Annex 03 Calculation Methods

7/27/2019 Annex 03 Calculation Methods

http://slidepdf.com/reader/full/annex-03-calculation-methods 1/107



This document is part of the work of the IEA Energy Conservation

in Buildings and Community Systems Programme

Annex I11 Residential Buildings Energy Analysis

Participants in this task:

Belgium, Denmark, Italy. Netherlands, Sweden,

Switzerland, Turkey and the United States

.

.

I

-

ISBN 91-540-3885-5

Swedish Council for Building Research, Stockholm, Sweden

Sp4ngbcrgsTryckcricr A B , Stockholm 1983



PREFACE .................................... 5

BASIS FOR CALCULATION OF ENERGY CONSUMFTION ANDSAVINGS IN RESIDMIAL BUILDINGS ............ 1 1Heat balance of the room air ............... 13

................eat balance for a surface 15

.................eat balance for a window 16

...........eat transfer in walls and slabs 19

..........................nternal loads 23

Solar radiation ......................... 23

References .............................. 25

........INVESTIGATION OF CALCULATION MEITDDS 27

2 . 1 Classification of methods ................. 27

.............. 2 Simulation methods investigated 29

............ 3 Correlation methods investigated 32

.........CALCULATIONS FOR THE ETLANDA HOUSE 37

3 . 1 The house ............................... 37

........................ 2 Calculation cases 39

......3 Predicted energy consumption and savings 41

......CALCULATIONS FOR THE TEKNIKERN BUILDING 51

4 . 1 The building ............................ 51

4 . 2 Predicted energy consumption . . . . . . . . . . . . . . 51

.........NALYSIS OF THE CALCULATION METHODS 53

Predictions of annual energy consumption ..... 53

Predictions of energy conservation ......... 56

Calculation of ventilation losses .......... 58

.........alculation of transmission losses 59

Calculation of internal load .............. 60

Calculation of solar gain through windows .... 60

....Utilization of free heat .................. 64



6 CONCLUSIONS AND RECOlLMENDATIONS ON CALCULATIONMETHODS ................................... 69

7 INFLUFXCE OF INHABITANTS ...................... 7

..................................1 Background 7.............2 Demands from the calculation methods 71

7.3 A Swiss investigation ........................ 73.................4 Conclusions and recommendations 75

APPENDIX A Participating analysts ................. 77................PPENDIX B Building specifications 79

.............PPENDIX C Weather data specifications .I05



PREFACE

1 r n W T I O N . L ENERGY AGENCY

1 n o r d e r t o s t r en g t h en co o p e r a t io n i n t h e v i t a l a r e a o f e ne rg y

po licy , an Agreement on an In te rn a ti o n a l Energy Program wasformulated among a number of indus t r ia l ized countr ies in Novem-

b e r 1974. The International Energy Agency (IEA) was established

a s an autonomous body w i th in th e O rgan izat io n f o r Economic Co-

op er at io n and Development (OECD) t o ad m in ist er t h a t agreement.

Twenty-one co un tr ie s a r e cur re n t l y members of th e IEA, w i th th e

Conmiss ion of the European Conmuni t ies par t ic ipat ing under spe-

c i a l a r r an g em en t .

As one e lement of th e In t e r na t io na l Energy Program, th e P a r t i c i -

p a n t s un d er ta k e c o o p e r at iv e a c t i v i t i e s in energy rese arch , de-

velopment, and dem on strat io n. A number of new and improved en-

e r g y t ech n o l o g ie s which h av e t h e p o t en t i a l o f m aking s i g n i f i -

c a n t c o n G i b u t i o n s t o o u r e n er gy n ee ds were i d e n t i f i e d f o r c o l -

laborative efforts . The IEA Committee on Energy Research and

Development (CRD), a s s i s t e d by a smal l Se cr e t a r i a t s t a f f , co-

or d i na te s th e energy research , development, and demons t ra tion

program.

ENERGY CONSERVATION IN BUILDINGS AND C ~ R R J I T YYSTMS

The International Energy Agency encourages research and devel-

opment i n a number o f a r ea s r e l a t ed t o energy . In one of the se

a r ea s , en e rg y co n s e r v a t i o n i n b u i l d i n g s , t h e IEA i s encouraging

v a r i o u s e x e r c i s e s t o p r e d i c t m ore a c c u r a t e l y t h e e ne rg y u se of

bu i ld ings , inc lu d ing compari son of e x i s t in g computer p rograms,

bu i ld ing moni to ring , compari son of ca l cu la t io n methods, e t c .

The d i f f e rence and s imi la r i t i es among these compar i sons have

t o l d u s much a bo u t t h e s t a t e o f t h e a r t in b u i l d i n g a n a l y s i s

and have l e d t o f u r t h e r IEA encouraged rese arch .



Energy conservation retrofits are in all IE4 countries an im-

owner as well as for the nation as whole, it is vital that cor-

rect evaluations can be made of the energy saving potential of

different retrofits. Since most of the analysis and installa-

tion of these retrofits are not done by architects and/or engin-

eers, there is considerable concernthat the retrofits will not

be properly selected nor perform up to expectations.

reliable calculation methods. The calculated recommendations

then need to be applied in houses and tested for validity.

Recommendations may include new heating, ventilation and air-

conditioning systems, new appliances, new insulating material,

new glazings, etc. Because of the large number of possibilities

of calculation types and recommended retrofits, international

cooperation will accelerate the resolution of the problems in-volved.

The main problem, common to all, is how to generalize experimen-

tal results from time to time, place to place, on the national

level. If this problem is solved findings in one country could

also be used in another, and consequently extensive national re-

search programs could be reduced and rationalized.

In order to generalize experimental results two things are

needed: A reliable technique to observe and measure the conser-

vation effect, and methods for converting these data to other

environments. The main effort in Task I11 has therefore been

made at finding the limitations and the best use of a number of

calculation models that are currently used for predicting the

energy consumption of dwellings (Subtask A), and at collecting

and summarizing guiding principles concerning the design of ex-

periments, instrumentation and measuring techniques (Subtask B).

Finally the results of these two subtasks have been usedwhen

studying the energy conservation effect of a night-temperature

setback in dwellings (Subtask C) .



The pa rt ic ipa nt s i n th e Task are : Belgium, Denmark, I t a ly , th e

Netherlands, Sweden, Switzerland and the United Sta tes.

This report documents work carried out under subtask A of th i s

task . The coopera tiv e work and re su lt in g repor t is described in

the following section.



INTRODUCTION

The ex pl ic it purpose of t h i s work was t o fi nd - i f poss ib le a t

a l l - a sp ec if ic ca lc ul at io n method t o recomend when predi ct-

ing energy conservation in re si de nt ia l building s and how the

influence of inhabit ants should be tre ate d in diff ere nt calcula-

t ions.

The par tic ipa nts have used di ff er en t ca lcu lat ion methods t o pre-

dict energy consumption and energy savings by different conser-

vat ion measures. The re su l t s of th e ca lcul at io ns were compared

with each other and in one case with measurements from a real

building. The influence of inh abit ant s has turned up t o be too

great a task t o handle i n proper d et ai l within t hi s subtask.Some of the demands from the calculation methods and a Swiss in-

vest ig at io n on the energy consumption i n 60 similar houses has

been discussed. I t should also be noticed th at the question of

influence of inhabitants to a great extent has been investi-

gated within the subtask covering guiding principles concerning

design of experiments, inst rumentation and measuring techniques.

The report has been prepared by the Lead Country in subtask A.In th i s preparation t he ana lys ts from the part ic ip at in g coun-

t r i e s have given gre at support by attendi ng ana lyst s' meetings

where different aspects of the report have been discussed in

detail. The draft report has then been reviewed by the annex

participants for approval.

The report i s intended f o r both the non-expert as well a s fo r

th e ex pert on ca lc ul a ti on methods f o r energy consumption inbuildin gs. For th e non-expert t he inte nti on i s to give an intro-

duction t o the f i e l d and give him a po ssi bil i ty t o follow and

use th e more det ai le d analyses exhibi ted here. The ben ef it s of

th i s r epo rt i s a better understanding of some calculation

methods and thcir limitations in dif fe re nt situ ati ons. The re-

po rt could thus be usef ul when se le ct in g a calc ul at io n method

t o be used during genera lizat ion of the r es ul ts from a spe cif ic

energy conservation experiment.



Chapter 1 gives a br ie f intro ducti on t o the physical and numeri-

cal basis used in this field and chapter 2 gives a presentation

of the inves tigate d ca lc ul at io n methods. These were chosen t o

cover th e most s imple a s well a s t he most complex methods in

use. Chapters 3 and 4 describe the calculation cases which have

been used for prediction of energy consumption and conservation.

The calculation results are then analyzed in chapter 5 and chap-

t e r 6 gives th e conclusions and recommendations. Chapter 7

covers the work ca rr ie d out on influence of inhabi tan ts.

The reader should note that this investigation is l imited t o a

few examples of bu ildings and conse rvation measures, thus the

resu l t na tura l ly i s limited and the conclusions and recomenda-

tions should be read with care.

The major conclusions arising from work reported can be s m r -

ized as:

For pre dic tio n of energy conservation, a s well a s en-

ergy consumption, th e s im pl if ied methods seem t o be

a s good as t he more complex computer program, a s long

as t he method handles th e f r ee heat from int er na l and

so la r loads i n a proper way.

As the a cc es si bi li ty of the methods var ies , and di f-

ferent types of retrofits do not always have their

corresponding rep res ent ati on i n th e methods, the

methods t o be used when studying a spe ci fi c r e t r o f i t

have t o be chosen from case t o case.

The work carried out and reported under Annex I of the IE4 im-

plementing agreement on Energy Conservation in Buildings and

Community Systems - Computer modelling of Building performance

should a ls o be reviewed t o ge t a more general review of th e

s t a t e of the a r t of calc ulat i on methods.

Finally, I w i l l express my acknowledgements t o a l l an al ys ts fo r

th ei r valuable contr ibutions t o the report, to M Arne Boysen,

Operating Agent of Annex 111, for his support during the work

with t he re por t and t o th e involved st a f f of the Department of

Building Science for their kindly help.



1 BASIS FOR CALCULATION OF ENERGY CONSUMPTION AND SAVINGS IN

RESIDENTIAL BUILDINGS

This chapter w i l l give a br ie f overview of some ba si s f o r cal -

culation of energy consumption in buildings. These basis w i l l

form a background for the description in the next chapter of

the ca lcu lat ion methods in vesti gate d in th is task . As we re-

s t r i c t ourselves to the heat ing of res ident ial bui ldings , there

w i l l be no discu ssion about a i r conditioning systems.

I f , f o r a moment, we disr ega rd tr an si to ry heat stora ge in the

build ing mass, th e suppli ed energy i s always equal to theen-

ergy losses (see Figure 1.1). This i s a lso the s i tua t ion i f a

longer period i s studied as the storage heat then is neglected

compared t o the t o t a l los ses .

The energy losses are of three types:

1 Transmission los ses

2 Ventilation losses

3 Sewage losses

Transmission los se s depend on the hea t conduction through the

build ing envelope. They can, a s a simpl ifi ca tio n, be determined

by multiplying th e ar ea by the U-value by th e temperature di f -

ference. Ventilation losses depend on the ventilation rate and

the temperature diff eren ce i n the a i r entering and leaving the

building. The sewage losses depend on the amount of water used

inthe building and the temperature diff eren ce of the water en-

tering and leaving the building.

The t o t a l s upplied ne t energy can be divided int o four area s,

namely

For the household (including domestic hot water)

For the heating system

From personsFrom solar radiation



Qvent.

Q

Qtran

Qheat.

Figure 1 . 1 Schematic energy balance of a building.



Energy fr6m persons and solar radiation i s of ten referred t o as

free heat, but sometimes the household energy i s also con-

sidered a s f r ee .

The ca lc ul at io n methods di scussed in th is repor t a re res t r ic ted

t o cal cula tio n of energy los ses by transmission. and ve nti lat ion ,

thus neglecting th e sewage losse s. Furthennore, the calc ul ati on

methods assume schedules f o r the energy from household and per-

sons going t o th e room a i r and inner surfaces . Thus, the se

methods normally deal only with a part of the total energy bal-

ance of a bui ld ing. Each method contains a more or l e s s complex

heat tr an sf er model of the building .

For each par t of a building, such a s room a i r , room surfa ces ,

wall s, et c. , a hea t t ra ns fe r equation can be given. These equa-

tio ns give, as a t o t a l system, the heat balance, or the heat

t ra ns fe r model, of a bui lding. W w i l l here give some examples

of par ts of such models, but have t o leave out a l o t of d et ai ls .

For more de tai led s tu di es of heat tr ans fer , see e.g. F.Kreith,

Principles of Heat Transfer (1965). More applied techniques can

be studied in e.g. ASHRAE, Handbook of Fundamentals, 1981.

1.1 Heat balance of the room a i r

The heat balance of th e room a i r i s of major interest when dis-

cussing heating of a bu ildin g. In most cases, heating is pro-

vided i n o rde r t o maintai n t he room a i r a t a minirmun tempera-

ture level , the se tpoint . If the room a i r tends to exceed t h i s

set po in t, t he supply heat (i n cas e of use of a room thermostat)

w i l l be cut off and the room temperature i s an unborn variable

in the problem. On the o ther hand, i f the room a i r tends t o

drop below th e se tp oi nt , the temperature w i l l be s e t t o t h i s

point and the supply heat i s the m h o r n var iable . This dis-

cussion of a controlled heating system includes some approxima-

tio ns. In practice , a thermostat i s sensitive to a combination

of room a i r and inne r sur fac es ' temperatures and the cont rol. .. of

the supply heat i s more or less accurate.



Normally there are variations in the room air temperature, e.g.

c lose to the ce i l ing the a i r can be 3-4 OC wanner than c lose t o

t h e flo or. In a model fo r energy calculati on, t h i s normally

w i l l be neglected and t he room a i r temperature w i l l be assumed

uniform over the whole volume. Another comon assumption i s

tha t a l l s ur fa ce s can be assumed isothermal. Under th i s condi-

tion , th e hea t balance f o r the room a i r can be writt en

where

C~

= Heat ca paci ty of room a i r

T~ = Room a i r temperature

t = Time

qfree,conv= Convective, "free" energy t o th e room

%upply, conv= Convective supply energy t o t he room

C~= Specif ic heat of a i ri = Mass flow of ventilati on and in fi lt ra ti on

a i r

T o

= Outdoor a i r temperature-

'=ci= Convective heat fi lm coe ffi ci ent fo r

surface iAi = Area of surface i

Ti = Temperature of surface i

In many energy rel at ed cal cul ati ons , the heat capacity of the

room a i r can be neglected, thus giving a zero on the l e f t si de

of the equation.

The equation indicates some questions, e.g.

The convective "free" hea t in residential buildings

i s mainly dependent on the individual use of the

building. For an exi st ing building th e usage patt ern

e.g. the use of energy fo r cooking, domestic ele c-

t r i c i t y , number of in habi tant s present i n the house

during di f fe re nt hours of the day may be con tr ol le d

o r monitored. This usage pa tt er n normally gives a



reasonably accu rat e ba si s f or the assumption of the

f r ee hea t. For non-existing buildings a genera lized

usage pa tt e rn , based on broad stud ie s, has t o be

used.

The total venti lat ion i s made up of two a i r flows,the inf i l tr a t i on and the venti lat ion. The inf i l tr a-

t ion i s really a leakage through the building envel-

ope due t o temperature di ff erence and wind pressu re.

The ve nti la ti on is,on the ot her hand, control led by

th e in ha bi ta nt s and due t o window openings, th e use

of fans, et c. Therefore, a usage pa tt er n has to be

monitored o r assumed.

The convective heat f il m co ef fi ci en ts can be tr ea te d

in many 'd if fe re nt ways and th e l i t e r a tu re normally

gives a temperature dependent value in the format

w =a(T. T )b, where a and b vary depending on whichc i R

type of s urf ace i s involved. These constants ar e

given di ff er en t values by diff er en t authors. Min e t

a1 (1956) g ive, f o r example, a=0.14 and b=0.25 f o r

a warm ceiling, while Davis and Griffith (1922) give

a=1.31 and b=0.25 f o r the same kind of sur fa ce , a

value almost ten times greater. This i s an extreme

example and f o r oth er types of surfaces th e values

nonnally a r e much clos er t o each oth er.

1.2 Heat balance f o r a sur fac e

For each inne r sur fac e ( or pa rt of a surfa ce), the following

equation i s val id :

where

qw,in

= Heat flow from th e wall t o the inner sur fac e

qabs = Radiation absorbed on the surface



=c i= Convective heat f ilm coe ffi ci ent- -

T~= Room a i r temperature.

Ti,Ti Surface temperatures-5 . J = Radiative heat film coefficient between

surface i and surface j

The radiation absorbed on the surface, mainly comes from two

sources, int ern al sources and sol ar radiation. The int ern al

sources ar e people, li ght ing , et c . , which ra is es t he same type

of quest ions as alr eady mentioned in the text concerning the

heat balance of the room. The so la r radia tio n w i l l be discussed

later and the convective heat film coefficient has been dis-

cussed above.

The radiative heat film coefficient depends on the geometry of

the room, the emittance of th e surfac es and th c i r temperatures.

These coefficients can be calculated for empty rooms assuming

dif fus e emittance on the surfaces. In re al it y, furni ture , etc .,

w i l l give deviat ions from the values f o r the empty room, bu t

t h i s i s normally neglected.

The equation for inner surfaces i s also val id fo r outer sur-

faces, but in t h i s case the room a i r temperature i s changed t o

th e outdoor a i r temperature and t he surrounding surf aces (T.)I

a re ground and sky. In t h i s case, the convective heat fil m co-

e f f i c i en t may be dependent on th e wind speed a t t he sur fac e.

1.3 Heat balance f o r a window

Figure 1.2 shows the pri nc ip le s of heat tr an sf er f o r a double-

glazed window. For th e inne r pane, following equation can be

used:

where

= Radiation absorbed in the inner pane, mainly

qabs'2 solar radiat ion



~ l a rr a d i a t i o n\Long waver a d i a t i o n-onvec t ive

h e a t t r a n s f e .-i g u r e 1 . 2 Hea t ba la nc e o f a window.



OLI-= Radiative heat f ilm coef fic ie nt between the

panes

Uc= Convective heat fi lm coef fi ci en t between the

panes

TI ,T2= Temperature in pane 1 and pane 2

"LCi= Convective heat fi lm coef fi cie nt on the ins ide

of the window

T~= Room a i r temperature

T.I

= Temperature of inner surface j

%2j= Radiative heat film coefficient between inner

pane and surface j

For the outer pane a similar equation can be used, but with the

outdoor air, the ground and the sky as surround, as discussed

for outer surfaces.

In the above equation, the heat capacity and the heat resis-

tance of th e pane a re neglected. The ea r l i e r discuss ion on con-

vective and rad ia ti ve heat f i lm coeffici ents i s s t i l l valid .

For the radi at ive f i lm coeff icie nt i t is sometimes difficult to

get proper data f o r odd types of coat ings, films, et c.

In or de r t o ge t a s im pl if ied model of a window, i t i s common t o

use constant values fo r a l l heat f i l m coefficients and then by

weighting according t o the heat re si st an ce , determine how much

of th e absorbed s o l a r rad ia ti on goes i nt o the room. The

simplest model of the window i s obtained by assuming the ground

and sky temperatures as equal t o the outdoor a i r temperature

and a l l inner surf ace temperatures as equal t o t h e room a i r tem-

perature. These assumptions together with constant heat film co-

e f f i c i e n t s give a U-value f o r the window.

A cu rt ai n o r vene tia n blind s covering most of the window w i l l

change the heat balance of the window. In some ca se s they can

be t re at ed a s anot her pane i n the window, but with proper op-

t i c a l d ata. In a more exact anal ys is the a i r exchange between

the space enclosed by t he c ur ta in and the room a i r should be'

considered.



The equation above gives the heat balance for the transparent

par t of t he window. For th e framing a U-value is nonnally as-

s a d . Some window manufacturers give ove ra ll U-values for

t h e i r windows. These include framing as well as th e gla zing and

sometimes the a i r leakage as wel l. Such ov er al l U-values ar e

primarily intended for simp lifi ed calculations.

1.4 Heat transfer in walls and slabs

In t h i s section we w i l l b r i e f ly d iscuss some common ways t o

ca lc ul at e one-dimensional heat flow in walls and slab s.

Steady s ta te calculat ion----- ------------------

In case s where t he heat capa city of the wall i s neglected, the

heat flow through the wall i s described by the steady state

equation

where

q = Heat flow

Ti = Surface temperature

r = Heat r esi sta nce of the wall

In more det ai le d models, the heat ca pacity i n build ing pa rt s

must be taken i n to account.

In energy ca lc ul at io n methods the most common pr in c ip les t o

t r ea t unsteady s t at e heat f low in wal ls are the f i n i t e d if fe r-

ence methods and the response factor method.

Finite difference methods.........................

The expression fo r one-dimensional heat conductions i n sol id s

i s given by the Fourier equation ...



and

T = Temperature

t = Time

x = length

a = Thennal diffusivity

= Thennal conductivity

c = Specific heat

p = Mass density

In order to solve the above equation we must convert the infi-

nite differentials into finite differences. In the explicit

method, the following approximation of the Fourier equation is

used:

Given the initial temperatures, the successive use of this dif-

ference equation gives the new temperatures step by step. The

increments cannot be chosen arbitrarily, the expression a.at/2

( A X ) must be less than 0.5. In order to avoid this restriction

other forms of differences can be chosen but then the method

becomes implicit, requiring sinultaneous solution of all tem-

peratures. One example of an implicit method is the Crank-

Nicholson's method using the following equation

Response factor method--- ------------------

The response factor method utilizes the fact that a. wall can be



ch ara c te r iz ed by th e way i t r es p on d s t o a t e m p e r a tu r e p u l s e . Ap u l s e ca u se s a h e a t f l o w h i c h v a r i e s w i t h t h e t im e . F o r a

' b i t p u ls e" w i t h g i v en s h ap e and s i z e , t h e h e a t f lo w a t d i f f e r -

e n t t i m e s is c a l l e d t h e r e s p on s e f a c t o r f o r t h e w a l l . 'Ihe r e -

sponse fac to rs depend on the the rmal p roper t ie s and the d imen-

s io n s o f t h e w a l l . M ethod s t o d e r iv e t h e s e f a c t o r s c an b e f ou ndin many pa pe rs, e .g. Stephen sen and M ita la s (1967) and M ital as

and Stephensen (1967).

A r e a l t empera tu re v a r ia t i on can be approx imated a s a sum of

u n i t p u l s e s and, f o ll o w i n g t h e s u p e r p o s i t i o n p r i n c i p l e , t h e re -

s u l t i n g h e a t f lo w c an b e c a l c u l a t e d by a d d in g t h e c o r re s p o n din g

r e s p o n s e f a c t o r s .

The re sp o ns e f a c t o r m ethod r e q u i r e s t h e o r e t i c a l l y i n f i n i t e

s e r i e s . I n p r a c t i c e t h e s u m m a t io n s must be t ru n c a te d a f t e r a

l im i te d number o f t e rms bu t even so , th e amount o f c a l cu la t io n

work can be co ns id er ab le . There ex i s t , however, some methods t o

r e du c e t h e c a l c u l a t i o n work w h i l e m a in t a in in g an a c c e p t a b l e a c -

c u r a c y .

The Fourier-method (Heind l-Procedu re)....................................Th is method assumes, th a t every thermal process in t h e b u i l d i n g

is rep resen ted by a se r ie s o f harmonic func t io ns in t im e , a s

The Fou r ie r equa t ion reduces then f o r each f requency t o

The s o lu t i o n o f t h i s e q u a t io n c a n b e r e p re s e n t e d by h y p e r b o l ic

s i ne - and cos in e func t io ns . The advan tage o f th e method is ,

t h a t f o r t h e c o n n e c tio n betw ekn th e i n s id e and o u t s id e t em pe ra-

t u r e s u a n d h e a t f l u x e s 4 o f a l a y e r , a s im p le matrix rep resen -



tation can be found

2 .with w =lwk>p

When a small construction is composed by different layers, the

effective matrix can be found by multiplication of the. atrices

of the different layers.

The final solution of a problem can be found by superposition

of the solutions of the partial frequencies. (Linearity of the

Fourier-equat on) .2- and 3-dimensional heat flows...............................

The above discussion is limited to one-dimensional heat flow.

In practice we have a lot of two- and three-dimensional heat

flows in a building, for example around studs in a wall, in

comers and in the foundations of buildings. To include a

proper calculation of these types of heat flow in an energy con-

sumition calculation would require too mch computation work

and large computers, thus making the calculation too expensive.

In energy calculation methods we normally have to assume all

walls and slabs to have only one-dimensional heat flow.

Heat transfer in ground----------------- -----A detailed calculation of'heat transfer in the ground around

the foundation of a building is a rather elaborate task. In en-

ergy calculation methods, the foundation must be treated in a

simplified way and normally, as mentioned above, with one-dimen-

sional heat flow. However, by precalculated coefficients, two-

or three-dimensional heat flow can be taken into account to

some extent. Usually, a ground temperature, assumed acc-ording

to experience, is used as the outside temperature experienced



by the construction. Often the average outdoor temperature over

a longer period i s taken as the ground temperature.

1.5 Int ern al loads

Every method used t o pr edi c t th e energy consumption in a bui ld-

ing mst in some way predict the internal loads caused by the

use of the building. In a re si de nti al building, the int ern al

loads may be caused by occupancy, cooking, l ight ing, h ot ta p

water, boilers , etc.

In pri nc ipl e every ca lc ul at io n method has t o use schedules giv-

ing values and durati ons f or a l l the se loads. Necessary de ta il s

in these schedules depend on how de tailed the model used is ,

for example an hour-by-hour calculation mst have hourly infor- -mation while a method based on monthly periods might use aver-

age values.

Most of the internal loads are caused by the inhabitants and

t h e i r use of equipments i n the buildin g. Subtask B of th is

annex has de a l t with t h i s question and the guide- lines produced

by that group w i l l give extensiv e de ta i l s . (IEA Energy Conserva-

t ion in Building and C o m i t y Systems, 1982).

1.6 Solar rad iat ion

Solar heat gain i s often an important part of fr ee heat i n a

residential building and i t i s important t o calcula te t hi s gain

in an accura te way. However, de ta il ed calc ula tion of s ol ar heat

gain is rather elaborate and a l o t of uncertain factor s w i l l

influence the so la r gain obtained in rea l buildings. In t h i s

paragraph, some of the phases in t he cal cu la ti on of s ol ar heat

gain w i l l be briefly discussed. More detailed information can

be found from the work of IEA Solar R&D. The report "An in t ro-

duction t o Meteorological measurements.. . " (1980) gives some

basis and several references.



A common s t a r t i n g point is weather data giving direct solar

rad iat ion and di ff us e sky radiati on. These values have t o be

transformed i nt o the t o t a l rad iatio n indicent on the dif fe ren t

surfaces of the b uilding by tak ing in to account sol ar angles,

horizon angles, surrounding buildings, th e shape of the t ar ge t

building and reflections from surrounding buildings and the

ground.

If the absorptivity for the opaque surfaces of the building is

known, th e absorbed pa r t of t he r ad ia tion can ea si ly be ob-

tained and, in the t o t a l heat balance, used in the equations

for the walls , etc .

In order t o ca lcu lat e transmission of so la r radiation through

windows, ba si c o pt ic a l theory can be used f o r gl as s and th in

layers. This requires basic physical data, such as refr acti on

coefficients and absorptions in t h e gla ss mass. This type of

data is sometimes d i f f i c u l t t o ge t and one has t o use sim-

pl if ie d ways t o t r e a t the window.

One connnon way t o descri be s o l a r transmission through a window

is by th e shading co ef fi ci en t, which gives the r a t i o between

the transmission for the window in question and a standard

single-pane window. Thi s r a t i o is normally assumed t o be a con-

s t a n t and give s no po s s ib i l i t y t o examine how much so l a r rad i-

at ion is absorbed in each pane.

Inside th e room, th e distr ibu tio n of dir ec t so lar radiation is

an elaborate geometrical problem, normally avoided by assuming

only dif fu se rad iat ion in the room. The di st ri bu ti on of t he di f-

fuse radia tion can be calcula ted assuming di ffus e r efl ec tio n in

each inner surface of the room. In th i s case, the view fac tors

have t o be calc ula ted and then combined with t he r e f le c ti v it y

of th e sur fac es. Some ca lc ul at io n methods simplify t h i s dis -

tri bu tio n according t o th e ar eas of the in ner surfaces and some

methods make th e simp lifying assumption th a t a l l o r par t of th e

transmit ted sol ar radiat ion is a convective heat gain. t o the.- .

room a i r .



1.7 References

ASHRAE, Handbook of fundamentals, American Society of Heating,

Refrigerating and Air-conditioning Engineers. Inc.Atlanta,

G.A., current edition, e.g. 1981.

Davis and Griffith, The Transmission of Heat by 'Radiationand

Convection, British Department of Sci. and Ind. Res.

(1 922).

IFA Energy Conservation in Building and C o m i t y System,

Guiding Principles concerning Design of Experiments, In-

strumentation and Measuring Techniques, 1982.

IFA Solar Research and Development, An Introduction to Meteoro-

logical Measurements, 1980. .Kreith,F., 1965, Principles of Heat Transfer, International

Textbook Company, London. Reprinted 1972,ISBN 0 70020118 1.

Min et al, 1956, Natural Convection and Radiation in a Panel-

heated Room, ASHR4E Transactions, Volume 62 (1956).

Mitalas and Stephensen, 1967, Room T h e m 1 Response Factors;

ASHR4E Transactions 73 (1967), Part I, paper 2019.

Stephensen and Mitalas, 1967, Cooling Load Calculation by

T h e m 1 Response Factor Method, ASHR4E Transactions 73,

(1967) Part I, paper 2018.



2 INVESTIGATION OF CALCULATION METHODS

2.1 Classification of methods

,

Two main cat eg ori es of methods t o cal cu la te annual energy con-

sumption in buildings can be distinguished, namely simulationand co rr el at io n methods. Simulation methods ar e based on th e

sol uti on of more o r l e s s de ta il ed thermal models of the build-

ing in sh or t tim esteps , e.g. hour by hour. Cor relation methods,

on the ot he r hand, give the energy consumption as a simple re-

lation between the thermal losses of the building and mean

weather data for longer periods.

The classification of the methods i s summarized in the follow-

ing figure.

Fini te differences

Thermal

Dynamic I Response factors

Simulation

methods I L Other methods

I Weighting

factor methods

r Steady-state heat balance

Degree-day o r correc ted

degree-day methods

Other correlation methods

Calcu lation of savings due t o some conserva tion measure i s nor-

mally ca rr ie d o ut a s a d if fe re nc e between two annual consump-

ti on lev el s. This implies t ha t the method must be able t o

handle the measure in question w i t h an accuracy gr eat enough to

use t he d if fe re nc e between two annual consumption fi gu re s. On

the other hand, e rr or in some par t of the ca lc ul at ion might

equal out when taki ng the d if fe re nc e, thus i n some cases a

method can be more acc ur at e when used t o pred ic t savings than

when used t o pr ed ic t consumption. .



G a m i c simulation methdds_ .......................In simulation methods the thennal behaviour of the building is

simulated s te p by st ep , usual ly with hourly inter val s. This

allows one t o ca lcul a te the energy requirement and/or the room

temperature, i f th e va ri ab le in te rn al 'and extern al boundary

conditions a re h o r n . Using weather-data fo r a whole year, the

annual energy requirement then can be predicted. In addi ti on ,

the simulation methods can be used with selected weather data

for a "design-day" t o s i z e hea ting equipments.

Two ccmon ways t o ca r ry out a simulation ar e th e thennal bal-

ance method and th e weighting fact or method.

In the thermal balance method, the equation fo r a l l par ts of a

building, combined w i t h the exci tat ions due t o int ernal loads,

so la r rad iat ion and outdoor temperature give the t o t a l heat

balance of a bu il di ng . This balance can be descr ibed more or

l es s complexly, depending on how many of the building p ar ts a r e

included i n the thennal model. Thus, some methods only model

the t o t al bu ildi ng a s a si ng le space, while other methods take

care of a l l heat t ra nsf er between rooms by modelling a l l rooms

and a l l int ern al walls and slab s of the building.

The thennal balance method can be characterized by the numeri-

ca l method used t o t r e a t the heat conduction i n so lid s. As de-

scribed i n chapter 1, we can use f i n i t e difference methods, re-

sponse factors , th e Fourier-method o r some ot he r method.

The weighting fact or method gives a di f ferent approach. Here we

des cr ib e, the hea t requirement and/or room temperature as func-

ti on s of the i nt er na l and ex te rnal boundary conditions. (This

is somewhat sim il ar t o th e response facto rs f or a wall). I f

these functio ns ar e hown, th e calc ula tio n work f or each time-

step i s less th'an with the thermal balance methods. On the

other hand, th e output information cannot be so deta il ed as .with the balance methods, par tic ula rly as f a r as the various

temperatures ar e concerned. Moreover, th e functions must be.. ..precalculated using a thennal balance model. Another restric-



t i o n w i t h w e i g h ti n g f a c t o r meth od s i s t h a t o n l y l i n e a r s y s t e m s

c a n b e t r e a t e d ; th i s can somet imes limit t h e i r u s e f u ln e s s .

Co rre l a t ion me thods-------------------In some cases i t i s n o t n e c e ss a ry t o h o w t h e v e r y d e t a i l e dt r e n d s o f t h e rma l p a ra me t e r s o f t h e room, i f o v e ra l l r e s u l t s o f

e ne rg y re q ui re m e n ts a r e s u f f i c i e n t . I n t h e s e c a s es a c o r r e l a -

t i o n i n v o l v i n g b u i l d i n g c h a r a c t e r i s t i c s and mean a n d /o r o v e ra l l

c l i m a t i c pa r am e t e rs , c a n be u se d t o c a l c u l a t e t h e t o t a l e n e rg y

requ i remen t ove r a g iven pe r iod . D i f fe ren t complex i ty l eve l s

a re pos s ib le , ranging from th e we l l hown "degree-day" method

t o more o r l e s s d e t a i l e d m e th od olo gie s f o r t a k i n g i n t o a c co u nt

t h e u t i l i z a t i o n of " f r e e e ne rg y" g a i n s .

2 .2 Simula t ion me thods in ve s t ig a t e d

Al l the s imula t ion methods involved in t h i s i n v es t ig a t io n a r e

computer programs working with one hour t imesteps. Some of the

main c h a r a c t e r i s t i c s o f t h e se m ethods a r e s m r i z e d below.

S i z e o f Ma in Inc ludest h e m 1 model ma t he ma t ic a l a i r c o n d i t i o n in g

method system

BA-4 Simple F in i t e Nod i f f e r e n c e s

NBSLD Com plex Response Yesf a c t o r s

DYWON-2 Simple F i n i te Nod i f f e r e n c e s

KLIMASIM Complex Response No -- -W A V a r i a b l e F i n i t e No

d i f f e r e n c e s

DOE- I I Complex W eighting Yesf a c t o r s



BA-4-his program handles one room or a whole building treated as a

unit. The thermal network is reduced to one thennal capacity

and four thermal resistances. The method is intended to give an

inexpensive method f or a f u l l year s h l a t i o n and i s well

su it ed f o r making spe ci al ve rsi ons by adding subroutines.

Ref.: Hans Lund, BA-4 User' s Guide, Thennal In sulation Labora-

tory, Denmark. Report No.44, 2 ed., 1979.

Hans Lund, The model and theory behind the BA-4 program,

Thennal Insu lat ion Laboratory, Inte rnal r epor t, 1977.

NBSLD-his program, o r ig ina l developed by National Bureau of Stan-

dards, US, us es a detai led hea t balance model of each room i n

the building. One of the few simp lif ica tio ns i n th i s program i s

the use of th e shading coe f fi ci ent s when tr ea ti ng th e windows.

Ref.: NBSLD: the computer program fo r heating and cooling loadsi n bu ilding s, National Bureau of Standards, U.S.,

NBS-BSS 69, 1976.

In t h i s program the b uildi ng is divided in to a ground f lo or

with one heat capac ity , a si ng le o r double glazed window area ,

an opaque oute r s urfa ce representing the exte rnal wall s and

roof with a weighted mean U-value and a hea t ca pa ci ty and th e

inner walls and slabs represented by a third capacitive con-

struction. Radiative heat exchange is calculated by an approxi-

mating f o m l a i n which the r at io of the areas in question and

the glob al shape of the building are taken a s parameter values.

The thermal balance is ca lcu lat ed each hour. The so l a r rad i-

at ion on vert ical surfaces i s calculated with special subrou-

t i n e s.Ref. : H.A.L.van Dijk, Basic study re su lt in g from the r ev is io n

of the Dutch requirements concerning thermal in su la ti on



of dwellings. Collected Papers of the 9th TWL-RIO sem-

inar, Delft, October, 1979.

H.A.L.van Dijk, Gebruiksbeschrijving computerprograrmna

DYWON, berehing van energieverbruik woningen. (User'sguide computer program DYWON, calculation of energy con-sumption in dwellings), 2nd ed., internal report TPD,January 1982.

KLIMASIM

The calculation procedure concerned is based on calculating

momentary events by solving heat balance equations for inside

as well as for outside surfaces of each single room enclosure

element, and for the room air in every single room. Dynamic

heat transfer through the fabric of each enclosure element of

different composition is represented by precalculated time

series of response factors for temperature as well as for radi-

ation excitation at the boundary. By applying convolution al-

gorithms momentary events of temperature and heat flow at the

surfaces are calculated. Room temperature deviations from the

reference level of 20 OC in every single room are calculated by

applying a weighting function concept, which includes the re-

sponse of the entire room concerned. The weighting factors are

precalculated.

Ref.: R.S.Soeleman, see Appendix A. (Documentation not pub-lished).

JUUYrrA

JllLOTTA is basically an RC-circuit analysis algorithm. This

means that while the user must ordinarily manually create a

network analog of the building to be simlated, JULO?TA gives

him at the same time the ability to easily manipulate the net-

work and to deal with a wide variety of different problems in

extreme detail. The program has no Heating, Ventilating and Air

Conditioning (WAC) system modelling capabilities included; how-

ever, a special subroutine, S Y m l , allows the user to build

HVAC and control routines and access the simulation with appro-

priate alterations of modelling conditions while it is actually



underway. The program also provides the user with routines for

detailed calculation of shading, absorption and transmission of

solar radiation in windows etc. In this subtask, the buildings

have been represented in detail, modelling all convective film

coefficients, radiative exchange in rooms and spaces between

window-panes, etc. The heating system has been modelled in the

above mentioned routine SYSTEA1 as a "perfect" controlled system.

Ref.: K.Kallblad, F.Higgs, Building Energy Use Modelling inSweden by JULOTTA, Proceedings of the Third InternationalConference on Energy Use Management, Berlin (West), Octo-ber 26-30, 1981.

WE-I1

This public domain program was developed by Lawrence Berkeley

Laboratories. The basic principles used in this program are the

ASHRAE weighting factor methodology based on a rather complex

thermal model of the building. The program uses an extensive

Building Description Language to simplify the input prepara-

tion.

Ref : M.Lolonanhekim, F.Winkelmann and A.Rosenfeld, W E , A NewState-of-the-art Computer program for the energy utiliza-tion analysis of buildings, The Third International SF-posium on the Use of Computers for Environmental Engin-eering Related to Buildings, Banff, Canada, May 10-12,

1978.

2.3 Correlation methods investigated

Common to all correlation methods involved in this subtask is

a steady-state calculation of transmission losses through the

building envelope. For ventilation losses mean values for a

particular time-base are used and none of the methods include

any type of air conditioning system.

The main characteristics of the methods are summarized in the-..

following.



Required calculation Time-base forequipment weather data

LPB-4 Computer Week

EFB- Pocket-calculator Month

SECC Pocket-calculator Month

NEVACA Pocket-calculator Month

Degree-Day - Year

BKL Pocke -calculator Month

J W Computer Day

The heat gains in this method are assumed to be totally convec-

tive to the room air. The boundary conditions for the outer

surfaces are the solar air temperatures, thus taking into ac-

count solar radiation on outer surfaces. All meteorological

data are processed by another program on hourly basis, taking

shading etc. into account. A fixed temperature for the ground

is used.

Ref.: F.Lorenz, Static computation of thermal loads, Technical

report E.E.C./78.12/A2 TCOl Contract No.615-78-1 EEB,University of Likge, 1978.

J.Lebrun, Y.Delonne, Donnees climatiques utilisablespour le calcul des charges thenniques appliquees auxbatiments, Laboratoire de Phisique du Batiment, Univer-sity of Ligge.

EFB--he method uses monthly mean temperatures and monthly solar

radiation through actual windows in a room or building. It can

take into account the radiation on outside' alls and roof, how-

ever, that is normally not done. Two extremes are calculated,

the energy consumption for a "heavy" building where all "free"

heat is utilized if needed and for a "light-weight" building

where only a fraction of the, "free" heat is utilized. Other

buildings with the same heat losses are then computed by inter-



polation between the results for the extremes in each month.

Ref.: A.Nielsen, A method for calculating the energy consump-tion by means of a desk calculator, CIB Symposium,Copenhagen, May 28 - June 1, 1979. Thermal Insulation

Laboratory, r ep or t no.81, 1979.

SMECC-ven th i s method uses the so la r a i r temperature approach fo r

ou te r opaque sur fa ces and precal culat ed values f or shading of

windows. "Free heat gains" a re taken i n to account by means of

a "coefficient of util iz at io n" as a function of suit abl e "aver-

age" ov era ll heat t ra ns fe r co ef fi ci en t, the mass of the build-

ing per uni t f lo or are a and the r a t i o between over all losse s

and ov er al l gain s. This fu nct ion has been determined by a l o t

of parametric runs with the NBSLD-program.

Ref. : L.Agnoletto, P.Brunel10, R.Zecchin, Simple methods f o rpredicting thermal behaviour and energy consumption ofbuildings, ASHRAE-WE Conference, Orlando, Florida, U.S.,1978.

NEVACA

For out er opaque su rf ace s the s ol ar a i r temperature is used and

solar gain on outer surfaces and through windows should be

given (in cludin g shading) f o r c l ea r days of each month. The

monthly free heat is calculated from the number of clear and

cloudy days, where th e cloudy days a re assumed t o give 25% of

solar gain compared with clear days. Days which are neithercl e ar nor cloudy a r e assumed t o give 65%. The los ses ar e calcu-

l a ted by use of t he monthly mean temperature. All so la r and

int ern al gains ar e assumed to be fu ll y util iz ed .

Ref.: A-C.Andersson, Enkel metod fo r be rahing av vamebalans

i byggnader, Rapport NBH-7040, Department of Bu ild ingTechnology, Lund In s t i t u t e of Technology, 1978.



Degree-Day

The Degree-Day method used in this investigation has been based

on the Swedish definition of degree-days. These are based on an

indoor temperature of 17 OC and on a heating season defined by

days with daily mean outdoor temperature below 17 OC during

November-March and 12, 10, 10, 10, 11, 12 and 13 OC for April-

October. Other days are not included when the degree-days are

calculated. The real mean indoor temperature is in fact some-

what higher, but the temperature step from 17 OC is achieved by

the free heat. Of course this temperature step in fact differs

from situation to situation.

This method can (by definition) not handle conservation meas-

ures such as night temperature setback. Still, this report

shows results for this method, which has been carried out by

recalculating the degree-days for an indoor temperature lower

than 1 7 OC.

BKL-This method uses daily solar radiation data, but by givingthese in a simplified way, the energy requirement can be calcu-

lated month by month. The method neglects solar radiation on

outer opaque surfaces. By the use of precalculated transmission

coefficients for different oriented and shaded windows, the

solar gain through windows can be estimated. The heat losses

are calculated by the use of the monthly mean outdoor tempera-

ture. By comparing these losses with the distribution of "free

heat", the monthly energy requirement for heating can be calcu-

lated.

Ref.: K.Kallblad, B.Adamson, Hand calculation method for esti-mation of heat consumption in buildings, CIB Symposium,May 28 - June 1, 1979 in Copenhagen, Denmark.



JAEW-his method uses the solar air temperature approach to outer

opaque surfaces. The transmission of solar radiation through

windows can be defined in the input according to estimates for

the individual case. To estimate the influence of shading,

shading diagram are used. Transmission losses are calculated

for day- and night-time sepaiately and the utilization of free

heat is determined by the use of a utilization function which

assumes decreasing utilization at increasing total heat gain.

Required weather data for each day are mean outdoor temperature

for day-time and nigth-time, total daily radiation on the hori-

zontal and vertical planes in the 4 cardinal directions and the

average wind speed for day. The latter is used to estimate in-

filtration.

Ref.: J.Gass, R.Sagelsdorff, Heizenergieverbrauch von Wohn-bauten, EMPA 39200, February 1980.



3 CALCULATIONS FOR THE VETLANDA HOUSE

3.1 The house

The Swec!ish As soci at ion of Pr efabr ica ted Home Manufacturers

bu i l t in 1975 a low energy one-family house i n the town of

Vetlanda, Sweden. The house (Figure 3.1) was designed by Bengt

Hidemark, a rc hi te ct and professor a t the Royal I ns t it ut e of

Technology, Stockholm, and professor Bo Adamson, Lund I n s t i t u t e

of Technology, Lund.

The low-energy house has one storey, no basement, and was given

qui te an ordinary ex ter ior a t t he request of the Association.

The plan i s shown i n AppendixB.

The floor area within the2outer walls is 110 m . The house consists of three bedrooms,

one l ivi ng room, a dining room, a kit chen, a laundry , WC, and

a bathroom including a sauna. The facade with the large windows

faces 15' east of south. On that facade the roof has an over-

hang of 1.8 m in o rder t o shade the windows during the summer,

when solar heat gain is undesirable. In the spring and autumn

the al t i tu de of the sun i s so low that a considerable amount of

insolation can be utilized for heating.

The windows i n th e so uth wall co ns is t of fou r panes in s ea led

units. These windows cannot be opened, i n contrast to the few

ot he r windows with only t hree panes, wfiich can be opened. In

th e window wall facin g south, th e thermal in su la ti on i s not

more than 95 m. On the ot her hand, ther e ar e only small par t s

which are not windows or doors. As only a few of the windows

are openable and as a well-insulated door is much better than

a window with re spect t o both in su la ti on and speedy a ir ing , -the

low-energy house has six external doors. These doors w i l l dur-

ing th e s m e r provide an easy and good contact between t he

house and the garden.

The othe r walls have a thermal in su la ti on of 190 & mineral

wool. Below th e con crete f loo r sl ab (160 m ) , the f loor is in-

sulated with 120 nun mineral wool of a heavy quality. The th&-

ma1 insulat ion in the cei l ing i s 300 nun of mineral wool.



Figure 3.1 The Vetlanda house .



The ventilation i s mechanical, with fans both to supply and

exhaust air. The ventilation is divided into two parts. The

bedrooms and the living room are continuously supplied with3

20 and 40 m /h fres h a i r , respectively. The used a i r i s ex-

hausted from the kitchen, bathroom and WC. I f more fre sh a i r

is needed i n the l iv ing room, dining room o r kit chen, one can

push a button i n any of th e rooms, which re su l t s i n a fr es h a i r

3supply of 250 m /h to ta l . This forced vent i la t ion i s on for a

sh or t period, f o r exemple 15 minutes, and then automatically

cuts of f . I f fur ther forced vent i la t ion is needed, one has to

push the button once again. The a i r supplied i s preheated t o

16 OC whenever the outdoor a i r temperature is lower than this

leve l .

The building has been monitored for almost two years, but for

t h i s study we have se le ct ed a 6-month period with well con-

trolled conditions. This enables comparisons not only between

ca lc ul at io n methods but a l so t o some ext ent between estimated

and measured energy consumption. During the monitored period

the building was not occupied, thus avoiding the influence ofinha bitan ts. However, some simple occupancy pat te rns were simu-

lat ed by use of time cont roll ed el ec t ri ca l heat sources.

3 . 2 Calculation cases

The detailed specifications of the building used for the calcu-

lations are given in Appendix B. These were made as close as

pos sib le t o t he r ea l bu ildi ng t o get a good comparison. The

weather tape produced fo r t h i s ta sk contai ns si x months' dat a

and Appendix C give s some of t h i s dat a.

The study has covered six cases, namely the original case and

fi ve parametric stud ies. The d et a i l s of these ar e described i n

Appendix B, and a sh or t pres enta tion of the main cha rac ter is -

t i c s of the cases a re given below.



The original case_____---_------

The original building i s used t o enable comparison between pre-

dicted and measured energy consumption and together with the

other cases t o examine th e di ff er en t calculation methods' abi l-

i t y t o handle vari ous type s of commonly used r e t r o f i t s and of

heat transfer phenomena.

Case (a)--------

In t h i s case, a l l windows ar e assumed t o have only two panes

ins tead of th ree o r four panes. The predi cted consumption makes

i t poss ible t o s tudy changes i n so la r gains combined with

changes i n thermal in su la ti on achieved by i ncrea sing th e number

of panes i n th e windows.

Case (b)--------

In th is case the in sula tion in walls and ceili ngs is decreased

compared t o th e o ri gi na l case. This case w i l l , together with

the original case, i l lustrate an increased insulation from a

ra th er nonnal l ev el t o an extremely well insulate d building.

The U-values f o r th i s cas e and the or igi nal case are:

2For th e main wal ls : 0.57 and 0.23 W/m , K , respectively

2For th e ce il ing : 0.40 and 0.14 W/m , K , respectively

Case (c)- - ---- - -This case differs from the original case by employing commonly

used le ve ls and times f o r nigh t temperature setback from e.g.

20 OC to 16 OC between 22.00-06.00 hours a l l days of t he week.

(The setpoin t f o r th e preheater i s not al tered).

Case (d)---- -- --This case is a combination of t he night temperature setback i n

.. ..

case (c) and an ins ul at io n of th e concrete flo or by a carpet



2with a heat resistance of 1.0 m K/W. The insu lat ion of the

f loor i s thus increased and the heat capacity inside the house

i s decreased a t the same time. This ca se was intended t o show

that the effect of the night time setback is potential ly

greater i n l igh ter houses.

Case (e)--------In th is case the forced vent i la t i on r a te i s increased from 100

m3/h t o 200 m3/h. Thi s allows a check of th e ab i l i t y of the

calc ulat ion methods t o handle ven ti lat io n losse s.

3.3 Pre dic ted energy consumption and savings

The monthly and t o t a l energy consumption predic ted by the in-ve st ig at ed methods ar e given i n Tables 3.1 and 3.2a - 3.2e. The

values include energy for the preheater as well as for the radi-

ator s. The setpoi nt. of the preheater is 16 OC and some methods

do not distinguish between preheating and the remaining heating

of a i r by the radia tor s . Thus, thes e methods can normally only

pr ed ic t the t o t a l consumption. The degree-day method i s one of

these, but in t h i s i nves tiga tion an extr a s e t of degree-days

has been used by th e ana ly st and hence the sepa rat ion mentioned

above was also possible in the degree-day method.

The pred icte d energy conservation, f o r a l l methods, is the d i f -

ference between predicted energy consumption in case (a) - case

(e) and the or ig in al case. These diff eren ces ar e given i n

Tables 3.3a - 3.3e.

The predicted energy consumptions and savings are further il-lus t r a ted i n chapter 5, wfiere a n arralysis of the results i s

car r ied out .



Table 3.1 The Vetlanda House, orig. case. Heating Loads (MJ)

Preheater Meas.

DD

EFB-1

smccNEVACA

BKL

LPB-4

JAEAVBA-4

DYHOB-2

DOE-2

JULOTTA

KLIHASIM

Radiators Meas.

DDEFB-1

smccNEVACA

BKL

LPB-4

JAEHY

BA-4

DYHON-2DOE-2

JULOTTAKLIHASIM

Total Meas.DD

EFB-1

smccNEVACA

BKL

LPB-4JAENV

BA-4

DYWON-2DOE-2

JULOTTAKLIMASIM

NOV DEC JAN FEB HAR APR TOT

1464 1801 1801 1698 1413 1283 9460



Table 3.2a The Vetl and a House, Case ( a ) , H eati ng Loads (MJ)

NOV DEC JAN PEB HAR APR TOT

Pr ehea t e r DD

EPB-1

smccBEVACA

BKLLPB-4

JAENV

BA-4

DWON-2

DOE-2

JULOTTA

KLIHASIM

Radia tor s DD

EPB-1

smcc

NEVACA

BKL

LPB-4

JAENV

BA-4

DWON-2

DOE-2

JULOTTA

KLIKASIM

Tot a l DD

EFB-t

smccNEVACA

BKL

LPB-4

JAENV

BA-4

DYWON-2

DOE-2

JULOTTA

KLIMASIU



Table 3.2b The Vetlanda House, Case (b), Heating loads (MJ)

NOV DEC JAN FEB W APR TOT

Preheater DDEFB-1

smccNEVACA

BKLLPB-4JAENVBA-4DYWON-2DOE-2

JULOTTAKLIMASIM

Radiator8 DDEFB-1

SHECC

NEVACABKLLPB-4JAENVBA-4DYWON-2

DOE-2

JWOTTAKLIMASIM

Total DDEFB-1SKECCNEVACA

BKLLPB-4JAENVBA-4DYWON-2

DOE-2JULOTTKLIKASIM



Table 3 . 2 ~ The Vet land a House, Case (c ) , H eati ng Loads (MJ)

NOV DEC JAN PEB MR APR TOT

Prehea te r DD

EFB-1

SMECC

NEVACA

BKL

LPB-4

JAENV

BA-4

DWOR-2

DOE-2

JULOTTA

Kmu s I n

Radia to rs DD

EFB-1

SHECC

NEVACA

BKL

LPB-4

J A E N V

BA-4

DYWON-2

DOE-2

JULOTTA

KLIMSIM

To ta l DD

EFB-1SMECC

NEVACA

BKL

LPB-4

JAENV

BA-4

DWOR-2

DOE-2

JULOTTA

KLIKASIM



46

Table 3.2d The Vetland a House, Case ( d ), Heating Loads (M J)

p r e h e a t e r DD

EFB-1

smcc

N EV A CA

BKLLPB-4

JAENV

B A-4

DYWON-2

DOE-2

JULOTTA

KLIHASIM

Radia to rs DD

EFB-1

smcc

N E V A C A

BKL

LPB-4

JAENV

BA-4

DYWOA-2

DOE-2

JULOTTA

KLIHASIM

T o ta l DD

EFB-1

smcc

NEVRCA

BKL

LPB-4

JAENV

BA-4

DYWON-2

DOE-2

JULOTTA

KLIHASIM

N O V DEC JAN FEB IWL APR TOT

1309 1905 1814 1750 1516 1270 9564



Table 3.2e The Vetlanda House, Case (e) , Heating Loads (MJ)

P r e h e a t e r DD

EFB-1SMECC

NEVACA

BKLLPB-4JAENV

BA-4DWON-2DOE-2JULOTTAKLIHASIM

R a d i a t o r s DD

EPB-1SMECC

NEVACABKL

LPB-4J A ENV

BA-4DWON-2DOE-2N L O T T A

KLIHASIM

T o t a l DD

EFB-1

SMECCNEVACA

BKL

LPB-4J A E N V

BA-4DWON-2DOE-2JULOTTA

KL1MAS.CN

NOV DEC

2462 3603- -2405 35212462 3603- -2504 36502547 37012506 36542514 36652608 3800- -2506 3653

4873 6895- -3957 7054

3849 6130- -4063 63764548 67673832 59263399 57183459 5910- -3976 5377

7335 104986928 10197

6362 105756311 9733- -6567 100267095 104686338 95805913 93836067 9710- -6482 9030

JAN

3434-33483434-34843534348734993630-3488

6610-6220

5404-56666152528951 0 35033-5044

100449347

95688838-91 509686877686028663-8532

FEB

3305-32273305-335834053362337 t

3495-3361

6350-9581

5067-51 215621479144534299-4526

96558673

88088372-84799026815378247794-7887

IWL APR TOT



Table 3.3a

DD

EFB-1

smcc

NEVACA

BKLLPB-4

JAENV

BA-4

DWON-2

WE-2

JULOTTA

KLIllASIM

Table 3 . 3 b

DD

EFB-1

SMECC

NEVACA

BKL

LPB-4

JAENV

BA-4

DYWON-2

DOE-2JULOTTA

KLIMASIM

Energy Conservation i n Case ( a ) , (MJ)

NOV DEC JAN FEB MAR APR TOT

Energy Conservation i n Caae (b), (MJ)

NOV DEC JAN FEE MAR APR TOT



Table 3 . 3 ~

DD

EFB-1

smcc

N E V A C A

BKLLPB-4

J A E N V

BA-4

DYWON-2

DOE-2

JULOTTA

KLIMASIM

Table 3.3d

DD

EFB-1

SHECC

N E V A C A

BKL

LPB-4

JAENV

BA-4

DWOR-2

DOE-2JULOTTA

KLIPIASIM

Table 3.3e

DD

EFB-1

SMECCAEVACA

BKL

LPB-4

JAENV

B A-4

DYWON-2

DOE-2

JULOTTA

umsm

NOV

648--674-708

552

395

586

549-1120

DEC J A N

713 713- -- -739 726- -728 732

599 585

550 515

726 709

759 725- -779 1040

FEB MAR

648 674- -- -661 713- -661 733

521 532

460 379

650 635

650 556- -616 897

APR TOT

660 4056- --661 4174- 3270

710 4272

520 3309

445 2744

575 3881

441 3680- 3110

953 5405

Energy Conservation i n Case ( c ) , (MJ)

NO V DEC JAI FEB MAR APR TOT

Energy Conservation i n Case ( d ) , (MJ)

Energy Con aemati on i n Case (e ) , (MJ)

N O V D E C J A N FEB M APR TOT



4 CALCULATIONS FOR THE TEWIKERN BUILDING

4.1 The building

This building i s an 8-storey apartment building, typical for

many pa r t s of Sweden and northern Europe. Each st or ey has about

2665 m of living area. The walls are concrete sandwich-elements

with 70 nun of insula tion and the ce il ing has 200 nun of insula-

2t io n giv ing approximately a U-value of 0.49 W/m , K fo r the

walls and 0.19 w / ~ ' , K for the ceiling. All windows have two

panes and the building i s equipped with an exhaust a i r vent il a-

ting system. The building is a hypothetical building and the

sp ec if ic at ion s, given i n Appendix B, are t o a great extent sim-

p l i f i ed in order t o avoid dif fer enc es caused by analyst in te r-

preta t ion .

The main purpose for including this building in the comparisons

between calculation methods is t o study the dif fe re nt methods'

cap abi lit y t o predi ct the yearly heating requirement of big

res ide nt ia l build ings.

4.2 Predicted energy consumption

The energy consumption for heating has been predicted for a

whole year using weather data from Stockholm 1971. The predic-

ti on s of t he di ff er en t methods are given in Table 4.1 and w i l l

be f urt her il lu st ra te d in connection with the analysis in chap-

t e r 5.



Table 4.1 The Teknikern Building, Total Heating Loads (GJ)

JAN PEB MAR AP R NAY JUN JUL AUG SEP OKT ROV DEC YEAR

SMECC 1 6 5 1 5 2 1 6 6 7 7 0 0 0 0 2 4 3 1 2 7 1 3 9 8 7 1

NEVACA 201 187212115 1 1 0 0 0 11 6 8 1 6 6 1 7 3 1 1 4 4

BKL 170 153 175 77 0 0 0 0 2 13 13714 3 870JAENV 171 15 71 75 92 20 1 0 0 19 58 13 81 45 976

NBSLD 161 140 159 64 8 0 0 0 3 30 123 131 819DOE-I1 1 6 5 1 4 9 1 6 2 70 4 0 0 0 3 4 0 1 3 1 1 3 8 862

JULOTTA 162 150 161 68 4 0 0 0 4 41 121 137 848



5 ANALYSIS OF THE CALCULATION METHODS

5.1 Predic tions of annual energy consumption

The predic ted energy consumption of the ori gi na l case of th e

Vetlanda house i s shown in Figure 5.1 together with th e meas-

ured value. In th i s fi gure, only fi ve months have been se le ct ed

as a l l methods showed g re at di ffere nces from th e measurements

dur ing November 1976. This led t o th e discovery tha t t he thermo-

s t a t s in the building were erroneously giving too high indoor

temperature compared with the calculation specifications.

The differences from the measured value can depend on many

thing s, e.g. th e cal cul ati ons a re based on spe cif ica tio ns which

might not have been fulfilled in the real building, such as

cold bridges, et c. I f the calculati on methods react i n di ffe r-

ent ways t o these deviations, the re su lt s w i l l deviate from

each other. This shows that a single energy cocsumption figure

i s de fi ni te ly inadequate when one t r i e s t o verify a calcu latio n

method. To make a complete ve ri fi ca ti on of a ca lc ul at io n method,

many more parameters have t o be monitored, thus giving a possi-

b i l i t y t o check a l l par ts of t he calculati on. Even fo r a method

which predicts a consumption close to the measured value, one

cannot be sure th at t h i s i s not a re su lt of opposing err or s.

From Figure 5.1 one can observe th a t the pred ic ted energy con-

sumption varies around the measured value. The relation between

the highest and lowest prediction i s 1.43 t o 1.0. Looking a t

ca se (b) of t he Vetlanda house (decreased in su la tion ) a s shown

i n Figure 5.2, the same va ri at io n i s only 1.27 t o 1.0.

The observation of the ratio between the highest and lowest

prediction might indicate a problem with the handling of free

heat from people and solar radiation and i t s u t i l i z a t ion fo r

heatin g. This f r ee he at becomes le s s important when th e lo ss es

are greater , thus the smaller ra ti o in case (b) . This indicates

th e need of a deeper analys is of how the f r ee heat is taken in-.to account in the different methods.



- - I

BKL IJ W IEFB- 1

NEVACA I

KLIlWIM IDYWON- 2

Measured

Figure 5.1 The Vetlanda house, original case. Five months'heat consumption.



EFB- 1 -IBKL II SMECC I

BA- 4 I

F i g u r e 5 . 2 T h e V e t l a n d a h o u s e , c a s e ( b ) . Six m o n t h s ' h e a tc o n s um p t i on . ( D e c re a s ed i n s u l a t i o n ) .

EFB- 1 .I

DnVON- 2JULOTTA

NBSLD

F i g u r e 5.3 E s t im a t e d h e a t i n g l o a d s f o r t h e T e kn ik e rn b u i l d i n g .



The yea rly consumption estimated fo r t he Te h i k e m building i s

shown in Figure 5.3. Most of the resu l t s are ra ther c lose t o

each ot he r, but th e degree-day method shows an unacceptab le de-

viat ion from the othe r re sul ts. This i s cle ar ly caused by the

fac t tha t t h i s method cannot take i nto account the r el at iv e

high leve l of f re e heat in t h i s example. Another observation i s

that the relation between the same methods changes then they

ar e used fo r d if f e r ent types of buil din gs, e.g. th e EFB-1 and

the BKL-methods a r e ra th er cl os e t o each oth er f o r t he Vetlanda

house, case (b) (Figure 5.21, but di ff e r s about 13%for the

T e h i k e m building. This again, r equire s more det ai le d examin-

at ions in order to f ind the reasons.

5.2 Predic tion s of energy conservation

The fi v e parame tric runs f o r th e Vetlanda house form, togethe r

with the original case, five examples of predicted energy con-

servation. In these cases , a l l six months have been u sed-as we

no longer have any po ss ibi li t y of cpmparisons with th e measure-

ments.

Figure 5 .4 give s an ove ra ll p ic tu re of t he estimated energy con-

servation. The following observations can be made:

The change from four t o two pane windows in case (a) in-fluences two parameters, namely the transmission losses

and th e so l a r gai n. The increased transmission lo ss es

w i l l increase the load while the increased solar gain

w i l l decrease the load. Methods that over-estimate the

so la r gain might the re fo re under-estimate th e change inenergy requirement. The re su lt s i n th is c ase var ied

widely, the gr ea te st estimated change i s 2 .1 times the

lowest.

In case (b), only the transmission losses are influenced

and the re su lt s come' sl igh tly cl os er t o each other than

in case (a) . Note also the rather large changes i t - h e

ordering of the results.



DOE-

EFB-

JAENV JSMECC

DD

NEVACA

KLIMASIM

NEVACA

BKL J

EFB-

DYWON-

KLIMASIM

LPB-4

BA- 4 1

Case (a)Double-glazing

Case (b)Change of insulation

1 KLIMASIM IBA- 4

JAENV

BKL

NEVACA-l DOE- JI NEVACA I

DYWON-

JULOn'A

I BKL I

Case (c)

Night set-backCase (d)Set-back & carpet

Case (e)Change in ventilation

Figure 5.4 The Vetlanda house. Estimates of energy conservation.



The night setback in case (c) again seems t o give a wide

range of r esu lt s . The pr in ci pl e used in EA-4 makes t h i s

method less suited for calculations of such conservation

methods a s night temperature setback. This i s indica ted

by th e low value f o r EA-4 i n cas e ( c ) .

In case (dl the nigh t setback i s combined with a carp et

which decre ases th e U-value of t he fl oo r. The va ri at io ns

in t he r esu l t s a re l ess in case (d) than in case (c) .

Only i n th e case (e) - change i n the ven tila tion ra te -most methods predict about the same energy conservation.

This is , on the othe r hand, not s urpr isi ng as the ven ti-

la tio n losses a re well defined.

The order between the methods depends on what type of r e t r o f i t

t ha t i s studied, e.g . DOE-2 predicts higher savings with change

of in su la ti on le ve l than KLIMASIM, but i n the cas e of ni gh t

temperature setback the situation i s th e opp osite. Note however,

th at the changing i n th e order overdramatizes the r e al changes

in absolute or relat ive terms.

The deviations in the predicted savings arise the same ques-

ti on s a s when looking a t t he energy consumption. In ord er to

understand the se devia tion s, the methods' way to cal cu la te

losses, in terna l and solar heat gain and i t s ut i l i za t io n, w i l l

be discussed in the following.

5.3 Calculation of ventilation losses

The energy consumption for the preheater in the Vetlanda house

i s a value r ef le ct ing the dif fe re nt methods' cap abi lity t o pre-

di ct venti lat ion los ses. For the orig ina l case (Table 3.1) the

greatest predicted consumption to the lowest is 1.04 t o 1.00,

reflect ing only the fact that there i s no standardized way to



assume the yearly mean value of the heat capacity of the air.

A s imilar p ic ture is shown by Table 5.1, where the specif ic

heat los se s per degree a re given f or some of th e methods.

Together with the earlier observations in section 5.2, a l l

methods can be assumed reliable i n the prediction of venti la-

.tion losses.

5.4 Calculation of transmission losses

For some of t he methods, th e sp ec if ic heat lo ss es through th e

flo or and through a l l other surfaces are specified in Table

5.1. A diffe rence of 1.0 W/'C i n specific transmission losses

corresponds t o about 0.6% changes i n t o t a l los ses and, with

respect t o the fr ee heat in the building, t o about 1 % of the

predic ted energy consumption. With t h i s as a background, the

only noticeable deviations are the transmission losses through

the other surfaces for the BA-4 and th e EFB-1 methods. This

might be a r e s u l t of th es e methods' use of a combined a i r and

surface temperature instea d of a di st in ct room a i r temperature.

Some methods use a fixed ground temperature instead of the out-

door temperature when calculating the energy consumption. A

0change of t h i s ground temperature by 1 C w i l l change the "de-

gree-hours" fo r th e fl oo r (by 4344 in t h i s example). For spe-

c i f i c losses of 25 W/'C this also causes a change of the same

magnitude as the above discussed for errors in the specific

losses.

The two above explanations f or di ff er en t t ransmission l osse s do

not expla in why the re su lt s f or transm ission through the fl oo r

seem t o f a l l in to t hre e groups. There i s no apparent reason f or

t h i s .

Using only the total specific losses i n Table 5.1, the EFB-1

method should give 1 2% higher energy consumption than theDYWON-2 method. According t o Table 3.1, the di ff er en ce i s 20%



and the dif ference s i n calculated losses cannot explain the

whole difference. Another example which requires an explanation

i s the difference between DOE-2 and JULUITA, where the differ-

ence i n to t a l speci f ic losses is about 1%and i n energy consump-

tion about 24%. This w i l l be discussed further in following sec-

t ions.

5.5 Calcula tion of in te rn al load

The degree-day method i s the only investigated method which

does not take dif fe re nt int er nal loads into account. This can,

in cases with high in te rna l lo ads, give great e rro rs . The heat

consumption pre dic ted by t he degree-day method fo r the T e h i -

kern building (Figure 5.3) il lu s tr at e s th i s problem. For a l l

the oth er methods, the re e xi st s no reason t o expect any di f fe r -

ences caused by di ff er en t cal cul at ion of t he amount of in te rn al

load. On the other hand, the ut il iz at io n of the interna l load

ky i f f er and th i s is discussed together with util ization of

solar gain in section 5.7.

5.6 Calculation of solar gain through windows

This section w i l l deal with s ol ar gain through windows; th e

solar gain through opaque surfaces is here totally neglected.

As described in chapter 1, the problem of calculating solar

gain through windows can be divided in t o th re e pa rt s:

calc ulat ing the radiati on incident on ve rt ic alsurfaces

calculating the effect of shading- calculating the transmission through the windows

These thr ee s teps have been examined for some of th e inve sti -

gated methods and ,some results are shown in Table 5.2 for the

Vetlanda house and in Table 5.3 for th e Teh ik er n building. The

re su lt s from a l l s ix months for the Vetlanda house a re als o i l-



lu st ra te d i n Figure 5.5, t o which the following discussion w i l l

r e fe r .

In Figure 5.5 t he r es ul ts of t he rad ia tio n on the south facade

of th e Vetlanda house without considering shading ef f ec ts a re

2shown. The values l i e between 638 and 946 MJ/m from the 6

months period. Two di f fe ren t circumstances can influence th e

re su lt s; namely, i f the methods take th e var iat ion of the

ground r e f l ec t i v i t y i nt o account and i f th e methods use average

values in the calculations. These deviations i n res ult s a re un-

acceptably large. One would think that there should be more uni-

formity, s ince t h i s s tep of the calculation is based upon wide-

ly accepted principles of geometry.

Figure 5.6 shows th e e f f ec t of shading according t o Table 5.2.

In cas e of BA-4, the r a t i o between columns 4 and 3 has been

used a s the shading r a t i o . The redcution fa ct or s vary from 0.39

t o 0.83. Four of the seven r e su l t s a r e grouped between 0.71 and

0.79 and according t o qu al it at iv e considerat ions t hi s band

seems reasonable for the shading effects. Figure 5.7 represents

the values of the transmission factor in the same manner.

Values range between 0.48 and 0.63, which is too wide a band t o

be accept able . Four of th es e values a re found i n the narrow

band between 0.53 and 0.55, which seems t o be an acceptable re -

s u l t fo r a four-pane window. To take th e most f requently used

val ues cannot be accepted a s th e proper method, but on the

other hand, i t seems t o be reasonable from a qu al it at iv e point

of view.

Fina lly, Figure 5.8 shows the s ol ar gain ins ide t he house and

the values vary from 2 .2 t o 1.0, a to ta ll y unacceptable vari-

at ion.

The great est ef fe ct i s t h a t of shading. This i s exemplified by

the results from the SKCC method which gives the greatest radi-

at io n on the outs ide but, due to t he great influence of shading,

has the lowest so l a r gain in sid e the house. See Figure 5.8..

Reasons f or t h i s la rge e f f e ct nay be not only a si mpli fie d geo-



SMECC

M E - 2 IDWON - 2 IBA-

4I

I J ULOTTA I-UIMASIM

0

F i g u r e 5 . 5 S o l a r r a d i a t i o n o n u n s h a d e d s o u t h f a c a d e .

M E - 2

J.GW

1 J ULOTTA I

I (% II I

0

J

2 0 4 0 60 8 0

F i g u r e 5 . 6 I n f l u e n c e o f s h a d i n g .



DOE- 2J A W 1BA- 4

DYWON- 2

I SMECC 1

F i g u r e 5 . 7 I n f l u e n c e o f tr a n s m i s s i o n .

IDYWON- 2KLIMASIPI

BA- 4

JULOTI'A

SPECC

( M J I ~ ~ )I I t I

01 -

100 200 300 400 500

F i g u r e 5 . 8 S o l a r g a i n t h r o u g h s o u t h w i n d o w s .



metrical calculation, but also inadequate 'howledge of the dis-

tribution of diffuse radiation. The wide variations in the ef-

fects of shading and transmission can, to some extent, be a

result of how solar radiation on the south facade is split in-

to direct and diffuse components.

5 . 7 Utilization of free heat

The calculation of the net heat consumption can always be

represented by

-.Q .n e t - Q o s s g a m

where Qloss is the sum of steady state heat losses exactly pro-

portional to the time integral of the inside-outside tempera-

ture difference, and Qgain the total available free heat in the

building. The free heat can be utilized for heating only for an

extent?, a factor partly representing the dynamic treatment of

the building by the calculation method.

In Table 5 . 4 , the steady state losses have been estimated ac-

cording to the specific losses in Table 5.1 and the method's

normal procedure of using sol-air temperature, etc. The total

sum of free heat is the solar gain not included in the steady

state losses and the internal loads. The net' eat requirement

is taken from Table 3.1 and the above equation is then used to

calculate the utilization of the free heat.

Table 5 . 4 shows a wide variation in the utilization factor

which can be explained partly by the different methods' math-

ematical models of the building.

5 . 8 Summary

In this chapter, deviations in the results of the investigated

methods have been pointed out and possible causes for these de-

viations have been discussed.



Although a detailed description of the building envelope was

given, the analysis ended up with a 15%variat ion in the loss

calculat ion. A much bigger spread would be expected i f each

analyst was t o ex tr ac t t he b ui ld in g data from drawings and

other basic sources.

The diff ere nt estimations of the s ol ar gain cause a variati on

of 10% in the t o t a l energy consmption of the building, a

spread which seems caused by d if f e rent ca lcu lat ion algorithms

rath er than by the anal ysts ' inte rpre tat io n of data.

The ut i li z a ti o n of fr ee h eat shows a var iat ion of 40% which

also results in another 10%variation in the total energy con-

sumption. This variation i s normally caused ent ir e ly by th e

mathematical model and cannot be inf luenced by th e use r .

Energy conservation effects are often smaller than the vari-

ati on re sul tin g from a use of di ff er en t calcul ation methods.

This underlines the fact that comparison of different conser-

vation measures must be based on estimations using the same

ca lc ula t io n method. The ana ly st must i n each case make h i s ownchoice, but t he ana ly si s in t h i s chap ter may give some guidance



Table 5.1

LPB-4

BA-4

EFB-1

SMECC

DYWON-2

KLIHASIM

JULOTTA

JAENV

DOE-2

Sp ec if ic Heat Losses (W/~egc ) ,

The Vetlanda House, Original Case.

Vent. Transmiseion Losses Total

Losses Floor A l l Other Losses



67

Table 5.2 S o l ar Ra dia tio n o n South Facade, Vetlanda House (Mj/MH2,Month)

December

BA-4SMECC

DWON-2

J A E NV

DOE-2

JULOTTA

KLIMASIM

A p r i l

BA-4

smccDWON-2

J A ENV

DOE-2

JULOTTA

KLIMASIM

6 Month

BA-4

SMECC

DYWON-2

JAENV

DOE-2

JULOTTA

KLIMASIM

1: Unshaded on Outside

2: Shaded on Outside

3: Unshaded t roug h Vindow

4: Shaded tr ough Window

5: Shading = 2/1

6: Transmission = j / l



Tab le 5.3 S o l a r R a d ia t i on o n S o u th e a st Fa cad e, Tek ni ke rn ( M J / ~ ~ 2 , ~ o n t h :

1: Unshaded on Out s id e 4: Shaded trou gh Windov

2: Shaded on Ou ts id e 5: Shading = 2/1

3: Unshaded tr ou gh Windov 6: Tra nsm iss ion * 3/l

December

BA-4

SMECC

DYWOA-2

JAEINDOE-2

June

BA-4

SMECC

DYUOA-2

JAEAV

DOE-2

Year

BA-4

SKECC

DYUOA-2

JAENV

DOE-2

Table 5.4 U ti l i za t i on of Pree Heat i n the Vetlanda House

(Original Case, Period December-April)

Heat Solar T o t a l Net Heat U t i l i e ~ t l o n

Losses Gain Fre e Heat Requirement. of Fre e Heat

BA-4 43279 565% 13968 29277 1 O'O

EPB-1 451 42 57418 14056 32892 0.87

SKEC 431 24 7328. 15643 33706 0.60

DTYON-2 41555 7732, 16047 n 5 5 3 0.87KLIIUSIM 39055 5084, 13399 28789 0.76

J A W 45008 7780 16095 33997 0.68

DOE-2 43951 91 39 17454 26849 0.98

MEAS. 33889.......................................................................# Sola r Gain through Wi n d . ~ ~nd o ut er Opaque surface s. '

So la r Gain throu gh Windov + lo% f o r O uter Opaque S urf ace s... ..



6 CONCLUSIONS AND RECOhMENDATIONS ON CALCULATION METHODS

After the analyses summarized in the previous chapter, the par-

ticipants have agreed on the following conclusions and recom-

mendations. It was also noted, that the analyses in this task

confirm the conclusions and recommendationsin

the report "Com-parison of Load Determination Methodologies for Building Energy

Analysis Program, prepared for International Energy Agency, En-

ergy Conservation in Building and Community Systems" by U.S.

Department of Energy (WE/CE/20184-I), 1981.

In one case, the annual heat consumption predicted by the dif-

ferent calculation methods has been compared with measured con-

sumption. Even if some of the methods give a result close to

that of the measurements, this might be a result of opposing

eroors. Thus more detailed verifications of calculation methods

have to be carried out.

The parametric runs on the Vetlanda house have shown the var-

iety in the methods' response to different conservation meas-

ures. These measures have been chosen as rather normal conser-

vation measures. In order to test the calculation methods, the

variations could have been chosen with greater liberty in order

to really probe the range of values resulting from the methods.

Calculation methods are of course important when planning field

experiments. This is because the results from the experiment

should be generalized. Thus the interaction between the calcu-

lation method and measured quantities has to be carefully ob-

served. The data required by the calculation method as input

have to be measured, nd the calculation method has to be able

to handle the conservation measure in question.

The explicit purpose of this subtask was to find - if at all

possible - one calculation method to recommend when predicting

energy consumption and energy conservation in residential build-

ings. However, no such method can be clearly recognized, and we

have instead discussed different qualities of the calculation



methods and how well they f u l f i l l t hese q ua li ti es . The conclu-

sions regarding these q ua li t i es are based on:

- Resu lts from th e work i nsi de t h i s subtask

Information from th e author s of th e met.hods

The analysts ' experiences with different calculation

methods

The following table gives the cross-reference of chosen qual-

i t i e s , methods and conc lusio ns. Carefully examined, th e table

could be useful when selecting a method for a certain purpose.

Confidence inenergy conser-vation by

The methods' treat- Otherment of qualities

Method ClassificationI I according tosection 2.1

IWCC 1'0ther correlation I H H H L M 1 C C S S S S S S G G M

LPB-4

BA-4

EFB-1

LO

2 : ;> m nW L X" 2 2g . 2 2 5 . 2 z . 2. 4 u . 4 w u L O u

- o m 3m 4

4 . 4 u u . 52 U O Z D

" c ; . ? E L

Iw I Degree-Day I H H M L L I s S - - - - - - 1 G G -

C.d

m E LOC M W.4

m - 5 2 - cM - 2 5 2

LOL E 3 mm w 4 A Y l w

W L O c - L O U C L O

L O W 0 O C L . 4Yl LO .4 LO .d L 40 LO LO c . d : E- o L O W W o n L L

4 X M . 4 0 0u : ,O ,Og c S . 5 E c c w . 4

2 : 5 . 2 . 2 5 2 2 2m m . 4 m

$ L ' s z s s m n Dm m

N N L C U UL L .4 .d

w0 .r l .4

g 2 z z . 2 z z 4 4w w u u E g g"P- 3 z z u e > 1 8 8 x z F u < <

Other correlation

Finite dcfferences

Other correlation

DYLI'OY-2

N , I b L S I M

NC\'AU\

G G -

H H M L M

H H H M H

Finite differences

Response factors

Corrected kgree-Day

BN,

JUW l TA

J A F N

WE-2

- - G

- G -

G G G

S C - C S S C S G

C C C C C C C S G M G

H H H M M C C S S S S S - G G -

H H H H H

H H H H H

H H M L L

Other correlation

Finite differences

Other correlation

Keighting factors

C C C C C C C C

C C C C C C C C

C C - - S - - -H H H M M

H H H H H

H H H M M

H H H H H

H: HighM: EkdiumL: Low

C C - S S S S S

C C C C C C C C

C C S S S S S -C C C C C S S .C.C: Complex5 : Simplified-: Not at all



7 INFLUENCE OF INHABITANTS

7.1 Background

The energy consumption of residential buildings is normally

influenced by the ir?habit.ants. Thus the inhabitants and their

behaviour must be taken into account when generalizing results

from f i e l d experiments on re t r o f i t s of buildings. I t i s also

hown, th at l arge spreads in energy consumption are shown even

f o r id en ti ca l houses. The influence of inha bitan ts may also

have a s pe ci fi c nationa l char acte r which requir es a howledge

in order t o generaliz e from fi e l d experiments.

The ori gin al idea f or t h i s task, was t o tr y t o model the in-

fluence of inha bita nts as a si ng le parameter representing the

di ff er en ce between an unoccupied and an occupied bui lding. Thi s

idea should allow t o simply transform the r es ul ts from experi-

ments on r e t r o f i t s between di f f e re nt coun tri es. However, the

question of inhabitants' behaviour i s a complex one, and wi th

exi sti ng experimental data only one attempt t o es ta bl is h such

a f ac to r has been ca rr ie d out and i s re ferred t o i n sec t ion

7.3.

The quest ion of th e inf luenc e of inh abi tan ts has, a s mentioned

before, al so been tr eat ed in subtask B of t h i s annex (IEA En-

ergy Conservation i n Buildings and Community Systems, Guiding

Pr in ci pl es concerning Design of Experiments, Instrumenta tion

and Measuring Techniques, 1982).

7 .2 Demands from the calculation methods

Every method used t o pr edi c t th e energy consumption i n a bu ild -

ing must take into account the management and use of the build-

ing, i .e .

Int e rna l loa ds produced by: occupancycookingl ight ing



hot tap watere t c

Management of th e hea tin g boi ler s

and ve nt il at in g system: setp oint se t c

Factors which change the mechanical ventilation

heat losses: a ir i ng .shut te rs & blinds

e t c

In princ ip le every method has t o use schedules and/or formulas

giving values and durations f or a l l these parameters. The

necessary de t a i l s i n th ese schedules and formulas depend on how

detailed the model i s used, f o r example an hour-by-hour ca lcu-

la ti on might (but not nec es sar ily ) need hourly da ta, whereas a

method based on monthly periods might use monthly average

values. The po ssi bl e combinations between the groups depend

pa rt ly on the building , e.g. without thermostats no setp oint

adjustment can be performed.

The influence might a l so be divided i nt o the following groups

according t o th e manner which each fac to r can be t re at ed i n

cal cul ati on methods:

Unconditional management measures

heating operationbui l t - in se tpoin ts

fixed mechanical ventilationtime-controlled systeme t c .

"Common pat ter ns" o r schedules

occupancy

cookingshu t t e r s

e tc .

"Conditional p atte rns " or formulas

l igh t ing

se tpoin ts

a i r i n ge tc .



The unconditional measures a re easy to handle i n a model, as

they a l l ar e functi ons of time or outdoor conditions, et c. and

do not depend on th e i nhab itants; they ar e bu il t- in p a r t s of

the system. Such measures can be l e f t out side our discuss ion of

the inhabitant factor.

The "common pa tt er ns " inc lude the use of a building caused by

tradi t ions in the cul tural area in question, e.g. i t seems t o

be a reasonable assumption t hat occupancy and cooking i n each

ar ea o r country fo llow some cormnon pa tt er n. This might a lso be

va li d f or t he use of sh ut te rs during the night, and some types

of airi ng, et c. Also so ci al facto rs, family si ze , et c. might

inf luence thes e pa t terns . Cormnon patterns may be given as time

schedules.

The "condit ional pa tterns" include the behaviour of the in-

hab itants r el at ed t o the indoor climate. For example, a high

indoor temperature can result in the opening of windows and

turning down the thermostat. Another example is the use of

l igh t ing and bl in ds which depends on the amount of daylight

and the sol ar ins olat ion reaching in to the building. Condi-

ti on al pa tt er ns may be given a s funct ions of indoor climate,outdoor climate, thermal comfort, e tc .

The above discussed pa t terns may a lso change in time (changes

in family si ze , et c. ) or by r e t r of it s of the building. One

example of the l a t t e r type of change is when extra insulation

i s used t o obta in hig her comfort from increased indoor tempera-

tu re inst ead of saving energy by maintaining the lower indoor

temperature.

7 . 3 A Swiss investigation

Looking fo r a si ng le family build ing t o monitor and t o obta in

dat a f or va li da ti on of simulation programs, M A hose a pre-

fabricated type, which has been built 100 times in a more or

less identical version. A t the same time the discussion about



user influence came up. They decided as a result to enlarge

this experirent of measuring an empty house by a survey of as

many of these house owners as possible. With the advantage of

knowing very well the behaviour of the empty house, they ex-

pected more quantitative information about the user influence

itself. They obtained the agreement of about 60 house ownersto cooperate in this investigation.

The occupant influence on heating energy consumption is here

defined as the difference between the energy consumption for

heating of an unoccupied house kept at 20 OC during the day and

setback 2 OC at night, and the energy consumption for heating

of the same house with occupants.

The experimental investigations with the unoccupied house have

been used to check and adjust the program JAENV, which was then

used to calculate the theoretical energy consumption of each of

the other 60 houses individually under unoccupied conditions.

The difference between the measured effective value and the

calculated value was then established as user influence. The

total sample of occupant influences was divided into classes

regarding different parameters. The distributions within the

classes were then analyzed, whether their mean values are sep-

arated significalntly (i.e. the parameter under consideration

has some influence) or not.

One of the major results of this investigation is the finding

that the mean value of the occupant influence is practically

zero, but with a spread of 250% of the consumption of the un-

occupied house as defined above. According to these experiments

the conclusion is that in simplified calculation methods no in-

habitant term has to be introduced, but.one has to be aware of

the spread the occupants may cause.

This interpretation has some limits. The special kind of rela-

tively cheap single family houses under consideration prese-

lected a certain class ~ f ' ~ e o ~ l eo buy them. These may not be

representative of the Swiss population in general. Moreover it



is t o be expected t ha t th e occupant influence associ ated with

people living i n rented apartments, or i n other socia l or

cl im at ic circumstances, could be much di ff er en t.

For some parameters a correlation with a narrow margin above

th e confidence l ev el was found; fo r some oth ers t he cor re la ti on

was l os t in th e er ro rs . From these r es u lt s we can lea rn t ha t

in order for an investiga tion of t h i s kind t o bring more sig -

ni fi ca nt re su lt s, th e sample has t o be enlarged by about a fac-

t o r of 10. Knowing the spread of t he occupant behaviour one can

eval uat e the necessary s iz e of th e sample in case one wishes t o

investigate the interaction between the occupants and different

types of r e t r o f i t s .

Ref.: DPA report 41643.

7.4 Conclusions and recommendations

After discussions of t he material in th i s chapter, the part ic i-

pant s i n t h i s annex reached some conclusions of how th e i nf lu -

ence of i nh ab ita nt s can be tr ea ted .

F i r s t , a l l pa r t i c ipan t s agreed on the fa c t tha t t h i s is a com-

plex problem and a complete investigation could not be included

in t h i s annex. Nevertheless, it is a very important f i e l d and

furthe r invest igat ion s ought to be carried out .

Second, the approach to describe the influence of h a b i t a n t s

a s common pa tt er ns and cond iti onal pat te rn s is very useful when

one needs t o ge ner aliz e re s ul ts from fi e l d experiments on en-

ergy-saving measures. These patterns are clearly indicated as

s ta t is t i c al data and have to be t reated a s such. I t i s there-

fo re of gre at importance i n a f i el d experiment th at ei th er th e

det aile d patte rns ar e monitored, or t ha t th e experiment i s de-

signed so t h a t "rep res ent ativ e patterns " can be assumed. I f

t h i s is done, a calculation method ca be used with clirnatlg

data and "representative patterns" fo r another location, thus



allowing the results to be generalized to other places and

countries. How detailed these patterns need to be partially

depends upon the calculation method used.

Third, with the limited number of available experiments, it

has not been possible to evaluate the original idea of treat-

ing the inhabitant factor as a constant. The Swiss investiga-

tion shows however, that further experiments in this direction

ought to be carried out.



APPENDIX A

PARTICIPMS

Method Analyst Organization Telephone

LPB-4 D.Marret CSTC 02-6538801Rue de Lombard,411000 BmellesBelgium

Yves Delorme Laboratoire de Physique 041-520180du Batiment ext .367

UniversitE de Lisge15 ,avenue des Tilleus-BAT.DlB-4000 Likge

Belgium

BA- H. undEFB-

Technical University of 02-88351Denmark ext .5286Building 118DK-2800 LyngbyDenmark

SNECC R. Zecchin Padova University (049)653533NBSLD 1st .Fisica Tecnica

1-35100 Padova

Italy

DWON-2 H.A.L.van Dijk 1nst.of Applied Physics 01 5-569300TNO-TH, P.O.Box 155 ext .2442600 AD-DelftThe Netherlands

KLIht4SIh1 R.S. oeleman TNO-Research Org . 01 5-569330P.0.B.214 ext.24132600 AE-DelftThe Netherlands

NEVACA A.-C.Andersson Lund 1nst.of Technology 046-107000DD Dep.of Building Technology

?.O.Box 725

S-220 07 LundSweden

BKL K.Kallblad Lund Inst .of Technology 046-107000JULQTTA Dep.of Building Science ext.7358

P.O.Box 725S-220 07 Lund

Sweden

JAEMi J.Gass FMPA 01 -823551DOE-I Uberlandstrasse 129

CH-8600 DiibendorfSwitzerland



WPENDIX BBUILDING SPECIFICATIONS

1 SPECIFICATION OF THE VETLANDA HOUSE

1 . 1 L o c a t i o n, o r i e n t a t i o n a nd s u r ro u n d i n g s

The house is s i t u a t e d i n t h e town o f V e tl an d a w i t h

L a t i t u d e 5 7 .4 2 d e g r e e s N o rt h

Long i tude 15 . 0 8 d e g r e e s E a s t

(Degrees and dec imals o f deg ree )

T he l o n g i t u d e o f t h e s t a n d a r d t i m e m e r i d i an i s 15 .00 degrees

e a s t ( Sw e d is h S t a n d a r d T i m e ).

The house is t o som e e x t e n t s h ad e d f ro m t h e sun by t h e s u r r o u n d -

i n g s a nd some n e i g h b o u r i n g b u i l d i n g s , b u t a s t h e s o l a r an d s ky

r a d i a t i o n a r e measured a t t h e s i t e t h i s s h i e l d i n g c an b e ne-

g l e c t e d an d t h e h o u s e c a n b e assu med t o be p l a c e d w i t h a f r e eh o r i z o n .

T h i s may i n t r o d u c e some s m a l l e r r o r , e s p e c i a l l y a t low s o l a r

a l t i t u d e s d u r i n g t h e a f t er n o o n s , b u t t o overcom e t h i s o n e h a s

t o t ra n sf o rm t h e m easured s o l a r r a d i a t i o n t o r a d i a t i o n w i t h a

f r e e h o r i z o n . T h i s t r a n s f o rm a t i o n w i l l i n t ro d u c e new e r r o r s an d

t h e r e f o r e t h e m ea su re s v a l u e s a r e u se d a s t he y a r e .

The ground i s assumed t o h av e a d i f f u s e r e f l e c t i v i t y t o a l l

s o l a r r a d i a t i o n o f 0 . 2 w i t h o u t s now a nd 0 . 8 w i t h s now . The

p e r i o d s w i t h snow a r e g i v e n i n t h e w e at he r t a p e s p e c i f i c a t i o n .

T h e n o r t h f a c a d e i s f a c i n g 1 5.0 0 d e g r e e s w e st o f n o r t h a s s h o m

i n F i g u r e 1 l .



1 . 2 Dimensions

Figures 1.2 - 1.6 show th e plan, a sect ion , elevat ions and win -

dow and door d e t a i l s of th e house. These fig ures provide a l l

dimensions of t he building. Note th a t the sca le s might be e r -

roneous, so no dimensions should be measured from the figures.

The d e t a i l s of th e house a re somewhat simp lif ied t o avoid mis-

understandings, but care in any case should be taken t o inc lude

a l l pa rt s t ha t influence the energy consumption of the building.

1.3 Thermal pr op er ti es

1.3.1 Fabrics

The thermal pro pe rt ie s of th e materi als a re given in Table 1 . 1 .

In case of composite la ye rs , e.g. insu lat ion between stud s,

weighted properties are used.

1.3.2 Outer opaque su rf ac es

The emissivity for long wave radiation is 0.93.

The abs orp tivi ty fo r so la r radiation of a l l outer opaque sur-

faces i s 0.80.

If more detailed calculation is not carried out, following film

co ef fi ci en ts must be used:

Convective 11 .7 w/mL, KRadiative

24.3 W/m , K

1.3 .3 Inner opaque su rfaces

The emissivity for long wave radiation is 0.93



No inner surfaces i n the a t t i c i s exposed t o s o lar rad ia t ion .

For the rooms' surfaces, the absor pti vit ies fo r sola r radiat ion

are:

Ceilings 0.20

Floors 0.70

Inside ex te rn al doors 0.80

Other in te rn al surfac es 0.50

If more deta ile d calcu lat ion is not carr ied out, following fi lm

coeff icient s ar e t o be used:

Convective a t bottom f loo r 2.0 w/m2,g2

Convective a t othe r surfa ces 3.0 W/m ,K

Radiative2

5.3 W/m ,K

1.3.4 Windows

A l l windows are sealed units with 4 nun cl ea r glass with 12 m

a i r gaps between them. The south f acing windows ar e q~ ~ ad ru p l e

glazed, a l l others t r i pl e glazed, except fo r door unit s, which

are double glazed.

The cha rac ter is t ics fo r solar radiat ion of n o m l incidence on

a single sheet of glass are:

Transmitted 0.85

Reflected 0.07

Absorbed 0.08

The heat r esi sta nce and capacity of the gla ss ar e neglected.

The emi ss iv it y f o r long wave radia tion is 0.93 f o r a l l g la ss

surfaces.

If more de ta il ed ca lc ul at io n of heat tr an sf er through windows

i s not carri ed out, the following film coef fic ien ts and resi s-



tances must be used:

Outsideconvective 11.7 I Y / ~ ' , K

Outside ra di at iv e 4.3 w / m 2 , ~

Inside convective2

3.0 W/m , K

Inside radiative2

5.3 W/m , K

The thermal resi st an ce of each a i r gap 0.17 mL K / W .

1.4 In f i l t ra t io n and vent i la t ion

In fi l t ra ti on and vent ila tio n f lows are given i n Table 1 . 2 with

the flows referr ed t o a 20 OC a i r temperature.

The infiltration flows are assumed as independent of the venti-

lation system.

Forced ventilation is used between 12.00-13.00 each day, othe r-

wise nomal ven til ati on i s used.

The in le t ai r is preheated t o 16 OC hlhenever th e outdoor a i r i s

below th is level, othenvise the i nl et a i r has outdoor a i r tem-

perature and th e preheater i s completely cut of f.

The a i r flows between rooms passes the inter nal doors and

should be calculated only from the ventilation flows without

regard t o the in f i l t r a t i on . No flow is assumed between room 1

and 3. The fan powers are totally excluded from the calculation.

1.5 Heating system

The heating system consists of the preheater i n the vent i la t ion

system and di re ct el ec tr i ca l ra dia tor s i n each room, except f or

room 10, which has no heat ing system. In the ca lcula t ions we

assume "perfec t" systems in each room-

i . e .



Each room-thermostat has a se tp oi nt of 20 OC and the sys-

tem provides each room with the exact amount of heat

needed t o maintain th e a i r temperature in th at room. I f

th e indoor a i r temperatu re i n a room exceeds 20 OC, a l l

heat t o t h at room is c u t o f f .

1.6 Domestic ho t water

No domestic hot water i s used.

1.7 Inte rnal loads

A l l in te rn al loads a re assumed as convective heat t o th e room

a i r . The loads and schedules are fo r a l l days:

Room 1 800 W between 12.00-13.00, otherwise 300 WRoom 6 100 W " 07.00-08.00, ,( zero

Room 7 200 W " 07.00-08.00, ) I zero

Room 8 100 W " 07.00-08 . OO , ( 1 zero

Room 9 875 W " 20.00-21 .OO, I, 275 W

1.8 Specif icat ions for the parametric studies

These stu di es include fi c e cases, each case spec ifi ed by the

or ig in al sp ec if ic at io n of the Vetlanda house and the changes

given in the l i s t below. Note t ha t t he only changes in a spe-

c i f i c case a re those g iven in the l i s t , e.g. case (c) has the

ori gin al type ofwindows, wall ins ulatio n etc., and the only

change from the original specification is the setpoint schedule.

Case (a)

All windows are double glazed units with 4 m c le ar g lass with

.1 2 m a i r gaps.



Case (b)

External walls (not south):

Insulation layer changed from 190 nun t o 50 nun.

Inner roof:

Fibe r gla ss of 130 nun is excluded

Insulation layer changed from 170 m t o 100 nun.

Case (c)

for the preheater 's se tpoin t! ) .

Case (d)

The setpoints in a l l rooms a re 16 OC between 22.00-06.00 during

a l l days of t h e week, and 20 OC the remaining hours. (No change

Floor slab:

The t h e m 1 esistance of the carpe2

t o 1 .OO rn , /W .

: t changed from 0.08

Setpoints:

As in case ( c).

Case (e)

Double a l l normal ven ti la ti on and exclude the forced ve nt il a-

t ion. Use in fi l t ra t i on as i n the original specification. Thus

we have f o r 24 hours/day:



3Room I n f i l t r a t i o n ' Ve nt il at io n (m /h)

Inlet Exhaust

To tal 24.1 200.0 200.0

Attic 265.0

1.9 Simplifications

Some of the in ve st ig at ed c al cu la ti on methods can not handle

many d e t a i l s . For the se methods th e following sim pl if ic at io ns

were recommended by the analysts.

Number of rooms---------------As a f i r s t s te p th e house can be assumed t o have f i v e rooms as

follows:

a) Room 2, 3, 4, 5 and 10 a r e t r ea ted as one room

b) Room 7 and 8 are treated as one room

C) A l l p a r t i t i o n wall s i nsi de these new rooms ar eneglected

As a second step the whole house can be treated as one room

without any pa r t i t io n wa l l s ' a t a l l .

All room numbers a r e accord ing t o F igures 1 .1 and 1.3.



The att ic---------In order t o simplify the treatment of the at t i c , i t s ventila-

ti on, th e outer roof and the outs ide film coe ffi cie nt can be

2modeled with a thermal resistance of 0.30 rn ,K/W between the

top of the inner roof and the outdoor air. The given value canbe discussed, but t o avoid problems a l l pa rt ic ip an ts were re-

commended t o use e i t h e r t h i s value or t he or igi nal spe cif ica -

t ions.



Figure 1 . 1 Orientatio n and room numbers.

Figure 1.2 Section of the Vetlanda house.



Figure 1.3 Plan and inner dimensions of the Vetlanda house.



T o t a l s i z e G l as s s i z e

I$ ; GI\' GI{

S o u t h 1200 1700 1090 1630

Nor th 1100 570 91 0 380

Yorrh 600 570 410 380

Nor th 1 1 0 0 8 7 0 9 1 0 6 8 0

E a s t - N e s t 1100 1270 910 1080

Framework

FT RI' HI FS

180 55 35 20

1 0 0 9 5 9 s 2 0

100 95 95 20

100 35 . 9 5 2 0

100 95 95 20

F i g u r e 1.5 I \ ' indow dimensions of t h e V e t la n d a h o u s e .



Door

INSIDE

F i g u r e 1.6 E x t e r n a l d o o r s ' d i m e n si on s o f t h e V e t l a n da h o u s c.

.. ..



Table 1.1 T h e m 1 propert ies of bui ld ing fabric

Thick- Conduc- Density Spe cif ic Thennal

ness t i v i ty heat res is tance~~

m W/m, K kg/m3 Jlkg, K m2 , K/W

INTERNAL WALL

Gypsum 13 0.46 900 1090

Cavity 45 - - - 0.05

GYPsum 13 0.46 900 1090

INTERNAL D03R

Wood 25 0.14 500 2720

EXTERNAL WALL

(SOrnOInside

Board 13 0.14 600 1340

Insul. layer 95 0.070 150 880

Asph . oard 12 0.050 340 1340

Cavity 20 - - - 0.05

Wood panel 40 0.14 500 2720

Outside

EXTERNU \$ALL

(NOT SOUTH)

Inside

Gypsum 13 0.46 900 1090

1nsul.layer 190 0.055 150 880

Asph. board 12 0.050 340 1340

Cavity 20 - - - 0.05

Wood panel 40 0.14 500 2720

Outside

EXTERNAL D03R

Wood panel 20 0.14 340 1300

Insulation 50 0.050 100 600

Wood panel 20 0.14 340 1300



Table 1.1 Thermal p ro pe rt ie s (cont 'd)

Thick- Conduc- Density Sp ecific Thermal

ness t i v i ty hea t res i s tance

m W/m, K kg/m J/kg,K2

m ,K/lV

INNER ROOF

Top

Fiber glass

Insul. layer

Cavity

'3'Ps'Jm

Bottcm

OUTER ROOF

TOP

Roof bricks 20 1 .74 2000 880

Cavity 20 - - -

Wood panel 20 0 .14 500 2720

Bottom

FLOOR SLAB

Carpet - - - - 0.08

Concrete 160 1 .74 2300 880

Insulation 120 0.040 150 750

Ground - 1.15 140 0 1670

FRAM3VORK

Wood See f igures 0.14 500 2720

ATTIC IVALLStWood 20 0.14 500 27 20- -

* and the a t t i c fl oor extending beyond the south facade.



Table 1.2 Infiltration and ventilation in the Vetlanda house

~ o o m Infiltration3

Ventilation (m /h)

N o m l Forced

(m3/h) Inlet Exhaust Inlet Exhaust

TOTAL 24.1 100.0 100.0 250.0 250.0

Note: for forced ventilation given values should be used,

not sum of nonnal and forced.



2 SPECIFICATION OF M TEKNIKERN BUILDING

2 ; 1 L o c a ti o n , o r i e n t a t i o n an d s u r r o u n d in g s

T h e b u i l d i n g i s s i t u a t e d i n S to ck ho lm w i t h

L a t i t u d e 5 9.3 3 d e g r e e s N o r th

L o n g i t u d e 1 8 .0 3 d e g r e e s E a s t

( D e g r e e s a n d d e c i m a l s o f d e g r e e s )

T he l o n g i t u d e o f t h e s t a n d a r d t i m e m e r id i an i s 1 5 . 0 0 d e g r e e s

e a s t ( Sw e di sh S t a n d a r d T i m e) .

T h e b u i l d i n g i s a ss um e d t o be p l a c e d w i t h a f r e e h o ri z o n a n d no

s o l a r s h a d in g f r om n e i g h b o u r in g b u i l d i n g s .

The g round ha s a d i f f u s e r e f l e c t i v i t y t o a l l s o l a r r a d i a t i o n o f

0 .2 0 d u r i n g t h e w h ol e y e a r .

The N /E f a c ad e o f t h e b u i l d i n g i s o r i e n ta t e d e x a c t l y t o t h e

n o r t h - e a s t .

2 .2 Dimens ions

F i g u r e s 2 .1 - 2 . 3 s h o w p l a n s , s e c t i o n s and e l e v a t i o n s o f t h e

b u i l d i n g . A l l d i m e ns io n s a r e p r ov i d ed on t h e s e f i g u r e s . N o te

t h a t t h e s c a l e s m ig h t be e r r o n e o u s an d t h e r e f o r e , n o d im e ns io ns

s h o u l d b e m e a s u r e d f r o m t h e f i g u r e s .

T h e b u i l d i n g s i s t o a g r e a t e x t e n t s i m p l i f i e d t o a v o id m isu nd er-

s t a n d i n g a nd t o r e d uc e t h e c a l c u l a t i o n w ork, b u t i t i s a c c e p t -

a b l e i n t h i s form s i n c e i t i s u s e f u l w hen s t u d y i n g t h e c a p a b i l -

i t y o f d i f f e r e n t m ethods t o h a nd le b i g b u i l d i n g s.

I n t e r n a l d o o r s a r e a ss um ed t o h av e t h e s am e t h e r m a l p a r a m e t e rs

a s t h e w a l l s i n wh ic h th e y a r e p l a c e d .

A l l f ra m ew or ks a r e n e g l e c t e d , t h u s t h e d im e n si on s on F i g u r e 2 . 3



give gla ss and extern al do ors t areas.

2.3 Thermal pr op er ti es

2.3.1 Fabrics

The thermal properties are given in Table 2 . 1 .

2.3.2 Outer opaque su rf aces

The emissivity for long wave radiation is 0.93.

The abso rpt iv i t ies fo r so la r radia t ion are :

Roof 0.80

Walls 0.50

Constant values used f o r fi l m coe ffi cie nts are:

2Convective 1 1 . 7 W/m , K

Radiative2

4.3 W/m , K

2.3.3 Inner opaque su rf ac es

h i s s i v i t y fo r long wave radiat ion: 0.93

Absorpt iv i t ies for so lar radia t ion:

Ceilings 0.20

Floors 0.70

Other 0.50

If constant values are used, the f i lm coeff icien ts are:

2Convective 3.0 lV/m , K

Radiative2

5.3 W/m , K



2.3.4 Windows

All windows have 2 panes of 3 mm clear glass with 20 mm air

gaps between them.

The characteristics for solar radiation of normal incidence on

a single sheet of glass are:

Transmitted 0.86

Reflected 0.07

Absorbed 0.07

The heat resistance and capacity of the glass are neglected.

The emissivity for long wave radiation is 0.93 for all glass

surfaces.

If more detailed calculation of heat transfer through windows

is not carried out, the following film coefficients and resis-

tances have to be used:

2Outside convective 11.7 W/m ,K

Outside radiative2

4.3 W/m ,K

Inside convective2

3 .0 W/m ,K

Inside radiative2

5.3 W/m ,K

2Heat resistance of each air gap 0.17 m ,K/W.

2.4 Infiltration and ventilation

Infiltration and ventilation' flows are given in Table 2.2 with

the flows referred to a 20 OC air temperature.

The infiltration is assumed to be the same during the whole

year and independent of the ventilation flows.

The ventilation system is an exhaust air system.



2.5 Heating system

In each room in th e basement we assume a "perfect" heat ing sys-

tem - i . e .

Each room thermostat has a s et po in t of 20 OC and the sys-

tem provides each room with the exact amount of heat

needed t o maintain the a i r temperature in tha t room. I f0

th e indoor a i r temperature i n a room exceeds 20 C, a l l

heat t o th at room is cu t o f f .

2 .6 Domestic hot water

No domestic hot water i s used.

2 . 7 Internal loads

The basement and the a t t i c have no in te rn al loads. Each room

has the in te rn al load given in Table 2 . 3 . These loads are

tre ate d a s convective loads t o the room a i r and w i l l be the

same a l l days of the yea r.

2.8 Treatment of the basement

To avoid the problems associated with calculating the heat

tr ans fer with t he ground, th e basement has been excluded andthe bottom surface of th e s la b between basement and f i r s t f l oo r

is kept a t 20 OC. This sim plifica tion i s included in the re-

s u l t s in th i s repor t .



F i g u r e 2 . 1 P l a n s of t h e T e k ni ke rn b u i l d i n g .



Figure 2.2 Secti on of the Teknikern building.



F i g u r e 2 .3 Facades of t h e T e kn i ke r n b u i l d i n g .

Section A - A (South-east facade)

1SO, 3200 .I400 . 3250 ,ASO7

Section B - B (North-west facade)



Table 2.1 Thermal pro per tie s of building fa bri c, the Telaikern

building

Thick- Conduc- Density Sp ec if ic T h e m 1

ness t i v i ty heat res is tance- .- ~ p ~ ~INrERNAL WALL

150 m

Concrete 150 1 50 2300 880

I N r W WALL

100 m

Gypsum 13 0.46 900 1090

Cavity 74 - - - 0.05

Gypsum 13 0.46 900 1090

E X T E W WALLConcrete 40 1.74 2300 880

Insulation 70 0.040 150 880

Concrete 4 0 1 . 7 4 2300 880

BASEMENT FLWX

Concrete 250 1.74 2300 880

Ground - 1.15 1400 1670

SLABS

Concrete 200 1 .50 2300 880

IhWER ROOF

TOP

Min .wool 200 0.040 1 50 880

Concrete 200 1.74 2300 880

Bottom

OUTER R03F

To?

Asph.roofing 0.03

Board 20 0.14 34 0 1300

Bottom



O

5

PP

-O

E



Table 2.2 Infiltration and ventilation in the Teknikern

building

Infiltration Ventilation

(m3/h) (m3/h)

Attic 300.0-

Per room 10.0 50.0

Basement 85.0 425.0

Table 2.3 Internal load per room

in the Teknikern build-

i n g

Hour Load

(W)

0 - 06 250

06 - 07 750

07 - 16 250

16 - 17 1000

17 - 20 400

20 - 24 250



AF'PrnIX C

WTHER DATA SPECIFICATIONS

Two se t s of weather dat a have been used i n t h i s work. I t shouldbe noted, th at both of these were prepared fo r th i s spec ial

task, and the data should not be used without care for other

purposes.

The main c ha ra ct er is ti cs of the weather data a r e given i n Table

1 and Table 2 . In these t ab le s, the degree-hours ar e defined by

A l l values given i n the tables are calculated from the hourly

values on the weather tape specified i n this appendix.



Table 1 Win ch ar ac te ri st i cs of the weather data in Vetlanda,

November 1976 - April 1977

Month Mean Degree-hours So la r radiat ion onoutdoor the horizo ntal plane

temp.(OC).

1 6 OC 20 OC Diffuse Global

Nov 1.57 10390 13270 8904 12046

Dec -4.36 15148 18124 5589 6158

Jan -3.44 14462 17483 8377 10909

Feb -4 .74 13935 16623 17448 21367

March -0 .27 12105 15081 30282 52954

April 1.95 10114 12994 44073 70835



Table 2 !&in characteristics of the weather data in Stockholm

1971

Month Mean Degree-hours Solar radiation onoutdoor the horizontal planetemp.

(OC) (OC.h/month) (Vh/rn2,onth)

L o r n =

16 OC 20 OC Diffuse Global

Jan

Feb

March

April

hky

June

July

Aug

Sept

Oc t

Nov

De c



\VEA?HER TAPE SPECIFICATION

?he weather tape contains four files:

1 Weather da ta f o r the Vetlanda house. November 1 , 1976

- Apri l 30, 1977, measured on s i t e (181 blocks, a

t o t a l of 4.344 recor ds).

2 Weather data for the Teknikern building. January 1,

1971 - December 31, 1971 from Stockholm (365 blocks,

a t o t a l of 8.760 record s).

3 I de nt ic al t o f i l e 1 .

4 Ide n ti c al t o f i l e 2.

The tape contains some calcu lat ed values in add itio n t o th e

measured values. The tape should not be used for purposes other

than the work i n t h i s annex.

Record f o n a t

9-Track, Odd Par i ty , 800 bp i, no labe l .

Data i s encoded a s card images, one byte pe r EBCDIC ch ar ac ter ,

one 72 char ac te r card pe r hourly record.

The records a re blocked in to days wit h 24 records per block.

A standard tape mark i s recorded a t t he end of each f i l e a n d

two at the end of the fourth file.



Card format

Field

12

3

4

5

6

7

8

9

10

1 112

13

14

15

16

17

18

19

Columns

1-4

5-8

9-1

12-14

15-16

17

18

19-20

21-25

26-30

31-35

36-40

41-45

46

47-62

63-66

67-68

69-70

71-72

I tem

Dry bulb temp.

Dew point temp.

Wind direction

Wind speed

Weather (w)

Weather l a s t hour (W)

Cloudness (N)

Sunshine duration

Global radiation

on horizontal

Diffuse radiationon horizontal

Direct normal radiation

Solar a l t i tude

Solar azimuth

Snow

Blank

Year

Month

Day of month

Hour of day

Unit

0.1 O c

0.1 O c

Degree from north

0.1 m/s

Eights

0.1 h/h

0.1 Wh/h,m2

0.1 1Wh m2

0.001 radians

0.001 radians fromnorth

Common remarks

A l li t e m are writ ten as integers, with negative values signed.

Notes on th e usage of u n i t s

0.1 OC means t ha t the l a s t di g i t in the four columns fi e l d

i s tenth of degrees.

Except fo r the f i e l d s without ava ila ble data , see below, values

are given for each hour and no action for missing values ..should.be necessary. "From north" i s counted over ea s t - i . e . e a s t i s

90' etc .



Remarks on weather dat a f o r Vetlanda

Fields 2 , 3, 5, 6 , 7 and 8 ar e not av ai la bl e and blanked out .

Fields 1 and 4 are instantaneous values.

F ie ld 14 i s se t to 1 i f the ground i s covered by snow, e l s e i t

is s e t t o 0.

Latitude i s 57.42 degrees north.

Longitude i s 15.08 degrees east.

Reamrks on weather dat a f o r Stockholm

Field 14 i s not available and blanked out.

Fields 1 , 2, 5 and 7 are observations close before indicatedtime .Field s 3 and 4 a r e mean values from 10 minutes preceding i nd i-

cated hour.

Latitu de i s 59.33 degrees north.

Longitude is 18.03 degrees east.

Remarks on solar data

Fields 8 , 9 and 10 a r e values in teg rat ed during 60 minutes pre-

ceding the indicated hour.

Field 11 i s calculated fromv alues in f i el ds 9, 10 and 1 2 .

Fields 1 2 and 13 ar e calculated fo r the indicated hour. In case

of negative a l t i t ude s both f i e l ds are s e t t o zero.

Remarks on hour of day

The hour of day goes from 1 t o 24 and re fe rs t o mean European

time, equal t o Swedish Standard Time.

Time neri dia n i s 15.0 degrees ea s t .

Remarks on wind direction

The wind direction i s defined a s t he dir ecti on from which the

wind blows. North di re ct io n i s ind icat ed as 360, d ir ec ti on 0

i s used for calm weather.

Annex 03 Calculation Methods

Documents

Transcript of Annex 03 Calculation Methods