Dawson Co. Engineering Community Blog

Flash Tanks for Steam and Boiler Systems

2007-03-27T09:17:00.000-07:00

Low in initial cost, flash tanks are easy to apply and aid in condensate drainage of steam equipment

By Roy C.E. Ahlgren
Associate Member ASHRAE

In the old days, steamfitters and plant engineers knew the advantages of having a properly designed flash tank in the right place and for the right purpose. In trying to reduce steam system operating costs today, many are turnmg once again to this surprisingly versatile and energy efficient device.

This article describes how flash tanks work, how they can be employed to improve system operation and save energy, and how they should be designed for safe and trouble-free operation.

Operation
Flash tanks can be an important part of many steam systems. They get their name from the sudden evaporation or "flashing" that occurs when hot water at some higher pressure is suddenly released to a lower pressure. We can think of the heat that causes this flashing as the difference between the enthalpy of the liquid at the higher pressure and the enthalpy of saturated liquid at the lower pressure.

The enthalpy of evaporation (the heat required to change the liquid to steam at the lower pressure) determines how much of the liquid can be converted to steam by the heat available In other words,

% of flash steam =
heat energy available due to change in pressure
energy required for change of phase

% flash =
high presure liquid enthalpy - low pressure liquid enthalpy
enthalpy of evaporation at the lower pressure

For example, one pound of boiling water (saturated liquid according to the steam tables) at 100 psig/338°F (689 kPa/170°C) is flowing froma pipe into a tank at atmospheric pressure. What fraction, X, of the water will flash into steam?

From the steam tables, we know that each pound of water at saturation conditions has a given maximum heat content (liquid enthalpy) or total enthalpy for steam. The data from the example and the steam tables can be summarized as shown in Table 1.

If we ignore heat losses, the enthalpy of the high pressure water must be equal to the total enthalpy of the low pressure water and steam entering the tank:

This says that 0.113 lb (0.06 kg) of steam at 0 psig/212°F (0 kPa/100°C) will be formed from each pound of water as it drops in pressure from 100 psig/338°F to 0 psig/212°F (689 kPa/170°C to 0 kPa/100°C). In general terms,

% flash =
high pressure liquid enthalpy - low pressure liquid enthalpy
enthalpy of evaporation at the lower pressure

So, if a given flow of water, 2,000 lb/hr (0.25 kg/s) at 200 psig/338°F (689 kPa/170°C), is flashing to steam at atmospheric pressure, the flow of flash steam will be:

0.133 x 2,000 lb/hr = 266 lb steam/hour (0.03 kg/s, or 13.3%)

and the remainder of the flow will be:

0.867 x 2,000 lb/hr - 1734 lb water/hour (0.22 kg/s), or 86.7%

The production of flash steam is influenced by other components in the system too. For example, thermostatic steam traps open only after the condensate has subcooled, or dropped below the saturation temperature, so the amount of flash steam for a given drop in pressure will be reduced. For one pound of condensate, each Farenheit degree drop in temperature below saturation reduces the enthalpy of the condensate by approximately one Btu.

Uses for the flash tank
Flash tanks provide one or more of the following: a common lower pressure point for collecting condensate from steam equipment operating at different pressures; a means to cool hot condensate to allow the use of low temperature rated pumping equipment; and a source of low pressure steam for heating or process use.

One of the simplest flash tank applications uses high pressure condensate to raise the average temperature of a mixture of low pressure condensate and make-up feed water in a vented receiver. The high pressure condensate should be introduced by means of a perforated pipe below the water line so the flash steam can be completely condensed as it mixes with the volume of water. If the flow rates and temperatures result in an average temperature below 212°F (100°C), vent losses will be minimal.

Often several steam condensing units are operated in parallel, each with it's own temperature regulating valve. The steam pressure in each unit will vary widely, depending on the setpoint and load. If these units were simply connected to a common condensate return pipe, back pressure in the pipe from the unit discharging condensate at a higher pressure would cause condensate to back up in the unit operating at lower pressure. Installation of a flash tank vented to the atmosphere reduces the possibility of back-pressure and allows all of the units to drain, increasing their productivity and minimizing the potential for heat exchanger damage due to condensate flooding.

Although the flash steam is often vented directly to the atmosphere, many have recognized the significant waste of energy and water associated with this practice and installed another heat exchanger (called a vent condenser) to condense the flash steam recovering the heat and water.

Flash tanks usually drain by gravity to a receiver and pump combination. Because condensate flashed to atmospheric pressure is cooled immediately to 212°F (100°C), only a little more cooling is required to allow use of conventional, low temperature rated, condensate pumps. These pump receivers are usually vented to the atmosphere and rated to handle condensate below 200°F (93°C). A vented flash tank is shown in Figure 1.

Flash legs are really small flash tanks that have been built into the steam piping to dispose of the condensate from high pressure steam mains. The flash leg is an oversized pipe with end-plates welded to it. High pressure condensate passes through a steam trap into the flash leg where it immediately flashes, not to atmospheric pressure, but to the pressure of the low pressure steam main. The flash steam is vented directly to the low pressure main and the remaining condensate passes through another trap into the low pressure condensate main. A flash leg is shown in detail in the ASHRAE Handbook - Systems.(1)

If several steam loads are operating at constant pressures, we may connect them to a flash tank operating at a pressure lower than the load's but higher than atmospheric. In this way, the flash steam can be recovered under pressure and the flash tank becomes a source of low pressure steam.

Pressure in the flash tank may be maintained by connecting the steam vent from the flash tank to a pressureized system such as a deaerator or low pressure steam main. In that case, a check valve must be installed to prevent backward flow if the flash tank pressure should drop. A back-pressure valve could be used to control the maximum pressure in the tank and a relief valve is always required to protect the system. The steam piping between the flash tank and the application point should be generously sized to minimize velocity and carryover of water droplets.

Ideally, the system should be designed so that the maximum rate of flash steam flow will not quite provide all the low pressure steam required for the application. Under these conditions, a pressure reducing valve (PRV) from a steam supply main would be used to make up the total amount needed and carry the entire load when the flash tank is not in service or when high pressure condensate is not available. When flash steam is available, the PRV will close, reducing load on the boiler. A pressurized flash tank is shown in Figure 2.

In designing a pressurized flash tank, be aware that the water leaving the flash tank may flash again in the piping as pressure on it is reduced. Because of flashing, the slection of condensate pipe size involves some complexity.

One way to size this pipe is to calculate the amount of flash steam that would form if the condensate went immediately from flash tank pressure to destination pressure. Then assume that this flashing occurs all at once, and size the condensate pipe as if it were a stam pipe carying only the flash steam. The pressure drop due to the liquid portion is negligible.

If we fail to consider flash steam in the condensate pipe, the pipe may be badly undersized. this further condensate flashing results in a significant waste of energy if vented pump receivers are used, so special closed condensate systems have been developed to keep pressure on the condensate as it leaves the flash tank to prevent additional flashing and, thereby, save energy.

The receiver in a closed condensate unit operates at the same pressure as the flash tank so condensate flows to it by gravity. Because there is no pressure reduction and no vent, there is no loss through condensate cooling or vented flash steam and steam traps between the flash tank and the closed condensate equipment are not required. Specially designed pumps, which can handle the high temperature condensate without cavitating, are required for this kind of equipment.

In all of these applications, the steam traps at the condensing equipment must be selected and installed to be able to handle their condensate load under the worst case condition of highest back-pressure and lowest steam pressure.

Another common use for flash tanks is to recover heat from the continuous blowdown or surface blow from the boiler. An additional liquid-to-liquid heat exchanger can extract still more energy from the boiler water leaving the flash tank by cooling it to an environmentally acceptable temperature before discharging it to waste. Boiler water is introduced into the flash tank by a perforated pipe below the waterline to aid in reducing carryover of boiler contaminants.

Designing the flash tank
Calculate the amount of condensate entering the flash tank. This will be the sum of the steam condensing capacities of all equipment to be connected to the flash tank. Do not use trap capacities because traps are usually sized by multiplying the condensate load by a safety factor to allow for unknowns and flow variation. For boiler blowdown heat recovery, multiplying the maximum blowdown percentage by boiler capacity and the number of boilers.(5)

Calculate the flow rate of flash steam formed by using the formula and data from the steam tables. Multiply this percentage by the flow rate of water into the tank to get the flow rate of flash steam.

The tank steam volume must be large enough for the instantaneous flash. We can calculate the required volume using specific volume figures from the steam tables (3) In this example, one pound of steam at 0 psig will occupy 268 cubic feet.

For example, 2,000 lb/hr (0.25 kg/s) of condensate at 100 psig (689 kPa) flashes to atmospheric pressure. As before:

Steam flow rate = 0.133 x 2,000 lb/hr = 266 lb/hr (0.03 kg/s)

Steam volume =
266 lb/hr x 26.8 ft³/lb x 1 hr/3,6000 sec = 1.98 ft³/sec

In addition to providing adequate volume for expansion of the steam, a vent pipe of proper size must be installed to avoid pressure build up. use steam flow tables, formulas or a nomogram to select a pipe that will vent the steam at a velocity of about 4,000 ft/min (20.3 m/sec) or less. This will limit flow noise and allow the flash steam to vent off withouit seignificant increas in pressure.(3)

In no case should the vent size be less than 2 in. (50 mm). The ASHRAE nomogram for steam pipe sizing(2) at 0 psig indicates that a 2-1/2 in. (65 mm) vent pipe could carry 266 lb/hr (0.03 kg/s) of flash steam at low velocity and without bnuilding up significant pressure in the tank.

The tank water volume depends on the condensate flow rate and density. For the example above:

2,000 lb/hr condensate - 266 lb/hr flash steam = 1,734 lb/hr water

The density of water decreases a bit at typical flash tank temperatures. In this example, it's about 59.8 lb/ft³ (957 kg/m³), and:

The total volume entering the tank each second will then be:

Therefore, this flash tank should have a volume of approximately two cubic feet.

In this example, no allowance has been made for varitations in condensate flow, or the fact that flash steam separating from the condensate at high velocity may carry over significant amounts of the liquid or boiler contaminants. Some sources have recommended that arbitrary safety fators be applied to the steam and water volumes, but these are hard to justify and they lead to wide differences in the tank size recommended.

A better method for finding the flash tank size is available. it depends on the disengaging area as well as the volume of the tank. The disengaging area is the surface through which the steam must flow to separate from the liquid, so this concept is especially useful when liquid is introduced to the tank below the water line.

The disengaging area is interpreted as the rectangle that is formed by the intersection of the water surface and the inside of the tank when the tank is exactly half-full, or the length of the tank times its diameter for a horizontal cylindrical tank. For a vertical cylindrical tank, the disengaging area is simply the area of the circle formed by the intersection of the water surface and the inside of the tank. The disengaging area is shown in Figure 3.

The flash tank disengaging area required is shown in Figure 4. you can use the disengaging area curves without looking up values from the steam tables to ensure that a tank will have a certain maximum steam separation velocity. these curves take into account the amount of flash steam generated under different conditions of pressure and pressure drop as well as different steam separation velocities.

Use high velocities for lowest intitial cost or when some provision for liquid carryover (such as steam separator) has been provided in the low pressure steam main. Lower velocities will limit the amount of carryover and are particularly important for boiler blowdown applications. Use of lower velocities will lead to larger tan ksizes, an advantage when the flow of liquid entering the tank is expected to be highly variable.

Enter at the bottom of Figure 4 with the initial pressure of the liquid entering the tank. Follow a vertical line up to the intersection with the flash tank pressure curve in the desired velocity range, then across to read the disengaging area required per 1,000 lb/hr of high pressure liquid.

For our earlier example, 100 psig (689 kPa) condensate flashing to 0 psig requires about 0.5ft² per 1,000 lb/hr to maintain separation velocities below 2 ft/sec (0.6 m/s). that is approximately the separation velocity commonly used in packaged boilers to ensure dry steam.

For our example:
Disengaging area = 0.5 ft²/1,000 lb/hr x 2,000 lb/hr = 10 ft²

Knowing A, the area required, we can select a commercially available tank diameter and calculate the minimum length by using the formula:

or we can calculate the diameter required in a vertical tank by:

Some flash tank designs are specially suited for high separation velocities. In a tank wher ethe high pressure liquid is introduced above the waterline, the actual disengaging area is the sum of the surface areas of all the water droplets spraying into the tank. So, tank size can be reduced by using the curves that allow higher steam separation velocities, up to a maximum of 10 gt/sec (3 m/s).

Some manufacturers supply a vertical cylindrical tank with a tengentially-mounted condensate inlet pipe to give the water a whirling motion that aids in separating the steam and water and also allows higher velocities. For our example, the disengaging area could be reduced from 1.0 ft² to 0.2 ft² simply by allowing the higher separation velocity.

Flash tank shape is not critical so horizontal or vertical cylindrical tanks may be used. notice that the level of the water is the vertical tank is not as important as it is in the horizontal tank because the same disengaging area exists from top to bottom. That's the reason for the discharge piping loop around the horizontal tank - it ensures that the water level will remain at half-full.

On the other, a tank with adequate disengaging area does not necessarily have sufficient volume to maintain the tank pressure within acceptable limits.

The tank volume required is shown in Figure 5. This figure allows us to quickly estimate the tank volume required without using steam tables or arbitrary safety factors. Returning to our example, enter the chart at 100 psig, then across to read 0.99 ft³ per 1,000 lb/hr.

For our example:

Volume = 0.99 ft³ per 1,000 lb/hr x 2,000 lb/hr = 198 ft³/sec, as we calculated before.

Using the disengaging area chart in Figure 4, we determined that a minimum area of one square foot is required for low velocity. The actual volume and disengaging area of the tank will depend on the selection of the other tank dimensions as shown in Table 2 for some readily available tank sizes.

At low flow rates, standard saize and shape tanks that meet the volume requirement will usually meet the disengaging area requirement too. At higher flow rates, or where a tank is to be built to order, applying the disengaging area criterion becomes more important.

Size the low pressure condensate piping using standard water pipe sizing methods if the condensate will not be subject to further flashing. Condensate from a pressurized flash tank will flash again as pressure on it is reduced and vent losses will occur if a closed condensate system is not used.

Additional considerations
Because of heat losses from the piping and tank, the actual amount of flash steam generated will always be less than the calculated amount. Because it is good practice to provide a pressure reducing valve to augement the flash steam from a pressurized flash tank anyway, this small discrepancy need not be of concern. The condensate piping to the flash tank and the tank itself should be well insulated to reduce these losses.

In determining the feasibility of using flash steam, the timing of its availability and its requirement in the process should be considered. if flash steam will not always be available, then the augmenteing pressure reducing valve must be capable of handling the entire low pressure steam demand. If the peak production of flash steam occurs at the time of minimum requirement, then another applicaiton must be found, or some way to exhauset the steam to a vent condenser or the atmosphere must be provided.

Finally, there are several features that should be included in every flash tank installation:

**A condensate bypass around the flash tank should be provided to return condensate directly to the pump receiver when the flash tank is out of service. Adequate provision for venting the receiver and cooling the condensate must be made or a suitable closed condensate recovery system must be provided.

**All pressure tanks should be ASME stamped and equipped with relief valves of proper size and capacity as well as a vacuum breaker if cold water could be introduced in to the hot tank.

**A pressure gauge should be provided on the flash tank.

**Properly sized steam traps should be provided on a pressurized tank.

**Each condensate feed branch to the flash tank should be equipped with a check valve to avoid backfow. The steam piping may also need a check valve if it is connected to pressurized equipement.

**The flash tank should be equipped with a thermostatic air vent to aid in start-up and to remove any non-condensable gases that may accumulate during operation.

Conclusions
Flash tanks operate with no moving parts, using simple thermodynamic principles. Therefoer, they are low in initial cost and easy to apply in many condensate handling situations to aid in condensate drainage and proper operation of steam equipment.

By simultaneously generating flash steam and cooling the liquid, the flash tank can supply steam for useful purposes, thereby reducing load on the boiler and allowing condensate transfer by use of readily available pumping equipment.

As energy costs increase, flash tanks as an addition to the steam and condensate system will become increasingly popular.

This article was offered in the hope that it will serve to help you design better steam systems. Although a great deal of research and study has gone into the article, no guaranty or warranty either written or implied is given for the fitness or usability of these ideas.

Local building codes and regulations should be carefully checked to ensure that the information agrees with the codes and regulations. Local conditions regarding atmospheric and other conditions should be checked to ensure the usability of the dtails and design data.

References
1. ASHRAE 1980 Handbook-Systems. Chapter 13, figure 38. Atlanta, Georgia.

2. ASHRAE 1989 Handbook-Fundamentals. Chapter 33, figure 10. Atlanta, Georgia.

3. ITT Fluid Handling Division. 1978. Basic Steam Pipe Sizing Charts. Training Manual TES 378. March. Morton Grove, Illinois.

4. ITT Fluid Handling Division. 1982. High Temperature Condensate Return Systems. Training Manual TES 582. May. Morton Grove, Illinois.

5. Kendrick, Lee. 1986. Design Manual for Heating, Ventilation, Plumbing and Air Conditioning Systems. 4th ed. Arlington, Virginia: Technical Standards Publications.

Printed in ASHRAE Journal, August, 1991

SPECIFYING ARI STANDARD 400-2001

2006-11-08T12:04:00.000-08:00

SPECIFYING ARI STANDARD 400-2001
For Plate And Frame Heat Exchangers

Why should you specify and install Bell & Gossett GPX plate and frame heat exchangers?

Bell & Gossett GPX plate and frame heat exchangers offer maximum efficiency, a small footprint, and exceptional application flexibility. The modern plate design allows the Bell & Gossett GPX plate & frame heat exchanger to perform with less than one-third the surface area required by conventional U-tube heat exchangers designed for the same application. Finally, and most importantly, the Bell & Gossett GPX line of plate heat exchangers are ARI-400-2001 certified, guaranteeing you the performance you expected and designed for.

What is ARI Standard 400-2001?

ARI Standard 400-2001 is a certification program established to define test requirements, rating requirements, and conformance and marking conditions for liquid to liquid heat exchangers. The certification requires annual testing by an independent ARI approved testing lab. These test results are then compared to the manufacturer’s published performance ratings.

ARI Certification is only granted to units that meet or exceed the following manufacturers’ published thermal performance ratings.

• Total Heat Transfer Rate: >95% of published

• Tested pressure drop: <110%>

Failure to meet the test requirements requires re-rating or ceasing of labeling of the failed product as ARI certified.

ARI Standard 400-2001 provides a common method for evaluating the thermal performance of liquid to liquid heat exchangers. Specifying ARI Standard 400 certification allows buyers and end users to make equal comparisons between manufacturers.

Why do you need an ARI Standard 400-2001 Certified Heat Exchanger?

• To ensure "alternate" products do not take "liberties" with their published performance data (defined in more detail later)

• It ensures all three main components in commercial HVAC systems are independently certified. The three main components of a commercial HVAC system are the cooling tower (CTI Certified), the heat exchanger (ARI Certified), and the chiller (ARI Certified). These components work together as a complete system. Their relationship has a large effect on energy savings since a chiller’s efficiency improves with colder water. Since energy prices are increasing continuously, it is very important to look at every component in a system to ensure that each component performs as originally specified and assured by the manufacturer and that each one is as energy efficient as possible. If your cooling tower is CTI certified and your chiller is ARI certified, shouldn’t your heat exchanger be ARI certified as well?

• ARI certified components, including the heat exchanger, may assist in obtaining LEED Certification. One category that HVAC systems can impact LEED certification in is Energy & Atmosphere. This category requires a reduction in building energy consumption and has the greatest number of potential points towards LEED certification. It is estimated that a chiller’s energy efficiency increases 2% for every degree cooler the supply water is to the chiller. An ARI certified plate heat exchanger ensures that the chiller receives water at the temperature specified. Therefore the engineer can then specify closer temperature approaches and be confident that what is specified and expected is actually what the plate heat exchanger will provide.

• An ARI certified plate heat exchanger will result in cost savings for the end-user. Undersized (non-ARI certified) plate and frame heat exchangers will increase operating costs by requiring the chillers to start sooner and operate for a longer period of time.

What do we mean by "liberties"?

Due to the ability of plate type heat exchangers to achieve close temperature approaches with high heat transfer rates, altering the design temperatures by even tenths of a degree or understating the actual pressure drops can significantly reduce the amount of surface area required and therefore, cost of the heat exchanger. Many times these "liberties" are taken by our competitors but are not disclosed to the buyer or end-user. To counteract this practice, members of ARI’s Liquid to Liquid Heat Exchanger committee developed Standard 400.

Though members included delegates from ITT, Alfa Laval, Tranter and FlatePlate, only ITT and one other manufacturer obtained their certification. Please help even out the playing field by specifying ARI Standard 400. This will help ensure an "apples to apples" comparison at bid time and procurement time.

How to specify ARI-400-2001:

By adding the following wording to your specifications, you can be assured that what you specify is what will be provided. Thereby guaranteeing the performance you demand

• The manufacturer shall provide written guarantee to the accuracy of the heat exchanger thermal design.

• The manufacturer shall be listed with the ARI Liquid Heat Exchanger Certification Standard 400 for the model being supplied.

• Should the heat exchanger not perform to the specified conditions as defined in ARI 400, the manufacturer is responsible to replace or repair the heat exchanger in order to achieve the stated performance.

• If the manufacturer is not ARI 400 certified, a witnessed factory performance test must be completed and documented per the testing specifications of ARI 400.

Please feel free to contact your local Dawson Company representative for a complete ARI-400 formatted Bell & Gossett GPX specification, or to assist you with sizing of your plate and frame heat exchangers. For more information on ARI-400 please visit www.ari.org.

Copper Finned Tube Boilers and Primary/Secondary, Why?

2006-11-08T11:56:00.000-08:00

"Hey Jim, I need to bring something to your attention. This copper finned tube boiler isn’t piped primary/secondary. Unfortunately we won’t be able to complete our start-up until the boiler piping is corrected".

"Dave, what do you mean you can’t complete start-up, we have to be on line tomorrow for the grand opening!"

"I’m sorry Jim; this boiler should be piped primary/secondary to insure trouble free long lasting operation. If it’s not piped primary/secondary, you may get a low flow condition or condensation may form. Both will cause tube bundle failure and it will not be covered by warranty"

"But Dave, this is the way it’s shown on the plans…"

This is one of the most uncomfortable situations I’ve been faced with as a manufacture’s representative and, unfortunately, one that occurs all too often. The use of copper finned tube boilers has grown tremendously in the last ten years through plan/spec, design build and replacement projects. The recent improvements of copper finned tube products, their low cost and ease of use have increased their use. In the past, the consulting engineer or contractor has designed/installed copper finned tube boilers on the job as they always have using a steel tube boiler with primary pumping. What they don’t realize is that copper finned tube boilers differ from other boiler types and special consideration must be taken when piping them, especially in comfort heating applications.

The biggest difference between the two technologies is the amount of water they hold. Steel tube boilers are high mass units, typically holding many gallons of water. They also take longer to reach set point and are slower to respond to system demands. Due to the amount of mass they hold, they can handle variations in flow well as variable speed drives and 2-way control valves respond to system demands.

Conversely, copper finned tube boilers are low mass units and hold a small amount of water. The water they hold is based on the amount of water the heat exchanger tubes can hold, which depending on the size of the boiler may only be a few gallons. They basically heat the water via convection and gas radiation by passing the products of combustion through a series of baffled finned copper tubes, thus heat transfer is very effective and warm up time is fast, almost too fast. In fact, copper finned tube boilers are often referred to as "quick-start" boilers because they are able to heat the water inside the tubes up to 9 times faster than cast iron or steel tube boilers.

Copper finned tube boilers are designed to handle high "heat flux" or heat transfer, and rely upon higher tube bundle velocities to keep the tubes free from scale build-up and to "unload" all of the heat they can transfer. Therefore, a basic element of design for all copper finned tube boilers is the flow rate that the heat exchangers require. Piping these boilers using direct primary pumping can pose a myriad of problems and in most installations will shorten the life of the boiler drastically. Problems that may occur include:

• Over pumping: if the pump isn’t matched to the boiler properly, excessive velocities may occur in the tube bundle leading to tube erosion and premature failure. When higher flow rates are required, such as in domestic water heating applications with high hardness levels, cupro-nickel finned tubes that have greater resistance to erosion may be ordered instead of the standard copper finned tubing. I must stress that even cupro-nickel finned tubes have limits on internal velocities and direct primary pumping may still pose a problem.

• Low/no flow over heating: in a 2-way valve variable volume and/or variable speed system, the flow constantly varies to match the system’s load. As the valves begin to modulate closed or the pump speed is reduced, the flow through the boiler may not be enough to take away all of heat the boiler is producing, causing excessive heat flux, liming inside the tubes, and ultimately, tube bundle failure. Even if the boiler turns off, there still may be enough residual heat in the combustion chamber to cause damage. Thus the pump must continue operating until the heat is dissipated.

• Condensation: In today’s world, equipment manufacturer’s have been forced to design equipment to operate at the highest possible efficiency. For boilers, this may or may not pose additional design considerations. As boiler efficiency increases, the temperature of the products of combustion will decrease. If the temperature of the products of combustion falls below the dew point, condensation will form and fall back into the boiler. It turns out that this condensation is mildly acidic, about like Coca-Cola and prolonged exposure over time will cause oxidation of the copper fins, increasing the potential for soot build-up which will obstruct hot flue gasses from rising, as well as serious corrosion to all ferrous materials it comes in contact with. Premature tube bundle failure is sure to follow. (Boy, I guess I better stop drinking so much Coca-Cola!)

One way to prevent condensation from forming is to keep the return water temperature at some minimum temperature, typically 110F–130F, depending on the boiler of course. This is especially important in low temperature return systems such as water source heat pump or swimming pool applications.

Primary/secondary piping provides a good method of preventing all three scenarios I have mentioned above. I always advocate to the customer to specify/purchase a boiler with a factory mounted pump to serve as the primary pump. In some cases the factory furnished pump is not large enough, thus an external field mounted pump would be required.

Figure 1 shows typical primary/secondary piping for a single boiler comfort heating system. The boiler is piped directly to the secondary loop by placing two tees 6-12" apart. A pump dedicated to the boiler and two tees, it’s that simple.

Figure 2 shows a multiple boiler system where the primary loop is piped in reverse return fashion, assuring each boiler’s inlet temperature is the same.

Additional attention must be given to low temperature return systems because they typically operate below 100F. Figure 3 shows an additional blending loop with a balancing valve within in the primary loop to assure some of the leaving water is blended back to the boiler, raising the inlet temperature to the minimum required by the boiler.

The boiler line that Dawson represents, Laars Heating Systems, has designed a header for the Laars Pennant that includes a factory mounted 3-way mixing valve specifically designed for low temperature return system (see below). The factory mounted three way valve also comes in handy in large systems where initial system warm-up is slow, thus condensation may still be a factor.

Primary/secondary piping is not the solution to every installation, but more times than not it will keep the manufacture’s representative and contractor out of trouble and the engineer and building owner happy. Primary/secondary is a simple low cost method that works well and I strongly recommend that it be considered as the starting point of every copper finned tube installation.

Cal-Code Sight Glass is no longer required for Externally Pressurized Expansion Tanks in California

2006-11-08T11:41:00.000-08:00

Subchapter 2. Boiler and Fired Pressure Vessel Safety Orders
Article 4. Installation
§763. Low-Pressure Boilers.

(g) All hot water heating systems shall be equipped with a suitable expansion tank that will be consistent with the volume, temperature, pressure, and capacity of the system as required by the Code. All such expansion tanks shall have an allowable working pressure at least equal to the maximum allowable working pressure of the boiler with which they are used, and the maximum allowable working pressure shall be stamped on a nameplate visible after installation.

All expansion tanks connected into systems having boilers designed for more than 30 psi working pressure shall be constructed, inspected, and stamped according to the Code, Section VIII, unless it can be proven to the satisfaction of the Division that the design and construction will provide equivalent safety. Expansion tanks connected into systems having boilers designed for 30 psi or less shall be designed, constructed, and stamped according to the Code, Section VIII, or according to good engineering practices with a factor of safety of at least 4.

All expansion tanks shall be fitted with either: (1) a water gage glass or other means for indicating visually the water level in the tank, or (2) a bladder-type expansion tank provided the tank is fitted with an airtight bladder inside the tank and it is provided with a means of determining the presence of air cushion in the tank. The hot water heating system shall be installed, inspected, and equipped with the required safety relief and shut-off devices in accordance with the Uniform Mechanical Code, Chapter 10, February 1997 Edition.

Note: Authority cited: Section 142.3, Labor Code. Reference: Section 142.3, Labor Code.

The above information is provided free of charge by the Department of Industrial Relations from its web site at http://www.dir.ca.gov/.

Steam Control & Condensate Drainage for Heat Exchangers

2006-06-27T11:14:00.000-07:00

General
Heat transfer units that use steam to produce hot water are known as indirect heaters. They are often shell and tube type heat exchangers and are generally referred to as converters, hot water generators, and instantaneous heaters.

The ASME Code for Unfired Pressure Vessels is the nationally recognized authority prescribing their construction for given temperatures and pressures. The term used varies with the heating medium and the manner of application. When these heaters use steam as the heat source they are usually called steam to water converters. In steam heated converters, the water to be heated circulates through the tubes and steam circulates in the shell surrounding the outside of the tubes. This results in condensate draining to the bottom of the heat exchanger shell as the steam gives up its latent heat.

Steam to Water Heat Exchangers
The operation of the shell and tube heat exchanger is as follows. Steam enters the heat exchanger shell through the top vapor opening and surrounds the outside of the tubes. As energy is transferred through the tubes it heats the water inside the tubes. The heat transfer condenses steam inside the shell forming condensate that drops to the bottom of the heat exchanger shell. The condensate flows through the bottom condensate outlet and into a steam trap.

The steam pressure in the heat exchanger shell has a direct correlation to the temperature of the condensate formed in the shell. The properties of saturated steam are such that the steam temperature varies with steam pressure (See Table 1 below). When the latent heat of vaporization is removed, the resulting condensate will be close to the saturation temperature.Depending on the system load, slight sub-cooling may occur from the bottom of the heat exchanger and the inlet piping to the steam trap.

The heat exchanger should be selected to operate at the minimum possible steam pressure. This allows the lowest possible condensate temperature to discharge from the steam trap and reduces the amount of flash steam in the return system. When heating fluids up to 200°F, the heat exchanger should be selected based on 2 psig steam pressure in the shell for the most efficient system operation.

TIP:
When heating fluids up to 200°F,
select heat exchanger
with 2 psi steam pressure in shell
for best efficiency.

This may require a slightly larger heat exchanger than one operating at higher pressure, however it will result in a smaller less expensive low pressure steam trap and a smaller steam regulating valve. The low pressure selection will also limit the maximum temperature that can occur inside the tubes, should the temperature controller fail in an open position.

It is standard practice to add a fouling factor in the heat exchanger selection. This fouling factor adds additional tube surface area to assure adequate heating after normal scale and corrosion deposits on the tube surfaces. A standard .0005 fouling factor will add 20 to 25% additional tube surface area. When the heat exchanger is new and the tubes are clean and shiny, the heat exchanger will operate at lower than design pressure even at full system load. For example, a new heat exchanger designed for 15 psi steam to heat water to 160 degrees will generally heat the full system load with 0 psi steam in the heat exchanger shell.

TIP:
Fouling factor assures adequate heating.
A .0005 fouling factor adds 20-25% addtional tube surface.

Heat Exchanger Selection
The heat exchanger should be selected for operation at the minimum pressure to provide the most efficient operation. The properties of saturated steam tables show a larger amount of the latent heat is available at low pressure. Less energy remains in the condensate reducing the flash steam losses. A reasonable guide would be to select a steam pressure that has a saturation temperature approximately 30°F higher than the required outlet temperature of the fluid being heated in the tubes. For fluid temperatures up to 200°F, 2 psig steam is recommended.

TIP:
Steam temperature should be selected 30°F
over heat exchanger outlet temperature.

When a high steam pressure source is used, the pressure should be reduced by installing a steam pressure regulating valve or by using a combination temperature pressure regulator. After selecting the heat exchanger, the next step should be planning the installation. The heat exchanger should be mounted high enough to allow gravity drainage of the condensate from the steam trap into a vented gravity return line. If a gravity return line is not available, a condensate pump should be installed. The heat exchanger should be mounted with a pitch toward the condensate outlet. A minimum 1/2 inch pitch per 10 foot length should be provided. The heat exchanger should also be located such that removal of the tube bundle is possible.

Steam Traps
The steam trap must be capable of completely draining the condensate from the heat exchanger shell under all operating conditions. On a heat exchanger using a modulating temperature regulator to heat fluids under 212°F, the steam pressure in the shell can be 0 psig. To assure condensate drainage, the steam trap must be mounted below the heat exchanger outlet tapping and it must drain by gravity into a vented condensate return unit. When possible, the trap should be located 15 inches below the heat exchanger outlet. The 15 inches static head to the trap inlet will provide 1/2 psig static inlet pressure to the trap when the shell steam pressure is at 0 psig.

TIP:
Trap should be located at least
15 inches below heat exchanger outlet.

The trap should be sized based on this 1/2 psig differential pressure. A safety factor of 1.5 times the calculated full load capacity should be used to handle unusual start up loads. A float and thermostatic trap is normally the best selection for a heat exchanger. The thermostatic element quickly vents the air from the heat exchanger shell. The modulating float element provides continuous condensate drainage equal to the system condensing rate.

Failure to provide complete condensate drainage will lead to poor temperature control and possible water hammer. Any lift in the condensate return piping after the trap discharge requires a positive pressure to develop in the heat exchanger shell to provide condensate drainage. For this to occur, condensate must back up in the heat exchanger shell until enough tube surface is covered by condensate to build a positive steam pressure. When the positive steam pressure develops to move the condensate through the steam trap and up the vertical return line, over heating can occur on the tube side of the heat exchanger due to the positive steam pressure remaining in the shell. This results in a wide range of outlet fluid temperatures from the heat exchanger.

A lift or back pressure in the steam trap return piping can flood the heat exchanger shell and cause severe water hammer as steam enters the flooded shell. The resulting water hammer can damage the steam trap, the steam regulating valve, the heat exchanger tubes and cause the gasket in the heat exchanger and trap to fail.

Trap Installation
The trap should be located below the heat exchanger shell to allow free flow of condensate into the trap. A strainer complete with a screen blow down valve should be installed ahead of the steam trap. A shut off valve should be provided in the trap discharge return line to isolate the unit for service. Unions should be provided to allow trap service or replacement. The return line from the trap discharge should be pitched into a vented condensate return unit.

Vacuum Breakers
Most steam to water heat exchangers provide a tapping in the shell to allow installation of a vacuum breaker. The vacuum breaker allows air to enter the shell when a induced vacuum occurs. Failure to install a vacuum breaker will allow the heat exchanger shell to operate at a negative pressure which may cause condensate to be held up in the shell. During light load, the heat exchanger will have a layer of steam at the top and air under the steam to provide just the right amount of heat. The vacuum breaker should be mounted on a vertical pipe 6 to 10” above the topping to provide a cooling leg. This will protect the vacuum breaker from dirt and extreme
temperatures.

TIP:
Install a vacuum breaker
on all steam converters.

Steam Regulator
The choice of the temperature regulating valve includes self contained temperature regulators, pilot operated regulators and pneumatic regulators.

The steam inlet pressure to the regulator must be higher than the required heat exchanger operating pressure to allow flow. The available steam pressure should be at least two times the heat exchanger operating pressure to provide modulation of the regulator for good temperature control. This will also provide the smallest size steam regulator.

TIP:
For good control,
take at least a 50% pressure drop
across the control valve.

The steam regulator should be sized based on the maximum lb./hr. of steam required by the heat exchanger. To properly size the regulator, the available inlet steam pressure and the heat exchanger design operating pressure must be known. The steam regulator should not be oversized. Oversizing the regulator may cause the temperature to overshoot and the regulator will hunt more than a properly sized regulator. The steam regulator is normally smaller than the connecting inlet and outlet steam piping.

Regulator Installation
A steam drip trap should be installed in the steam piping ahead of all steam regulating valves. Failure to install a drip trap will allow condensate to collect in the steam piping ahead of the regulator. As the regulator opens, the mix of condensate and steam passing through the regulator may cause water hammer that can destroy the diaphragms or bellows used to operate the regulator.

A steam strainer should also be installed ahead of the regulators to prevent dirt from entering the valve. Dirt can deposit on the valve seat and not allow it to close tight. The steam strainer should be installed with the screen pocket horizontally. Installation with the screen down, as commonly piped for water service, will allow a condensate pocket to form in the steam line. This condensate pocket can carry into the main valve and cause water hammer or sluggish operation.

Shut off valves, pressure gauges, a manual bypass and unions should be installed to allow proper servicing of the valves and strainers. When possible refer to the manufacturer’s installation manual for proper installation.

The temperature sensing bulb should be installed as close as possible to the heat exchanger outlet. It is important that the full length of the temperature sensing bulb be inserted in the system piping. Any portion of the bulb installed in a no flow area will reduce the accuracy of temperature control. When the sensing bulb is installed in a separable well, heat transfer compound must be installed between the well and the sensing bulb to aid heat transfer. The tube side of the heat exchanger should have a continuous running recirculation pump to provide continuous flow past the sensing bulb. A minimum 20% recirculation should be provided.

Pilot operated regulators with a pressure pilot require a downstream pressure sensing line. The pressure sensing line connection should be connected in a nonturbulent area downstream of the main valve; a minimum 10 pipe diameters downstream of the main valve is recommended. The steam pressure sensing connection can also be connected directly to the heat exchanger shell.

Condensate Coolers
When heat exchangers operate at high pressure, consideration should be given to the addition of a condensate cooler. The justification will be depend on the size of the heat exchanger and the actual number of hours per day the unit will be in operation.

With a condensate cooler, the discharge from the steam trap on the steam heat exchanger outlet is piped through a water-to-water heat exchanger. A second trap is then installed on the discharge of the water-to-water heat exchanger to maintain saturation pressure and prevent flashing and water hammer from occurring in the condensate cooler. A separate thermostatic trap is installed to allow direct air venting of the steam heat exchanger into the vented return line downstream of the condensate cooler.

The water-to-water heat exchanger design differs from a steam heat exchanger. The water-to-water heat exchanger has internal baffles to direct the water flow across the tubes to improve heat transfer. Water-to-water heat exchangers are externally distinguishable as the shell inlet and outlet tappings are the same size; steam heat exchangers have a large vapor opening in the top of the shell and a smaller condensate outlet in the bottom.

The fluid in the condensate cooler tubes may be the inlet water to the steam heat exchanger tubes. When the initial temperature of the fluid is too high to cool the condensate below 212°F, a separate fluid may be heated. Preheating domestic hot water or preheating boiler make up water are two possibilities.

Heat exchanger installations depending on
operating pressures and the of type of return pump.

Low Pressure (2 psig or less) with Standard Condensate Unit

Heat exchangers operating at 2 psig or less can be drained into a standard low cost floor mounted condensate returned pump.

High Pressure with Flash Tank and Low NPSH Condensate Unit

Heat exchangers operating at higher pressure require a flash tank to vent the flash steam. An elevated condensate pump unit equipped with 2-foot NPSH pumps are required to handle the condensate at saturation temperature.

High Pressure with Condensate Cooler
and Standard Condensate Unit

When it is necessary to operate a heat exchanger at high pressure (above 15 psig) a condensate cooler can be added to sub-cool the condensate below 212°F. The illustration shows the proper steam trapping for a condensate cooler.

The incoming fluid to the steam heat exchanger may be used to sub-cool if the temperature is low enough. A separate cooling source may also be used.

Low Pressure and Pressure Powered Pump

A pressure powered pump unit may also be used to return condensate. The installation shown would be used on low pressure heat exchangers. The receiver tank is vented to atmosphere on this unit.

High Pressure and Pressure Powered Pump

Heat exchangers operating at higher pressures may use a closed pressure powered pump system. The installation shown will allow condensate to discharge directly through the steam trap when the pressure on the heat exchanger is higher than the return line pressure. When the heat exchanger pressure is not sufficient, the pressure powered pump receiver will fill and operate to discharge condensate. The F&T steam trap can be sized based on the differential pressure from the pressure powered pump discharge pressure less the return line back pressure.

Condensate Drainage - Recommended Piping

2006-06-27T11:01:00.000-07:00

When using heaters with a modulating control valve, it is most important not to attempt to lift condensate from the float and thermostatic traps. Attempting to lift condensate leads to poor temperature control. Any condensate lifting will be from the bottom and the lift line will always be full of condensate. If the lift line is higher than the submerged coil, the coil will flood with condensate whenever the modulating control valves closes. This will result in temperature overshoot, as the hot condensate will transfer heat through the coil and over heat the water.

It is necessary for the designer/installer to make provisions to drain the condensate by gravity or into either an electric or pressure powered condensate pump. All Cemline heaters are
furnished with modulating control valves and as these valves close, the steam pressure in the coil is reduced to zero psig.

An example follows. A heater designed to operate on 15 psig steam is installed in an improperly designed system with 10 feet of condensate lift. A 10-foot column of water will have a pressure of 4.5 psig. The pressure drop across the valve will typically be 30%. This means the maximum pressure in the coil head is 10.5 psig. The steam condensing in the coil will further reduce the pressure in the coil to a pressure less than the 4.5 psig required to lift the condensate. The heater will stall until no more steam can enter because the back pressure from the 10-ft head of the condensate is greater than the available pressure in the coil. The valve, sensing the temperature in the heater is dropping, opens wider which allows more pressure into the coil and overcomes the 4.5 psig and lifts some of the condensate. This influx of steam will bring the heater to the set and as the control valve closes, the pressure in the coil goes to zero psig. As soon as the pressure drops below 4.5 psig, no more condensate is removed from the coil and the 10-ft head of condensate will then back feed through the trap and flood the coil with 250 degree F condensate. The coil transfers the heat from the 250 degree con condensate to the domestic water, resulting in an over temperature condition in the heater.

It is the responsibility of the designer/installer to properly drain the condensate. Cemline Corporation is represented by trained, expert sales agents in steam systems. Please contact the factory or one of our sales agents for assistance.

Remember, Do not attempt to lift the condensate from a water heater or Unfired Steam Generator.

SOURCE: www.cemline.com/liftcond.asp
posted by permission of Cemline

Energy Efficient Selection of Steam-to-Liquid Heat Transfer Systems

2006-06-27T08:41:00.000-07:00

System Solutions from ITT Residential & Commercial Water Group ...

Engineered Steam-to-Liquid Heat Transfer Systems

Traditionally, the engineering of steam-to-liquid heat transfer systems has been done by using the maximum available steam pressure to select the smallest and least expensive heat exchanger. The control valve, steam trap and other ancillary equipment would then be selected using the operating conditions of this heat exchanger. While this method works, there is a better way.

It is now possible to reduce both initial equipment costs and operating costs. This is done by selecting the components of the entire system at the same time. Your ITT R&CWG Representative can now optimize your steam-to-liquid heat transfer systems with each component working together in a manner that optimizes the system as a whole.

Optimizing The System As A Whole:
Let’s take an example. Say we have 90 psig steam pressure available for a heat transfer application and use a low pressure drop control valve—10 psig ÆP—to select a heat exchanger with minimal size and surface area.

At a high steam operating pressure, there is less latent heat available for the heat exchanger, while downstream of the heat
exchanger, a higher percentage of flash steam is produced. This flash steam results from re-evaporation of the condensate when it is exposed to lower pressure or vented return lines.

In this type of system, a flash tank may be required to handle this re-evaporation, and high temperature condensate return systems may be required to pump the condensate discharging from the flash tank.

The net result of using high steam pressure at the heat exchanger is flash steam losses causing wasted energy ($), lower system efficiency ($$), and extra equipment to handle higher temperature returns ($$$). But we did save initial costs by selecting the smallest heat exchanger....? Not necessarily!

When we re-select the same application but use a steam pressure regulator to reduce the steam pressure at the heat exchanger to 5 psig, the size of the heat exchanger will increase. However, the heat exchanger cost is only part of the system cost.

By reducing steam pressure at the heat exchanger, the steam control valve may decrease in size, lowering its initial cost. Also, the steam trap may become lower in cost because of the reduced operating pressure. By lowering your steam pressure, the result may be a net savings on your initial equipment cost.

But there’s more. By using the reduced steam pressure of 5 psig, we have more latent heat available in the heat exchanger, and thus a lower percentage of flash steam. This may allow for elimination of the flash tank and conversion of the high temperature condensate return unit to a lower cost, conventional condensate return unit.

The net result: lower initial costs—and most importantly, lower operating costs yielding annual cost savings that greatly increase payback. See the example below for a typical
comparison.

Your ITT R&CWG Representative can select an engineered steam-to-liquid heat transfer system that operates efficiently, effectively, and possibly at a lower initial cost because they fully understand all of the critical components involved.

Your ITT R&CWG Representative handles Bell & Gossett heat exchangers, pumps and air control equipment; Hoffman steam control valves, regulators, safety valves, steam traps, vacuum breakers and strainers; and Domestic condensate return and boiler feed pumps. Count on your representative for the right combination of training, expertise and tools to pull it all
together in a system that suits your application exactly, using the award-winning ESP-PLUS system evaluation and equipment selection program.

Using our example steam pressures, let’s assume our heat exchanger requires 3,000 lbs/hr of steam, operates 14 hours a day for 250 days per year and the condensate goes to a vented condensate receiver at 0 psig atmospheric pressure. We will assume our steam costs $6.50 per 1000 lbs.

NET ANUAL SAVINGS FROM REDUCING PRESSURE
1,071,000 lbs of steam or $6,961.00

Specifications For:
Furnish and install according to manufacturer’s instructions, one ITT Fluid Handling energy efficient steam-to-liquid heat transfer component system, which shall have the capacity to heat _________ GPM of _________ (fluid) from _________ °F (temperature) to _________ °F (temperature) when supplied with _________ psig saturated (or degrees superheated) steam to the steam regulator. The heat exchanger shall be sized for maximum _________psig inlet pressure. The system is to have a maximum of _________ % flash steam. Energy loss calculations shall be furnished to the engineer for approval and shall include annual dollar operating costs at design conditions.

The energy efficient steam heat exchanger component system shall be piped in the field with all necessary valves, pipe and fittings, according to plans and specifications and shall consist of the following major components:

Hoffman Series 2000/1140 (pneumatic or self-contained) modulating steam control valve.
Hoffman F&T trap and “Y” strainer for the drip leg.
Bell & Gossett “SU” type heat exchanger with _________ fouling factor, ASME constructed with signed U-1 form per heat exchanger specification.
Hoffman vacuum breaker for the heat exchanger.
Hoffman "Y" Strainer and F&T Trap for the heat exchanger. (F&T Trap size based on 0.5 psig differential pressure with 1.5 min. safety factor.) Trap installed a minimum of 15" below the heat exchanger.
Optional components
• _________ steam safety relief valve
• _________ gauges, high pressure cocks and pigtails
• _________ thermometers
• _________ Bell & Gossett circulating pumps for liquid (primary/secondary) system with flow measuring and balancing valves
• _________ ASME relief valve for system liquid.
• _________ factory piped and frame mounted construction or individual components shall be specified.
__________ Duplex Domestic/Hoffman condensate return unit with accessories.

Single source system responsibility requires all major components to be supplied by a single source manufacturer.

©1992 ITT Corporation Printed in U.S.A. 11/92
Posted by permission from ITT Residential & Commercial Water Group
**NOTE: This article was originally published under the name ITT Fluid Handling. Where appropriate, the name has been edited to ITT Residential Commerical & Water Group, reflecting the current name of this subdivision of ITT Corporation.

Pump Start Up & Seal Failure

2006-06-26T16:54:00.000-07:00

I’ve started my pump, and now it’s leaking . . .
Or even worse “I’ve started my pumps, and they are all leaking.” The next question is “when can this be fixed?” but the real question should be “why did the seal fail?” There are a variety conditions that can lead to seal failure, and the majority are preventable.

First, lets examine the conditions that are not preventable in the field. This could include a bad seal. It does happen, just not very often. Mean failure rates of the mechanical seals used in Bell & Gossett pumps are extremely low. Another possibility is an improperly constructed seal chamber in the pump itself. Out of thousands of pumps sold over the last few years, I am aware of one occasion wherein this was the case. While material failure is always a possibility, this is an extremely rare occurrence in our company’s experience.

The article will address typical start-up conditions and cleaning procedures for hydronic systems, basic water treatment of these systems, and the effect that the procedures and treatment will have on the pumps mechanical seals.

Start-up conditions, procedures and water treatment
Changes in new construction, particularly with regards to scheduling, have virtually guaranteed a pump seal failure soon after start-up. The rapid pace of new construction does not allow for use of flush pumps. Flush pumps require a pipe side stream with the appropriate valving to operate, and this does take time to construct and then remove. Please take a moment to consider when you last saw a flush pump being used on new construction, retrofit or T.I. work? When did you last see the use of a flush pump specified? In some informal surveying, five years was the most popular answer, and many people answered ten years.

This means that the system pumps have been used for conditions other than what they were designed for, namely system flush and cleaning. These conditions, and the accompanying fluid characteristics, present significant problems for pump seals.

To begin, one of the first things that will occur in a new system is a pressure test. Water is introduced into the system piping through the pump. This water is dirty, with high levels of sediment and construction debris, and is passed over the pump seal. This fluid contains high levels of dissolved and suspended solids. Mechanical seals are not tolerant of dissolved solid levels above 1000 ppm or suspended solids above 20 ppm, but we will address that in greater detail later.

System cleaning and pipe treatment are typically done with alkaline cleaners. Alkaline cleaning uses chemicals to raise the pH level in the system to somewhere between 9.5 and 10.5, with this level being most beneficial to the pipe.
A quick review of pH –

By definition pH is the measure of free hydrogen activity in water, and can be expressed as : pH = -log[H+]. A more practical statement for our purposes is that pH is the measure of acidity or alkalinity. Measured on a scale of 0-14, solutions with a pH of less than 7.0 are acids while solutions with a pH of greater than 7.0 are bases (alkaline). This is a logarithmic scale, and as such each unit represents a 10-fold increase or decrease relative to the next unit. For example, seawater with a pH of 8 is 10 times more alkaline than fresh, clean water, with a pH of 7. This is noted to illustrate how a pH increase or decrease of 1 or 1.5 is very significant with regards to the fluid properties.

Corrosion levels increase in copper and steel as pH falls below 7. This is acidic, and should make sense on an intuitive level. Since we don’t fill our systems with acids, this is a non-factor. Corrosion occurs readily at pH levels between 7 and 8.9, but falls off dramatically at 9, so the alkaline side of water treatment is where our attention should be focused.

The pH target is 9, with significant corrosion consequences if the system pH should fall below this number. As such, the most common treatment pH target range is between 9.5 and 10.5. Spikes in pH can occur, but for the most part this is not harmful to the piping. It can however be extremely harmful to the pump seal.

Mechanical Seal Tolerances
Internally flushed mechanical seals include a stationary face and a rotating element. The materials from which the seal are made have specific tolerance ranges with regards to temperature, pH and particulate levels. A standard seal, which will have carbon stationary element and a ceramic rotating element, has the following limitations:

· Maximum temperature: 225ºF
· pH range: 7 – 9
· Total dissolved solids: 1000 ppm
· Total suspended solids: 20 ppm

Based on what has already been outlined regarding start-up and water treatment, it can be reasonably assumed that the seals have already been subjected to conditions that exceed there limitations, and probably for an extended period. The water chemistry during cleaning will quite likely have a negative effect on the seal a few months later.

At elevated pH levels, the binder in the ceramic material is leached out, leaving the surface slightly softened and porous. As this element rotates, pieces of the ceramic embed in the carbon face and act like a grinding wheel on the weakened mating face. This will lead to seal leakage. Once the pH spike occurs, the damage is done regardless of subsequent adjustments.
The flushing water that cools and lubricates the seal evaporates due to the frictional heat generated between the seal faces, leaving behind whatever chemicals are dissolved in the water. For this reason, chemical concentrations of dissolved solids above 1000 ppm will act as an abrasive on the seal faces. This will erode the face, and the seal will leak. Similarly, with suspended solids, they can work between the faces and be trapped by the evaporation of the system fluid. The seal will be damaged as consequence.

If high pH levels and elevated concentrations of solids are typical of the installation, other seal materials are available as an option. Like anything, this will come at a price. These seals cost more, are more susceptible to thermal shock, and will crack quite readily if run dry even for a moment. The most common alternative seal would be of EPR/tungsten carbide construction. EPR, ethylene propylene rubber elastomer, is the stationary element. Tungsten carbide is the rotating element. This seal will tolerate pH levels of up to 11 and temperatures to 250ºF. Since the materials are harder, they will handle solids more effectively, but the exact amount is not published.

While we have not discussed temperature in depth, this is not typically an issue on standard hydronic heating systems, but may be a factor in some cogeneration facilities. Please bear in mind that at elevated temperatures, the effects of high pH levels become more pronounced, as well as the evaporation rate of the fluid between the seal faces. These seals will fail more rapidly as a result.

Conclusions and Solutions
How to cure the problem? The most cost effective solution is to simply change the seal after substantial completion of the project. While there are options regarding water treatment that will not elevate pH levels, they come at a steep price increase. Given our competitive market place, this is unlikely to be a regular alternative. Monitoring the water chemistry during start-up and system cleaning will provide an early indicator of seal replacement necessity. Please bear in mind that because the system is corrected does not mean that damage has not been done.

This is a topic about which very lengthy articles have been written, and it is impossible to discuss all the factors of seal failure at any length in this newsletter. I would welcome you comment and input on this topic.

by Mike Caffrey
Branch Manager - San Diego
Dawson Co.

References
· B&G Bulletin 4976 “Mechanical Seal Selection Guidelines”
· The Burgmann Dictionary at http://www.burgmann.com/
· The Chesterton web site at http://www.chesterton.com/
· John Crane Company, article “Avoiding Premature Seal Failure”

Thanks to Roy Ahlgren, Director of Training and Education, Bell & Gossett.
Thanks to Mark Pondel, Manager of Field Service, Bell & Gossett.
With Special Thanks to Michael Burns, General Manager, Aquatec.

Simple Yet Sophisticated

2006-06-20T13:24:00.000-07:00

It isn’t often that a new product is introduced and one could argue that the method used in its development is far more interesting than the product itself. How many of us wish we were asked to give our input in the early design phase of the products we select and specify on a daily basis? For example, I just purchased a Toyota Sequoia and found that it really has some nice features, like both the driver and passenger front windows will automatically roll up or down when the driver’s side buttons are fully depressed. But does it make any sense that if my four year old wants the dome light on, I have to take my eyes off the road and reach behind my head and flip the switch with my hand? Every car I’ve ever owned had an easily accessible switch on the dash. I curse those Toyota engineers every time! Research and development is the basis for any new innovation that comes along, but how innovative can a good old fashion centrifugal pump be? Well, I can tell you that Bell & Gossett used some very simple-sophisticated techniques in the development of their new VSX double suction pumps.

They simply started their research by asking the customer what features and benefits were important to them. Utilizing a technique known as The Voice of the Customer, Bell & Gossett sent their centrifugal pump Product Line Manager, Steve Schmidt on an extensive information gathering assignment all over the country and the world for that matter.

Some of you might have even been part of the process about two years ago when Steve conducted in-depth interviews with more than 300 consulting/specifying engineers, contractors, building managers, wholesalers, distributors and other industry experts. He focused on learning about their business, their long and short term goals, and any problems or concerns that might have developed over the years. As a result, he did a lot of listening and little talking to flush out the true requirements of the customer, compiling a list of 600 legitimate "needs". Incredibly each "need" was listed on a "Post-It-Note" and displayed in his office where he could be constantly reminded of the task at hand. These "needs" were then broken down into groups and entered into an elaborate concept matrix that ranked the importance of each, in order to help with their implementation in the design phase of the project.

They sophisticatedly developed the new line of pumps by utilizing Computational Fluid Dynamics (CFD). In the past, pumps had been "improved" by either "tweaking" an existing impeller design or modifying the volute to enhance performance. Until now, a completely new double suction pump had not been developed by Bell & Gossett in over forty years. By implementing CFD technology, impellers were hydraulically matched with pump volutes, allowing pump engineers to concentrate on the desired "sweet spot" the pump curve was to incorporate.

Determining the curve’s "sweet spot" required taking a step back and looking at how pumps are being applied today and what prime movers are being offered in today’s market. Hydraulic targets were identified for several market applications and design parameters were then entered into the computer. This process was no simple task and required a bank of 18 computers working up to 30 hours to solve equations for just one set of hydraulics! During the course of the design process, liquid flow was electronically visualized and pressures generated along the pump surfaces were studied for optimization. When a design concept was narrowed down, and prior to moving to large scale CFD analyses, rapid prototype parts were created to either validate or disprove the computer model. Foam blocks of the pump parts were then created and served as stepping stones until a desirable design was finally developed.

The result of this tedious process is a product that will offer exceptional reliability and versatility to the HVAC industry. It truly is an exciting time to be associated with this industry. By the way, does anyone have Dr. Phil’s number? I’m really hung up on this dome light thing. HAPPY PUMPING!