detailed discussion of the specific chemistry regimes follows. AVT (All Volatile Treatment) AVT is defined as the exclusive use of volatile conditioning agents. Volatile chemicals evaporate from the water into the steam in a gaseous form. When steam condenses, the chemicals dissolve into the water. They do not form a solid phase and thus they do not form a scale or deposit on heat transfer surfaces. Common volatile conditioning agents are ammonia, amines, and hydrazine (or hydrazine substitutes). With AVT, feedwater pH ranges from 8.8 to 9.8. Low-level AVT has a pH between 8.8-9.3 (especially in plants with copper alloys), and high level AVT has a pH between 9.2-9.8. Even though high pH AVT provides better corrosion protection of steel, it has also its disadvantages: questions of waste water treatment, chemicals consumption, exclusion of ion exchange resin to run in H+ form. AVT (R) is defined as AVT that employs a reducing agent such as hydrazine or other oxygen scavengers. This results in a low (highly negative) ECP (electrochemical potential). Thus the highest possible oxidation state in the oxide layer is magnetite. It has the disadvantage to form thicker and more porous magnetite layers resulting in a higher iron content in the water. Subsequently ripples can be formed. It is beneficial in plants with copper heat exchangers in the feedwater train because it reduces copper corrosion and therefore the release of copper into the feedwater. In oncethrough systems, the cycle does not however include copper or its alloys. AVT (R) also favors Flow Assisted Corrosion (FAC) in the high pressure feedwater system if it contains carbon steel alloys therefore, it is not used as a water treatment regime. The use of hydrazine for wet lay-up is not relevant for operating conditions and therefore possible. Environmental aspects on the use of hydrazine, however, have to be considered.
9
AVT(O) excludes the use of an oxidizing agent and therefore, the ECP will be substantially higher than with AVT(R). AVT(O) favors the formation of hematite layers on top of the magnetite, which are less soluble and hence more stable than the magnetite layers of AVT(R). As a result, the oxide layer is thinner and denser and gives more protective margin against FAC and minimizes orifice fouling.
steam chemistry conditions. Cycle chemistry guidelines would be updated accordingly. A research need that readily comes to mind is the behavior or volatility of the various oxides at these higher temperatures and pressures. It has also become evident through limited use of these alloys that development of new chemical cleaning procedures or solvent mixtures will be required to effectively remove the oxide scales or deposits should the need arise.
OT (Oxygenated Treatment) OT will provide a high ECP that provides the formation of hematite layers, which are less soluble and hence more stable than the magnetite layers of AVT(R) and the mixed magnetite / hematite layers of AVT(O). As a result, the level of iron oxide in the feedwater is much lower. It gives excellent protective margin against FAC and minimizes orifice fouling. As an additional bonus, it has already been successfully used with very low-level AVT, i.e. feedwater pH 8.0-8.5. The related small ammonia concentrations permit a very long life of H+ mixed-beds, and ease wastewater questions. Of course, OT can also be applied with higher pH AVT. OT is not without problems. Oxygen, when coupled with anions, will be corrosive. OT therefore requires a strict control of feedwater impurities (> condensate polishers!) Conclusions Large capacity, supercritical units will continue to play a major role in fulfilling power generation needs in the global marketplace. The fundamental need for improved cycle efficiency capable of variable pressure operation will require increase in steam temperatures and pressures, aiming at the development of a cycle with steam conditions at 5440 psig/1290ºF/1330ºF (375 bar/700ºC/720ºC). The boiler waterwalls will need to be constructed of tubes made of higher-strength, corrosion resistant martensitic steels. The high-pressure outlet headers, piping, and the final stage of the superheater tubes will need to be fabricated of Ni-based alloys. Turbine components will include similar alloys as well as titanium in the LP section. The new applications of these alloys will in turn require an evaluation of water and
10
References
[7]
Deposit and Water Chemistry Studies with Rifled Tubing, F. Gabrielli, N. Mohn, B. Teigen, TIS # 7530
[1]
VGB-D (1988-97) “ Availability of thermal power plants”
[8]
[2]
KEMA-NL (1997) “ Comparison of subcritical and supercritical units.
[3]
The Supercritical Steam Power Plant: Operational Success and Technological Advancement, Edward S. Sadlon, Guenter Scheffknect.
Komet 650 – Coal fired Power Stations with Steam Temperatures up to 650° C – Findings from a Successful Ten-year Field to Examine Materials for Boiler Tubes, Pipes, Turbine and Valves. Helmut Meyer, Dieter Erdmann, Peter Moser, Sabine Polenz, VGB PowerTech 3/2008
[9]
Highly efficient Steam Turbine Technology for Coal-fired Power Plants, VGB Symposium Steam Turbines and Steam Turbine Operation. A. Tremmel, D. Hartmann, ‚June 2004, Regensburg, Germany
[10]
EPRI 8th International Cycle Chemistry Conference. Calgary, Canada. 20-22 June 2006
[11]
EPPSA, FDBR and VGB PowerTech Guidelines for Feedwater, Boiler Water and Steam for Power Plants/Industrial Plants, VGB-R450Le, Second Edition 2004
[4]
[5]
[6]
Analysis and Summary of Rifled Tube Heat Transfer Date, Mark Palkes, J. H. Chiu State-of-the-Art Large Capacity Sliding Pressure Supercritical Steam Generators, Mark Palkes, Edward S. Sadlon, A Salem NERC-US (1989) “ Boiler tube failure trends” State-of-The-Art Sliding Pressure Supercritical Steam Generators, Mark Palkes, John Banas, Gerhard Weissinger, and Werner Kessel
11
Appendix
STANDARD SPECIFICATIONS FOR ONCE-THROUGH UTILITY BOILER SYSTEMS (Water Chemistry Requirements for Normal Operation) 1. Demineralized Water (at Demineralizer Water Plant Outlet) Parameter
Unit
Specific conductivity at storage tank outlet *** Silica as SiO2 Sodium + Potassium as Na+K Iron as Fe TOC Oxygen
μS/cm μS/cm ppb ppb ppb ppb -
A1*
S** ***
N
A1*
A2
S**
An
< 0.10 <1 < 10 <5 < 20 < 300 satur.
0.1-0.2 ≥1 10 - 20 -
-
≥ 0.2 ≥ 20 -
c c c m m m
- indicates that there is a problem - actions shall be taken to bring values to normal - operation may be continued as long as feedwater/steam specifications are not jeopardized do not put this water into the demineralized water storage tank conductivity as measured at demineralized water storage tank outlet, including carbon dioxide
2. Condensate (at Hotwell Outlet) Parameter Specific conductivity Conductivity after cation exch. pH-value Iron as Fe
Unit
N
A1
A2
μS/cm μS/cm --ppb
2.5 - 11 < 0.20 9.0 - 9.6* < 20
--0.2 - 0.5 -----
--> 0.5 -----
* for oxygenated treatment, values in accordance with those in Table 4
12
S
An
---
m c m m
>1
-----
3. Feedwater (AVT Treatment - measurements at economizer inlet) Parameter
Unit
N
A1
A2
S
An
Specific conductivity Conductivity after cation exch. pH-value
μS/cm μS/cm ---
4.0-11 < 0.20 9.2 - 9.6
--> 0.5 > 10
-->1 ≤7*
c c c
Sodium + Potassium as Na+K
ppb
<5
--0.2 - 0.5 < 9.2 > 9.6 5 - 10
10 - 20
> 20
Silica as SiO2 Iron as Fe Hydrazine Oxygen
ppb ppb ppb ppb
< 20 < 20 *** < 10
≥ 20 ** ---
-----
-----
≥ 50 ****
---
---
m or c m m m c
*
if dosing is limited during system refilling, ammonia will be below normal. In such a case, pH ≥ 5.5 is tolerated ** time permitted above A1 see “Main steam” *** hydrazine or equivalent should not be used if O2 < 10 ppb **** no cumulative time limit for exceeding A1
4. Feedwater (Oxygenated Treatment instead of AVT) Parameter Specific conductivity Conductivity after cation exch. pH-value Sodium + Potassium as Na+K Oxygen **
Unit
N
A1
A2
S
An
μS/cm μS/cm ppb
2.5 - 7.0 < 0.20 8.5 - 9.0 <5
--0.2 - 0.5 * --5 - 10
--> 0.5 * --10-20
-->1 --> 20
ppb
50 -150
---
---
---
c c c m or c c
* discontinue oxygen injection ** target for oxygen injection
13
5. Main Steam and Reheat Steam (at OTSG Outlet) Parameter Specific conductivity Conductivity after cation exch. pH-value Sodium + Potassium as Na+K Silica as SiO2 Iron as Fe
Unit
N
A1
A2
S
An
μS/cm μS/cm ppb
2.5 - 7.0 < 0.20 9.0 - 9.6* <5
--0.2 - 0.5 --5 -10
--> 0.5 --10 - 20
-->1 --> 20
ppb ppb
< 20 < 20
≥ 20 ** ---
-----
-----
m c m m or c c m
* see Section 4 if on oxygen treatment ** time permitted above 20 ppb with an Alstom steam turbine: [hours]*[ppb] <105. If this time has expired, turbine pressure and efficiency must be measured. If turbine capacity/efficiency has degraded, the turbine should be cleaned.
6. General Remarks All conductivities are referred to 25°C. Possible contributions from carbon dioxide may be excluded. Operation is desirable at the lowest achievable impurity levels, with the shortest and least frequent excursions. The specification is related to the following conditions: • There are no copper alloys in the system • Conditioning is done with ammonia and oxygen injection. Note: This is a general specification, valid for the plant type mentioned only. The criteria will be reviewed for a specific application and at commissioning. Except where defined otherwise in the footnote of a table, the following definitions apply:
N
Normal Value. Values are consistent with long-term system reliability. A safety margin has been provided to avoid concentration of contaminants at surfaces. A1 Action Level 1. There is a potential for the accumulation of contaminants and corrosion. Return to normal values within 1 week. Maximum exposure is 336 cumulative hours per year, excluding start-up conditions. A2 Action Level 2. The accumulation of impurities and corrosion will occur. Return to normal levels within 24 hours. Maximum exposure is 48 cumulative hours per year, excluding startup conditions. S Immediate Shutdown. Immediate shutdown of the concerned system is required to avoid damage.
14
Once-Through Boiler Cycle - Guidelines for Initial Start-up and Restarts After Long Outages
1. Flush to waste until suspended solids in the condensate, feedwater and boiler water are less than 3ppm (a). The feedwater (b) can be pumped through the boiler into the start-up separator, flash tank (if present), and then to waste. 2. When this limit is achieved, circulate water through boiler, start-up separator, flash tank and back to condenser. Place condensate demineralizers in service. 3. When total iron concentration drops below 1 ppm, silica is less than 100 ppb, and cation conductivity is less than 1uS/cm (c), unit firing can commence. Boiler and start-up system operation should be as per instruction manual. 4. Proceed to normal feedwater control limits (c). These should be obtained before exceeding one-third unit load. (a). (b). (c).
Permissible value depends on demineralizer performance/constraints. Feedwater pH > 9. Parameters measured at economizer inlet.
15
