Chapter 3: Economizer Components Supplemental Information

Most economizer and mixing sections include some or all of the following components.

1 Dampers to control the flow and mixing of the air streams and regulate minimum outdoor air.

3 Sensors that provide the inputs to the control loops, interlocks, and safety circuits that control the temperatures, pressures, and flows associated with the economizer process.

3.6.1. Dampers

3.6.1.1. Damper Selection

Dampers are provided in the mixing section to modulate the outdoor air and return flows as required by the economizer cycle. There are two common blade arrangements used for these control dampers, opposed blade and parallel blade, shown in Figure 3.9.

· The blades of a parallel blade damper remain parallel to each other throughout the rotation cycle. This arrangement allows the damper to direct the air as it moves through them, which can be an advantage in a mixing situation. However, the dampers require a higher pressure drop through them to achieve a linear control characteristic compared to opposed blade dampers.

· Opposed blade dampers have a linkage arrangement that causes pairs of blades to rotate towards each other as the damper actuates. As a result, the air stream experiences very little change in direction as it passes through the damper. When compared to a parallel blade damper with identical dimensions and the same air flow across it, an opposed blade damper will achieve a linear flow characteristic with less pressure drop.

In both arrangements, the damper blades are mounted on a shaft that allows the blades to rotate around their long axis. For multi-blade assemblies, there is typically one blade or shaft that is designed to be the drive blade or shaft to which the actuator is connected. The other blades in the damper are driven by a linkage system from the drive blade. There are typically two arrangements for this linkage system.

· The linkage drives the blades by rotating the shaft and is completely concealed in the damper frame. This arrangement keeps the linkage and pivot points out of the air stream, which can be an advantage from a pressure drop standpoint as well as a maintenance stand point if the air stream is dirty. However, if linkage maintenance is required, it is often more difficult due to the concealed location inside the frame.

· The linkage drives the blades through a linkage that extends from blade to blade. In this arrangement, the shafts provide support and a pivot point, but generally do not transmit power from the actuator to the blades. The drive linkage is more accessible in this arrangement, but is also exposed to the air stream, which can add pressure drop and reduce the life of pivot points if the air stream is dirty.

In addition to the blade-oriented considerations discussed in the preceding paragraphs, the actual configuration of the blade itself should be considered when the damper is selected. There are two general blade designs available in the HVAC market currently.

· Flat Plate This configuration is the standard offering of most manufacturers and consists of a flat plate secured to the damper shaft, usually with some sort of bend or break folded to the length of the blade to ensure rigidity.

· Airfoil In this configuration, an extruded blade with a streamlined profile is used. An airfoil provides a rigid assembly that is resistant to flutter at high velocities and helps to ensure good blade seal compression. Thus, these dampers are often have a low damper leakage rating. The streamlined shape results in a significantly lower pressure drop as compared to the flat plate design; often by as much as half for identical damper geometries at identical flow rates. This lower pressure drop can be an advantage or a disadvantage depending upon how the damper is applied as will be seen in the next paragraph.

3.6.1.2. Damper Sizing

For most control applications, it is desirable to achieve some sort of linear relationship between damper position and damper flow. For example, the damper should reduce the flow through it by 50% when the damper has rotated 50% closed. To achieve this, it is necessary for the damper to have a significant pressure drop relative to the system that it serves. Figure 3.10 illustrates the damper characteristic curves for a parallel blade damper at a variety of pressure drop ratios (a) where a is the ratio of system pressure drop to damper pressure drop. Opposed blade dampers have similar characteristic curves, but require less pressure drop to achieve the same linear control.

· Ideally, the mechanical designer or controls contractor should have sized the dampers to achieve a nearly linear characteristic. Unfortunately, this step is often neglected and commissioning problems are often the result. Generally, the problems will fall into the following categories: Poor mixing due to lack of velocity of the air streams. This topic is discussed in Chapter 3, Section 3.1.2: Economizer Free Cooling.

· Poor controllability due to a non-linear relationship between damper position and flow.[1]

Alpha is the ratio of system pressure drop to damper pressure drop. An alpha of 200 means the damper pressure drop is 1/200 of the pressure drop of the entire system. As seen from the graph, a parallel blade damper needs to have a significant pressure drop across it relative to its system in order for the control to be nearly linear (50% stroke = 50% flow). The curves for opposed blade dampers are similar, but require less pressure drop to achieve the same linearity of control. (Developed from data in the ASHRAE Handbook of Fundamentals)

As a general rule, achieving a linear damper characteristic in most systems will require damper face velocities in the range of 2,000 to 2,500 fpm for flat plate type dampers and 2,500 to 3,000 fpm for airfoil type dampers. Based on this information a commissioning agent can quickly assess a system for potential mixing and economizer control problems by simply dividing the design flow rate by the damper face area. If the result is a velocity outside of the ranges indicated above, the commissioning agent should spend extra time on the testing and tuning of the economizer cycle.

As can be seen from the preceding discussion, the details of damper blade configuration can have a significant impact on the performance of the system the damper is installed in. Other details related to damper design, installation and construction with significant efficiency and/or other operational implications include:

· Document the details to successfully achieve design intent An amazing number of commissioning problems occur simply because the dampers are not installed correctly. There have been instances where multi-section parallel blade dampers were assembled with different blade rotations for each section; in other words, the blades in one section were horizontal and rotated downward as they closed, in the next section horizontal but rotating upward as they closed, in the next section the blades were vertical and rotated to the left. In other instances, it was virtually impossible to maintain actuators and linkages due to the installed configuration of the damper assembly. To avoid this sort of confusion, mixing damper configurations should be detailed in the design, preferably by the designer but as an alternative, by the control contractor, prior to installation. The detail should show configurations, number of sections, blade rotations, actuator locations, actuator linkage arrangements, blank-off panel locations and all other items required to install the dampers as required to achieve the design intent.

· Provide adequate access Each control damper (not just fire and smoke dampers) should be provided with an adequately sized access panel (or panels) that is unobstructed by surrounding systems, equipment, and building structure. This panel should be located so that when a technician is working through it, they can reach all damper components that are inside the duct, including linkage, bearings jamb seals and blade seals. (A 12" x 12" access panel in a 72" x 24" duct located 25 feet in the air over a motor control center and 3 feet upstream of the control damper is an access panel in name only.)

· Vertical blades need to have thrust bearings Usually, but not always, the manufacturers catch this. Without them, the operation of the damper will wear out the jamb seals very quickly at a minimum. It is also possible that the blades will start to push into the degraded seals as they wear leading to other operation problems.

· Limit blade length Blade lengths in excess of 48 inches are undesirable, especially for non-airfoil, non-extruded blades. At high velocities, the longer blades can flutter, and getting good blade seal compression as required to achieve the rated leakage rates can become a problem.

· Select dampers that securely attach the shaft to the blade There are a variety of techniques used to secure damper blades to shafts. Approaches vary from set screws to keyed arrangements. As a general rule the more positive connection provided between the damper blade and shaft by a slot and key or hexagonal shaft in a hexagonal hole is highly desirable. Less positive arrangements like set screws can often become loose over time, resulting in loss of damper functionality.

On occasion, the configuration of the air handling unit and the building make it desirable to combine some of the code-dictated fire and smoke isolation damper requirements with the economizer functions. This approach can have several advantages including:

· Reduced energy requirements due to the elimination of one pressure drop generating element from the air stream.

As a result, some of the economizer damper assemblies are fire and smoke rated assemblies and the control signals to them must accommodate these functions in addition to the more conventional economizer functions. These applications deserve special attention from the commissioning agent to:

· Ensure that all life safety control functions take precedence over environmental control functions, regardless of operating modes. For example, the smoke control cycle should have priority over the economizer cycle.

· Ensure that all life safety functions are implemented in a manner that does not threaten to harm the air handling system due to excessive positive or negative pressures or air hammer effects. Air hammer is presented in Chapter 11: Distribution.

· Ensure that the integrity of the fire and smoke damper assemblies are maintained during the installation and start-up process so that the agency listing of the dampers is not violated.

3.6.2. Actuators

The actuators used for HVAC control dampers generally fall into the following four categories, which are discussed in detail in this section.

· Gear train actuators use an electric motor with reduction gearing to rotate and output shaft that then moves the damper via crank arms on the actuator and damper shaft and an intervening linkage system.

· Gear train actuatorsuse an electric motor with reduction gearing arranged to mount with their output shaft concentric with the damper shaft, thereby eliminating a crank arm and linkage system.

· Linear actuators use an electric motor with reduction gearing to drive a jack screw which drives a crank arm on the damper shaft directly with out an intervening linkage system.

3.6.2.1. Piston Actuators

From a mechanical standpoint, piston actuators are generally the simplest, consisting of a housing, a piston and diaphragm, a spring, and a shaft through which force on the piston is transferred to the damper crank arm. They are typically actuated by pneumatic air pressure, with a maximum pressure of 30 psig or less in commercial applications. The spring constant determines the range over which the damper will actually stroke as the actuating pressure varies.

There are variations on the pneumatic actuator concept seen in the commercial HVAC market that are electrically powered. One fairly common design uses a small hydraulic pump to circulate oil from a reservoir to a chamber behind the piston. A variable size orifice controls how quickly this fluid can bleed back into the reservoir from the piston chamber. The orifice size is controlled by the control signal. As the size of the orifice is decreased, pressure builds up in the piston chamber and the piston extends. As the size of the orifice increases, pressure bleeds off and the actuator spring causes the actuator to retract. In another variation developed for remote sites with no control air source, a small self-contained compressor generates air pressure for use by a conventional pneumatic actuator. A fairly rare actuator design that may be seen on older projects uses a wax that has a high thermal coefficient of expansion and a heater. The heater output is varied by the control signal, which causes the wax to expand and move the piston.

Pneumatic piston actuators are often sequenced by selecting spring ranges appropriately so that one actuator strokes fully before the next actuator begins to stroke. However, forces generated by airflow acting on the damper blades in the HVAC system can feed back through the actuator shaft and work against the spring. The forces can shift the spring range significantly and significantly alter the sequencing, which can lead to energy waste.

For example, if a control system designer wanted to sequence a chilled water valve with the economizer dampers based on discharge temperature, the designer might specify a spring range of 5-8 psig for the economizer dampers with a normally closed outdoor air damper and a normally open return damper. To sequence the chilled water valve properly, they might specify an 8-13 psig range for the valve actuator with a normally closed valve. On paper, the outside air damper would be fully open by the time the signal reached 8 psig. At this point, the chilled water valve would start to open, thus guaranteeing the full utilization of free cooling using the outdoor air prior to using chilled water.[2] However, if the forces on the damper blades shifted the spring range 1 or 2 psig, the chilled water valve might start opening before the outdoor air damper was fully open, thus using chilled water for cooling when it may have been possible to serve the load with outdoor air. To solve this problem, a device called a positive positioner or positioning relay is installed on the actuator.

The positive positioner is a position controller that has its own source of supply air. The device senses the position of the damper shaft with adjustments for start point and span. The positioner applies pressure to the actuator piston in order to get the piston to move as the control signal varies in the specified range. For example, if you wanted to be sure that a damper moved exactly over a 5-8 psig range, then you would install a positioning relay on the damper and set it to have a 5 psig starting point and a 3 psig span. In operation, when the control signal to the positioning relay reached 5 psig, the positioner would begin to apply main air pressure to the actuator piston until it detected shaft motion, even if it took 7 or 8 psig to make the damper start to move. The positioner would then proportionately move the damper shaft in response to an increase in control signal to ensure that the shaft was fully extended by the time the control signal reached 8 psig (the sum of the 5 psig start point plus the 3 psig span value).

3.6.2.2. Gear Train Actuators With Crank Arm Drives

These actuators represent older electronic technology in which a small single-phase motor (usually the shaded pole type) is used to drive a crank arm through a gear train. The gear train actuator can interface with a variety of signals including electric floating contact type controllers, variable resistance type controllers, two position signals, and common automation system outputs like 4-20 milliamps, 1-10 vdc or pulse width modulation. The type of interface is usually determined by an interface circuit board in the actuator or an interface module mounted to the actuator. Settings on the circuit boards or independent modules allow these actuators to be sequenced in a manner similar to that described for pneumatic actuators discussed in the piston actuator section. Most models can be equipped with a spring return feature to guarantee a fail-safe position on power failure.

3.6.2.3. Gear Train Actuators With Shaft Concentric Drives

These actuators represent newer technology that developed in response to the widespread use of DDC control systems in the commercial buildings industry. The actuators avoid some of the potential linkage problems associated with piston actuators and gear train actuators driving through crank arms by mounting the actuator directly to the damper shaft, with the output torque applied directly to the shaft. Since the DDC market drove the development of these actuators, they can usually be interfaced directly to common DDC outputs. Sequencing with other actuators is usually accomplished by settings incorporated directly into the actuator. Most of these actuators are offered with spring return options for fail safe positioning on power failure.

Several manufacturers combine the actuator in the same housing as their VAV terminal unit controllers. This tends to make the installation costs lower with less start up problems as compared to systems that require that the VAV unit damper actuator be mounted and wired independently from the controller. But, it can increase the replacement cost when the unit fails since the entire assembly needs to be replaced, not just the failed component.

3.6.2.4. Linear Actuators

These actuators are more common in the industrial market than the commercial market but are sometimes found serving applications such as inlet guide vanes where a lot of actuating power is required. They typically consist of a jackshaft coupled with a motor of some sort and can accept most of the common output signals available from a DDC system.

3.6.2.5. Installation and Commissioning Issues

Generally, there are several commissioning issues related to installation of damper actuators as follows:

· Sizing By reading the fine print on the damper leakage curves, one may discover that the preload torque required to achieve the leakage rate is higher than the torque required for actuation. Actuator selection, sizing, and set-up need to take the pre-load torque into consideration and the commissioning process should include verification that the field installation meets all applicable requirements.

· Linkage Arrangement Piston and linear type damper actuators need to be mounted a manner that allows the linear motion of the actuator shaft to be converted to rotation at the damper shaft. Usually this is accomplished via a linkage system that includes crank arms, extensions and swivels. Actuators should be mounted to maximize the linearity between actuator stroke and blade position and also maximize the torque available to the damper from the actuator through the damper stroke. The kinematics associated with the linkage arrangement are often not well understood by the field personnel installing the actuators, thus issues related to this can show up as a commissioning problem. Figure 3.16 illustrates how the relationship between actuator stroke and damper bland position varies for a piston actuator.

Note 1: A + indicates that the percentage of stroke leads the percentage of blade rotation.

Note 2: Only the actuator force applied perpendicular to the crank arm creates rotation and torque.

Similar considerations apply to linear actuators as well as gear train actuators that drive the damper via a crank arm and linkage. Actuators that concentric to the shaft and drive the damper shaft directly eliminate these issues by virtue of their mounting and drive arrangement.

Multiple section dampers with multiple actuators need to have identical actuators with identical linkage arrangements for all sections to make the characteristic for all of the sections consistent. Consistent damper section installations also make the performance of the entire assembly predictable based on the performance of one section. If the outdoor air damper and the return damper are multi-sectioned dampers controlled by the same signal, then all actuators for both sections should be identical with identical mounting arrangements for the same reason.

· Actuating Speed and Power Most electric actuators have operating times that can vary from 30 to over 90 seconds to move through a full stroke. The slow speed is due to the small motors, which must be geared down significantly to deliver the necessary torque required to actuate common HVAC devices. Pneumatic actuators, on the other hand, can have full stroke actuating times that are extremely fast and can deliver tremendous power due to the amplifying effect of air pressure over a large diaphragm area. For instance, a six inch diameter piston damper actuator with 20 psi air acting on it can easily deliver over 500 pounds of force at the shaft, even after compressing the damper spring. These characteristics can be advantages or disadvantages depending on the application.

· Rapid actuation speeds are desirable and even necessary if tight, reliable, and stable control is to be achieved on some systems. Usually, this is expensive and difficult to achieve electrically using commonly available technology in the HVAC industry. This item is worthy of review by the commissioning agent in the design phase. During start-up or in a retro-commissioning environment, changing from electric to pneumatic actuation may represent a viable solution to a start-up or operational problem related to system response time.

· Rapid actuation can be undesirable in cases where air or water hammer could be the result of a rapidly closing valve or damper.[3] In situations where this is a possibility, the slower actuating times associated with electric actuators may be an advantage.[4] If pneumatic actuators are installed at locations where they could generate air or water hammer, then it may be necessary to slow them down by using restrictors or metering valves in the circuit serving them. The commissioning agent needs to be aware of these issues and requirements and design functional test sequences to ensure that they are properly addressed.

· Sequencing As described earlier in this section, proper sequencing of actuators is often essential for efficient system operation, but cannot be solely guaranteed by the specifications and requirements of the contract documents. The functional testing performed during the commissioning process should include verification of proper sequencing under all operating modes. If sensors are installed at the proper locations in the system, trending can be used to allow the commissioning agent and operators to verify the persistence of the original sequencing. “Smart alarms” triggered by conditions that are not logical (a temperature rise across a heating coil that has its valve commanded closed for instance) may also be desirable.

From a historical perspective, pneumatic actuation and control systems where quite common in the commercial buildings industry. In the past decade, the wide acceptance and implementation of DDC control systems coupled with the development of shaft centerline drive type actuators have made electronic/electric actuation system more common in the commercial HVAC market. The electric/electronic actuators often can interface directly with the DDC system where as pneumatic actuators require some sort of signal converter[5]. In addition, pneumatic actuators require a source of clean, dry air. On sites where this is not already available, the use of pneumatic actuation will require that this equipment be installed, operated and maintained. However, if a large operating torque is required and/or quick (less than a minute) actuation speed is necessary, then pneumatic actuation often represents the most cost effective and reliable option.

3.6.3. Sensing Elements

Sensing elements perform critical tasks in ensuring proper economizer operation. The following paragraphs discuss sensing requirements particular to economizers.

3.6.3.1. Temperature Sensors

The mixed air sensor is in the red circle. The spiraled sensing element is the freezestat. This single point sensor will have a difficult time averaging the temperature in this 6 ft by 10 ft mixing plenum.

Selection and proper configuration of the temperature sensors used by the economizer cycle can have a very significant impact on its ability to perform as intended. Since even a well-designed mixing plenum can have some temperature stratification over its cross section, selecting and installing a temperature sensing system that averages the temperature across the entire cross section is an important aspect of the overall system design. There are a surprising number of projects encountered by commissioning agents where the first step in solving the economizer operating problems is the replacement of a single point sensor with an averaging sensor. The picture in Figure 3.17 is from one such project. In this instance, the poor sensor configuration coupled with independent heat transfer element control loops was causing the system to do 10°F of unnecessary preheating.

Specifying averaging sensors for economizer applications is a good first step, but it does not fully address the issues related to measuring temperatures in a mixed air plenum. The following items should also be considered:

1 The temperature that will ultimately be achieved by the complete mixing of a stratified air stream is a function of both the temperatures and the mass flow rates of the different temperature layers. The information sensed by averaging temperature elements is purely temperature and is effectively immune from the influences of mass flow rate[6]. If the nominal face velocity over all portions of the averaging element is the same, then the average temperature indicated by the element is probably a reasonable measurement of the temperature that will be achieved in the air stream when it is completely mixed. However, if there are significant differences in the velocity over certain areas of the averaging element, then the average temperature indication sensed by the element may not accurately reflect this final mixed temperature. For example, when mixing a teaspoon of water at 200°F water with a gallon of 100°F, you would not expect the mixed temperature to be 150°F, even though this is the average of the two temperatures being mixed. Similar considerations apply to airstreams in mixing plenums.

· If the true mixed temperature is higher than what the averaging element predicts, there may be comfort control problems since the system will not be delivering air at the required temperature. In some systems, the refrigeration equipment may operate to make up for this difference even though the system could achieve the required setpoint by using additional outdoor air in the economizer cycle, thereby negating some of the energy benefits of the cycle.

· If the true mixed temperature is lower than the value indicated by the averaging element, then there could be problems with nuisance freezestat trips or coil failures. In some systems, the preheat coil will be activated to warm the air up to the required setting despite the fact that the system could achieve this temperature by using less outside air on the economizer cycle. Again, this wastes energy and negates some of the benefits of the cycle.

2 Stratification patterns and their associated velocity profiles can change with the operating point of the economizer. As the outdoor air and return air dampers modulate due to changes in setpoint, return, and outdoor air conditions, the temperature and velocity profile produced at the discharge of the mixed air plenum can change. Systems that mix well under some operating conditions may stratify under other conditions. Other systems may exhibit similar profiles regardless of the operating point. The Temperature Traverse Test and the High Turndown Ratio Test help identify temperature and flow patterns in the mixed air plenum and target solutions to related problems.

3 Large plenums may require multiple sensors to accurately reflect what is occurring in the plenum. The mixed air plenum stratification test included in this chapter can be used to determine the best configuration for the mixed air sensing elements under different operating conditions. It may be necessary to perform this test as part of new construction or retro-commissioning to optimize the configuration of the sensing elements so they provide repeatable, reliable information under all operating modes.

4 The mounting system used for the averaging elements should allow them to be easily reconfigured during the commissioning process if necessary. It should also support the sensing elements in a manner that allows them to accurately reflect the air stream conditions, free from other influences such as heat conduction from coils or other supporting equipment and radiant temperature effects from high temperature heat transfer elements like steam coils. Radiant temperature effects can be felt upstream as well as downstream of a heat transfer element. If you can stand at the mixed air sensor location with the palm of your hand facing a nearby high temperature coil, and can feel the radiant heat from the coil with your hand, then the element is probably being influenced by the coil.

3.6.3.2. Freezestats

Freezestats are limit controls that are installed to protect the coils in an air handling system from being subjected to sub-freezing temperatures. Typically, they are arranged to shut down and lock out the system and a manual reset is required to resume normal operation. Contrary to popular opinion, the long sensing elements that they are provided with are not really averaging elements, even though they are visually similar. The sensing elements on most freezestats will respond to the coldest temperature seen by any one or two foot segment. This is desirable because it only takes a small stream of subfreezing air to freeze and rupture a tube in a coil.

Even though freezestats are not averaging elements, the same considerations outlined previously for the installation of averaging elements apply to the installation of sensors on freezestats. In fact, it is often desirable to string the freezestat element along with the mixed air temperature-averaging element so that they both are subjected to the same conditions. This allows the mixed air sensing element to reliably generate a pre-freezestat trip alarm that can be used to prevent an unscheduled outage. The advent of DDC technology has lead to the development of software-based freezestats. At first glance, this approach appears to offer some advantages in terms of lower installed cost and ease of operation. However, there can be some problems with this approach that can very quickly negate any of its benefits.

1 Usually the first cost savings is achieved by eliminating a device; i.e. the independent freezestat and its associated wiring to the starter. This means that the mixed air sensing element becomes the input to the freezestat control code and, in a sense, the mixed air sensor ends up trying to protect itself from its own failure.

2 Since the operation of the freezestat is now dependent on the operation of the DDC controller, a controller failure eliminates the freezestat protection. In some cases, a controller failure will not necessarily shut down the unit, so it is possible for a system to end up on line and operating without the intended protection in place.

3 Since the freezestat function in this approach is part of the controlling software, it is a part of the automation. Usually, the automated output that starts and stops the fan is wired to the “Auto” position of the “Hand-Off-Auto” selector switch on the fan. If the fan is placed in “Hand”, then the freezestat protection is eliminated from the circuit. An independent device can be wired into the common side of the selector switch so that it provides the necessary level of protection regardless of the selector switch position.

For these reasons, the best approach is to simply not use software based freezestat (or a software based safety of any type for that matter). If the software-based system is employed on a project, it needs to be very carefully designed and commissioned to prevent the types of problems outlined above from occurring.

On one project, the owner wanted their mechanics to be able to rapidly identify a freezestat shutdown of the air handling equipment and demanded that the freezestat be wired into the control circuit only on the auto side of the hand-off-auto selector switch. This would allow the mechanics to quickly place the unit in hand and see if it ran when it was off for some unexplained reason. If it ran, then they knew they had a freezestat trip, the anticipated most likely problem. If the unit didn’t run, they would focus their troubleshooting efforts elsewhere. The consultant reluctantly agreed with the owner’s demand with the stipulation that the hand position be designed to spring return to the off position to prevent an operator from inadvertently walking away and leaving the system operating in hand with no freezestat protection. During the first winter of operation, the designer was called to the project site to investigate and determine why a coil had been frozen in one of the air handling units equipped with the spring return type switch. When they walked up to the starter, the cause of the problem became immediately obvious. A creative mechanic (who didn’t understand the control system) had solved the problem with that pesky selector switch that wouldn’t stay in the hand position by clamping the jaws of a crescent wrench to the actuating tab on the selector switch.

Most people think of a freezestat as providing protection for systems with water coils and would not consider installing one on a system that is equipped with direct expansion cooling and gas fired furnaces for heating. Generally, this is a reasonable assumption. But there have been instances where the plumbing and sprinkler piping in a building was frozen because such an air handling system remained in operation with an economizer that failed wide open during extreme weather over a weekend. Since there are no water coils to protect in this situation, the freezestat can be installed in the discharge duct where it is relatively immune from the effects of poor mixing and the related problems with nuisance freezestat trips, yet will provide protection against freezing the building plumbing and sprinkler piping.

3.6.3.3. Enthalpy Switches and Sensors

To perform the enthalpy-based economizer changeover function described in Chapter 3, Section 3.5.6.2: Enthalpy Based Interlock, enthalpy switches or transmitters may be used.

Two Position Enthalpy Switches Figure 3.18 illustrates a typical two-position enthalpy switch. The different scales on the switch correspond to different near-constant enthalpy states. By coordinating the scale selected with the requirements of the project, the switch can be used to provide a relatively low cost enthalpy based economizer change-over signal. The switches need to be mounted in an accessible location that subjects the sensing element to a free-flowing outdoor air stream yet protects the element from exposure to water and direct and reflected sunlight. This location can introduce some operating anomalies into the system due to lack of air flow when the unit is not in operation. For instance, minor air leakage back out of the unit from the building when the unit is off may cause the sensor to “think” the ambient conditions are suitable for cooling in the summer when they are not. When this unit first starts, it may try to operate on the 100% outdoor air cycle until the outdoor air flowing through the intake system can bring the local environment at the switch location into alignment with the actual outdoor conditions. This process usually happens fairly quickly, but it can be misleading to the commissioning agent during testing.

Protecting the enthalpy switch from dust is also desirable, but usually difficult to accomplish because most recirculating systems do not filter the air until after the mixing plenum. For sites with a 100% outdoor air unit, the enthalpy switch can be protected by locating it downstream of the first set of filters but upstream of any heat transfer elements. This enthalpy signal can be used to pilot the other units on the site. For this approach to be successful, the unit where the switch is installed must be one that runs at all times that any of the other units piloted by the signal run. Without air moving across the filters and to the enthalpy switch, the signal from the switch will not necessarily reflect the ambient conditions.

The switch settings correspond to a set of near-constant enthalpy lines on a psychrometric chart. Many of the manufacturers include a small chart with the operating curves as a part of the switch installation literature or they may print the chart on the switch housing. However, in the event that the installation instructions are not available, then it is fairly easy to construct the operating curves from a blank psychrometric chart and the tables included in the catalogs. Figure 3.19 is an example of a chart constructed in this manner. Another option is to go to the internet and download the installation instructions since many manufacturers now provide this information in the form of .PDF files. The following link contains the information for the switch depicted in the curves below; https://hbctechlit.honeywell.com/request.cfm?form=60-2301.

If the installation information with the enthalpy switch operating curves is not available, operating curves can be drawn on a psychrometric chart using the information typically included in the supply catalog. Once you have made the curve for one type of switch, you can just keep a copy with your test equipment. For this set of curves, if the switch was set on scale D, then it would be in economizer mode for any condition left of curve D and out of the economizer mode for any conditions to the right of curve D.

Notice how the operating curves in Figure 3.19 are really not constant enthalpy lines. They parallel the lines reasonably well for the upper portion, but fall away from running parallel, from the 50% point to their lower end. This curvature should be taken into consideration when determining the appropriate switch setting.

Enthalpy Transmitters Figure 3.20 depicts an analog enthalpy transmitter. These units provide an analog output that is directly proportional to enthalpy. They can be wired as an input to the control system and the information they provide can be used for various enthalpy-based control decisions such as enabling or disabling the economizer mode. In general terms, the same mounting and servicing considerations apply to enthalpy transmitters as apply to enthalpy switches, although there are some designs available that are suited for direct or nearly direct exposure to the outdoor environment.

Enthalpy Calculations Today’s control systems generally are capable of performing mathematical operations on their variables. This opens the door for calculating enthalpy using psychrometric equations contained in the ASHRAE Handbook of Fundamentals or the ASHRAE Brochure on Psychrometrics. Both of these documents contain psychrometric equations that can be solved for enthalpy if other parameters are known. Usually, the required inputs include dry bulb temperature, relative humidity or dew point, barometric pressure and elevation. For most applications, assuming a constant elevation and barometric pressure for the site will provide a satisfactory level of accuracy. As a result, the calculation can be reduced to one based on dry bulb temperature and humidity or dew point. For instance, outdoor air temperature and humidity are measured, then outdoor air enthalpy can be calculated and used for enthalpy based economizer interlocks assuming a return air enthalpy. Space or return air temperature and humidity sensors can be used to calculate space or return air enthalpy, and economizer changeover strategies based on differential enthalpy can be implemented.

3.6.4. Pressure Sensors and Switches

Pressure switches are also used in some systems to provide a safety interlock to prevent over or under pressurization of the duct or air handling unit casing in the event of a control failure. This is discussed in greater detail in the Chapter 3, Section 3.5: Control Strategies.

While pressure sensors are not used directly by the economizer cycle, they are used in some control strategies to modulate the relief air dampers, which relates to the economizer function. This is discussed in greater detail earlier in Chapter 3, Section 3.1.3: Building Pressure Control. When this approach is employed, building static pressure is measured and the system is arranged to modulate the relief dampers to maintain a slightly positive pressure in the building.

Minor variations in static pressure can cause interesting system problems if detected and processed by the control system. On one project, the minor building pressure changes associated with the employee entrance door opening and closing were driving the fan speed control loop for a 40,000 cfm air handling system into instability during shift changes. When the door opened, the building pressure dropped several hundredths of an inch. The change in building pressure was telegraphed to the supply duct static pressure transmitter, located in a branch duct near a diffuser near the door where it showed up as a change in duct static pressure of several hundreds of an inch (the setpoint was 0.5 inches w.c.) One or two door operations were not a problem, but when 40 or 50 employees left the building within a 10-15 minute interval, the pressure pulses drove the system into instability for several hours as the fan control system tried to follow them and the inertia of the large DWDI airfoil fan kept it from catching up. Filtering the input so that minor pressure pulses were not processed by the control system solved the problem.

One of the most important things to remember about measuring building static pressure is that very low differential pressure levels must be measured and controlled, typically less than on tenth of an inch water column. In most cases (but not all as will be discussed later), this pressure needs to be converted to some sort of force for an instrument to detect it and produce a signal as a result of it. The force needs to be great enough to move whatever mechanism is used to generate the signal in response to its variations magnitude. The impact of the mechanism activated by the force on the signal needs to be minimal. Frictional forces, springs, and other mechanical resistances imposed on the force generated by the measured signal in converting it to an output all work against generating a repeatable useable signal. For instance, if the break-away torque for a bearing is 1 foot-pound, then the shaft it serves will not move until the torque applied to it exceeds this value. If the force that was being applied to the shaft in normal use never ran much above this, then the mechanism would be fairly useless in transmitting the input force to the output shaft.

Instruments that measure the low differential pressures associated with building and air handling system static pressures typically use a diaphragm to actuate the signal producing mechanism. The differential pressure is applied across the diaphragm, causing it to deflect. The deflection is then used by a system of levers, pivots and springs or by discrete electronic components to generate an output.

To put this into some practical terms consider a 4-inch diameter diaphragm measuring a signal of one tenth of an inch water column.[7] In this instrument, the force available from the diaphragm (assuming it is a perfect diaphragm with no mechanical losses of its own) to move the signal generating mechanism is less than one tenth of an ounce. A first class letter weighs approximately one ounce. That is not much force to move the levers, pivots and springs associated with this type of mechanism in a repeatable, reliable fashion for years in an environment that can see large temperature swings, dust, moisture, and vibration. Anything that can be done to improve this situation, like providing more force via a larger diaphragm or eliminating moving parts or increasing sensitivity to motion via electronics, or using some other approach to measuring pressure will result in an improved signal, a more robust system, and better control and efficiency. Thus a transmitter with a large diaphragm area will tend to be superior to a transmitter with a small diaphragm area, all other things being equal. Transmitters that are based on electronics rather than mechanics and kinematics will tend to be superior to those with many moving parts.

The transmitter on the left serves as an input to the controller on the right. Because this type of controller “receives” its input from an external source rather than a built-in sensor, it is referred to as a “receiver controller” and can be used with a variety of input sensors rather than being segregated to one particular function. (Image courtesy of the Siemens web site)

Another important point to remember is that even though the desired result for a building static pressure control system is a positive building pressure, in the real world operating environment, it may be beneficial to be able to detect and indicate both positive and negative pressures. Thus a sensor with a range of -0.1 inches w.c. to +0.1 inches w.c. might be better than a sensor with a range of 0 to +0.1 inches w.c.

There are a variety of technologies used to sense pressures in this range for building static pressure applications. Generally, the fall into the following categories:

1 Pneumatic differential pressure transmitters that sense via a diaphragm and create a force based on the sensed pressure - These transmitters typically use a diaphragm to sense and create a force based on the measured pressure. Figure 3.21 presents a typical example along with a controller that it would be coupled to. The force generated by the diaphragm in the transmitter is used by a system of linkages, levers, and springs to generate a nominal 3-15 psig pneumatic signal that is proportional to the sensed pressure. This signal is used as an input to a controller that allows it to be compared to a setpoint and then generates an output. The output is used to modulate a final control element such as an actuator or variable speed drive as required to make the sensed pressure match the setpoint. In some cases, the pressure sensing function is combined with the control function in one package.

2 Electronic differential pressure transmitters sense via a diaphragm and create a force based on the sensed pressure Electronic components use the force to generate a standard electronic output such as 2-10 vdc, 1-5 vdc, or 4-10 ma. The output is proportional to the measured pressure. As in the case of pneumatic transmitters, this signal is then used as an input to a controller that, in turn, generates a control output to modulate the system.

This particular sensor is designed for monitoring hospital isolation room pressure, but the technology can be adapted to other low differential pressure applications. (Image courtesy of the TekAir web site)

3 Electronic differential pressure transmitters that measure the velocity of a flow stream created by the pressure difference between to pressure zones as an indicator of pressure. If there is a pressure difference between two areas, then there will be flow between the two areas if there is an open flow path between them. This phenomenon is used to advantage by some transmitter technologies to measure pressure difference. Generally, these transmitters consist of a tube with a hot wire anemometer or similar thermal based flow-measuring device located in it.[8] One end of the tube is connected to the location where the desired pressure is to be monitored. The other end is connected to the reference pressure. If there is a difference in pressure between the two ends of the tubes, then air flows through the tubes. This flow is detected and measured by the hot wire anemometer and an output in terms of pressure is generated based on the flow rate. The output is typically a standard electronic output such as 2-10 vdc, 1-5 vdc, or 4-10 ma and this signal is then used as an input to a controller that, in turn, generates a control output to modulate the system. Figure 3.22 illustrates a typical transmitter of this type.

3.6.4.1. Pneumatic Pressure Transmitters

Though less common due to the advent of DDC technology, there are still numerous existing systems that use pneumatic pressure transmitters to provide inputs to pneumatic controllers for building static pressure control. While the systems can be made to function, the resolution and repeatability associated with modern electronic-based transmitters will provide a better solution on new projects, thus a pneumatic based system should not be the system of choice for this application in new construction if at all possible. However, there are often situations where commissioning agents find themselves confronted with making this approach work reliably and repeatedly. In this case, the following issues should be considered and evaluated.

1 Differential Pressure Measurement Capabilities Unfortunately, many of the transmitters available in the commercial pneumatic product lines are not very well suited for measuring the low pressure differentials typically associated with building static pressure control.

· The stated accuracy of the transmitter is often a significant percentage of or even equal to the signal that must be measured.

· Diaphragm areas tend to be small and thus the forces produces to actuate the measurement mechanism.

· The measurement mechanisms tend to be mechanical with the associated frictional losses and hysteresis problems.

A one pipe transmitter creates its signal by varying the flow of control air through a fixed orifice. The sensing mechanism is arranged to continuously bleed air from the down-stream side of the orifice with the bleed rate controlled by the process variable. As a result, the pressure downstream of the tee will vary with the process variable. The term “one pipe” comes from the fact that the sensing element or transmitter has only one pneumatic line connected to it. A two pipe transmitter has a supply line connected to it as well as a signal line and does not continuously bleed air to generate a signal. Instead, it uses a feedback mechanism to add or remove air from the signal line so that the pressure in the signal line is proportional to the variable sensed.

2 Signal Transmitting Technology Transmitting the pneumatic signal is further complicated by the fact that most of the commercial grade units use one-pipe technology. This approach is illustrated in Figure 3.23 and is prone to the following problems:

· The tubing between the restrictor tee and the transmitter can become a source of pressure drop that affects the signal from the device but is not accounted for in its calibration. If a long run of tubing is installed between the transmitter and the restrictor tee, the calibration may not match the factory specifications. Most manufacturers state a length of run limitation for their transmitters to address this issue. It is important that this length is the length of the tubing run, which may be different than the distance between the transmitter and the restrictor tee on a plan view of the building.

· Most restrictor tees include a foam filter element to protect the very tiny hole that forms the orifice in the tee from being obstructed with grit. As this filter becomes dirty, it also adds an un-calibrated pressure drop to the system. Most manufacturers recommend changing restrictor tees out annually (or more often) in order to address this issue. However, most building maintenance staffs are unaware of this requirement and may even be unaware of the presence of restrictor tees in the first place. So, when faced with commissioning or re-commissioning an existing system that uses this technology, a good first step is to perform the restrictor tee replacement recommended in addition to checking calibration of the instruments.

· Most commercial one-pipe transmitters are factory calibrated and cannot be calibrated in the field. (If you decide to try it you will probably wish you hadn’t. Barely detectable movements of the zero and span screws can cause output changes of 10% or more.)

All of these issues with pneumatic transmitters can result in an input system that does not provide reliable, repeatable data for use by the control system. Fortunately, there are several options available that can help address these problems.

1 Install a Process Grade Transmitter Pneumatic pressure transmitters are still available in the product lines of many process control manufacturers. Often, the available ranges are not much better than the ranges available in the commercial product lines, but the overall accuracy can be significantly better (0.5% of full scale vs. 5% of full scale). Some product lines include a device called “draft pressure transmitters” which were originally designed for boiler combustion control systems. These devices can have spans as small as 0.25 inches w.c. with accuracies of 2% of full scale. This is a significant improvement over any of the available commercial products. Generally, these improvements are achieved by using larger diaphragm areas and high precision mechanisms with better bearing systems and lower resistance to movement and hysteresis. In addition, the process grade instruments use two-pipe technology, thereby eliminating all of the problems associated with one pipe technology.

The down side is that the process grade products are much more expensive than the commercial products, often by a factor of 10 or more ($150 vs. $1,500). This cost can often be very easily justified in light of the energy and comfort benefits gained by making a non-functioning system function as intended and the durability and improved service life that can be expected. However, the cost premium can come as quite a shock to an Owner who is unfamiliar with the problem and the requirements of the solution. The other problem with using process grade transmitters is lack of familiarity with the product in the commercial HVAC marketplace. Many times, this can be easily addressed by training, a primary component of the commissioning process. But, in buildings where the operating staff undergoes frequent changes or where the maintenance is handled by a service contract, this lack of familiarity can be a significant issue.

2 Install an Electronic Sensor and Convert the Signal to Pneumatic for Use by the Pneumatic Controller While slightly cumbersome technically, this approach can provide a significant improvement in the input signal quality to an existing pneumatic control system in some cases. The input signal improvements are accomplished using readily available commercial products, combining a standard low range, high accuracy electronic pressure transmitter with an electronic to pneumatic signal converter. The accuracy, repeatability and hysteresis characteristics of the signal converter are key to achieving good overall accuracy with this approach. A low quality converter can quickly erode away any benefits gained in using an electronic sensor. The most significant disadvantage of this approach is that the existing pneumatic tubing from the transmitter to the controller location must be replaced with wire to accommodate the electronic signal or wiring must be run to the transmitter location to power the transmitter and electronic to pneumatic signal converter.

The controller pictured here is typical of the single loop stand alone controllers manufactured by many process control companies. While it may appear somewhat intimidating to the uninitiated, the operation is usually very well documented in the manuals. If you can work a VCR remote, you can probably master one of these devices (Image courtesy of the Partlow web site)

3 Replace the Pneumatic Controller and Sensor with an Electronic Controller and Sensor and Convert the Output to a Pneumatic Signal For a relatively modest incremental cost increase over option 2, it is often possible to replace the entire pneumatic control system through the controller with an electronic system using a stand-alone process controller (Figure 3.24). This approach allows all of the advantages typically associated with DDC technology but can be easily interfaced to the existing pneumatic actuation system via a standard, readily available electronic to pneumatic converter. In installations where the existing pneumatic signal is being converted to an electronic signal for interface to a variable speed drive, this approach can actually simplify the system since the controller will most likely be able to interface to the drive without a signal converter. The disadvantage to this approach is that wiring must be installed to replace the pneumatic tubing that provides the input signal in an all-pneumatic system. Once installed, the system will generally provide much better performance and can be readily adapted to DDC technology and controllers if and when the rest of the control system is upgraded.

3.6.4.2. Electronic Force Based Differential Pressure Transmitters

These particular transmitters use a capacitance type pressure cell. (Picture courtesy of the Kele web site)

In principle, electronic force based transmitters function in a very similar manner to the pneumatic transmitters discussed in the preceding paragraphs. The difference is that the motion of the diaphragm (created by the pressure difference across it) is converted to an electronic signal rather than a pneumatic signal. However electronic components are used to make the signal conversion rather than a mechanical system of levers, pivots and springs. This takes advantage of the inherent sensitivity of electronics and their lack of moving parts to provide a more repeatable signal with higher resolution. Techniques for making generating the electronic signal typically involve using the diaphragm to cause some change in capacitance or resistance with is then amplified and conditioned by an electronic circuit to provide a standard output. It is not unusual for the diaphragm to function as one plate of the capacitor in capacitance-based instruments. In other instruments, strain gauges attached to the diaphragm generate a variable resistance as the diaphragm flexes. There are products where the diaphragm is actually an etched out part of the semiconductor chip upon which the integrated circuit for the sensor is fabricated.

All of these factors combine to create sensors that provide more robust performance as compared to pneumatic products in the same price range. However, even better performance in terms of resolution, can be achieved with the flow based pressure sensors discussed in the next section.

3.6.4.3. Electronic Flow Based Differential Pressure Transmitters

As indicated previously, transmitters in this class indicate a pressure difference based on and input generated from the flow created by that pressure difference. The technology was originally developed to provide monitoring and control for hospital isolation rooms where very low-pressure differentials must be monitored, maintained, and alarmed if infections control requirements are to be met. The technology has also found application in application specific controllers targeted at laboratory fume hood control systems. Both systems can be adapted for use to monitor and control low-level static pressures associated with other HVAC processes such as building static pressure control. These transmitters are generally capable of resolving pressures as slow as one ten-thousandth of an inch water column and can be set up for a full scale range of several hundredths of an inch with an accuracy of ±2% of that range, making them ideal for this application. This performance comes with a bit of a cost penalty relative to the other electronic technologies; typically by a factor of 2 or 3 ($150 - $200 vs. $600-$900). However, in the context of the task to be performed and the benefits in terms of energy and comfort that can be attained by a persistent, successful implementation of the strategy, the benefits almost always outweigh the costs.

3.6.4.4. Transmitter Installation and Sensing Line Considerations

Regardless of how the static pressure signals are measured, the method used to extend the sensing connections from the transmitter to the location sensed as well as the method used to terminate the tubing ends at the sensing location can have significant impacts on the system’s ability to perform. The issues to be considered are as follows.

1 Transmitters are More than Signal Converters One of the functions that a transmitter performs is to convert a low grade, noise susceptible, non-standard signal such as the resistance or capacitance variation in a differential pressure sensing element to a higher grade, industry standard, noise immune signal such as a 4-20 ma current loop. This function:

· Allows a nearly infinite variety of process sensors to be readily interfaced to a few industry standard inputs.

· Protects the relatively low level (e.g. milivolt, micro-ohm, micro-farad, hundredths of an inch water column) sensor signal from the effects of electrical noise between the sensing location and the input location.

· Allows the signal to be transmitted over large distances without degradation due to wiring resistance.

This figure shows the impact of an open calibration port on the input signal measured by a pressure transmitter located at the control panel location instead of at the point where the signal is measured

Many times, pressure transmitters are installed with a plugged tee on their input sensing lines where they connect to the transmitter to allow a separate meter to be connected in parallel with the inputs for calibration purposes. This figure illustrates what happens to the input signal to the transmitter if one of these ports is unintentionally left open in a situation where the transmitter is located at the control panel rather than at the location where the pressure is being measured. At a minimum, the open connection at the equipment room tends to reference the transmitter to that location rather than to the desired point. If there is a significant pressure difference between the two locations, then air will flow through the input tube, creating a false reading due to the pressure drops associated with flow through the tube. While the magnitude of the pressure drop may not be significant in absolute terms, it can be a significant factor when the pressure being measured is only several hundredths of an inch w.c. in the first place. Leaks in the tubing due to nicks or loose fittings can create a similar situation.

It is not uncommon to find transmitters located in the auxiliary panel immediately adjacent to the controller and hundreds of feet from the location of the sensor. While desirable from a maintenance standpoint since all of the transmitters are easily accessed for calibration or replacement when necessary, it defeats some of the most significant benefits of having a transmitter in the first place; to protect the sensor signal from noise and degradation. In the case of the transmitters used for building static pressure monitoring and control, locating the transmitter next to the control panel and extending the tubing to the sensing location means that the very low signals associated with building pressurization (usually no more than ten hundreds of an inch) are sent through hundreds of feet of tubing. At first glance, this distance may not seem like much of an issue, since the tubing does not carry flow (except for a few fractions of a second when there is a pressure change). The pneumatic tubing is dead ended at the transmitter location and therefore is immune to the effects of pressure drop due to flow as a matter of normal operation. However, the pressure drop is not immune to the effect of flow created by leakage due to branch connections that were un-intentionally left open, loose fittings, and nicks introduced during the construction process and subsequent life of the building. Figure 3.26 illustrates the impact of a calibration port that was left open to the equipment room environment. Other unintended openings in the sensing tubing runs can have a similar effect. Even a minor leak can have the effect of neutralizing the signal from the intended location and referencing it to a different pressure condition.

When properly installed and applied, the electronic output signals from pressure transmitters are relatively immune to the effects of distance and the conditions in the spaces they traverse. Thus, the best input system is provided by locating building static pressure sensors (as well as other sensing elements) near the signal they are measuring and allowing them to act as signal transmitters as well as signal converters. Additional information regarding signal transmitters and their impacts on input accuracy can be found in the Control System Design Guide, Chapter 3 Selection and Installation of Control and Monitoring Points.

2 A Single Sensing Tube Connecting the Pressure Sensor Input to Multiple Spaces at Different Pressures does not Provide an Input that is the Average of the Different Space Pressures Often, it is desirable to measure several different building pressures and use the average value as an input or indication of building static pressure. Many times, an effort is made to minimize the cost to accomplish this by installing one static pressure sensor and then connecting its reference ports to multiple spaces. Unfortunately, this results in a situation very similar to what is depicted in Figure 3.26 above. The input created is some function of the different space pressures, but it does not adequately represent the mathematical average. The resulting pressure input is the result of a complex set of flow patterns and relationships set up in the tubing system by the pressure differences in the rooms that the tube is connected to and the pressure drops through the tubing system due to the flows created by the pressure differences. While it may be possible to make the system perform reasonably well based on this information, the signal is not equivalent to the average pressure or any actual pressure but as a potentially non-linear pressure coefficient that is a function of the various pressures.

Approaches that would provide a true indication of average pressure value include:

· Provide a transmitter in each space wired to an individual input and use the system software to average the signals. This approach makes the most information available to the operators and the allows greatest flexibility since the software can be modified to do other things besides average the inputs, like select the highest or lowest if necessary. However, the approach requires an input for each space referenced.

· Provide a transmitter in each space wired to an independent signal-conditioning module that provides an output that is an average of its inputs (See Figure 3.27 for an example). This approach minimizes the number of inputs required but is less flexible because it provides less data and the averaging calculation is fixed in hardware rather than software.

3 The Technique Used to Terminate the Static Pressure Reference Sensing Lines can Impact the Signal Measurement The ends of the sensing lines used to measure static pressures need to be located and arranged so they are immune to the effects of wind, space air motion, duct air flow, and other velocity pressure influences. A 5-mile per hour breeze has a velocity pressure of 0.01 inches w.c. associated with it. This error could be a significant portion of the value registered by a building static pressure probe.

A well-designed outdoor air static pressure probe will provide a reference point that is relatively immune to this type of effect if properly located. Location is an important factor because a probe that is located in a low-pressure zone created by the motion of the air around the building will still provide a bad reference and cause operational problems. Finding a good location for a static probe can be tricky and takes some time during the commissioning process. The benefits are usually worth in the effort in terms of more stable system performance.

There are a wide variety of probes manufactured by different companies targeted at measuring different static pressures including outdoor air static, duct static and space static. Figure 3.28 illustrates several designs by one manufacturer. In general, the devices make use of baffles and pulsation dampening chambers to eliminate some of the spurious effects associated with air motion.

4 Provisions that Filter Out Minor, Random, Static Pressure Variations can Often Provide Significant Performance Improvements Many of the electronic pressure sensing technologies used to measure building static pressure are capable of detecting and reporting very small and rapid pressure changes. This can be both a good thing and a bad thing for the commissioning agent, as can be seen from the previous side bar. Detecting and attempting to respond to the minor variations in building static pressure associated with a door opening every once-in-a-while may not be necessary and could create more problems than it solves. On the other hand, responding to a small but sustained loss of building pressure created by a door that is blocked open for some reason might be highly desirable in terms of controlling building pressure relationships and comfort.

Because the static pressure signals measured in air handling applications are relatively small and easily influenced by other factors such as air motion, they tend to be very “noisy” when viewed in terms of the desired average value. There are a variety of ways to mitigate this signal noise. Implementing one of the following techniques to correct a noisy signal identified in a functional test may yield very significant improvements in the performance of the system.

· Many modern DDC control system can apply a filtering factor directly to any analog input on the system. If available, this approach is probably the easiest to implement.

· If a standard filtering factor is not available, most systems will allow the user to write some software that averages the input based on a moving window of some sort. The output of this calculation can then be used as the actual input to the control or monitoring process.

· Resistor/capacitor and resistor/inductor networks can be wired across the input connections on most system to stabilize a noisy signal. The companies that manufacture the signal conditioning modules shown in Figure 3.27 also typically manufacture modules that will provide this function as a packaged solution.

· Simply creating a pulsation chamber with a restricted inlet in the input sensing lines to the transmitter can often solve a noisy signal problem[9]. Any pressure change that occurs must first change the pressure in the entire volume of the pulsation chamber before it actually impacts the reading at the transmitter. The restriction in the input prevents minor pulses from affecting the chamber pressure significantly but will allow a sustained pressure change to eventually be reflected in the chamber, and thus, at the input to the transmitter.

3.6.4.5. Flow Sensors

In economizer systems, flow sensors are typically encountered in the minimum outdoor airflow regulation portion of the system. The setpoints of the loops controlling outdoor air flow may be optimized by a variety of factors including CO₂ level, schedule, occupancy; however, the primary control loop is typically based on measuring and controlling for a flow. As a result, verification of flow measurements and flow sensing equipment often must be accomplished as a part of testing the operation and performance of the mixing section of an air handling system.

A general discussion of flow measurement can be found in the Control System Design Guide, Chapter 3: Selection and Installation of Control and Monitoring Points. In addition, since many flow measurements are based on velocity pressures, many of the guidelines in the preceding section on static pressure measurement also apply. The following additional points should be considered when commissioning flow measurement systems for minimum outdoor air control.

The round devices on the plenum are air control valves that include an averaging type flow measurement ring as a part of the assembly. (Image courtesy of the Trane web site)

1 Obtaining a Uniform Velocity Profile Under All Operating Conditions is Critical to Ensuring Good Control This point is true in any flow measurement application, but is often neglected in the sensing arrangements used for measuring the minimum outdoor air flow into a unit. In addition, the close-coupled configuration of typical intake systems can often make it difficult to find the physical space to meet the typical rules associated with establishing a uniform velocity profile. Finally, the modulating operation of the maximum outdoor air dampers can have a variable impact on the performance of the minimum outdoor air flow sensing and control system depending on the current operating point of the economizer.

Some manufacturers now offer packaged products that incorporate flow measuring and control elements to accurately measure and control outdoor air flow using technology similar to what is applied on a VAV terminal unit. Figure 3.29 is one example of this type of product.

2 The Pressure Relationships Between the Outside, the Mixing Plenum, and the Return System Play a Critical Role in System Performance The flow through the minimum outdoor air flow control system is driven by the pressure difference that exists between the outdoors and the mixing plenum. On economizer systems, this pressure can vary as a function of the position of the economizer dampers, especially if they have not been sized for linear performance. If the system is a VAV system, then the pressure relationship can also vary with system flow rate. All of these issues need to be considered when selecting and commissioning the flow control system and its sensing elements. If the pressure required to drive the necessary flow through the sensing system is not sufficient, then the system will not be able to achieve the required flow rate, no matter what its nameplate rating is.

3 When the System is Operating on the Economizer Cycle at a Flow Rate Above its Minimum Outdoor Air Flow Requirement, Then the Operation of the Minimum Outdoor Air Flow Control System May Be Irrelevant Operation of the minimum outdoor air flow measurement and control system is critical for controlling indoor air quality when the system is operating in the minimum outdoor air mode. The system must be arranged and controlled in a manner that prevents it from delivering less than the minimum outdoor airflow requirement. Usually, this occurs during extreme winter and summer weather.

If the system is using significant amounts of outdoor air for an economizer cycle, then the performance of the minimum outdoor air flow control system may not be relevant since the system is bringing in outdoor air in excess of the minimum flow requirements. Thus, the functional testing of this system can often be targeted at ensuring its performance during extreme conditions with little if any attention paid to its performance under the economizer cycle, unless there is some requirement for documenting the minimum outdoor air flow at all times.

3.6.5. Blank-off plates

Most control dampers (like control valves) will be smaller than the duct they are in. Usually, this means that blank-off plates are installed to make up the difference. Two important factors need to be considered with regard to blank-off plates.

· The location and configuration of the blank-off plate/damper assembly needs to promote mixing.

· The pressure loss associated with the reduced damper area relative to the duct need to be taken into account in the sizing of the fan system. Some manufacturers include a factor for this in their pressure drop tables. Others allow the designer to determine the impact through an independent assessment.

Blank-off plates are also used in troubleshooting and retro-commissioning applications to deactivate portions of an existing, oversized damper to improve its performance characteristics. Disconnecting blade linkages to deactivate the blades and then screwing the blades in the closed position can achieve a similar effect. These procedures will be discussed in more detail in the following section.

3.6.6. Air Blenders and Baffle Plates

In a mixing plenum with dampers that has been optimized to promote mixing, there will be some distance required for mixing to occur. As a general rule, there should be at least 3 or 4 feet between the downstream side of the mixing dampers and the location in the unit where complete mixing needs to have occurred (usually the first coil in the unit). There are a variety of ways to accomplish full mixing including:

· Use the space required for filter banks and operator access to provide the necessary distance for mixing. In some instance, the filters can help to promote the mixing process due to the turbulence they create as the air flows through them and because they tend to promote a uniform velocity profile.

· Locate the mixing box some distance from the air handling unit so there is duct work and, ideally, an elbow or two between the outside and return dampers and the components of the air handler.

There are many instances where there simply is not enough space allocated in the mechanical room to provide the distance necessary for good mixing to occur. In these instances, air blenders or baffle plates can provide a mechanism to enhance the mixing of the outside and return air flow in a shorter distance.

The picture to the left is an air blender module. The picture to the right is a similar module under test, demonstrating its ability to mix the two colored air streams.

Air blenders are manufactured devices that are designed to promote mixing and eliminate stratification in very short distances. A well-designed mixing box with properly sized and arranged dampers and sufficient distance for mixing can often achieve a 3-5°F spread between the warmest and coldest spot in the plenum discharge. For many HVAC applications, this level of mixing will be satisfactory. But if more uniform temperatures are required, or the available space does not provide sufficient distance for mixing, air blenders can be applied to meet or possibly exceed this level of performance. On existing projects with stratification problems, air blenders can be retrofitted into the system as a corrective measure, assuming the system has sufficient static pressure capability to handle the added pressure loss.

On many occasions, commissioning agents, Owners, facilities engineers and other people concerned with system operations find themselves faced with a system that has severe stratification problems that cause nuisance freezestat trips (or worse) without the luxury of time or budget necessary to allow air blenders to be retrofitted into the system. In these situations, baffle plates can offer a low cost, easily improvised, field solution to stratification problems that would otherwise be handled with air blenders. Baffle plates are sheet metal baffles installed downstream of the mixing dampers in an arrangement that will cause the airflow through the unit to change direction and/or induce turbulence. These directional changes can often improve the mixed condition enough to eliminate the problem.

Baffle plates can be solid metal or perforated metal. The solid metal plates rely on directional changes to achieve the desired effect. The perforated metal plates cause some directional changes but also induce mixing due to the jet effect created by the air that passes through the holes in the plate. If sufficient free area is maintained so that the average velocity in the airflow path is below 800-1,000 feet per minute, baffle plates can produce a significant improvement with little increase in pressure drop.

Baffle plates can be especially effective when combined with other improvements like adding a mixed air low limit cycle and improving the performance of oversized mixing dampers by disabling damper blades.

On the VAV unit in Figure 3.31, a stratification problem that was causing nuisance freezestat trips when outdoor air temperatures dropped below the mid 20°’s F was solved by blanking off one of two damper sections on the outdoor air and return air dampers and adding a baffle plate ahead of the filter bank. Blanking off one damper section increased the velocity through the remaining section, improving its linearity and adding momentum to the air stream, which promoted mixing. The baffle plate forced the air to turn and then re-expand at the filter bank, which further improved the mixing. The velocities through the area below the baffle plate were still well below 1,000 fpm, thus little additional pressure drop was added. Increasing the damper velocity through the blank-off plates added some pressure drop, but running the motor into its service factor on design days and changing the filters slightly sooner accommodated this. The VAV system spent very little time at full capacity, thus running into the service factor had minimal impact on the motor life expectancy.

[1] To gain some insight into this, consider two parallel blade dampers, one sized for an a = 10 and the other sized for an a = 200. By looking at the curves in Figure 3.10, it can be seen that when the first damper is at 20% stroke, the flow through it will be approximately 28% of maximum. While not perfectly linear, this control is much better than the second damper, which at 20% stroke allows nearly 77% of maximum flow. The over-sizing of the damper causes the controller to have to control over a limited portion of its span, making it difficult to achieve tight, stable control.

[2] This sequence would be typically be overridden when outdoor conditions were no longer suitable for free cooling to return the system to minimum outdoor air.

[3] See the Chapter 13 Distributionfor a discussion of air hammer and its effects.

[4] Note that there still may be a maximum allowable actuation time for life safety related functions like smoke or fire dampers that is dictated by the governing codes.

[5] Some manufacturers have developed devices that combine an electronic to pneumatic signal converter in the same package as a pneumatic positioning relay.

[6] Mass flow rate will in fact influence the sensed temperature to some extent in that higher mass flow rates will have better convective heat transfer coefficients between the sensing element and the air stream. Sensing elements in air streams with higher mass flow rates (and thus higher velocities) will generally display a quicker response to a change and a closer approach to the actual air stream temperature. These effects are relatively insignificant in the context of the accuracy of the temperature measurements made in a mixed air plenum. The improved response characteristic associated with the higher flow rates can make the control loop easier to tune because time lags are reduced.

[7] Four inches in diameter is a fairly typical size for many of the pneumatic pressure transmitters in the commercial market. One tenth of an inch water column is probably the high end of the control range for building static pressure; i.e. you will control for that pressure or less and need to be able to detect and respond to changes that are at least one order of magnitude smaller (one one-hundredth of an inch water column).

[8] Hot wire anemometers and other thermally based flow measuring technologies are very sensitive to low flow rates. They work by measuring the cooling effect of airflow over a heated wire. The amount of cooling is directly related to velocity and thus to flow. By measuring the energy used to heat the wire, the devices can detect and accurately report very low flow rates.

[9] There are commercially available products that perform this function, but one can be easily fabricated in the field from a 6” long piece of 1/2” or 3/4” PVC pipe, a couple of end caps, an in line pneumatic control system restrictor fitting, a barbed fitting. and some 10 minute epoxy. Since the pressures are in terms of inches w.c., the pressure rating of the assembly is not critical. The caps are simply glued to the end of the tube and drilled to accommodate the inline restrictor fitting at one end and the barbed fitting at the other. The fittings are then epoxied in place. The pulsation chamber is then simply installed in the line ahead of the transmitter.