Take a snapshot of the indoor climate

Take a snapshot of the indoor climateFigure 1: Thermal measurement set-up at a work station.PMV stands for Predicted Mean Vote. Standard ISO 7730 describes how to calculate PMV andpresent the interpretation of the number values found. Input for the calculations:Operative temperature, air velocity, air humidity, insula-tion value of clothing and the person’s activity level. Result:A PMV value between -3 and 3 indicating the expectedlevel of satisfaction with the thermal environment. Seethe comfort scale.Measurement set-upAll of the probes are mounted on a stand and the wholeWhile poor thermal comfort is just one of many indoorclimate problems, it generates the greatest number ofworkers’ complaints. If the indoor climate is already onthe agenda or if preventing indoor climate problems ispart of company policy, then you should consider aninvestment in a snapshot of the thermal environment. It takes only 3 days to take a snapshot of the thermalindoor climate in an office building with 80 work stations.Two days are used for taking measurements and oneday for the report generation. A snapshot gives a factualbasis for discussionss regarding the acceptability of indoor thermal comfort.The methodThe starting point for an assessment of the indoor climate is the international standard ISO 7730 /ref. 3/.This standard is also accepted as a national standard inmany countries.The standard is used when calculating a person’smost probable vote on a comfort scale (PMV) and therisk that the person will be dissatisfied with the thermalenvironment (PPD). In addition, using ISO 7730 one canestimate the percentage of people who will be dissatisfied with drafts (DR), vertical gradients in the air temperature and asymmetric temperature radiation. The most common complaints include: ”It’s too cold inthe room”, ”It’s too hot in the room” and ”It’s drafty”. Ourinvestigation was formulated to assess and quantifythese three major complaints. According to ISO 7730,the parameters we needed to measure were PMV, PPDand DR. The statistical analysis is based on a large number ofindependent measurements. One measurement wastaken at each work station, with locations systematicallyverified from the company directory. Procedure was as follows:1. Persons whose work stations were to be meas-ured were notified a couple of days in advance.2. Those chosen were asked to complete a question-naire prior to removing themselves form the area forthe study.3. The chair at the work station was replaced by themeasurement station. See Figure 1.4. The position of the work station was keyed into themeasurement set-up and measurement started.5. Measurement was completed after 5 minutes andthe measuring equipment was moved to the nextwork station on the list.measurement set-up is thus easy to transport from oneplace of measurement to another. The measurement set-up is arranged so that it can measure PMV/PPD anddrafts at the ankles and neck simultaneously. The set-upconsists of 3 omni-directional air velocity sensors, an operative temperature sensor and a humidity sensormounted on the stand.All of the probes are connected to the ComfortSensemain unit, which contains the electronics for the probes aswell as an A/D converter and USB 2.0 interface. From thePC the operator can communicate with the whole set-upat one time. The measurement is set up using a PC and applicationsoftware. The measurement itself is defined in a template, so the operator only needs to key in the identityof the measurement position and then press the start button at each measurement position.Measurement data is transferred to the PC where theoperator can monitor on-line data from the individual probes. The operator can also graphically follow the progress in the lapsed time.QuestionnaireA simple questionnaire was issued at every work station.See figure 3. If the person was not at the work station,the measurement operator filled in the employee’s nameand the last 2 lines of the questionnaire. The question-naire is designed for ease of use and processing.Starting the measurement- Clothing -- Level of activity -- Time of measurement -The above 3 parameters must be determined beforemeasurement begins.Take a snapshot of the indoor climatePMV = 3 Hot2 Warm1 Slightly warm0 Neutral-1 Slightly cool-2 Cool-3 ColdComfort scaleFigure 2: The thermal measurement set-up on its stand. Airspeed is measured at heights of 0.1 and 1.1 m, and operativetemperature, air velocity and humidity are measured at 0.6 mabove the floor.Figure 3: The questionnaire used.All PMV measurements are carried out using the samesetting for clothing - we chose a winter level of 0.85 clo.This closely resembles the average for the place of workand is a level everyone can adapt to. By choosing a uniform level of clothing it is easier forus to compare the thermal indoor climate at differentwork stations. The ability to compare measurements withthe individual questionnaire results, however, is sacrifi-ced. Since the expected vacancy of work stations whenmeaurements begins is approximately 50%, and valuesfor individual levels of clothing cannot be determined insuch cases, our method of choosing a uniform clothinglevel seems to be the most practical.Similarly, PMV measurements are performed with auniform level of activity for the same reason. The level ofactivity is set to 1.3 met. This value is perhaps a little lowconsidering that only 47% of staff were at their work stations when measurement started. An investigationcarried out by Tatsuo Nobe and others /ref. 1/ indicateseven lower occupation times at individual work stations.Chosen times for measurement should be carefullyselected in order to better evaluate the building’s thermal2regulation system. The presence of an external load onthe building in the form of cold or heat in addition to anormal internal heat load is preferred for such an eva-luation. The criteria should be pre-set and the start of the measurements delayed until the criteria are met. The criteria for performing the present measurements werean external temperature below 0°C and normal internalheat load.Take a snapshot of the indoor climateProcessing dataData from the measurements and questionnaire werekeyed into a pre-programmed spreadsheet that produced a result table as shown in Figure 6. Data processing consists primarily of calculations of statisticalvalues and distribution diagrams. Only processing of themeasured PMV values was slightly different.The assessment of the thermal indoor climate shouldcover a complete section of the building with many workstations, but the term PMV/PPD is only formed to assessa single work station. If we simply average the measuredPMVs and PPDs then we lose some information, so astarting point is taken in the distribution of the votes lyingbehind the calculation of both PMV and PPD. Differences between people mean that they do not allvote the same. The votes that, for example, give PMV =0 are distributed statistically as in figure 4. P.O. Fanger/ref.2/ has shown that the votes have a normal distribu-tion with a standard deviation of 0.75 of a vote step. Weuse this knowledge to calculate the predicted distributionof votes for the whole section of the building. ResultsFrom the processed measurement results given in figure6 it can generally be concluded that the thermal indoorclimate in the building measured is good. The estimatednumber of dissatisfied people (PPD) is 6%, which is aacceptable figure, and drafts are not a general problem.Complaints about drafts in some work stations can beexpected, and data results tend to substantiate theclaims. A detailed look at the results provides the follow-ing information:• The temperature in parts of the building (rooms 6,7,9and 10) should be lowered.• There are problems with drafts in about 10% of the work stations (DR > 20%).• The level of thermal comfort throughout the day is reasonably constant.• The employees experience the thermal indoorclimate as rather worse than the measurements show it to be.Differences between measurement and questionnaireWhen the distribution of the comfort votes from the measurement and the questionnaire are compared (seefigure 6), the distribution from the questionnaire is clearlythe broadest. Several theories may help explain this finding. For the study, we assumed that everyone in the build-ing section had the same level of activity and clothing.We know that this is not the case, but we supposed thatthe people would adapt their clothing to their level ofactivity. Figure 5 shows graphically which related metFigure 4: Statistical distribution of votes at PMV = 0. The num-ber of dissatisfied (PPD) is defined as the percentage of votesfalling into the scale values ’hot’, ’warm’, ’cool’ and ’cold’. Here PPD = 5%. This distribution is from P.O. Fanger /ref.2/.Figure 5: Set of (clo, met) values that result in the same PMVvalues.3and clo values result in the same experience of comfortas the values used in the measurement. If the peopleworking in the building did not want to adapt their levelof clothing to the activity or did not have the opportunityto do so, the estimated percentage of dissatisfied employees from the measurement would be too low. Another phenomenon that has arisen in a number ofthe field measurements is differences in expectationsand preferences, see e.g. R.J. de Dear /ref. 4/. If weenter a fully air-conditioned building, our expectations forindoor thermal comfort are different from those in a naturally ventilated building. What we assess as ’warm’in a fully air-conditioned building we assess perhaps asonly ’slightly warm’ in a naturally ventilated building. Thatwe apparently vote differently from situation to situationmeans that there is always a quite large degree ofuncertainty connected with the results of questionnaires. DiscussionThe thermal comfort level at all 85 work stations in asection of a building was measured. The findings suggested that the measured work stations offered “agood indoor thermal climate with a few faults”. The workinvolved was 3½ man-days during working hours and alimited amount of measurement equipment. The question is: ’What was the benefit?’If there are no known problems with the indoor thermal climate in the section of building, and if a measurement period with a reasonable load is chosen,then the measurements give a good picture of the performance of the heating and ventilation system.Other types of assessment and print out from HVACControl Systems cannot provide an assessment of equalquality. A good picture of the indoor thermal climate of thebuilding section can be obtained if the present wintermeasurement is supplemented by a similar summermeasurement under high heat load from the sun. If indoor thermal climate problems under special climaticor load conditions are suspected, supplementary measurements when these occur would be advisable.Literature1. Tatsue Nobe et al: Investigation of seat occupancy rate in office. Paper presented at Roomvent 2002, Copenhagen, September 2002.2. P.O. Fanger: Thermal Comfort, McGraw-Hill Book Company, 1970.3. ISO 7730: Moderate thermal environments - Determination of the PMV and PPD indices and specification of the conditions for thermal comfort, 1994.4. R.J. de Dear et al: Thermal comfort in naturally ventilated buildings: revisions to ASHRAE Standard 55, Energy and Buildings 34, 2002.Take a snapshot of the indoor climate 4Take a snapshot of the indoor climate 550.040.030.020.010.00.0Distribution of comfort votesCold Cool Slightly Neutral Slightly Warm Hotcool warmPredicted from measurements QuestionnairePercentageMeasured PMV atwork-place1.000.500.00-0.50-1.00Measurement timePercentage ofmeasurementsDraft RatePercentageAir temperature difference from ankle to neckMean of measured PMV values in roomRoom 10Room 9Room 8Room 7Room 6Room 5Room 4Room 4Room 3Room 2Room 1-1.00 -0.50 0.00 0.50 1.00Outside climateDate Air temperature Air speed Direction Sun13/12 -2 3 m/s Northeast No16/12 -1 3 m/s East No17/12 -1 3 m/s East SomeComfort:General thermal comfort Min Mean MaxPMV value from measurements -0.35 0.23 0.61Voting from questionnaire -3.00 0.52 3.00PPD from questionnaire 32PPD from measurement 6Local thermal comfort Mean MaxPercentage dissatisfied with draught measured as DR 7 29Percentage dissatisfied with draught (questionnaire) 18Percentage dissatisfied with temperature gradient 0Indoor climate Place of measurement: Dantec Dynamics, Tonsbakken 16-18, Skovlunde, DenmarkMeasurement carried out byDantec DynamicsData file:File: DMS-21587on Dantec Dynamics’ netclo value used when 0.85measuring PMV / PPDmet value used when 1.30measuring PMV / PPDFigure 6: The final report of the indoor thermal climate at Dantec Dynamics A/S.Dantec Dynamics, a Nova Instruments companyThe specifications in this document are subject to change without notice. Publication no. 303_v3www.dantecdynamics.com