Approach

In the discussion that follows, we assume that the preliminary steps outlined in Chapter 4 have been accomplished, including the initiation and planning phase and the historical review. For the process under consideration, the basic facts are presumed known, either by energy audits, calculations, or measurements. The basic items required are the various energy forms (fuel, electricity, steam, compressed air, etc.), quantities (liters, kilowatt hours, kilograms per hour), and states (residual oil, 480 V, three phase, 235°C at 1.5 MN/m² pressure, etc.), both entering and leaving the process. The process itself must be understood so the need for these various energy forms can be evaluated.

With this as background, our goal now is to determine how to manage process energy in the most efficient way and still get the job done.

The field of process energy use is so broad that it would take volumes to cover it in a comprehensive manner. In fact, entire books have been written about specific processes, e.g., energy use in foundries or plating operations. Therefore, our approach will be to review the fundamental aspects of some major areas, providing illustrative examples, and note sources of additional information where appropriate.

Potential Savings and Benchmarking

There have been a number of international studies to evaluate “best practices” in major segments of industrial energy use.^5,6 These studies examine specific energy use in various industrial applications (such as glass manufacture, petroleum refining, and iron and steel mills) at different plants in different countries. From these data, the most efficient plants are used to establish a “best practice,” or best available technology for each end use. Then, by comparing the best practice to the average specific energy required for that end use, an estimate of the savings possible by deploying the best practice can be made. Compilations of such data provide an “efficiency benchmark” for individual companies to compare their process performance against others in the same industry.

These studies indicate that savings in the range of 20–26% of current industrial energy use are possible.

Today there are many tools available to the energy manager to analyze or evaluate specific industrial processes. An example is the U.S. Department of Energy’s Advanced Manufacturing Office (AMO), formerly the Office of Industrial Technology. This office provides many information sources, case studies, and tools for improving industrial process energy efficiency. Some of the specific types of equipment addressed are boilers and steam systems, air compressors, motors, fans, pumps, and process heat technologies. The website is www.energy.gov/eere/amo/information-resources. Besides free tools that help energy managers assess the efficiency of various industrial processes, the website has links to other sources of information. Another example is a website hosted by the European Community at http://iet.jrc.ec.europa.eu/energyefficiency. Technical guides, training courses, manuals, and other documents can be accessed at this site in various languages.

Caution must be exercised in applying any of the principles that follow on a general basis. Process energy use tends to be site- and process-specific. What is appropriate for one process and one plant may not be appropriate for a similar process in another plant, depending on fuel types and costs, weather patterns, variations in material properties, and a host of other variables. Judgment, careful analysis, and sound engineering are prerequisite for any successful energy management program.

Heat Transfer Mechanisms

To reduce heat losses, it is important to recognize that heat is transferred (and lost) through three mechanisms: conduction, convection, and radiation. Refer to Figure 11.1 in connection with the following equations:

Conduction. Consider a section of material thickness $L$, area $A$, thermal conductivity $k$, and exposed to temperature ${\mathit{T}}_1$ at one side and $T_2$ at the other, such that: $\Delta T_{1-2}={\mathit{T}}_1–T_2$. The heat flow conducted through this material is

$\dot{Q}=kA[\mathrm{d}T/dx]\mathrm{W}$ [11.1]

[11.1]

where

(see Table 12.2 for typical values of thermal conductivity, k.)

Equation 11.1 can be rewritten as

$\dot{Q}=kA\mathrm{\Delta }{T}_{1-2}/L\mathrm{W}$ [11.2]

[11.2]

Thus, all other parameters being equal, the driving force for conduction heat losses is thermal conductivity and $\Delta T$.

Convection. Convective heat losses are described by a similar equation:

$\dot{Q}=h_{c}A\mathrm{\Delta }{T}_{2-3}\mathrm{W}$ [11.3]

[11.3]

where

Radiation. The third method for heat loss is radiation, given by this equation for a relatively small body in a large enclosure or space:

$\dot{Q}~A\mathit{\epsilon }\sigma (T_2^4-T_4^4)W$ [11.4]

[11.4]

where

Reducing heat loss. With Equations 11.1–11.4, it is possible to see how to reduce heat losses and to evaluate the benefits from doing so.

For a given situation, the driving force for conduction heat loss is the thermal conductivity of an insulating material and the temperature difference across its thickness. In the case of convection, the loss depends on the temperature difference between the surface of a heated material and the ambient temperature, as well as the transfer area (in m²) and the convection heat transfer coefficient (in W/m² K). Radiation heat loss is influenced by the area radiating heat, its emissivity, its temperature, and the temperature of its surroundings, where the temperatures are in Kelvin raised to the fourth power. At low temperatures, conduction and/or convection heat losses typically will dominate, while at high temperatures radiation becomes important.

To reduce conduction losses, either more or better insulation is required to increase insulation thickness and decrease thermal conductivity. To reduce convection losses, e.g., from a tank containing a hot fluid, an insulating cover should be added to reduce the convection heat transfer coefficient and the temperature difference. This would also reduce radiation losses. Radiation losses can also be reduced by adding radiation shields.

Combustion

Combustion is the starting point for almost all non-electric forms of process heat except nuclear or solar, neither of which is currently of great importance for manufacturing processes. In direct-fired operations, process heating is caused by radiant energy or is transferred by the products of combustion. In indirectly heated operations, the heat of combustion is first transferred to a working fluid (such as steam) that does the heating or that produces the electricity that does the heating. Thus, combustion is the first area to examine when considering energy management options.

Fuels can be rated according to their heating value, which is a measure of the available work they possess (see Appendices for typical values). In theory, nearly all the fuel’s potential to do work should be recoverable; however, due to practical limitations, there are losses due to the thermodynamics of combustion. The only current prospect for avoiding these losses is through the use of fuel cells that extract fuel energy without combustion. Size, economic, and lifetime constraints limit fuel cell applications for process energy use at present.

The next loss of available work in the combustion process results from the decrease in temperature from the flame temperature (on the order of 2000°C) to the process temperature. In the case of a steam cycle, the process temperature might be in the range of 200–500°C. In concept, efficiency could be improved if hot combustion products could be utilized at as high a temperature as possible, subject to the limitations imposed by useful engineering materials. Turbine technology permits temperatures of about 1100°C; this suggests that one improvement would be to use a topping cycle involving a direct-fired turbine to extract work. The turbine exhaust (typically at 600–650°C) could then be applied to a waste heat boiler to generate steam. The usual approach, to generate steam directly from fuel combustion, requires a drop in temperature from 2000°C to about 550°C due to material and corrosion limitations. This leads to a decrease in available work before the energy is even applied to a process.

While a combined cycle probably will not be feasible in an existing plant, there are several other EMOs that can be considered to improve combustion efficiency:

We discuss these opportunities below as well as in Chapter 8.

Fuel Quality and Excess Air

Fuel should be analyzed periodically to determine that it still meets requirements. For example, excess fines (small coal particles) in coal could lead to improper operation, including too much excess air, which in turn leads to energy losses up the stack. Excess air is best controlled by monitoring flue gas oxygen content, either by periodic measurements, or better yet, by permanently installed instrumentation. The proper setting for the type of fuel can be established by testing. Fuel savings of 2–19% are possible from correcting excess air. Reviewing the combustion process helps explains the science behind the excess air issue.

Combustion of carbonaceous fuels requires oxygen. Oxygen is normally supplied from atmospheric air, which has the following standard properties:

The combustion process may be outlined schematically as shown in Figure 11.2.

The following notation applies:

Subscripts have the following meanings:

If the theoretical amount of oxygen required for complete combustion is calculated, one obtains (for any fuel):

$w_{\mathrm{O}}=2.67w_{\mathrm{C}}+8(w_{\mathrm{H}}+w_{\mathrm{O}}/8)+w_{\mathrm{S}}\mathrm{kg}\text{oxygen}/\text{kg}\mathrm{fuel}$ [11.5]

[11.5]

where

The weight fractions for the various constituents in the fuel would normally be provided by the supplier or obtained by ultimate fuel analysis.

Using the properties of air, Equation 11.5 can be converted to an air basis:

$w_{\mathrm{a}}=11.5w_{\mathrm{C}}+34.5(w_{\mathrm{H}}+w_{\mathrm{O}}/8)+4.31w_{\mathrm{S}}\mathrm{kg}\text{air}/\text{kg}\mathrm{fuel}$ [11.6]

[11.6]

This is obtained by recognizing that O₂ is 23.2% (by weight) of air.

Combustion air calculations can be made on a mass basis, volume basis, or molar basis. The data in Table 11.2 are useful for such calculations.

Using the data in Table 11.2 and Equation 11.6 will provide the minimum amount of air (or oxygen) required for complete combustion of a given fuel. In reality, more air than this must be supplied, since some excess air is required for complete combustion. The amount of excess air depends on the particular system and fuel type. For natural gas, an efficient boiler might use about 15% excess air or roughly 19 kg per kg of fuel. Since this is 75.5% by weight N₂, it means that for every kilogram of fuel, there is nearly 15 kg of N₂ flowing through the boiler, being heated from ambient temperature to combustion temperature, and eventually carrying heat up the stack without bringing any benefit to the process!

Complete combustion requires that the following conditions be satisfied:

Too much airflow (excess air) leads to fuel waste by increasing the energy loss up the stack, while not enough air will lead to incomplete combustion.

Figure 11.3 schematically illustrates the relationship between heat loss and excess air. As the figure indicates, when the amount of excess air increases, the stack losses increase. For typical industrial type boilers, the correct amount of excess air is best determined by analyzing flue gas oxygen or carbon dioxide concentrations. Table 11.3 shows typical data for a natural gas-fired boiler.

On well-designed natural gas-fired systems, an excess air level of 10–15% is attainable. Higher amounts (e.g., 45%) will reduce boiler efficiency by 1–4%, depending on the stack temperature (Table 11.3).

Combustion Monitoring Equipment

On-going monitoring of combustion parameters will help identify and correct issues with the air-to-fuel ratio and will allow for assessment of the combustion efficiency. Measurements include the amount of O₂, CO₂, CO, and unburned hydrocarbons in the flue gas and combustion air and flue gas temperatures.

The O₂ measurements indicate levels of excess air and the CO measurements indicate incomplete combustion. The proper ratio of air and fuel can be maintained manually by an operator if care is taken. However, electronic combustion analyzers and controls can automatically maintain efficient operation. If not already installed, such a system would likely pay for itself in a very short period of time, even by reducing fuel consumption by as little as 5%.

The combustion products (O₂ and/or CO₂) and net exhaust gas temperature can be used to assess the combustion efficiency for a given type of fuel and burner using manufacturer’s literature or combustion efficiency curves and tables in handbooks. The net exhaust gas temperature is defined as the flue gas temperature minus the combustion air temperature (ambient temperature) if no air preheater or economizer is installed. If there is an air preheater or economizer, determine the temperature at its outlet. The combustion air temperature should be measured at the forced draft fan inlet.

Burner Maintenance Procedures

Proper maintenance and operation of burners, stokers, and other combustion equipment is important. For example, deposits on burners can reduce burner efficiency and, in oil-fired systems, the temperature at which oil is delivered to the burners contributes to proper atomization and combustion.

Waste Heat for Preheating Combustion Air and Feedwater

Efficiency also improves with increasing combustion air and feedwater temperature, since less energy has to be supplied to heat them. In some instances, waste heat from stacks can be recovered and used to heat the incoming air. Preheating combustion air from ambient temperatures up to several hundred degrees centigrade can save as much as 5–10% of fuel consumption. It may be feasible to use heat recovery to preheat feedwater.

Steam Generation and Distribution

Improvements in steam systems fall into two broad categories. The first applies to the steam system itself, while the second applies to the end uses of steam. For steam systems, consider:

For steam end uses such as process heating, operating steam driven equipment, or heating buildings, some EMOs include the following:

The subsections below summarize these boiler and steam system EMOs. See Chapter 8 for additional opportunities.

Steam leaks and defective steam traps. Small steam leaks resulting from defective traps or valves can lead to surprisingly large energy waste, as shown in Figure 11.4. These types of leaks may also cause significant loss of water, which is an increasingly expensive and scarce commodity particularly in drier climates.

Sizing and maintenance of distribution systems. As steam loads change over time, the distribution system may be used for purposes other than those for which it was originally designed. If the lines are too small, pressure drops may be excessive. If they are too large—i.e., supplying small loads—the losses may be disproportionately large. In addition, the importance of proper steam line insulation cannot be overemphasized. Figure 11.5 illustrates the magnitude of heat loss from uninsulated lines.

Management of condensate return. In many processes, after use, steam is condensed and the resulting fluid is returned to the feedwater supply to the boiler. Otherwise, lost feedwater must be replaced and heated to provide a constant supply of steam. Condensate return saves energy in a number of ways. Not only is less energy needed to heat feedwater, but also less energy will be expended in pumping and chemically treating makeup water. Condensate return also saves water.

Example. Benefit of condensate return. Consider a boiler producing 25,000 kg/h steam. The feedwater temperature is 15°C and the condensate temperature is 50°C. With no condensate return, the energy input to heat feedwater is 8.9 GJ/h. With 100% condensate return, the energy input to heat feedwater is 5.2 GJ/h or a savings of 3.7 GJ/h. Suppose the fuel is oil rated at 5.8 GJ/bbl and the boiler thermal efficiency is 70%. The theoretical savings is nearly one barrel of fuel per hour. In reality, this is an upper limit, since 100% condensate return is unlikely.

In addition to using condensate return to recover energy, if there is some other source of waste heat, a heat exchanger can be provided to preheat feedwater to the extent possible.

Maintenance of steam tracing systems. Steam tracing systems (used to heat pipes, tanks, etc.) can waste energy if not maintained properly. An obvious caution is to turn them off when not needed.

Steam supply pressure. Steam usually will be supplied at the pressure of the highest load. In industrial operations, this may be 1–3 MN/m² if steam turbine driven equipment is used. Steam is also used to transfer fluids by means of steam jets. For heating purposes, pressures are more typically in the range of 0.1–0.3 MN/m². If most of the loads are at lower pressure, steam should not be supplied at high pressure. Instead, a substitute energy source could be found for high-pressure loads.

Efficient alternatives to steam systems. In some cases, electric motors or some alternative drive system will be more efficient than steam turbines. This would be true in cases where the steam loads are small or are distant from the steam plant. Electricity might even be better for heating in a particular case, if line losses and the effect of more precise temperature control are included.

Cascading of steam use. Sometimes steam must be provided at high pressure. Instead of using pressure-reducing valves, look for opportunities to cascade steam use, within acceptable limits on steam pressure, temperature, and quality. For example, high-pressure steam can first be expanded through a noncondensing turbine to do useful work, and then the turbine exhaust steam can be used for process or building heating.

Hot Water and Water Pumping

The heating and transport of water and other fluids requires utilization of energy to raise the water temperature, to make up heat losses from pipelines and storage tanks, to pressurize fluids, and to overcome the resistance to fluid flow of pipelines. There are EMOs associated with each of these energy requirements.

Water heating. The energy input necessary to raise the temperature of water (or another liquid) is given by

$E_{in}=m{c}_p(T_f-T_i)/\eta \mathrm{J}$ [11.7]

[11.7]

where

This can also be expressed as a rate in J/s by using a mass flow rate in kg/s.

Therefore, for a given quantity of water, the main opportunities to save energy in the water heating process are to increase the efficiency of the heating technology or lower the required temperature rise. There are several types of technology advancements that heat water more efficiently, including condensing boilers, heat pump water heaters, solar water heaters, or tankless water heaters. There may also be opportunities to lower the temperature rise by using waste heat recovery to preheat the feedwater or to only heat water to the temperature absolutely necessary for the given application.

Water heaters range from small 50 gallon storage units found in residences to large boiler-fired systems found in commercial establishments and industry. Today, for residential and small commercial applications, tankless water heaters are available with flow rates of up to 10 gpm. Tankless water heaters have an “instant on” feature that avoids standby losses typical of storage tank units (see Figure 11.6).

Example. A seafood restaurant used hot water for cleaning, rest rooms, and to preheat water used in large 60 and 80 gallon kettles. (Preheated water would come to boiling temperature more quickly—important during the rush hour.) The conventional 150 gallon water heater proved to be unreliable so the restaurant installed three tankless units. Two served the restaurant loads and one was dedicated to the large kettles in the kitchen. Other restaurant applications have reported gas savings with tankless water heaters of as much as 40–50% compared to storage units.

When considering the hot water system as a whole, it is also important to reduce losses. Losses due to heat transfer to the environment can be reduced by the following steps:

A major source of loss in hot water systems is the standby losses that occur when tanks are continuously maintained at elevated temperatures or they are not well insulated. Water heaters should be maintained at the minimum required temperature. Lowering the water temperature setting 5–6°C will reduce annual fuel cost by 4–5%. For smaller applications, consider investing in high-quality units with extra insulation and anode protection to prevent tank corrosion. Water heater “jackets” or “blankets” are also available to reduce heat loss in existing systems. In addition, insulate all hot water pipes to prevent heat loss and install insulating lids on any open tanks containing hot water (or other liquids) to prevent both heat loss and liquid losses. Additional benefits of preventing these types of heat losses in conditioned spaces are reduced load on HVAC equipment and improved occupant comfort and safety.

As noted previously, heat recovery is another useful technique that can be used in industry. Waste process heat (e.g., refrigeration compressor cooling water) can often be reclaimed to heat or preheat water.

Water pumping. The power required to move an incompressible fluid through a piping system is given by

This may be written as

$pp=\dot{m}\left(\mathrm{\Delta }P/{\rho }_i\right)\mathrm{W}$ [11.8]

[11.8]

where

Or, in more convenient form, the power expended in pumping a fluid is related to the volume pumped:

$pp=\dot{V}\mathrm{\Delta }P\mathrm{W}$ [11.9]

[11.9]

where now

We also derived Equation 11.9 in Chapter 8 using the concept of head developed by a pump (see Equation 8.4).

The pressure drop depends on the system (pipes, channels, orifices, bends, etc.) and must be determined for each case. In the case of round pipes, it is given by the following equation:

$\mathrm{\Delta }P=\frac{f\overline{\rho }{v}^{2}L}{2D_e}\mathrm{N}/{\mathrm{m}}^2$ [11.10]

[11.10]

where

$f$ = friction factor, dimensionless

$\overline{\rho }$ = average density, kg/m³

$v$ = velocity, m/s

$L$ = length of pipe, m

$D_e$ = pipe diameter, m
The losses through bends, enlargements and contractions, and valves and fittings are similarly proportional to velocity squared.
For incompressible fluids, the volume flow rate $\dot{V}$ is related to the velocity $v$ by

$\dot{V}=Av{\mathrm{m}}^3/\mathrm{s}$ [11.11]

[11.11]

where

A = area, m²

solving

$v^2={\dot{V}}^2/A^2{\mathrm{m}}^2/{\mathrm{s}}^2$ [11.12]

[11.12]

Substituting these results in the equation for pumping power, we obtain

$pp=\frac{{\dot{V}}^{3}f\overline{\rho }L}{2A^2{D}_e}\mathrm{W}$ [11.13]

[11.13]

This is a very important and fundamental result. It indicates that once a pipe size is fixed, the power required for pumping increases as the cube of the volumetric flow rate. The practical significance is that if the flow rate is reduced 20%, the pumping power is cut in half. Conversely, for a fixed flow rate, the pumping power decreases in proportion to the fifth power of the diameter.

For a wide range of Reynolds numbers, f=0.022 for clean, commercial steel pipe. This holds true in either American/British or SI units, since f is a dimensionless ratio having units of length/length. If American/British units are used, a conversion factor (g=32.2 ft/s²) must be inserted in the denominator of the pumping power equation.

Pressure drops for complete piping systems require determination of losses through all components. This analysis is most easily performed with computer simulation models. Alternatively, handbooks provide data that convert flow resistance of typical pipe fittings to equivalent lengths of pipe.

Major pumping system EMOs can be summarized as follows:

It is a common practice to provide water at the pressure required to meet the highest pressure load. An alternate approach that will sometimes save energy is to provide water at the pressure needed by most of the load and provide booster pumps for high-pressure loads.

For additional energy management resources for pump systems, refer to the U.S. Department of Energy’s webpage: http://www.energy.gov/eere/amo/pump-systems.

Direct- and Indirect-Fired Furnaces and Ovens

Furnaces, ovens, dryers, and kilns have many similar attributes in terms of their basic heating functions as well as in their opportunities for energy management. For simplicity, we refer to this family of technologies collectively as furnaces and ovens. Direct-fired furnaces and ovens rely on heating directly by the products of combustion (fuel-fired) or by electric heating elements. Indirect-fired furnaces and ovens involve some type of heat exchanger for transferring heat from the heat source (e.g., steam or hot water) to the process. The majority of process heat is from combustible sources, although the share of electric process heat systems is likely to increase because of advantages related to ease of control, no emissions at the point of use, safety, smaller footprint, and widespread applicability.^7,8

There are three basic approaches for managing efficient use of process heat (Table 11.4):

The discussion below explains each type of opportunity in greater detail. (Refer also to Chapter 8 for the special case of heat recovery in HVAC systems.)

Reduce heat losses. Earlier in this chapter, we described three modes of heat transfer—conduction, convection, and radiation—and explained general ways to minimize heat loss through these mechanisms. Those general approaches apply to furnaces and ovens. Reduction of losses can be accomplished by insulation, reflective shielding, use of curtains, other methods to contain heat and isolate it from conditioned spaces, reduced temperatures, maintenance of heat transfer services, equipment shutdown when not in use, or improved designs. Generally, the economically optimal amount of insulation or other heat containment methods for furnaces and ovens (and associated piping or ductwork) depends on the temperature range and fuel costs, so there is no simple rule of thumb. Each case will usually have to be analyzed on its own merits.

Example. Improve insulation. A useful approach is to consider an analysis on a unit area basis. Consider an electrically heated oven or tank that loses 3.5 kW/m² from its uninsulated walls under normal operating conditions. Analysis indicates that this could be reduced to 0.7 kW/m² by adding 2.5 cm of Rockwool insulation. Energy costs are $28/GJ and annual operation is 2000 hours per year. Management requires a 1-year payback. The analysis indicates:

Unit energy loss without insulation:

$(3.5{\text{kW}/\mathrm{m}}^2)(2000\mathrm{h}/\text{yr})(3.6\times {10}^6\mathrm{J}/\text{kWh})=25.2{\text{GJ}/\mathrm{m}}^2\mathrm{yr}$

Unit energy loss with insulation:

$(0.7{\text{kW}/\mathrm{m}}^2)(2000\mathrm{h}/\text{yr})(3.6\times {10}^6\mathrm{J}/\text{kWh})=5.04{\text{GJ}/\mathrm{m}}^2\mathrm{yr}$

Energy savings:

$25.2-5.04=20.16{\text{GJ}/\mathrm{m}}^2\mathrm{yr}$

Cost savings:

$(20.16{\text{GJ}/\mathrm{m}}^2\mathrm{yr})(\$28/\text{GJ})=\$564{/\mathrm{m}}^2\mathrm{yr}$

Therefore, if the installed cost of the insulation is equal to or less than $564 per square meter, a 1-year payback will be obtained.

Other forms of heat losses that should be evaluated include:

More efficient equipment and processes. Some process heating applications may benefit from equipment retrofits or replacements to improve efficiency. The literature in this area is extensive and growing. Examples include improved controls for better regulating process temperatures and other parameters, and new more efficient equipment designs. More efficient equipment designs generally require improved heat transfer capability, although sometimes installing an entirely different process will lead to savings.

One thing to consider in improving a process is how to approach the theoretical maximum efficiency for accomplishing a task. Consider the drying process. Much of the energy used in drying is to remove water. Given that the specific heat of water is 4,180 J/kg·K, and since the heat of vaporization is 2.26 MJ/kg, the minimum theoretical energy input to evaporate water originally at room temperature (20°C) is 2594 kJ/kg. Real processes will not do as well, since some energy is lost, is used to heat other materials, or heats the container or dryer. However, since a great many heating processes involve drying, distillation, or related end uses, more efficient ways of accomplishing these tasks are of great interest.

There are so many potential possibilities for new designs that it is impossible to enumerate them here. Instead we provide few examples. The general approach should be to carefully examine the specific needs of the process and then attempt to provide heat only when and where its use is essential.

Example. Jet impingement heating. A unique approach is jet impingement heating, used in the metals industry.⁹ This allows the heat to directly impinge on the object being heated, thereby penetrating the surface film barrier and increasing heat transfer efficiency by as much as a factor of 3. Incidentally, the same technique can be used for cooling, as in gas turbines.

Process heat savings also result from converting batch-type processes to continuous operation. This conserves fuel by eliminating or reducing heat-up and cool-down periods. It often increases productivity as well.

Example. Continuous operation. One company replaced a group of batch-type kilns with a gas-fired walking beam kiln, and found the new kiln operated with 20% less fuel, and had 2.5 times the product throughput. Labor savings also resulted, since only two personnel (rather than ten) were required to operate the new unit.

Another process improvement relates to excess air. Excess air is provided to certain types of ovens for diluting the exhaust air. For example, in solvent drying ovens, air is introduced to create an air–gas mixture that is below the lower explosive limit (LEL). Many ovens use excessive amounts of dilution air, thereby wasting heat and fuel.

Typical industry practice is to operate in the range of up to 25% LEL concentrations. With automatic controls, operation up to 50% LEL limits is possible. Many ovens operate below 25% LEL and may be operating as low as 5 or 10% LEL. This is equivalent to about four times the excess air actually needed, or about twice the energy use actually required.

Refer to the “Electric Heat” section for examples of some promising electric process heat alternatives.

Heat recovery. As noted throughout this book, heat recovery is an important tool for the energy manager. Following the 1970s oil embargo, great emphasis was placed on identifying new methods to reclaim heat. Long practiced in industry, heat recovery is now being extended to agriculture and commerce. The technology can be applied to many waste energy streams, including hot gases, hot air, hot water, hot chemicals, and other hot process streams. The variables that determine the feasibility of heat recovery include the value of the heat, when it is available, the cost of installation, and the uses to which it can be put.

Typical uses for recovered heat include process heating and drying, preheating fuels or materials, space heating, and hot water. Sometimes heat recovery leads to other economies, such as reducing fan or pump operation, or permitting cooling tower operation to be reduced. The potential list of applications for this technique for process heating is too long to include here, but representative examples included throughout this book will illustrate the possibilities. A few examples include the following:

In evaluating heat recovery possibilities, the first step is to evaluate the temperature of and the quantity of heat available. If it is available at low temperatures (50–100°C) there is generally a use for it, if only for heating water and spaces. In the moderate temperature range of 100–200°C, there are many possibilities, including raising steam. Above 200°C, cogeneration and other high temperature applications begin to look attractive.

Example. Assume that 1.0 GJ of heat costs $10 in the form of steam. Thus for each 1.0 GJ of heat recovered per year, one could justify an investment of say $10 if a 1-year payback was required. Depending on local tax codes and credits, utility pricing and incentives, fuel cost escalation, and the cost of money, the allowable investment might be higher, say $15 or $20. After this determination is made, the next step would be to determine the cost of purchasing and installing heat recovery equipment capable of recovering 1.0 GJ/yr (nominal). From this analysis, a decision could be made as to whether or not the investment was justified.

Often the heat exchange equipment is a critical item, depending on the temperature range and corrosiveness of the hot effluent streams. The selection of the optimum heat recovery system size involves a balancing of the cost of the equipment compared to the benefit of recovering still more heat.

See Chapter 8 for more information on heat recovery applications and technologies for recovering heat, such as regenerative and recuperative systems as well as heat pumps.

Electric Heat Applications

Due to its relatively higher cost, electricity is not used extensively for process heat. However, there are some types of applications where electricity offers advantages for heating. Electric heat can take several forms:

Resistance heating. Resistance heating makes use of the i²R law; i.e., power dissipated is proportional to the square of the current times resistance. An example of this is a conventional residential electric water heater, which has two resistance heating elements, nominally rated at 3800 W and 240 V, single phase. This form of resistance heating has a high first law efficiency because all the heat is transferred to the material being heated; i.e., the water. Losses result from conduction through the tank walls and distribution piping.

Induction heating. Induction heating is similar to resistance heating in that the actual heating is caused by current flowing through resistance. However, in the induction heater, the heating current is induced in the work piece. An example is the heating of transformers, cores, and motor windings. Even though they are laminated to produce high resistance to the flow of such currents, transformers are in effect inductance heaters.

In an induction furnace, a coil surrounds the work piece, which must be a conductor. A variable frequency power source (oscillator) is connected to this coil, inducing eddy currents that in turn heat the work piece. The eddy currents exhibit a “screening” effect; i.e., the current density at the surface of the work piece is maximum and decreases exponentially with depth. A “penetration” depth can be defined, wherein the current has decreased to about 37% of the surface value. Approximately 90% of the heating occurs within the penetration depth. Since the penetration depth is inversely proportional to frequency, a low frequency would be used for heating a large piece and a high frequency for a smaller size.

Example. A forge heater. Billets of steel are brought by a conveying system into a water-cooled copper coil. The frequency is in the range of 1–10 kHz; specific power is about 300 kWh/ton. Advantages of induction heating include excellent temperature control and no surface decarburization. The disadvantage is a low power factor (typically 0.1–0.5), which can be corrected with capacitors.

Dielectric heating. Dielectric heating refers to the heating of nonconducting materials by an electric field. Basically, this is similar to the heating that occurs in the dielectric of a capacitor on which a high-frequency voltage is impressed. The electromagnetic fields excite the molecular makeup of material, thereby generating heat within the material. As a result, the heat is distributed uniformly throughout the work piece. Dielectric heating can be applied to wood, paper, food, ceramics, rubber, glues, and resins. The heating effect is proportional to the dielectric loss factor, the applied frequency, and the electric field strength.

Dielectric systems can be divided into two types: RF (radio frequency) and microwave. RF systems operate in the 1–100 MHz range, and microwave systems operate in the 100–10,000 MHz range. RF systems are less expensive and are capable of larger penetration depths because of their lower frequencies and longer wavelengths than microwave systems, but they are not as well suited for materials or products with irregular shapes. Both types of dielectric processes are good for applications in which the surface to volume ratio is small. In these cases, heating processes that rely on conductive, radiative, and convective heat transfer are less efficient.

Most industrial applications are below 200 MHz. Depending on the frequency, care must be given to design the system to avoid interference with telecommunication.

Electric arc furnace. The electric arc furnace has three electrodes connected to the secondary windings of a three-phase transformer. The principle is the same as in electric arc welding. When an arc is struck, the nearby gas is raised to such a high temperature (in excess of 5000°C) that it becomes highly ionized. In this state, it is a sufficiently good conductor to be maintained at high temperature by the resistive heating produced by the current. The high temperature of the plasma permits very efficient heat transfer. Arc furnaces with capacities in the range of a few tons to hundreds of tons are in use. The primary application of electric arc furnaces is for melting and processing recycled steel.

Microwave heating. Microwave heating (a form of dielectric heating) is a highly efficient technique for heating by high-frequency electromagnetic radiation. Typically, frequency bands are 896 or 915 MHz and 2450 MHz, corresponding to wavelengths of about 0.33 and 0.12 m. Energy is deposited in the work piece according to the same principles as the dielectric heater described above.

Furnaces can be designed to be resonant or nonresonant. The microwave oven found in many homes is an example of a resonant cavity device. Resonant systems have efficiencies generally in excess of 50%. Again, because the heat is deposited in the work piece, losses are minimized.

Infrared heating. Infrared heating is generated by i²R losses in heating lamps or devices, and this is a special case of resistance heating. The difference, however, is that infrared energy can be generated in a narrow bandwidth. This can be applied more efficiently in some cases than combustion energy that spans a broader bandwidth. To be most efficient, infrared heaters should concentrate their output at the peak of the absorption spectrum for the material being heated. For water, this corresponds to a wavelength of about 2.8×10⁻⁶ m. There are applications in papermaking, drying paints and enamels, and production of chemicals and drugs. Intensities in the range of 10–40 kW/m² are possible.

Heat pump. The heat pump is basically a refrigerator operating in reverse. An evaporator receives heat from a low temperature heat source (the air, waste heat, ground, water, etc.). This causes evaporation of the working fluid; the vapor is then compressed by the compressor. In the condenser, it gives up the heat collected at the evaporator as well as the heat of compression. As this heat is delivered, the vapor condenses, and the hot condensate passes through the expansion valve.

Heat pumps fall into the several categories, depending on the type of heating and the purpose. Those used for residential HVAC and water heating are primarily air-source or ground-source heat pumps, meaning they extract heat either from the air or from underground pipes. Therefore they use air-to-air or liquid-to-air heat transfer. Larger units for commercial and industrial applications employ liquid-to-liquid heat transfer.

Example. Potential for reduction in CO₂ emissions. We conducted a study for EPRI on the potential to reduce greenhouse gas emissions in the U.S. by 2030 by expanding end-use applications of electricity in the residential, commercial, and industrial sectors.¹⁰ For the industrial sector, we identified five electric process heating technologies capable of yielding net reductions in CO₂ emissions relative to their fossil-fueled alternatives, even considering losses due to electricity generation, transmission, and distribution. The technologies in order of highest to lowest potential for energy savings and CO₂ reductions are

The study timeframe was 2005–2030 and was based on the actual and projected generation mix in the U.S. for the four census regions (West, South, Midwest, and Northeast), as forecasted in the Energy Information Administration’s 2008 Annual Energy Outlook’s reference case scenario.

The results yielded a realistic potential of achieving CO₂ reductions of 45.1 million metric tons per year by steadily expanding use of these five technologies by 2030 across the U.S. The estimated technical potential was 73.1 million metric tons per year; however, the technical potential does not take into account cost-effectiveness or market acceptance, while the realistic potential attempts to factor in those key market barriers. The South and Midwest census regions had the highest potential for reductions because their generation mixes are more carbon intensive than the West and Northeast census regions.

This example helps highlight some of the benefits of electric process heating alternatives. In addition to electricity’s advantages in delivering heat when and where it is needed, as well as in its ability to extract heat from lower temperature sources (i.e., using heat pumps), some electric process heat technologies save considerable energy in certain applications compared to fossil-fueled counterparts—so much so that they lead to lower overall greenhouse gas emissions, even when generation, transmission, and distribution losses are factored in.

Three-Phase AC Circuits

Virtually all heavy industrial power is provided via three-phase circuits. The reason for this is one of size and economy. Larger blocks of power can be transmitted more efficiently and machinery can be more compact with three-phase systems.

Since copper and other metals have an important effect on capital costs of large equipment, there is also a trend to increase voltage, permitting lower currents, smaller size conductors, and lower i²R losses as well.

Three-phase voltages are generated with a generator having three sets of windings. Transformers likewise have three sets of windings. In a three-phase circuit, there are two types of connections: the wye and the delta. For either configuration, three-phase power is given by

$\dot{W}=\sqrt{3}{v}_{line}{i}_{line}pf\mathrm{W}$ [11.14]

[11.14]

where

Common nominal line-to-line three-phase voltages are 208, 230, 460, 575, and 2300 V.

Electric System Losses

Motors, transformers, and other electromotive and conversion devices make use of magnetic and electric fields for energy conversion. Mechanical energy is produced because of the fact that a conductor of length l carrying a current i in a magnetic field B experiences a force equal to

$F=Bli\mathrm{N}$ [11.15]

[11.15]

where

Transformers, motors, and other all electrical mechanical devices lose energy through various mechanisms, including

Internal resistance may be characterized as the losses caused by conductors (i²R losses). Hysteresis losses are those that are inherent to magnetic fields. When the magnetizing force is removed, not all the stored energy is recovered. An empirical relation for hysteresis losses is given by

$\dot{W}=k_{h}f{B}^n\mathrm{W}$ [11.16]

[11.16]

where

Eddy currents are losses that occur because the magnetic core material itself consists of materials that conduct electricity; as voltages are induced in this material by alternating magnetic fields, currents called eddy currents are produced. An approximate expression for these losses is given by

$\dot{W}=k_e{f}^2{B}^2\mathrm{W}$ [11.17]

[11.17]

where

When all types of losses are considered, the large transformers used in industrial processes typically have efficiencies at the high end of the 90+ percent range.

The energy lost in electrical distribution systems at voltages typical of industrial processes is predominantly in the form of i²R losses. These losses can be determined by knowing the length of the conductor, the conductor material and cross-sectional area, and its resistivity per unit length. With these conductor properties known, the losses can be calculated as follows:

$\dot{W}=i^2\rho l\mathrm{W}$ [11.18]

[11.18]

where

At very high voltages (experienced in transmission lines, but generally not in industrial applications), there are two additional sources of losses: radiation and dielectric heating. Radiation losses occur when the voltage is sufficiently high that the line acts as an antenna and radiates energy. Dielectric heating occurs when voltages are induced in insulating materials.

A secondary effect of transmission and distribution systems is voltage drop. Conductors need to be sized to minimize the voltage drop. There is a trade-off between the capital costs of larger conductors and reduced performance of equipment installed at the end of a long distribution line, where voltage drop causes the equipment to operate at a lower than specified voltage.

Conversion of Electrical Energy to Mechanical Energy

Electric motors are generally found as components of larger systems, e.g., pumps, fans, process cooling systems, air compressors, conveyors, elevators, and assorted equipment for crushing, grinding, stamping, trimming, mixing, cutting, and milling operations. In each of these systems, there are other losses in addition to those in the motor. These may be broadly summarized as follows:

The efficiencies of mechanical components such as clutches, gears, belts, and chains are typically high, in the range of 80–98% if used properly. However, when there are several of these components in the machine, the combined efficiency can be low. The wasted energy is primarily expended overcoming frictional losses and is eventually dissipated as heat. In some circumstances, this is useful. In air-conditioned spaces, this heat will add to the total air conditioning load.

Variable Speed Drives

Variable speed drives (VSDs), also called variable frequency drives, are a valuable tool for the energy manager. Typically the VSD system consists of a three-phase AC induction motor and a variable frequency power supply. The variable frequency power supply uses solid state components to produce a pulse-width modulated current that varies the power and frequency supplied to the motor. This enables accurate control of the motor speed over a broad range. VSDs are used in connection with pump and fan applications to vary the pump or fan speed according to demand, often with large savings in energy use.

EMOs with Motors

Table 11.5 summarizes EMOs with motors. The table lists a series of actions that users can employ to reduce energy use. (Note that many electric utilities offer incentives for installing efficient motors and VSDs.)

It is clear that there is not much one can do with an existing motor, so long as it is operated efficiently. The inherent i²R losses in the stator and rotor cannot be reduced without rewinding the motor or purchasing a new one. However, by understanding how these losses occur, the user can optimize motor use. Since running at no load wastes energy and results in a low power factor, it should be avoided. Operating motors as close to the fully loaded condition as possible will improve efficiency and power factor. If the power factor is very low (say less than 80% for a large motor or group of motors), it may be economically advantageous to install capacitors to correct it.

In the limited space available in this book, it is not possible to discuss all of the many applications of electric motors, so here (or in Chapter 8) we have focused on pumps and fans, refrigeration and process cooling, and compressed air systems. Additional details may be found in the references cited in this chapter.

Faraday’s Laws

Faraday’s laws form the basis for electrolysis. Briefly stated, Faraday discovered that 1.0 Faraday (26.8 ampere-hrs or 96,494 coulombs) would liberate one gram-equivalent of an element at an electrode during electrolysis.

Using Faraday’s law, it can be shown that only a small fraction of total mass of the storage battery reacts during charging/discharging (about 4 g equivalents for a typical automobile battery or about 0.4 kg).

Corrosion

Corrosion occurs as a result of oxidation–reduction reactions between a metal or alloy and a corroding agent. Corrosion can occur as a result of chemical reactions, which usually require high temperatures and a corrosive environment, or due to electrochemical reactions, which are more common. Note that corrosion is an important indirect use of energy.

The electrochemical reactions resemble the processes that take place in a battery. They can arise when dissimilar metals occur in the presence of an electrolyte or in the presence of external electric currents. A common electrolyte is water with trace amounts of dissolved salts, acids, or alkalis. The rates of corrosion reactions are dependent on the concentration of salts, acids, or alkalis in the electrolyte, and on the surface, temperature, and chemical constituents of the corroding metal.

Welding

Where possible, AC welders are preferred as they offer a better power factor and more economical operation. Automated systems reduce standby power losses compared to manual welding because they place the weld bead more consistently (less start/stop).

Electrowinning

An important use of electrolysis is the refining of metals such as aluminum. Basically the original process involved the electrolysis of a solution of aluminum oxide in molten cryolite, using carbon anodes and electrodes. In the electrolyte solution, aluminum oxide disassociates into aluminum and oxygen ions. As currents on the order of 10⁵ amperes pass through the cells (at potentials of 5.0–5.4 V), the aluminum ions migrate to the cell lining (cathode) where they are reduced to metallic aluminum. This process required 15–20 kWh per kg of electricity. New processes have been developed that reduce the amount of electricity required.

Plating and Anodizing

An electric current flows in a tank where the object to be plated or anodized serves as one of the electrodes. In plating, the plated object serves as the cathode and the anode has the material to be electrodeposited. Alternatively, the anode may be nonconsumable carbon and the plating material may be drawn from the bath. In anodizing (typical for aluminum), the object to be anodized is the anode and a direct current produces a buildup of aluminum oxide on the surface. By use of various organic acids, colored finishes can be produced.

Electroforming, Electrochemical Machining

Electroforming is a process whereby a thin layer of metal is deposited on an object to be coated or on a mold that is later removed. The classic example is copper plated baby shoes! Electrochemical machining is the reverse of plating; a high current is passed between an electrolyte and the part, removing metal. This process is used for fine, intricate parts or hard, difficult-to-machine metals.

EMOs in Electrolytic Processes

Table 11.6 summarizes typical energy management possibilities for electrolytic processes. The greatest users of energy in this field (aside from the large indirect use caused by corrosion) are in primary metals production, particularly aluminum and magnesium.

Agriculture and Food Processes

Modern agriculture is both a user of energy and a producer of fuels. In the post-World War II period, there has been a sharp decline in the number of family-owned and operated farms in the industrialized nations. Commensurately, the use of on-farm energy has increased dramatically. Sharp increases in farm yields per hectare (acre) have resulted. Today, many farms are both energy users and fuel producers. In addition, energy use in agriculture is both direct and indirect.

Energy Storage for Process Industries

Energy storage is an old concept. An early application was pumped storage, where water was pumped from a low reservoir to a higher one at night, and allowed to flow back during the day to generate electricity during peak hours. Batteries are another well-known method for storing energy. Experiments have been done to store energy as high pressure air underground or using high speed fly wheels.

For process industries, energy storage provides an important tool for the energy manager in efforts to increase overall process efficiency. Much of the waste of energy that occurs in processes takes place in the form of waste heat that is rejected to the atmosphere either directly from the process or indirectly via heat exchangers, cooling towers, stacks, and so on. There are a number of technologies available that permit energy storage at high and low temperatures. These processes will permit a better overall use of energy resources and can also help reduce the peak demand for certain energy forms, particularly electricity.

In the broadest sense, the quantity of energy stored in any media consists of four components: the sensible heat of the material in its solid and liquid phases, the latent heat of phase change, and any heat associated with chemical reactions.

Energy storage at high temperatures has been used for many years in industry. The metallurgical industries and their use of refractories is one example. The range of technologies that can be considered may be classified by operating temperature:

Due to the wide variety of thermal energy use in processes, it is not possible to provide specific guidelines. However, in addition to the metallurgical industries, there have been many applications in food processing, clay and ceramics manufacturing, and industries using steam, such as paper, and textiles. The special case of building heating and cooling is discussed in Chapter 8. Chapter 12 describes other options for thermal and electrical energy storage in integrated building designs.

Process Control

Control of industrial processes is an important determinant of energy efficiency. We’ve already seen how control of boilers, pumps, fan, and compressed air systems has a critical impact on energy efficiency. The same is true of a variety of other process controls.

In the original concept, process controls were designed to accomplish one or more of the following:

Except for the most energy-intensive industries such as aluminum or cement production, the efficient use of energy for many years was a secondary concern, since its value was much less than the value of the final product, or of the labor that went into making the product. Today the situation is different.

Now the optimization of process controls to use energy efficiently is essential. Many new types of controls and methods have been developed to do this. Some examples include simple timers, sensors, feedback devices, meters, programmable logic controllers (PLC), microprocessors, large-scale numerically controlled equipment, distributed control systems (DCS), supervisory control and data acquisition (SCADA) systems, programmable automation controllers (PAC), and web-based facility automation systems. These technologies and control strategies often overlap. As the sophistication of process controls has increased, so have their costs. The increased cost of controls has generally been more than offset by savings of rising energy and labor costs.¹⁶

Improved process controls save energy by reducing downtime and speeding up production. Often this requires improved instrumentation to sense, monitor, and measure process parameters. With the rapid development and advances in microprocessors, “intelligent” controls are now widely found in industry. These advanced devices can sense changing conditions (such as tolerance variations as a tool wears) and adjust the process to accommodate the changed circumstances. Numerically controlled machines can optimize part layout and manufacturing, minimizing scrap and waste of materials.

Example. Optimal scheduling. Process heat is a major source of industrial energy use. Industrial heating operations such as baking, tempering, annealing, and drying can use more energy in start-up, preheating, and standby time than is actually used in making the product. The simple act of optimal scheduling—i.e., planning production so the equipment is used as long as possible once started up—can materially reduce energy costs.