|
|
|
Links
|
Basics of
Steam System Design Steam Trap Application In order to effectively design a steam trap installation the designer needs to profile each trap requirement. Keeping in mind that the vast majority of trap requirements on a new project will be typical this reduces the amount of time involved by a significant amount. Most trap requirements on a project will pertain to drip traps and steam tracer traps (if the project is indeed using steam as its freeze protection medium). Consequently when you profile either of those requirements you are doing so for a group of traps. The same applies for typical process and utility equipment that requires steam trapping. When developing a *Steam Trap Profile Sheet provide space to list all steam traps, by identification number, that pertain to that particular profile. In other words, one sheet should be dedicated to one typical profile. All traps intended to match that profile should be identified on that sheet. Following is information needed to determine a steam trap specification:
By resolving these questions prior to specifying a trap or designing an installation the designer is much better prepared to do the job more effectively, more efficiently and correctly. For the engineer it means saving money by doing the job more efficiently at the design stage. For the owner it means saving money because the proper installation reduces steam waste, reduces the chance for having to make a correction during start-up and minimizes repairs and replacements after the system is on-line and the plant is in operation. Anyone that performs maintenance or steam trap surveys has had the opportunity to witness carelessness, disinterest and lack of knowledge at work. When design complacency or lack of understanding does occur it's the owner that usually gets stuck with the cost of having to correct the problem installation. As with most project design and engineering issues the owner needs to safeguard their interest by doing their homework. A large number of these design mistakes make it through start-up because they don't have an initial adverse effect on the operation that is obvious. After the design firm is long gone and the contractor has everything signed off and hits the road, these mistakes linger until, at some point later, they're discovered by design or by accident. These are conditions that usually don't get caught for months, or even years. However, until they are discovered these problems can cost the owner money. In such cases the owner isn't aware the problem exists until, in most instances, it's inadvertently discovered during a maintenance check or a trap survey. Once the problem is discovered and corrected the water hammer problem, that may have existed, disappears the system operates more efficiently and steam demand drops. These latent problems, when left undetected, can cost the owner thousands, tens of thousands and even hundreds of thousands of dollars. And by the time these problems are detected and corrected the owner usually has no recourse but to pay for the mistake out of their own pocket and go on. There is a way for the owner to reduce and possibly even eliminate these and many other mistakes from slipping through the checkout and start-up process. This will be covered in a later article. Right now we want to focus on determining the correct steam trap and installation design for various applications. For the infinite types and variations of steam heated equipment and their installation there are that many and more variations of steam trap applications. What this section will do is work with the primary steam trap applications and their basic installation criteria. Those applications are:
Before getting into specifics with the installation design of the above applications there are three issues that need to be mentioned. As a general rule it is poor practice to trap two or more equipment items with a single trap. In doing this, the potential exists for short circuiting one of the equipment items. A short circuit occurs when one of the equipment items has more of a steam or condensate load than another equipment item connected to the same trap. The equipment with the higher load, and this can be a very slight difference, will create a primary path to the single trap essentially blocking condensate flow from the equipment with the lighter condensate load. This effect can create a backup of condensate in the blocked equipment causing low heat transfer efficiency as well as damage to the equipment that is being short-circuited. Secondly, and these are not mentioned in order of importance, is consideration for the individuals that will be required to work on these installations after they are in operation. Their personal safety should be uppermost in the engineer's or designer's mind when designing any installation. The ease of access should be a definite consideration. This needs to be weighed against how frequent access is going to be required based on what service the trap will see and whether or not the trap is servicing a major piece of equipment. In particular, traps that will be servicing major pieces of equipment operating on a continuous basis should get the highest degree of consideration. Anyone who has had to peel back the insulation on a high pressure steam line in preparation for pulling a malfunctioning steam trap, while the system is still operating, understands the need for providing the necessary valving, pipe configuration and access to make this job as safe as possible. This is not only the responsibility of the engineering firm, but also the plant owners own project team. As a basic concept the owner needs to have maintenance personnel involved in reviewing installation drawings while the drawings are in design, not after the fact. This makes sense on two levels: 1. From a safety standpoint, plant maintenance won't inadvertently be placed in a situation where they have to deal with a potentially hazardous circumstance when maintenance is called for, 2. The cost involved in correcting a poor installation design after the fact is going to be relatively more expensive than the initial installation cost. It is more cost effective to put it through the proper reviews and get it right the first time before final approval, fabrication and installation. There will be further articles regarding project teams, planning, coordinating and execution. These issues go beyond the scope of the design of steam and condensate systems and the application of steam traps. And the third point of discussion is a maintenance program for steam trap stations. As an example, if a moderately sized plant produces an average summer/winter steam rate of 50,000 lbs/hr at a cost of $7.00/1000 lbs. the math tells us it is costing the plant $350.00/hr or $8400.00/day to keep this plant operating. At $3,066,000.00/yr the manufacture of steam in this plant is a major overhead. With that much associated cost operating the steam system and its users at a high degree of efficiency should be a top priority. At least you would think so. Too often plants are not aware of their steam trap population, how many traps they have, where they are located or the last time they were tested. Simple questions, but all too often unanswered questions. At least until something goes wrong. After discussing the various design applications we will expand on the profile questions and perform a sample trap specification routine. Boiler Main The following criteria is based on the assumption that the piping involved has been designed based on good and basic pipe design practices. As an example, the boiler main piping is normally contained within the boiler house itself with the plant distribution piping branching off of the boiler main inside the boiler house and then running outside to distribute steam to the plant. The purpose of the example is to make clear that in determining the number of traps and their placement on the boiler main, pipe run length should not be a factor. The boiler main (ref. Fig. 7) is the piping that connects directly to the boiler or boilers and feeds steam to the distribution piping. Steam distribution piping for the plant, or facility, feeds off of this boiler main piping to distribute steam throughout the plant. A portion of the boiler main piping should be at a low point of the piping system between the boiler connections and the plant distribution headers. Any carryover from the boilers will initially wind up at that low point. This low point is the section of piping where the traps should be located. If the steam in the boiler main, whether it is supplied from a single boiler or multiple boilers, flows in one continuous direction to its termination point only one drip trap station will be required and it should be located at the termination point of the header. If the steam in the boiler main is routed in multiple directions in order to arrange the distribution headers in a more efficient manner then a drip trap station will be needed at the termination of each of those boiler main runs. Referring to the Profile List, the boiler main should be considered a major utility in continuous service. This would suggest redundant traps and possible in-service trap repair/replacement. The installation design should reflect these requirements in a way that provides consideration for the safety of the individuals who’s job it will be to perform the repair while the system remains in service. Another consideration is possible water hammer (hydraulic shock) and variations in condensate load due to boiler carryover. The selected steam trap should be able to handle both conditions. The carryover, which could be a combination of condensate and entrained solids, has to be removed immediately before it gets out into the system. Because the specified traps may be required to handle these potential upsets the sizing will have to reflect that. At this point the developing trap station profile looks like this:
Fig. 7 - Boiler Main Trap Station Location
Gig. 8 - Boiler Main Trap Station Detail Referring to Figure 7 you will notice that the distribution segment of the boiler main is located in a pocketed section of the piping. The two boiler feed lines drop down into this section while the plant distribution mains feed up. Notice also, in Figure 8, that two trap assemblies are indicated at each location. This includes the necessary valving required to isolate each trap and strainer to allow for in-service repair or replacement. Since the frequency of preventative maintenance, or testing, would be on the order of once a year and the potential of repair/replacement would be on the order of three to five years plus, unassisted access is not necessarily required; only clear access. Unassisted access is a condition in which personnel can gain full access to a trap station simply by walking up to it at floor level. Unobstructed access is a condition in which personnel would require scaffolding or a portable platform to gain full and unobstructed access to the trap station. Obstructed access is a condition in which the trap station is in an elevated position with an obstruction such as piping, a cable tray, conduit, etc. between the station, the access floor or from the approach direction. It is not against good practice to locate the trap station itself a short distance from the drip leg if it facilitates access and simplifies support without creating additional problems. This applies to most applications. The concern involved with locating the trap station away from the drip leg is the added loss of radiant heat due to the added length of pipe. In calculating the condensate load requirement for the boiler main traps two factors need to be considered: radiant heat loss and transient carryover volume. Radiant heat loss can be calculated as follows:
Where: It is recommended that the transient upset condensate load be added to the radiant heat loss condensate load when sizing the trap. When factored in, the upset load is calculated based upon a percentage of the steam production. However, this percentage has to be tempered with the integrity and quality of the boiler system. If steam is produced at the rate of 20,000 lbs./hr and an upset factor of 10% is used, the additional load for the trap would be: 20,000 x .10 = 2000 lbs/hr. Based on saturated steam at 600 PSIG running through a 16"f pipe, 80 feet long, inside a heated facility the calculated condensate load for one of the trap stations would be: By using a safety factor of 2:1 (safety factors will be discussed later) the calculated condensate load would then be 340.8 or 341 lbs/hr. That gives us a total combined condensate load of 2000 lbs/hr (upset load) + 341 lbs/hr. (calculated load) = 2341 lbs/hr. If we elect to go with an inverted bucket trap one of the selections would be an Armstrong 3/4" model #983 with an 1/8" orifice and a capacity of 2425 lbs/hr. Another choice would be a 3/4" Spirax/Sarco disc trap model #TD42 with a capacity of 2200 lbs/hr. In sizing the trap for this application we are able to accommodate the added transient upset load by only marginally increasing the trap size that would have been selected had the upset load not been considered. However, this could change if steam production was 50,000 or 100,000 lbs/hr. Although the condensate load from radiant heat loss would remain the same, the increased production value would proportionately increase the volume of the upset load to 5,000 and 10,000 lbs/hr respectively. The determination as to whether or not to maintain that 10% factor or to make a compromise by reducing that factor has to be an educated one, a decision that should involve a discussion with the boiler manufacturer. Steam Distribution Mains Steam distribution mains are the segment of a piping system used to transport steam from the boilers to the various areas or departments throughout the plant. It is the designer's job to design a system that will accomplish this as efficiently as possible. The configuration for a distribution main will usually consist of risers, pockets and long distances with expansion loops. These are all conditions that are conducive to the formation and retention of condensate. In such conditions it is the designers responsibility to incorporate the use and placement of steam traps in strategic locations. In doing so it will prevent the accumulation of unwanted condensate by capturing and returning it.
Fig. 9 - Steam Distribution Main - Trap Locations Referring to Fig. 9, there should be a trap station located at each riser and on the upstream side of each expansion loop. There should also be a trap station approximately every three to four hundred feet on long horizontal runs containing no risers or expansion loops. Steam traps will be sized based on the amount of radiant heat loss occurring in a specific segment of the distribution main. Start-up conditions do not necessarily effect these traps. In starting up a steam distribution system there are two things you will need to consider in regard to steam traps. One is the start-up condensate load and the second is the displacement of air residing in the pipeline. The start-up load is short-lived and not that much in excess of the normal load. However, the necessity to displace air in the system along with the warm up of pipe and components as well as the additional condensate load at start-up demands specific consideration. Start-up of a system needs to be done in a very planned and controlled manner. This start-up consideration should be a separate issue from the trap station design itself. Which is based on the system operating parameters and not on the momentary conditions seen at start-up. This is another good example of a design situation that should include, whenever possible, discussions and input from the operations and/or maintenance personnel; the group that will ultimately be operating the facility. They may have an idea or two about how they would like to shut down and start up the system. What this type of input hopefully provides, when working with an existing plant site, is continuity and applied lessons learned. Continuity as to the methods and procedures used to perform typical plant activities from department to department. And the practicality of taking the lessons learned from previous plant installations and integrating them into current design. Although there are situations where automatic shutdown and start-up design is feasible the vast majority of main distribution systems are designed for manual, or supervised, shutdown and start-up. A typical automated start-up and shutdown system might be a steam tracing distribution system. This type of system would start-up and shut down based on the operation of an ambient sensing valve. With all sub-header and supply line valves closed, all drip leg drain valves closed and all drip traps valved off, steam is slowly bled into the system through the main header valve. (A future segment of the design series will cover check-outs and start-ups in greater detail. This segment on steam will cover only a few basics for clarification.) This can be done by cracking open the main valve or by providing a small, 3/4" or 1" by-pass around the main valve. This can either be an external or integral by-pass to the main valve. As steam fills the piping system the initial volume of steam will condensate at a higher rate due to the relatively cold pipe. However, as the supply of steam continues to move through the piping it will cause a large portion of the forming condensate to flash back to steam. This will be apparent by, what should be, a small amount of internal crackling and banging that takes place during this process. With the system now flooded with steam and the main valve fully open the closed system contains partial pressures of steam and air. The air that resided in the system before the introduction of steam is still contained in the closed system under a partial pressure. The theory of partial pressures is based on Dalton's Law of Partial Pressures. John Dalton observed that the Total Pressure of a gas mixture was the sum of the Partial Pressure of each gas, or P total = P1 + P2 + P3 + .......Pn. The Partial Pressure is further defined as the pressure of a single gas in the mixture as if that gas alone occupied the container. It is therefore determined that, in our case, steam, as a partial pressure, will have the same characteristics as its partial pressure. Let us assume that the entrapped air, contained within the closed system, now makes up 10% of the volume within that closed system and the newly introduced steam makes up the 90% balance of the volume. With an absolute pressure of 300 PSIA, the pressure of the air would be 30 PSIA and the pressure of the steam would be 270 PSIA. With an absolute pressure of 270 PSIA the steam's characteristics would be a temperature of 407.8 ºF with a latent heat content of 818.3 BTU's and a specific volume of 1.71 ft3/lb. Instead of the 417.4 ºF, 808.9 BTU's and 1.547 ft3/lb it would be at its original 300PSIA. Not only does the air reduce the overall heat content of the steam by volume but once the air is in the system it will have a tendency to plate out, or collect, on heat transfer surfaces further effecting the overall heating efficiency of the steam. It is therefore essential that the air be removed from the closed system before the sub-header valves are opened, compromising the efficiency of the entire system. Incidentally, this same process will be repeated with each sub-header. The overall checkout and start-up procedure should be outlined and accomplished in such a way that virtually all air is purged from the system before the final block valve, at each user, is opened. With the main system still closed off from the sub-headers, personnel, in a controlled procedure, will go to each drip leg drain valve and slowly open it. It is recommended to use a multi-turn valve like a gate valve rather than a quarter turn valve like a ball valve in steam service. The multi-turn valve provides much more control when opening and closing, unlike a quarter turn valve. The sequence of performing this blow-down should be in the direction of flow, starting with the furthest upstream drip leg. And due to the inherent dangers of steam, much caution should be exercised while performing this procedure. Each drain valve only needs to remain open for a brief period. With the valve fully open for approximately one minute this should be sufficient enough time to purge the steam containing air from each particular segment of the piping system. Once the purge is completed for a specific segment close off the drain valve and open the valves on both sides of the steam trap. Continue this process in the direction of flow. That basically touched on the high points of starting up a steam system. There is much more to the process that will be covered in a later section on check-outs and start-ups. These highlights were presented at this time to show that, when considering start-up loads and conditions, there are no special needs. And please do not take that as a general statement. There are certainly cases where a particular system may require that special consideration be given to the start-up and shutdown process of a system. Getting back to the placement of trap stations; In the process of locating the strategic positions of each trap the designer has to determine what segment of pipe each trap will be assigned to. In the case of the trap located at the base of the riser in Fig. 9, the designer should calculate the height of the riser plus the distance from the upstream side of the expansion loop (at Trap Station #2). This will be the segment of pipe assigned to Trap Station #3 at the base of the riser. As in the previous calculation for radiant heat loss:
Let us assume that the riser has a length of 80'-0" and the lineal distance from the upstream side of the expansion loop to the bottom of the riser is 160'-0". That gives us a total length of 240'-0" that trap station #3 has to be sized for. Based on 300 PSIA saturated steam running through a 10"f header in an outside pipe rack the calculation would be:
By using a 2:1 safety factor the condensate load would then be 677.94 or 678 lbs/hr. A practical trap for this application would be a thermodynamic or thermostatic type. One further note when designing trap stations for this application; address the issue of preventative trap maintenance. If you're in plant maintenance or operations and you have the opportunity to provide input during the design phase of a project, make this one of your discussion points. If you’re a designer, address this with the owner's team. There are two considerations we need to address with this particular application (actually it can be made to apply to virtually any application): the first is the need to facilitate testing of the trap for preventative maintenance, and the second is periodic strainer blow-down. In the first consideration, when designing trap stations for steam distribution mains a typical trap station for a drip leg would usually be installed in a pipe rack. This places the trap station in an elevated and very inaccessible location. When you consider the possibility of several hundred traps throughout a plant requiring lifts, ladders, scaffolding and tie-offs, just to gain access to, it can be a costly and time consuming exercise. What ultimately occurs is the trap stations are not maintained properly and in too many cases are even forgotten. The second consideration is an all too frequent problem. Strainers, whether a separate item or an integral part of the trap, are there to capture errant, entrained particles in the steam before they can reach and plug the trap orifice. Just the nature of the intended purpose of the strainer indicates the fact that particles will be trapped and will accumulate on the strainer mesh until they are removed. As they accumulate they progressively block off flow through the mesh until, if not periodically cleaned, they completely block the flow of condensate to the trap. Inaccessible trap stations are frequently ignored or forgotten. Dropping a trap station to grade or access level is one simple way to avoid this and should be a design consideration. In considering this option we need to develop a cost profile for the intended alternate configuration. This will determine what the initial installed cost for this alternative design might be and what the additional running or ongoing operating cost might be. This will provide management with the necessary information to help determine whether or not to make the change. This application we will use as an example will be an inside installation with a controlled environment. The initial installation differential will include an additional length of pipe, insulation, added fittings and labor. For this case we will assume the pipe support cost to be a trade off. In order to locate the trap station at an accessible level it will have to drop 15'-0" requiring 30'-0" of additional pipe, (2) 90B Elbows and 21'-0" of insulation. Using 3/4"f , sch. 80 c.s. pipe with 3000 lb socketweld fittings and 21'-0" of 1 1/2" thick fiberglass insulation. The condensate return leg from the trap station will only be insulated up to 7'-0" above the access level for personnel protection. If this installation was to be outside, the additional cost for heat tracing the condensate return riser along with insulating the balance of that return line would have to be added into the initial cost. Pipe, Fittings & Labor
$458.00 Additional cost consideration comes from any added operational costs due to relocating the trap station. In this case it would consist only of the added radiant heat loss in the additional 15'-0" drop to the trap station. Because the 15'-0" return section of piping is carrying condensate it is not a part of the heat loss calculation. The radiant heat loss can be calculated using the preceding radiant heat loss calculation as follows:
If we base this on a system that operates 24 hrs/day, 7 days/week the total annual, additional steam loss in that 15'-0" drop would be: 8760 hrs. x 2.07 lbs/hr = 18133.2 lbs/yr. Assuming an approximate cost of $7.00/1000 lbs to manufacture the steam, the total annual cost incurred by making the trap station more accessible would be: 18133.2 lbs/yr x $7.00/1000 lbs = $126.93. You can see by these figures that the ongoing operating cost differential incurred in providing improved access amounts to less than $11.00/mo. The advantage in doing this, however, will more than offset this added cost. By making these trap stations more accessible, and again this does not only apply to steam main applications, it makes the preventative maintenance effort much more efficient. Strainers can be blown down and traps can be tested in a much safer, less time consuming and a more controlled manner. This is an issue that needs to be tabled and discussed early in the design stages of a project. If agreement can be reached at an early stage it will save a great deal of time in trying to redesign a system to accommodate such changes. Steam Tracing Very frequently there is a gap between an engineers perception of priorities and the priority perception of the plant maintenance and operations personnel. In order to understand the importance of a well thought-out and well designed steam tracing system from the plant personnel's viewpoint a design engineer needs to get a phone call at 3:00 AM, with outside temperatures hovering around -20B F and winds gusting at 30 MPH. The maintenance shift supervisor says a steam tracer in a pipe rack froze and ruptured, creating a potential freeze-up of the pipeline it was protecting. The next thing you know, you're not in your warm bed any longer but are instead at the plant. You're 30'-0" up in a scissors lift in sub zero temperatures trying everything your experience allows to keep this line from freezing. Which, if you don't succeed, could ultimately shut down a production facility. Personal inconvenience aside, the potential capital loss incurred from a shut down of operations because a weak link in the steam tracing system failed to protect some piping is very real. That reality exists because all too often steam tracing, when called for on a project, is treated as an afterthought instead of as an integral part of the design process. At this time we will address only the trap station design for steam tracing. A later section in the design series will discuss the design of steam tracing systems in detail. To facilitate shutdowns, start-ups and isolated repairs, individual tracer supplies should be grouped together in a manifold arrangement whenever possible. With this intent the designer should plan the routing of the tracing in an effort to allow the tubing to congregate and terminate at common points for trap manifolding whenever possible. This is something that is relatively simple to do in a pipe rack situation but requires additional forethought when the tracing is required away from common groupings of heat traced pipe like a pipe rack. An identification system should be established in order to identify both the supply source and the termination of each common tracer. This will allow each tracer to be isolated safely for shutdown, start-up or repair when necessary. If this is not done it will not only cause unnecessary delays but could also place personnel at risk. When traps are arranged in a manifold configuration each tracer line should consist of (in the direction flow) a block valve, a strainer upstream or integral to the trap, the steam trap, a test valve, a check valve and a block valve. The check valve is required when the manifold, or condensate return header, is sized for expected flow only. There are other manifold arrangements where condensate is released into a collection chamber. The chamber is usually nothing more than a section of oversized pipe that allows condensate to be collected then removed through a dip tube by siphon effect. The tracer lines connecting to a collection chamber don't normally experience water hammer as a result of surges from other connecting lines. However, where a return header is sized for the flow load only, from several connecting lines using cycling traps, a discharge surge from one trap can create a reverse water hammer impact on the other connecting lines. The check valve installed downstream of the trap prevents and protects each trap from the hydraulic impact caused by the discharge of the other traps. Steam Separators A steam separator is an in-line piece of equipment that is used to remove entrained condensate particles. These condensate particles can originate from boiler carryover, or be the unwanted result of undersized pipe and equipment. A well-designed steam system minimizes the accumulation of condensate within the steam system and immediately carries off the condensate that does accumulate. With a properly sized and configured piping system, properly located drip legs and properly sized equipment a steam separator would not be required. However, when a plant experiences reduced heat exchanger efficiency, erosion at pipe directional changes, erosion to in-line equipment and water hammer, the installation of a separator is a consideration. All are possible indicators that the presence of entrained condensate particles and the accumulation of condensate exists in the flow of steam. Unless a system is designed poorly to begin with, these symptoms are usually the result of poorly planned, multiple expansions and modifications to a steam system. If a system is expanded to and operated at its capacity any weakness in the design and/or installation will begin to show. One of the first telltale signs that a steam system is at its capacity and/or has a weakness in its design is the above indications of condensate in the system. In order to alleviate the problem without going through the time and expense of resizing and replacing existing piping a separator could be installed. A strategically placed separator can't cure the problem but it can prevent erosion and other damage caused to pipe and in-line equipment due to the formation and build-up of condensate. The separator is designed to work in-line and should be expected to remove approximately 95 - 98% of the entrained condensate particles. With various designs of the same theme the separator functions as an enlarged section in a pipeline. Internally there are baffle plates designed to either impinge the particles, force the particles to the outer perimeters of the housing or both. As the steam, with the entrained condensate particles, enters the chamber of the separator it suddenly and momentarily looses some of its velocity due to the sudden enlargement of the separator chamber. The mass of the condensate particles propels them forward into the impingement baffles. Or, in the case of a cyclone type design, the particles will be forced by vanes or fins into a high velocity rotation inside the chamber. The particles will then be forced to accumulate either on the impingement baffle or the outer perimeter wall of the separator chamber and collect at a low point in the separator. Connected to the low point of the separator is piping that will lead to a steam trap where the condensate will then be carried off. The need for a separator, aside from installing one to safeguard the system from possible boiler carryover, indicates a system problem. The separator is installed to control certain aspects of the problem not correct it. Depending on the severity and magnitude of the problem it may be more cost effective to research the problem itself, determine the cause and correct it. Air Handling Unit Coils The heating coils in air handling units have a demand range governed mainly by outside temperature fluctuation. This demand range requires that the designer give added consideration in the selection process of a steam trap for each particular coil. In some cases there may be multiple sets of coils servicing one air handling unit. Each coil in the set will require its own steam trap. An excellent trap for this service, because of the wide range in pressure and volume, is the float & thermostatic trap. This trap allows condensate at low pressures and low flows to free flow through its orifice while also accommodating and regulating the higher flows at higher pressures. Two conditions can create a low pressure situation at the coil outlet. Depending upon what latitude an installation is located in the outside temperature could create a sufficient temperature differential across the heating coils to cause a coil outlet pressure of less than 1 pound. This can occur in either of two ways: 1. Either the demand across the coils is so great that the steam is condensing in the coils at a rate that exceeds the supply capability. Secondly the demand across the coils is so low that the steam control valve has throttled back to the point of providing less than 1 PSIG of steam. As an example, let us assume we are supplying a single coil with 15 PSIG steam. If this supply pressure were to be maintained through the coil and up to the trap it would have enough pressure to lift the outlet condensate 34.5 feet, or 15 PSIG x 2.3 (feet of lift per pound) = 34.5 ft. If enough of a demand, or the lack of demand, was placed on the steam in the coil it could create a 14 PSIG pressure drop across that coil or across the steam control valve. That would leave a pressure residual of only 1 pound at the trap. This translates into enough pressure to lift the condensate only 2.3 feet. You can see why additional, and early, consideration must be given to this application. The issue with how to return the condensate under such potentially low pressure needs to be determined early in the design phase of a project due to the impact that may be caused to other disciplines. In determining how the condensate is going to be returned there are basically two considerations: 1. Can the condensate return headers be run below (on a lower floor) the coil outlet trap for gravity drainage and 2. Does the condensate have to be lifted to overhead condensate return piping? If it is possible to run the return header below the trap outlet this would be the more practical method in regard to the flow of condensate. Regarding other concerns, dropping the trap discharge piping down normally necessitates floor penetrations. At the very least, grating or steel plate penetrations. Floor penetrations will require coordination with the architectural and/or structural group. In some cases it would be easier to plan on core drilling. If you're working with an unstable design due to equipment drawings that are not approved, lack of information on in-line equipment and any one of a number of other problems, don't attempt to resolve the location of the penetrations. Doing so only creates confusion, construction delays and added cost. Core drilling allows the penetrations to be made after their locations have been confirmed without regard to the flooring schedule. This allows the floor to go in on schedule without having to wait for approved penetration drawings. With the condensate dropping down from the trap discharge to the return header the designer doesn't have to be concerned with the lack of lift pressure. However, if there is no alternative but to return the condensate overhead then the designer is going to require a non-electric condensate pump. If it is at all possible combine the flow of two or more traps by routing a collection header and running it to a non-electric condensate pump. The non-electric condensate pump is a mechanical equipment item that allows condensate or other liquids to accumulate, by gravity, in the pump chamber under low pressure. The condensate then gets pumped to its destination by steam, air or inert gas pressure.
Fig. 10 - Spirax-Sarco's Pressure Powered Pumpt
Fig. 11 - Armstrong's Condensate Pumps Figures 10 and 11 represent examples of condensate pumps by two of the largest manufacturers of steam traps and other related equipment. Even though these condensate pumps vary in style and manufacturer they operate under the same basic principal. There is a check valve on both the inlet side and the outlet side of the collection chamber. As you can see in the cutaway in Fig. 10 there is a ball float attached to a spring mechanism. This spring mechanism is designed to trip at a predetermined point as the ball float rises. The spring mechanism is, in turn, attached by rod to two valve assemblies. One valve for the pressure supply port and the other for the vent port. In its empty, or idle, state the ball float is resting at its lowest point, the pressure side valve at the top of the unit is closed and the vent valve is in its open position. As condensate enters through the inlet port and accumulates in the pump chamber air, non-condensables and some flash steam are vented through the vent port (discharge of the vent should be piped to the outside) and the ball float begins to rise. The check valve on the discharge port prevents condensate in the return system from back-flowing into the condensate pump chamber. As the volume in the pump chamber increases the ball float eventually reaches its predetermined set point height, its high level point, and the spring mechanism trips. This pulls the valve rod down closing the vent port and simultaneously opening the pressure port. When the higher pressure, in the form of steam, air or inert gas, enters the pump chamber through this port it forces the condensate through the outlet port. The check valve in the inlet port prevents the condensate from flowing back through the inlet. As the level in the pump chamber lowers so does the ball float. When the ball float reaches the low level set point the spring mechanism trips pushing the valve rod up closing the pressure inlet port and opening the vent port. The pump is now ready for another cycle. When returning condensate to a return header located below the trap discharge configuration of the installation would require a drip leg, strainer (possibly integral to the trap), steam trap, test valve, block valve and header connection. The block valve on the steam side of the coil will be used to block the upstream side if the assembly needs to be isolated for maintenance. When returning condensate overhead with the assist of a non-electrical condensate pump the installation would require a drip leg, strainer (possibly integral to the trap), steam trap and test valve. If the condensate pump is servicing a single trap then the balance of the installation would include the condensate pump, block valve and header connection. If the condensate pump was servicing multiple traps with their discharge lines in a manifold configuration then the balance of the installation would include a block valve, sub-header connection, condensate pump, block valve and header connection. Since these heating coils, and that includes pre-heat and re-heat coils, operate on a seasonal basis there is ample time for maintenance to clean out or blow down each drip leg and to drop out the strainer screens for cleaning. This should be a scheduled preventive maintenance item. The trap for this application should be sized based on the specified BTU output of the coil. Let us assume that the specifications for the coil indicate a BTU output requirement of 400,000 BTU's/hr using 15 PSIG steam. To determine the anticipated condensate load divide the BTU output by the latent heat of evaporation for 15 PSIG steam. For this example the calculation would look like this:
Using a safety factor of 3 multiply the anticipated condensate load as follows: 423.2 lbs/hr x 3 = 1270 lbs/hr. The humidifier is an equipment item that introduces steam either directly into an area or into the airflow of an air handling unit in order to maintain an environment with a stable relative humidity. If the relative humidity is too low it causes discomfort to personnel and creates the potential for static electricity. In an explosive process area this cannot be tolerated for obvious reasons. If the relative humidity is too high it causes discomfort to personnel and is detrimental to office machines and operating equipment. The best engineered humidifier can be compromised by a poor installation design. It would be time well spent for the engineer and designer time to look into the installation and operation of humidifiers in some detail. This section will cover only the basics of the humidifier as a prerequisite to trapping the condensate. The humidifier assembly consists of a steam inlet connection with a strainer, a jacketed distribution manifold, metering valve with operator, separation chamber and steam trap. Some manufacturers choose to integrate the metering valve with the separation body and others choose to make them independent items. In regard to humidifiers with distribution manifolds and referring to Fig. 12, steam enters the assembly through the strainer, which removes most of the entrained particulate that may be carried by the steam. The flow of steam then enters the jacket of the distribution manifold. There may be multiple distribution manifolds determined by specific requirements. The humidifier is designed to remove moisture and sustain a dry steam up until the moment the steam is released into the area or airflow. The final protection against moisture build up is the jacketed distribution manifold. This prevents the airflow itself from condensing the steam prior to discharging from the manifold. After passing through the manifold jack the steam enters a separation chamber. Any condensate carried into the chamber by the steam is removed through the use of impingement baffles. The condensate flows to the bottom of the chamber and into a drip leg where it is trapped and returned to the condensate collection system. With the condensate removed in the separation chamber the steam flows into a second separation or drying chamber then finally into the distribution manifold. At this point steam is distributed directly into the airflow or out into a control space.
Fig. 12 - Humidifier When the humidifier is installed in an air handling unit the outlets of the distribution manifold should be directed against the airflow. This allows for a quicker and more thorough disbursement of the steam throughout the cross sectional area of the airflow itself. The type of trap used for this service could be an inverted bucket, a thermodynamic or thermostatic type. One of the key factors in cases where trap selection is a multiple choice is current plant spare parts inventory. Ideally what you want to do is minimize the additional new parts the plant is required to handle. If you can work with one of their current selections without compromising design integrity then this needs to be considered. With a large plant this is usually not a problem. Most plants will have a wide variety of trap types installed from a wide variety of manufacturers. As mentioned earlier, the more practical thing to do is to consult with plant personnel through the project team. Their may be a certain manufacturer, manufacturers representative or product that they have had difficulty with in the past. Don't lose site of the fact that the plant personnel will continue to live with this design long after the engineering firm is gone. Allow them to be a part of the process by providing know-how from their prospective. This is something that needs to be considered early in the design process. When sizing the trap the condensate load will be based on radiant heat loss. That is, the amount of heat loss that occurs through the jacket of the distribution manifold, and the upstream piping. There will be some heat loss in the body of the humidifier but we can factor that in by adding a foot to the length we calculate for the upstream piping. In calculating the condensate load there will be two separate calculations: one for the insulated upstream piping and one for the uninsulated manifold jacket. For this example we will use 15 PSIG steam in 1" pipe with an upstream length of 20'-0" from the last trap to the distribution manifold jacket. The jacket on the distribution manifold is 48" long. We can simplify the calculation process by replacing the jacket configuration with standard pipe. With the jacket having dimensions of 1 1/4" wide by 1 7/8" deep and one side open we can use a 1 1/2" pipe to represent the jacket in our calculation. Since the air-handling unit is in a controlled area we will assume an air temperature of 70B F. The calculation for the upstream piping would use the radiant heat loss calculation described earlier and would look like this:
The calculation for the jacket would look like this:
One difference between the manifold jacket and the supply piping, aside from the jacket having no insulation, is the airflow across the manifold. Since the manifold is mounted inside the duct it will be in the mainstream of the airflow. With 70B F air moving across the manifold jacket at a velocity of 1000 feet per minute that particular aspect would normally be considered in the above calculation. However, the operation of the humidifier basically negates that aspect. By injecting the steam against the flow of air it forces the steam to rapidly reverse its direction and flow back over the manifold jacket. This action actually creates a hot barrier between the jacket and the 70B F flow of air. In creating this barrier it virtually negates the tentative effect the air flow would have on the manifold jacket. In an attempt to simplify the calculation process while trying to maintain a balance between accuracy and efficiency we will allow the steam flow over the jacket to cancel out the airflow factor in the calculation. We have therefore omitted that factor from the calculation. With an accumulated condensate load of 0.95 lbs/hr + 1.10 lbs/hr = 2.05 lbs/hr x 3 (safety factor) = 6.15lbs/hr a 1/2" or 3/4" disc trap or an inverted bucket would be a good selection. Unit Heaters Unlike a central heating system that distributes heat throughout a building the unit heater is used to maintain a desirable temperature in a single, confined space. Depending on the configuration and size of the space to be heated multiple unit heaters may be required. Although standard convection heaters like baseboard heaters and radiators are considered unit heaters the only type we are concerned with in this section is the forced air type. This type of heater functions by forcing room air across coils heated by steam, hot water or an ethylene glycol solution. Radiant heat from the heater coils heats the air as it passes through the coils and out into the space to be heated. A later section will discuss the method for determining requirements of a unit heater. In regard to steam traps, while other types of traps will work in this application there are two types that are recommended for this service: the float & thermostatic and the inverted bucket. If a thermodynamic or thermostatic type trap is being considered for this application you need to consider whether or not the heater will be heating outside air. If the heater is positioned at an overhead door that is frequently opened, or remains opened for extended periods, that heater could see wide fluctuations in heat transfer requirements. Make certain to run the calculations on those variations to determine the load range that will be required of the trap at that location. Additionally, if a thermodynamic or thermostatic trap is specified for this application ensure that the trap is located sufficiently far enough away from the heater fins to prevent condensate from backing up into them, creating corrosion. As described earlier, these types of traps do not have an integral reservoir to collect the condensate until it discharges. In this type of trap the condensate backs up into the piping until it is discharged. Using the unit heater specifications to obtain the necessary information, the condensate load can be determined with the following calculation:
Where: C = Condensate in lbs/hr F = Cubic feet of air per minute Cp = Specific heat of air in BTU/lb/B F d = Dendity of air w .075 lbs/cu ft. DT = Temperature rise in B F H = Latent heat of steam As an example we will assume 15psig steam heating 9,300 CFM of air from 60°F to 112°F. The calculation would then be:
To illustrate the variance in loads when a heater is located near an open overhead door we will use the same criteria except we will change the inlet air temperature to 0°F. The calculation would then be:
This would indicate a condensate load for a unit heater, at an open door with sub-zero outside temperatures, at twice that of the units further inside the heated space. Using a safety factor of 3:1 the load range on the various units would be 1658 lbs/hr to 3570 lbs/hr. depending on their location. The only block valves required would be located on the upstream side of the unit heater and on the downstream side of the steam trap. Include a drip leg directly out of the unit heater. The strainer can either be independent or integral to the steam trap. Submerged Coils Submerged coils are heating or cooling coils contained inside a vessel and are designed to come in direct contact with the fluid to be heated or cooled. The depth, diameter, pipe size and length of the coil is determined by several factors that include, but are certainly not limited to: characteristics of the fluid to be heated or cooled, volume of the fluid, whether the fluid is agitated or static in the vessel and its expected retention time. Given the same fluid characteristics and volume the variation in the condensate load will depend on whether the fluid is agitated or static and what its retention time Depending on the type of vessel design coil configurations will vary. The inlet and outlet of the coil may be in the bottom out the bottom, in the top out the bottom, in the side out the side, in the side out the bottom, in the top out the side or in the top out the top. Special consideration needs to be assessed with an in the top out the top configuration when using steam as a heating fluid. This configuration creates a pocket in which the forming condensate has to be lifted up to the trap. Condensate at its equilibrium temperature has the potential to flash back to steam if either its temperature is increased or its pressure is reduced. In the process of being lifted from a lower elevation to a higher elevation the pressure of the condensate will be reduced at a rate of 1 pound for every 2.3 feet of rise in elevation. A lift of 12 ft. would translate into a 5.2 psig drop in pressure. The percent of flash steam created from a drop in pressure can be determined with the following calculation:
Where: SH = sensible heat of the condensate (btu/lb) at the higher pressure (before lift) SL = sensible heat of the condensate (btu/lb) at the lower pressure (after lift) H = latent heat of the steam (btu/lb) at the lower pressure To illustrate, lets assume we are supplying 50 psig steam to a top in, top out coil. The condensate that collects at the low point of the coil will be considered 50 psig condensate with a sensible heat content (heat of the liquid) of 267.6 btu/lb. The vertical distance from the inside bottom (invert) of the low point of the coil and the inlet to the steam trap is 10'-9". This is the height that the condensate has to be lifted. That vertical distance is equal to 10'-9" ÷ 2.3 (2.3 ft. of lift equals 1 lb) = 4.7. So the difference in pressure between the low point of the coil and the inlet to the trap is 50 - 4.7 = 45.3 psig. The sensible heat content at 45.3 psig is 262.2 btu/lb and the latent heat at that pressure is 915.4 btu/lb. Based on that information the calculation would be:
The accumulation of the flash steam will prevent the steam trap from operating properly and efficiently. As steam, the flash steam will prevent the trap from systematically collecting and discharging the condensate. It will bind the trap until the steam residing in the trap condenses. When the condensate contained in the trap is discharged additional condensate is lifted from the coil. As the new condensate is lifted a portion of it will flash. The trap will detect the flash steam and stop discharging before all of the condensate is removed from the coil. This process of removing only a portion of the condensate allows a large portion of the condensate to remain in the coil. This reduces heat transfer efficiency and promotes corrosion in the coil. There are designs that work effectively in trying to avoid these inherent problems. One such design utilizes a specifically designed steam trap by Armstrong. They call it their Automatic Condensate Controller as shown in Figure 13.
Fig. 13 - Armstrong's Automatic Condensate Controller This is an inverted bucket trap with a modified assembly allowing flash steam to by-pass the trap preventing the trap from binding. By continuously moving the flash steam through the system and into the condensate return line it allows the trap to remove condensate in a more efficient manner. Another method of removing condensate from this type of submerged coil involves the design of the discharge section of the submerged coil. This is the section of pipe at the end of the coil that rises straight up to the outlet nozzle. See Figure 14.
Fig. 14 - Coil & Discharge Dip Tube Detail This design, as shown in Fig. 14, is only practical when the coil size, or the pocket size is 2" and above. It consists of a low point trap for condensate to accumulate in and a smaller dip tube to serve as a siphon type discharge for the accumulated condensate. The cooler temperature further up the discharge pipe represents lower pressures causing the condensate to migrate up the discharge pipe toward the steam trap. The tee, as shown in Fig. 14, can attach directly to the coil without the two additional elbows as represented in Fig. 14. However, the above design provides added retention volume away from the coil for the accumulating condensate as it forms. Some coil designs extend a discharge pipe, the same size as the coil itself, vertically from the low point of the coil to the outlet nozzle. A dip tube is then extended down the length of the vertical discharge pipe to a location near the low point. Much like what is shown in Fig. 14. This type of design essentially creates a jacketed dip tube arrangement. The larger pipe will contain steam virtually the entire length of the dip tube. This effectively assists the condensate inside the smaller pipe to flash back to steam. We want to prevent flashing, or at the least not create a design that may assist the flashing effect as condensate moves up the discharge pipe. As mentioned, the lift alone will initiate flashing. The design in Fig. 14 eliminates the jacketing effect and allows further cooling of the condensate as it moves up the discharge pipe toward the outlet nozzle. Since the discharge pipe is not encased in the larger pipe the condensate in the discharge pipe is cooled by the surrounding product or utility fluid contained in the tank. Depending on the temperature differential between the condensate being lifted out of the coil and the surrounding fluid in the tank it may be enough to keep the condensate from flashing. If the design configuration in Fig. 14 is used then a standard inverted bucket or an F&T trap would be suitable for this application as well as the Automatic Differential Condensate Controller. In both discharge pipe configurations the situation exists for non-condensables to accumulate. To a greater degree where the discharge pipe remains the same size as the coil and extends to the outlet nozzle. In order to allow these non-condensables to escape an 1/8" diameter hole should be provided in the dip tube near the point where it passes through the cap, as in Fig. 14, or where it passes through the flange at the nozzle (not represented). This build-up of non-condensables, if allowed to accumulate, could eventually bind up the coil effectively preventing full use of the coil's heat transfer area.
|
|
HOME | POST AN AD | SITE INDEX | FUTURE
ARTICLES | ORGANIZATION
|
