Measurement Planning Setting the Goal

The planning of measurements is the first consideration to obtain infor­mation by the measurement approach. Why is it essential to make plans be­fore any action is taken? Could one not just take the instruments and carry out the monitoring? In very simple situations this approach might provide a satisfactory result, but it could result in failure as well. In complicated situa­tions failure, in terms of missing information, would be likely. Hence in order to obtain a sufficient quantity of high-quality information and to avoid the need to repeat any measurement or monitoring, and thus to save time and ef­fort, the planning of measurements is essential.

When setting the goals of a measurement project, it has to be asked, What ex­actly has to be determined? What are the final quantities required and what is the inaccuracy that can be tolerated in these quantities? Only when these factors are known can an analysis be made, where the quantities to be measured and the mea­surement accuracy of each quantity are defined. This analysis is based on the mea­surement method selected, and on the computation of measurement uncertainties. Usually the analysis of measurement uncertainties is made after monitoring; how­ever, making it beforehand is part of good planning practice. This approach en­sures that the correct information with the desired accuracy is achieved. Selection Methods

The measurement is always carried out according to a given method. The measurement of a single quantity may be simple, but complications may arise when measuring several quantities and the results require special consideration.

The goals and the final quantities to be obtained determine the method to be used. As a first approach, it is always worth checking if a documented and proven method exists. Standardized methods are often the best. They are devel­oped by experts in the field and are usually of a high quality and are well docu­mented. Frequently no standards can be applied and generally known methods have to be relied on. Sometimes situations occur when the whole of the mea ­surement procedure has to be tailored for the specific situation. This is a very demanding task and considerable experience is required. The Phenomena and the Quantities

The features of the phenomena to be investigated and the quantities to be mea ­sured should always be taken into account at the planning stage. Static phenom­ena, where the measured quantities do not change with time, are extremely rare in industrial ventilation applications. Usually everything moves and changes with time and from place to place. For this reason, it seldom is adequate to take only one sample of the measured quantity. An example would be mean air velocity mea­surements in an occupied space. The air velocity is changing all the time, and to ob­tain a value for a local mean velocity, several readings during a time interval have to be taken. Besides temporal variations, the velocity also has different values in different places. So local mean values should be measured in several places in the occupied zone in order to determine the mean value in the whole space.

Different kinds of disturbances, which distort the collected information, can also influence the measurements under examination. For example, opening a large door in an industrial hall will immediately influence the pressure distribu­tion in the whole building, causing changes in airflows and indoor temperatures as well. If the person carrying out the measurement ignores this fact the wrong information may be obtained. Such a disturbance is called an external distur­bance. An internal disturbance is due to the measurement itself, when the mea­sured quantity is changed by the existence of the measurement arrangements. Reducing the supply or exhaust airflow with a flow meter or changing the sur­face temperature with the temperature probe are internal disturbances.

Most of the measured quantities, like temperatures, velocities, and concentra­tions, vary from place to place. Hence, it is important to consider in advance the correct measuring positions. The probe location is different depending on the ap­plication, such as to determine the thermal boundary conditions for GFD computa­tions or to determine the reasons for complaints of thermal discomfort in the occupied zone. Another important factor to consider is the time when the measure­ments are taken, which can be critical in some cases. The riming normally relates to the state of the processes, the running mode of the HVAC systems, or the weather conditions. All these influence each other, the flow patterns, temperatures, energy’ use, and other measured quantities in the industrial environment. If the task is to determine the contaminant exposure of a worker; it might not be sufficient to mon­itor during the heating season, since the resulting flow pattern and associated con­centration levels might be totally different during the cooling season. Instrument Selection

The selection of instrumentation for a specific task is an important part of measurement planning. The instrumentation consists of measuring probes, the meters to convert the signals from the probes, and some intelligence to guide the

Monitoring and auxiliary equipment. The performance of the instrumentation depends on the quality of the probes and meters; hence these components must be selected according to the information required. The instrument measuring range has to be consistent with the scale that is of interest. The inacairacv ot the probes and meters should fulfill the demand of the preliminary error analysis. The equipment drift must be known in order to make necessary calibration-, during longer monitoring periods. The dynamics of the probes should be si lected to allow rapid changes of the measured quantities to be followed, if neces­sary. If, e. g., the velocity probe is slow in following the fluctuations of the airflow, the resulting value of turbulence intensity could be highly erroneous.

Organizing the measurements is also one part of planning and is closely tied to instrument selection. In a simple case there is normally no problem with organizing the work. The measurement is carried out manually. How­ever, in more complicated cases, such as determining the energy performance of a large industrial ventilation system, numerous monitoring points are re­quired. Manual measurement would be time-consuming, unpractical, and even impossible to carry out due to the constantly changing parameters. In this instance automation of the measurements is essential. The arranging of automatic recording equipment is time-consuming. The level of automatic re­cording and control has to be optimized so that the required information is achieved with the minimum time, effort, and cost.

One aspect related to automatic measurements is how the automation is achieved: it may be either centralized or decentralized. The centralized method is based on a data acquisition system consisting of a selection of measurement cards or units connected to a computer. The probes are placed in the moni­tored process and connected to meters with measurement cables. Provided the distance between the farthest probes is not too long, this approach is suitable. However, if quantities are to be measured simultaneously in different zones of a large building, the centralized approach may be impractical due to the very long wiring distances. In this case the decentralized approach, where several smaller data acquisition units are used, would be better. A wide selection of small data acquisition units are available and some of them are very powerful and high-quality instruments. Time Schedule and Other Aspects

It is essential to plan the time schedule of a measurement project-—the more detailed, the more important it is to carry out the measurements in a specified time. The time schedule is achieved by dividing the whole project into smaller tasks and assessing the time required to carry out these subtasks. A rough divi­sion ot a measurement project could contain the following elements:

Measurement planning,

Installing the instrumentation,

Testing of the instrumentation,

Taking measurements,

Analyzing results, and

Reporting and other documentation.

The testing stage of complicated situations can take a considerable time and, if underestimated, will delay the project. In simple routine measurements

Or when past experience of a specific task exists, the time planning can be con­trolled. However, when a large measurement project containing new elements is planned, special attention is required in assessing the time allocation. Sev­eral problems may occur. The support of the companies delivering the instru­mentation is essential in case of equipment failure or breakage. A good "rule of thumb” is to multiply the initial time estimate by a factor of 2. Analyzing and Presenting Results

It is not good practice to carry out the measurement analysis at the end of a measuring session. Some source of error not detected during the instrument test­ing period may influence the results and require repeating all measurements. To avoid such a situation, the first results analysis should be carried out during the measurement period and, if problems are noted, necessary action can be taken immediately. In case of long-term monitoring, the preliminary analysis has to be continuous to check the data quality and the equipment condition.

The final treatment of the data is then carried out when all results are avail­able. How this treatment and analysis is achieved depends on many factors. With regards to presenting the results, few general principles can be given. One should always consider the status of the reader of the report—the target group—and adjust the level of presentation accordingly. If large quantities of data are to be presented, a table is a poor choice. Graphs and figures are more illustrative than tables. The data can often be treated statistically with mean val­ues and standard deviations shown to the reader. The use of nondimensional groups of quantities, like Reynolds number or Biot number, is an ideal way of presenting results. Error analysis should always accompany the general analysis and treatment of data, showing the reliability of the measurements. Documentation

A report describing the measuring arrangements, instrumentation, and main results and findings is often the only information of the measurements carried out. With simple or routine measurements this is usually a sufficient level of documen­tation. In larger research and development activities or more scientific projects it is wise to expand the documentation to cover other material. Later on there may be a need to check some of the initial results, to treat the raw data according to a new idea, or give the data to somebody else for his or her own purposes. For this rea­son the raw data should be organized, labeled, and stored so that even after sev­eral years it can be reused. For the same reason, a sufficient amount of supplementary information like dates, places, persons, weather, running mode of ventilation and other systems, and status of doors and windows should be docu­mented and stored. It often is wise to photograph some details of the measure­ment arrangements, for reference. All this documentation can be stored on some powerful storage device, such as a CD-ROM disk of adequate storage capacity.