Steam Basics Part 4

In order to put into action what we have just discussed we are going to require a steam generator or boiler, which is normally fired by coal, gas or oil. As was mentioned earlier, the sensible heat of water at 32° F is taken as zero. The specific heat capacity of water is 1.0 BTU/lb/° F. Therefore raising the temperature of 1 pound of water from 32° F to 212° F will require a sensible heat of:

If there is 2000 pounds of water in the boiler brought to 212° F, the total sensible heat will be:

But, if the water in the boiler was already at 70° F, the enthalpy required to bring the mass to saturation would be:

It must be remembered that the total heat of the liquid is still 360,000 BTU. But, since the water at 70° F already had some heat content, we only had to make up the difference to get it to saturation. And that difference is 284,000 BTU.

As we did with the 1 pound container, we will close off the boiler and continue adding heat, allowing the water to evaporate to steam. As the volume of steam increases the pressure inside the boiler will continue to increase until the supply of heat to the water is stopped or the pressure inside the boiler released. Since we want to generate steam at a specific pressure/temperature, say 150 PSIG, then we need to continue adding heat and water to the boiler and releasing the steam that is formed, at 150 PSIG. By referring to the steam tables we can see that at 150 PSIG the temperature is 366° F, sensible heat is 338.6 BTU/lb, latent heat is 848 BTU/lb and specific volume of the steam is 2.76 cubic ft./lb.

Let us take into account the peak winter demands as well as the lighter summer requirements. We can average out those demands and assume a useage rate for a moderate size plant at 20,000 lb/hr of 150 PSIG saturated steam. The actual demand for this plant, in terms of energy requirements, would therefore be:

If all of the services and equipment requiring steam had the same steam pressure requirements of 150 PSIG, it would make life a little simpler. But life and the varying requirements of steam are never simple. When determining a specific steam pressure for a user; economics, efficiency, equipment design limits, production demands and other criteria will dictate steam pressure qualifications, not simply what steam pressure is available.

In order to determine what the initial steam pressure should be generated at and how much, in pounds per hour, will be required, some preliminary work will have to be done.

One of the first steps in this process is to determine what the main users will be. Process and mechanical engineering will have this information. Through the exercise of determining the process requirements and most efficient and cost effective mechanical specifications, the equipment will be sized and the steam requirements established.

Lets assume that the process and mechanical engineers have established that 150 PSIG saturated steam will be the highest pressure required. Therefore our primary steam loop will be 150 PSIG. There will also be a requirement for steam tracing, unit heaters, air handling units and utility stations. These will be supplied, through a let-down or pressure reducing station, with 50 PSIG steam from a secondary steam loop. Other users will require steam at pressures between 150 and 50 PSIG and below 50 PSIG. The individual pressure demand for these users will be met with pressure reducing valves (or let-down stations) also.

Now that the basic pressure requirements of the steam are known we need to calculate the quantity of steam that will be needed. In order to do this the required number of BTU/hr for all steam users will have to be calculated. This is done by multiplying the lb/hr rate of the amount of steam required times the latent heat of the steam at that pressure. If one of the users required 120 PSIG steam at a rate of 300 lb/hr the demand would be:

By calculating the requirements of all users in this manner, a total latent heat requirement of 17,160,000 BTU is arrived at. To determine the lb/hr rate of steam that will have to be generated at 150 PSIG it would be:

In trying to keep this fairly basic there are a few things that need to be touched on without going into great detail. In distributing steam throughout the plant, and particularly larger plants that may be generating 100,000 lb/hr, 1,000,000 lb/hr or more, line pressure loss is going to be a factor. In determining the steam generating pressure at the boiler you will need to calculate the pressure drop that will occur from the boiler to the most remote user. The result will be a nominal steam pressure somewhere in mid plant between where the steam is generated and the end user. In doing this we may be giving up a pressure loss of 5%. With a larger plant and higher production needs lets say the steam pressure required now is 600 PSIG. In order to meet the demands we just discussed we will generate steam at 615 PSIG so that mid plant would see 600 PSIG and the remote users would see 585 PSIG. There will also be line losses in the intermediate pressures such as 250 PSIG, 100 PSIG and 50 PSIG. However, since the distribution from the reducing station to the users will be more localized the runs should be shorter and pressure drop through friction loss less of a concern. With the boiler house running at a rate of 100,000 lb/hr or more, at 5%, the loss in energy is in thousands of pounds and millions of BTU's per hour.

One way to reduce this loss is in the use of superheated steam for transmission and distribution. If steam turbines are in use in the plant then superheated steam will be required in any case. The pressure and temperature of the superheated steam is chosen so that the moisture content of the steam, in the last row of turbine blades, is less than 10%. This is to prevent the erosion of the blades due to accumulated water particles. Superheated steam does not readily give up it's latent heat as does saturated steam. Although a pipeline pressure drop will still occur there won't be a significant loss in BTU's.

Another cause of pressure drop in pipelines is size. This is where the pressure of the steam has a major effect on the pipe size. If you will notice in the steam tables in the column for "Specific Volume of Steam" there is quite a difference in the volume of steam throughout the pressure range. The specific volume for 150 PSIG saturated steam is almost 4 times greater than that of 600 PSIG saturated steam. This has a significant effect on pipe sizing and operating pressure. In the design of the distribution system it should be considered that higher pressure steam can move more volume through smaller pipe size than lower pressure steam. Sometimes it is more economical to increase the initial steam pressure in order to reduce the pipe size.

What has to be considered here is the reduction of latent heat by volume as the pressure of the steam increases. As the latent heat content decreases with increases in pressure the volume of steam will have to increase in order to provide the same amount of latent heat.

A determination will have to be made about what pressure this stops being economically feasible. You will have to take into account the turn-down ratio on the pressure reducing valves, additional fuel costs for generating more steam, potentially higher cost in material and a higher cost in insulation. All of these and possibly more, are governing factors in sizing a steam system.

In laying out a distribution system the single objective is to get steam to the far reaches of the plant at as close to the generated steam pressure as possible. Doing so will require attention to line size, insulation, configuration, moisture separation and steam traps.

Line sizing is based on either pressure drop per 100 ft. or velocity in ft/sec or ft/min. Design parameters will sometime vary from plant to plant but, as a rule, allowable pressure drop is .5 lb/100 ft for runs 300 ft and less. For pipe runs over 300 ft long, .1 lb/100 ft is acceptable. Velocity is normally held at 6,000 to 10,000 ft/min or 100 to 167 ft/sec for saturated steam. Superheated steam can operate at higher velocities, 7000 to 20000 ft/min.

You have to maintain a balance between operability and economics. The larger the pipe the less pressure drop but also a higher cost for the pipe and insulation in addition to more heat transfer area for losing BTU's. Depending on the valving and pipe configuration, higher velocity rates may increase noise. And even though that might be tolerated, erosion to valve trim and piping becomes a definite consideration. This becomes even more of a concern where there is a high moisture content in the steam. With water droplets traveling at speeds of perhaps 12,000 ft/min they become very detrimental to the pipe internals.

Insulation is too broad a factor to go into with this report. It should be noted however, that this too hinges on economics. The cost to achieve a 100% insulation efficiency factor is both cost prohibitive and physically restrictive. Good design practice will usually apply a 75% efficiency factor when calculating insulation thickness.

In getting steam from the boiler to the user there is never a straight line. High points and low points are poor points when it comes to steam. As mentioned earlier, saturated steam is always on the verge of condensing back to water. And no matter what is done within the realm of economic restrictions some steam is going to condense. It's up to the designers to minimize the rate at which it does condense and to handle the condensate when it does. With that in mind lets take a look at what happens when steam does condense.

The feedwater used in generating steam will, of course, contain oxygen. It can also contain bicarbonate and carbonate alkalinities which, when broken down due to high temperatures, will produce C0₂. These two gases, O₂ and CO₂, alone or combined, when disolved in condensate are very corrosive. The oxygen causes oxygen pitting while the carbon dioxide, in solution with the condensate, forms carbonic acid. When combined, the oxygen accelerates the corrosive effects of the acid.

Deaerating the feedwater removes almost all of these gases. In general, as the feedwater enters the deaerator low pressure steam, typically 5 PSIG, is used to break up the water into a spray continuing across the spray carrying off the gases. There are also tray type deaerators and combinations of both the spray and tray types. They will typically have a residual deaerated feedwater storage tank combined with the deaerator.

Since deaeration does not remove all of the oxygen, an oxygen scavenger in the form of sulfites or hydrazine is employed. These scavengers are added to the already deaerated feedwater.

Even with the pretreatment of the feedwater there will still be carryover in the steam. Additionally there will be air in the system from start-ups and batch operations. The problem occurs when steam condenses. As it condenses these gases will separate out which will then have to be dispatched along with the condensate. Both the condensate and the gases need to be removed as quickly and as efficiently as possible. The oxygen and carbon dioxide are corrosive agents, the air acts as an insulator and the condensate can diminish heat transfer, create flow restriction, cause water hammer, moisturize the steam and corrode pipe and equipment. There are ways to alleviate a lot of the potential problems by using air vents, separators and steam traps in addition to good design practice.

Good design practice, in terms of configuration, should include, among other things, the following: proper slope, the elimination of pockets, proper trapping of condensate when pockets do occur, strategic location of steam traps and a configuration that integrates flexibility to keep the system piping itself within allowable stress ranges during expansion and contraction cycles.

An air film 0.04" thick has the same resistance to heat transfer as water 1" thick, iron 4.3" thick or copper 43 feet thick. As a film it acts as an insulator, in solution with steam it deprives the steam of its full heating potential. In other words, air will assume a part of the total volume or pressure that is available.

In gas mixtures, each gas assumes a part of the total volume or pressure. This is referred to as partial pressure. The partial pressure of each gas is dependent upon its proportion of the total mixture.

If we were to have a total steam line pressure of 50 PSIA consisting of 80% steam and 20% air the total steam pressure would be:

As a result the steam would effectively be 40 PSIA steam in a 50 PSIA line. In checking a thermometer in the line at that location we would find the temperature to be 267° F for the 40 lb steam and not 281° F for the 50 lb steam we would expect to find. That is a 14° F difference between the two pressures. In addition there will also be a change in BTU's. Although the 40 lb steam has more enthalpy of latent heat per weight than does the 50 lb steam it has less by volume. As a comparison, inside the pipeline, a 1 square foot area of the 50 lb line would contain 108.7 BTU; a 1 square foot area of the 40 lb line would contain 88.9 BTU. That is a difference of 19.8 BTU/sq ft.

In effect the air has displaced a portion of the enthalpy needed, by displacing a portion of the steam. In order to make the distribution system as efficient as possible it becomes necessary to remove any air before it can effect heat transfer by filming or becoming mixed with the steam.

Knowing that it is virtually impossible to keep air, oxygen and carbon dioxide from getting into a system, lets deal with getting them out of a system. As mentioned earlier, these gases become free when the steam condenses. All steam traps will pass air and other gases along with the condensate, as it forms. The only difference is that some will handle a good deal more air than others. Without getting into discussions of the various types of traps, which will be discussed at length a bit later, we will assume a worse case which would be a batch operation or a modulating steam control valve with a vacuum breaker downstream of the control valve. When the control valve shuts off steam to the heating unit, the steam between the control valve and the steam trap will begin to collapse due to condensing. As the steam begins to collapse a vacuum begins to form. When the vacuum overcomes the set point of the vacuum breaker it will open, allowing air to replace the volume of steam. Now, in the heating unit, we have a volume of air which we have to evacuate as quickly as possible when steam is introduced again.

Depending upon the configuration of the heating unit, it may require only a steam trap or it may require two or more additional air vents. If it's a piping coil of some sort it may only require a float and thermostatic trap. As the steam is supplied to the unit again it will purge the air ahead of it and out the thermostatic vent in the trap. When the steam reaches the trap the vent will close. In this circumstance, the air normally does not combine with the steam. The air is located in a dead leg separate from the steam. When steam is introduced again it will move the air out ahead of it. Only if the air can not be purged fast enough would it be forced to combine with the steam causing inefficient heat transfer.

If the heating unit has a large cavity, such as the shell side of a shell and tube heat exchanger, a pocket of air could develop that would require venting. A thermostatic air vent should be located at this point. The air vent will allow the air to be removed from the steam chamber before it has a chance to combine with the steam or film up inside the heat exchanger or piping.

A peculiar side issue to air vents and even the ambient sensing valves used to drain condensate lines automatically before they freeze up, is the fact that they are sometimes misinterpreted. Keeping in mind that the pocket of air in a steam line will be hot and humid when released to the atmosphere, what will happen when it is winter and 0° F outside?

The moisture content in the escaping air will have a tendency to freeze upon contact with the cold discharge piping. After several discharges the formation of ice will become apparent. This presents the illusion that the air trap is leaking. When in essence it is doing its job.

The same holds true for the ambient sensing valve. This valve senses the internal temperature of the pipe contents. When the temperature of the fluid inside the piping drops to a predetermined set point, about 8° F above freezing, the valve will open. This allows the fluid to be drained from the piping before it has a chance to freeze up. In doing it's job this valve has a tendency to be misinterpreted also. If it is 0° F outside and the valve discharges, some of the fluid will freeze upon contact with the cold discharge piping. With repeated discharges there is an excellent likelihood that a good sized icicle will begin to form off the end of the discharge piping. Again this gives the appearance of a leaking valve when, in actuality, the valve is functioning properly.

Now back to distribution design. As mentioned earlier, moisture in steam is both inefficient, in regard to heat transfer, and detrimental to piping and equipment. In identifying the fact that moisture is going to condense out of saturated steam we have to determine why it does and how best to control it.

The two main factors attributable to the formation of condensate prior to reaching the various users are; heat transfer and pressure drop. As stated earlier, it is not economically feasible, in most cases, to achieve 100% efficiency in insulation design. We therefore have to design more realistically for 75% efficiency. Other insulation inefficiencies involve poor design standards, poor installation and, after a period of time, poor maintenance upkeep.

Any time a fluid moves through a pipeline there will be a pressure drop. Just how significant that pressure drop is depends on the characteristics of the fluid, the condition of the internal walls of the pipe, the configuration and length of the piping.

The designer can't do anything about the fluid, which is steam at a predetermined pressure. Neither can the designer do anything about the internal roughness of the pipe wall. What the designer can do is determine the piping configuration and research the best type insulation for the job. Installing and maintaining the insulation will be the responsibility of plant maintenance.

In routing the main steam supply line a constant slope should exist in all horizontal runs, sloping down in the direction of the steam flow. A good rule of thumb, for slope, is 1 inch in 20 feet, although this is certainly open to variation. Using a rule of thumb in general is fine, only after that rule of thumb has been determined to work and provided the designer can recognize the marginal areas where that rule of thumb may not apply. In order to do that we have to understand the reason why and the basis for, sloping the lines in the first place.

As condensate forms on the inside pipe walls it will run down the walls to the bottom inside of the pipe. When the condensate gets there it has to be able to flow easily to a point where it can be removed; a drip leg. Knowing that it is going to be easier for the condensate to flow with the steam we will need to slope the line down in the direction of the flow. If the flow of condensate was going against the flow of steam the following would

Occur: With the steam moving at a velocity of 6000 to 10000 fpm the condensate would tend to eddy into pools trying to overcome the friction of the steam on its flow to the low point upstream. As the condensate builds up in this manner the velocity of the steam will cause it to be picked up and introduced into the steam as a mist. Essentially humidifying the steam. As the moisture level of the steam rises the heat content lowers. It is apparent by this that you never want to consider having to slope the steam line down toward the flow.

In determining whether or not our rule of thumb, 1" in 20 feet, will be sufficient slope we need to know the reason for the slope. What is behind this is the fact that we need to keep the condensate flowing so it can be trapped and carried off before it has a chance to accumulate. If the steam line was perfectly level then the condensate would accumulate at the inside bottom of the pipe and, with steam as it's motive force, flow in the same direction as the steam. But in reality, horizontal runs of pipe will deflect or sag btween supports. The amount of deflection dependent upon the pipe size, schedule and the span between supports.

Lets use, as an example, a 4 inch sch. 40 steam line that is supported every 20 feet. The deflection in this line would be approximately 3/16". If this line was supported at the same elevation at each support there would be a series of 3/16" pockets that would accumulate condensate. In order to alleviate the problem the pipe will have to slope in excess of 3/16" in 20 feet. So in this case the 1" in 20 feet would be sufficient.

Trapping condensate in supply headers is much like the difference between using a dress glove to catch a line drive baseball or using a fielders mitt; which would you prefer? Aside from the pain issue the obvious selection would be the fielders mitt. The larger catching area provides more opportunity to catch the ball. In most cases a ½" trap and condensate return line is all that will be required to handle the condensate at a drip leg. But if you are trapping a l0" or 12" header and run that ½" trap line directly off of the header, the chance that the condensate, all of it at any rate, will find that ½" opening is very slim. In order to try and capture all of the condensate, a branch connection large enough to allow the condensate to drain in to it while moving down the header, needs to be provided. Refer to Table 1 for recommended drip leg sizes

RECOMMENDED DRIP LEG SIZES
HEADER SIZE	DRIP LEG SIZE
1/2" Thru 4"	Same as header
6"	4"
8"	4"
10"	4"
12"	6"
14"	6"
16"	8"
18"	8"
20"	10"
24"	12"

If there is a situation where a drip leg is placed at the end of a horizontal run and there is a riser at that point always use a full size tee. For two very good reasons; using an o-let fitting, a half coupling or stubbing into the line will have to be done at the tangent of the elbow in order to install the drip leg at the lowest point. In doing this it may mean encroaching on the buttweld of the elbow, which, in most cases is not a good practice. Also, when a full size tee is installed it is more cost effective to install a full size section of pipe with a cap. Rather than install a reducer, a section of pipe and a cap in an effort to comply with a drip leg sizing schedule like the one shown above.

There should be a drip leg located at all low points or pockets. In long straight runs there should be a drip leg located about every 300 feet. This assumes a properly sized and insulated system is installed. If the piping is sized too small undue friction loss will create additional amounts of condensate. Likewise, with insulation that is not sufficiently thick enough. It will permit an additional amount of condensate to form due to added radiant heat loss.