Elsevier

Building and Environment

Volume 48, February 2012, Pages 77-83
Building and Environment

Development of simplified in-situ fan curve measurement method using the manufacturers fan curve

https://doi.org/10.1016/j.buildenv.2011.08.017Get rights and content

Abstract

A simplified in-situ fan curve measurement procedure has been developed using the manufacturers fan curve and one point (air flow and fan head) measurement. This paper presents the background theory, methodology, error analysis and step-by-step procedure developed for the practitioners. This in-situ method has been experimentally proven in full-scale air handling unit (AHU) systems. The results show that the fan curve identified using this simplified approach agrees with the fan curve identified using the point-by-point direct measurement method. Both the error analysis and the experiment show that the generated in-situ fan curve with least system resistance most closely matches the measured in-situ curve (cv-RMSE = 4.7%). The differences of the fan heads predicted by the fan curves are within the experimental error range.

Highlights

► The measured fan curve is different. ► A simplified fan curve measurement procedure developed. ► This method has been proven.

Introduction

It is very important to find an accurate and cost-effective way to measure air flow in the HVAC industry. Accurate measurement of air flow in all ranges is necessary and the only assurance is to maintain positive building pressure through volumetric fan tracking [1]. Air flow measurement techniques are also necessary to determine if there is enough ventilation air to the building [2]. The typical air flow rate measurement methods are:

A thermal anemometer—or hot-wire anemometer—has been widely used to measure air velocities in buildings. The air flow can be measured accurately when an adequate number of sensors are used across a duct cross section [3], [4]. However, the installation of thermal anemometers to measure air flow continuously is not practical because of the limitations:

  • ASTM Standard D 3464-75, “Standard Test Method for Average Velocity in a duct using a Thermal Anemometer,” specifies 4 to 20 sampling points, depending on the size of the duct. If we want to install sensors to measure the velocity continuously, the cost of installing an adequate numbers of sensors may be prohibitive [5], [6].

  • The unidirectional sensor must be carefully aligned in the air stream (typically to within ±20° rotation) to achieve accurate results. An unexpected change may occur to the sensor’s direction in the ductwork when we need to continuously measure the air flow rate. This will make the measurement inaccurate [6].

  • The velocity sensor must be kept clean because contaminant buildup requires periodic calibration.

For the above reasons, thermal anemometers have not been used in the volumetric fan tracking fan control [3], [4].

Pitot-static tubes are often used in air flow measurement for terminal boxes. The energy management control system (EMCS) can then sum all the air flow for the terminal boxes. The main drawback of using pitot-static tubes is that the accuracy falls off at the low end of the range [7], and the terminal box is frequently running at that range [8], [9].

To increase the accuracy, it is recommended to measure the air flow at the location with the highest velocity [10], specifically the fan inlet. Fan inlet technology can measure the air flow at the fan inlet mounted in the intake bell of the fan. However, the air flow profile in the fan inlet varies significantly with the total air flow, which results in high uncertainty and difficulty in measurement. An Air Movement and Control Association (AMCA) document (203-90) demonstrated six typical velocity pressure distributions in the measurement plane, which proves that the air flow velocity profile may produce big measurement error at fan inlet [11], [12]. This is also one of the reasons that air flow measurement at the lab and in the field for fan performance has been never conducted according to AMCA publications at the fan inlet. Therefore, this method cannot provide the accuracy required for volumetric tracking.

One of the most common methods to measure the air flow is using flow stations in the main supply duct and in the main return duct. The return fan speed is controlled by comparing the flow rates in the supply and return ducts. The fan is modulated to keep this difference at a constant set point, ensuring a constant exfiltration rate (assuming exhaust as a constant). For accuracy within 5%–10%, the air flow measurement station requires a straight duct for 6 to 10 duct diameters upstream and 3 duct diameters downstream [13]. Unfortunately, there are very few systems that have such duct runs in the main supply and return ducts [1], [14]. Moreover, the air dynamic head varies proportionally to the square of the air velocity. When the velocity decreased to a lower range, the pressure transducer often cannot provide adequate accuracy for the dynamic head measurement.

Fan air flow measurement continues to be challenging in HVAC systems. The fan performance curve can be represented as a multiple order polynomial equation. The in-situ method proposed in this paper makes use of fan performance curves to predict air flow. Under full speed, the fan curve equation can be represented as a second-order polynomial equation:H=a0+a1Q+a2Q2where H is a fan head [Pa] under full fan speed, Q is an air flow rate [m3/s] under full fan speed, and a0, a1 and a2 are fan curve coefficients.

Equation (1) works well at the normal operating region for most fans in AHUs. If the fan runs under partial speed by the variable frequency drive (VFD), the fan curve can be represented as Equation (2), which is derived by the fan lawsH=(ρρd)(a0ω¯2+a1Qω¯+a2Q2)where: ϖ is a current fan speed over 100% fan speed, Q’ is a calculated air flow rate under partial speed of the fan, and ρ/ρd: Air density ratio- the actual air density divided by the air density at the design condition. Air density is the mass per unit volume of the air.

The air flow rate at any measured fan head and fan speed can be obtained by solving Equation (2). The actual air flow rate is calculated by Equation (3).Q=a1ω¯a12ω¯24a2[a0ω¯2H(ρd/ρ)]2a2

Based on the above equations, an air flow control, VSD volumetric tracking (VSDVT) or fan airflow station (FAS) has been developed by Liu [16]. The fan air flow station includes a differential pressure transducer and fan speed transducer. It determines the air flow using the in-situ fan curve based on measured fan speed, fan head, and the pre-determined relationship of fan head and fan air flow (fan curve) under a given fan speed.

The fan speed transducer may be replaced by the control system command to the VFD. The fan air flow station can be implemented using a typical energy management and control system (EMCS). Fig. 1 shows the schematic diagram of the measurement set-up. The theoretical model of the fan air flow station has been experimentally tested, and it was found that the model closely agreed with the experimental values [14].

It is very important to find an effective way to measure air flow accurately. The accuracy of the fan air flow station primarily depends on the accuracy of the fan curve, because the fan speed and fan head can be measured accurately [1]. Typically, the fan curve under full speed can be expressed using second-order polynomial equations (a0, a1 and a2). The fan curve can be directly obtained from the manufacturer. However, the validity of the manufacturers curve is impacted by the actual air density, actual fan installation configuration and the actual location of the sensor for the fan head measurement. According to Liu’s study [1], the fan curve identified in a full-scale air handling unit (AHU) system has the same pattern as the design fan curve. However, the fan head projected using the in-situ measured fan curve is significantly less than the fan head predicated by the manufacturers curve. The differences are much larger than the possible measurement error. Using manufacturers fan curve could overestimate the fan air flow by 15% according to the measurement results. Stein and Hydeman [15] also discussed about the difference between the manufacturers fan curve and in-situ measured curve. Their research shows that the fan system model exceeds measured fan system energy by approximately 36%.

Therefore, an in-situ measurement is required. However, an accurate air flow measurement in-situ takes significant effort. An in-situ measurement method has been developed to measure the fan curve without interrupting normal system operations [1]. The measurements are conducted under partial fan speed conditions. The in-situ fan curve is based on the trended points at full speed, which were obtained from the projection of the measured points at partial speed. To get accurate fan curve, at least 20 measurements are recommended at different system resistances [1]. Otherwise, the measurement error may cause significant inconsistencies of the regression model.

The goal of this paper is to develop a method and procedure to generate the in-situ fan curve based on the manufacturers fan curve and one point measurement (air flow and fan head). The measurement procedure and verification are introduced and discussed in this paper.

Section snippets

Difference between lab test and in-situ fan performance

First, we should take a look at how fans are tested and catalogued. Most fans available in today’s market bear AMCA Certified Rating Seals [17]. The test procedures in the AMCA publication will be discussed below.

Air density adjustment

Our goal in this section is to obtain the air density ratio of the supply air (for supply fan) and the return air (for return fan). In our building pressure control application, the air densities we are interested in are the supply air density after the cooling coil and return air density before the fan. Therefore, the in-situ fan curve for supply fan should be measured downstream of the fan (after the cooling coil). The in-situ fan curve for the return fan should be adjusted to the inlet

Experiments

Experiments were conducted on a full-scale AHU to validate the theory and to compare the measurement results to the design fan curve. The unit is one of the two main AHUs serving a 22,947 m2 office building in Omaha, Nebraska. The design supply air flow was 66 m3/s). The design fan speed was 763 rpm. The design fan head was 1057 Pa. The in-situ fan curve was measured following Liu’s developed procedure [1].

Step 1: Obtain the manufacturers fan curve. Fig. 4 shows the dimensionless manufacturers

Conclusions

An in-situ measurement method has been developed to generate the fan curve using the manufacturers fan curve. The in-situ method can simplify air flow measurement if the manufacturers fan curve is available.

Error analysis shows that the less system resistance (more airflow) for the one point measurement, the higher accuracy the practitioners can achieve. The in-situ method has been experimentally validated in full-scale AHU systems. The results show that the fan curve identified using this

References (27)

  • Y. Cho et al.

    Minimum air flow reset of single duct VAV terminal boxes

    Building and Environment

    (September 2009)
  • Y. Cho et al.

    Correlation between minimum air flow and discharge air temperature

    Building and Environment

    (July 2010)
  • M. Liu et al.

    Development of in situ fan curve measurement for VAV AHU systems

    Journal of Solar Energy Engineering

    (2005)
  • J. McWilliams

    Review of air flow measurement techniques

    (2002)
  • H. Mueller et al.

    The Role of ventilation

  • M. Karimpanah et al.
  • ASTM Standard D 3464-75, “Standard test method for average velocity in a duct using a thermal...
  • D. Schwenk

    Air flow measurement accuracy, US Army Corps of Engineers, Construction Engineering research Laboratories

    (1998)
  • R. Klopfenstein

    Air velocity and flow measurement using a pitot tube

    ISA Transactions

    (September 1998)
  • J.P. Kettler

    Minimum ventilation control for VAV systems: fan tracking versus workable solutions

    ASHRAE Transactions

    (1995)
  • K. Wu et al.

    Numerical analysis of the internal turbulent flow of a diagonal fan including an inlet box of complex geometry and a rotating impeller

    Journal of Engineering

    (1998-03)
  • F. Liu et al.

    Numerical analysis of internal flow effect from inlet flow conditions in large axial fan

    Journal of Engineering Thermophysics

    (2006)
  • NEBB

    Testing adjusting balancing manual of technicians

    (1986)
  • Cited by (19)

    • A novel air flowrate control method based on terminal damper opening prediction in multi-zone VAV system

      2023, Energy
      Citation Excerpt :

      The amount of sample data and calculation will increase exponentially with the increase of the number of terminals, which have certain requirements for the calculation ability of controllers. This paper introduced a novel topological representation method of duct loop and several equipment models [27–29] to reduce the system nonlinearity and adapt to the application scenario of complex duct network. With this innovation, this paper proposed a novel damper opening prediction algorithm under demand air flowrate, and a novel air flowrate control method based on this prediction algorithm, which only needs a small amount of sample data and computing power to achieve the high-precision prediction of damper openings.

    • Rule-based fuzzy control method for static pressure reset using improved Mamdani model in VAV systems

      2019, Journal of Building Engineering
      Citation Excerpt :

      Liu proposed a supply airflow volume reset strategy according to the total supply airflow ratio measured by the fan airflow station [7]. On this basis, Liu further proposed a static pressure reset method based on the ratio between measured value and rated value of supply airflow volume [8]. Samira proposed a variable static pressure control method based on supply air static pressure of every terminal by comparing fan supply air volume of previous time and present time [9].

    • Experimental study on a novel fuzzy control method for static pressure reset based on the maximum damper position feedback

      2015, Energy and Buildings
      Citation Excerpt :

      Liu introduced a reset strategy based on total supply airflow ratio, which is measured by a fan airflow station [3]. Liu also presented a simplified in-situ fan curve measurement procedure using the manufacturers fan curve in order to improve the airflow rate measurement accuracy [4]. Shim compared three different strategies for resetting supply air static pressure through simulations and experiments.

    View all citing articles on Scopus
    View full text