1 THERMAL RENOVATION OF BUILDINGS WITH WORTH-PRESERVING FACADES

In the cases described in this paper insulations onhistorical facades without vapour retarder havebeen investigated. Capillary active insulationmaterials can transport occuring condensate andbuilt-in moisture away from the potentialcondensation layer to the warm side of theconstruction, thus keeping the moisture content ofthe original construction within safe limits.

1.1. "Gründerzeit"-house Dresden

The facade of the “Gründerzeit” test building(fig. 1) is typical for a large number of buildings.The material used for the inside insulation iscapillary active calcium silicate. The completerenovation and the installation of the measurementequipment were carried out in 1996.

The hygrothermal processes in the investigatedoutside walls of the ground floor, the first floor andthe ventilated yardside facade were numericallysimulated using the code DELPHIN with themeasured climatic boundary conditions(Grunewald, 1999).

The comparison between measured data and thecalculated temperature and moisture fields isshown for the wall in the ground floor (comp.Häupl et. al., 1999). Fig. 2 shows the application ofthe sensors, Fig. 3 and 4 contain the temperaturesand relative humidities in the layer between old in-

Fig. 1. Street facade of the "Gründerzeit"-house in Dresden

side plaster and the coupling mortar of theinsulation board. The correspondance between themeasured and the calculated values can beidentified as very good over the investigated timeframe of five years. Solely at the turn of the year1996/97 and in December 2000 some condensateoccurs. From the temperatures and the heat flux a

Inside thermal insulation for historical facades

P. Häupl, K. Jurk, H. PetzoldDresden University of Technology, Institute for Building Climatology

ABSTRACT: The development of thermal insulation systems with effective physical moisture protection isan important part of research at the Institute of Building Climatology at the Dresden University ofTechnology. The use of extensive measurement technique enables the monitoring of the hygrothermal effectsof energetic retrofitting measures. Additionally the validation of the physical models and simulation codesdeveloped at the institute for the coupled heat and moisture balance of building constructions becomespossible.

Selected results for buildings in Dresden, Senftenberg, Nuremberg, Ebersbach (East Saxony), andEdemissen (nar Hanover) are presented in this contribution.

U-value of 0.64 W/m²K results. The heat loss ofthis facade was reduced by approximately onethird.

630

150 435 15 30

Q_eg

K1_eg

T1_egphi1_eg

Topt100_eg

T1pt100_eg

T2pt100_eg

K2_eg

innenaußen

Fig. 2. Wall construction and sensor location in the groundfloor of the test building in Dresden

0

5

10

15

20

25

30

1.12.96 1.12.97 1.12.98 1.12.99 30.11.00 30.11.01

Tem

per

atu

re in

°C

measured

calculated

Fig. 3 Measured and calculated temperature between oldinside plaster and coupling mortar, time from Dec. 5, 1996 toJune 11, 2001

40

50

60

70

80

90

100

1.12.96 1.12.97 1.12.98 1.12.99 30.11.00 30.11.01

Rel

. hu

mid

ity in

%

measured

calculated

Fig. 4: Measured and calculated relative humidity betweenold inside plaster and coupling mortar, time from Dec. 5,1996 to June 11, 2001

1.2. Renovation of the buildings "Bahnmeister-gasse" in Senftenberg

The thermal renovation of the outside walls at thetesthouses "Bahnmeistergasse" in the city ofSenftenberg (fig. 5) was performed by theinstallation of seven different inside insulationsystems. Capillary active Calcium silicate, mineralfoam, cellulose fibre insulations as well aspolystyrene, gypsum and composite boards wereused. The measurement and numericalinvestigation has been carried out since summer2000 (Häupl et.al. 2001).

Fig. 5. Test building in Senftenberg

1.2

2

2

2

2.4

2.4

2.4

2.8

2.8

2.8

4.8

4.85.2

5.6

10

12

0 100 200 300 400 500 600 700 800

outside thickness (mm) inside

0

200

400

600

800

1000

1200

1400

firs

t fl

oo

r

hei

gh

t (m

m)

sec

on

d f

loo

r

moisture content (V-%)8.8+8 to 8.87.2 to 86.4 to 7.25.6 to 6.44.8 to 5.6

4 to 4.83.2 to 42.4 to 3.21.6 to 2.40.8 to 1.60 to 0.8

Fig. 6. Senftenberg: Moisture field around the wooden beamhead with calcium silicate inside insulation and heating pipe

heating pipe

measuredcalculated

measuredcalculated

Meanwhile the material properties of all usedbuilding materials are being measured by standard-ized laboratory experiments (eg. suction curve,sorption curve, water vapour diffusion, comp.Plagge et. al., 1999).

For a successful, damage-free renovation theconsideration of thermal bridges and connectionpoints is important. In this building the influenceof the room climate and heating pipes on themoisture balance of four timber heads ismonitored. One calculated moisture field onFebruary 19, 1999 can be seen in fig. 6. In fig. 7the measured relative humidity in the air space atthe front of the wooden beam are confronted withthe simulation results.

Fig. 8 contains the same comparison for thewood moisture content at the measuring pointHFBK1.2, which is located approx. 12 cm from theouter beam head. The shown results ofmeasurement and simulation correlate quite welland show safe moisture values for the time frameso far measured.

3 3.25 3.5 3.75 4 4.25

time (a)

30

40

50

60

70

80

90

rela

tive

hu

mid

ity

(%)

measured valuescalculated values

Fig. 7. Measured and calculated relative humidity in the airspace at the beam head with heating pipe, Jan. 19,1998 toMay 5, 1999

3 3.25 3.5 3.75 4 4.25

time (a)

3

3.5

4

4.5

5

5.5

6

6.5

7

mo

istu

re c

on

ten

t (V

-%)

measured valuescalculated values

Fig. 8. Measured and calculated wood moisture content 12cm from the beam head with heating pipe, Jan. 19,1998 toMay 5, 1999

1.3. Herrenschiesshaus Nuremberg

The so-called Herrenschiesshaus Nuremberg(fig. 9) contains a public educational centre withconference and office rooms. Together with theNuremberg Building Department in the summer2001 two rooms, one office and one conferenceroom, with massive sandstone outside wall wererenovated as an example project.

Fig. 9. Herrenschiesshaus Nuremberg

Additionally to making visible the originalwooden beam ceiling and the installation of a newfloor, an inside insulation of calciumsilicate wasattached to the outside wall. The thickness of thein-sulation is 50 mm, in reveals and window niches20 mm. Parallel a measurement system wasinstalled in each room: in the conference room atthe wall between two windows, in the office roomin the corner area. Surface and layer boundarytemperatures, relative humidity, heat flux as wellas outside and inside climate in both rooms arebeing measured in 30 min intervalls.

With the measured boundary conditions the heatand moisture balance of the walls is numericallyreproduced. The comparison of the measurementsand the simulation is shown for the relativehumidity at the cold side of the insulation infig. 10. Due to the infrequent use of the officeroom under low moisture loads the relativehumidity of the inside air is as low as 25...35%.Therefore behind the insulation no condensateoccurs, though the temperature decreases down to4°C. Thus, the capillary properties of the insulationmaterial are not exploited.

To demonstrate the usability limits of the wallconstruction, sample calculations with a constantinside relative humidity of 60% and 70% areshown in fig. 10 and 11. The actual climate

measuredcalculated

measuredcalculated

1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7Time (years)

20

30

40

50

60

70

80

90

100

Rel

. hu

mid

ity

(%)

Fig. 10. Herrenschiesshaus Nuremberg: Measurement data and simulation of the relative humidity between insulation and originalwall, starting from the application of the insulation boards

1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7Time (years)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Ove

rhyg

rosc

op

ic w

ater

mas

s (k

g/m

2)

Fig. 11. Herrenschiesshaus Nuremberg: Overhygroscopic water mass in the construction for several indoor air humidities

produces no condensation water, from a constant60% inside relative humidity during the cold period0.36 kg/m² would result.

Only with a constant RHi = 70% for the wholewinter the condensation period starts in Novemberand lasts until May, so that the allowed maximumcondensation mass of 1 kg/m² is reached. Thisshows the significant reserve of the capillarymaterial properties. For the renovation of thisbuilding a prize of the “Bund für Heimat und

Umwelt” (Association for Country andEnvironment) was awarded.

1.4. Framework house in East Saxony

At a 200 years old framework house in East Saxonybuilding physical measurements were performedfrom 1995 to 1999. Eight different inside insulationconstructions were tested at the massive woodenwall in the ground floor. In the framework wall ofthe first floor a new filling of light clay mortar was

Simulation with RH inside = 70 %

Simulation with measured climate

Measurement data

Simulation with RH inside = 70 %

Simulation with RH inside = 60 % Simulation withmeasured climate:

mw = 0

installed in three framework fields (fig. 12) andprovided with inside insulations of mineral wool andcalcium silicate boards (Fechner et. al., 1998).

Fig. 12. Installation of the new framework filling in the testbuilding in East Saxony.

In the remaining area a calcium silicate insulationwas investigated with the original straw loam filling.

From the numerous results, the effect of thecapillary activity of an inside insulation shall bepresented with the last alternative.

During the measurement time the building wasnot inhabited. Thus, the inside climate could beregulated to critical values (20°C, 60...65% duringthe heating period). With the measuring equipmentthe moisture content in the framework beams couldbe countinuously measured and compared to thesimulation results. Both measurement andcalculation (fig. 13) show maximum values of up to10 Vol%, this is approximately equivalent to the

critical wood moisture content of 18 Mass%, abovewhich a risk for the wood can be expected.

3.5 3.75 4 4.25 4.5 4.75time (years)

3

5

7

9

11

13

15

mo

istu

re c

on

ten

t (V

-%)

measuredcalculated values

Fig. 13. Measured and calculated moisture content in 13mmdepth in the wooden beam (calcium silicate insulation withstraw loam filling)

For a quantification of the capillary effect of theinsulation material a simulation of the moistureperformance shall be presented with and without thecapillary properties using the in-situ measuredclimate data. In fig. 14 (real capillary activity of allmaterials) the built-in moisture of the boardinstallation can dry out relatively quickly, and thecondensating moisture during the next winter periodis transported to the warm side of the construction. Ifthe capillary properties are not regarded – accordingto the current moisture protection codes – the built-in moisture cannot dry out before the condensationperiod starts (the installation took place in lateAugust), and a moisture content of more than45 Vol% occurs in the insulation material, leavingthe loam plaster wet (moisture field in fig. 15).

The direct comparison of the overhygroscopicmoisture content for these two cases in fig. 16underlines these statements. The incomplete dryingof the initial moisture leads to a water mass of6 kg/m² in the first winter. During the second winter,when water penetrates only in vapour form from theinside, still significantly over 2 kg/m² occur. Takinginto account the real material properties, the dryingtakes place quickly, and the construction containsonly a maximum of 0.6 kg/m² in the second heatingperiod.

straw loam

loam plaster(2 layers)

calcium silicate

Fig. 14: Water content profile in the center of the straw loam field (one-dimensional simulation) with measured capillaryconductivity for all materials

straw loam

calcium silicate

loam plaster

Fig. 15: Water content profile in the center of the straw loam in time (one-dimensional simulation) without consideration ofcapillary forces of all materials

installation

first winter

second winter

installation

first winter

second winter

3.5 3.75 4 4.25 4.5 4.75 5 5.25 5.5 5.75

Time (years)

0

1

2

3

4

5

6

7O

verh

ygro

sco

pic

wat

er c

on

ten

t (k

g/m

2)

with capillaritywithout capillarity

Fig. 16. Test building in East Saxony: comparison of theintegral overhygroscopic water mass in the construction withand without consideration of capillarity

Since 2000 measurements in an inhabitedframework house with inside insulation inEdemissen near Hanover (fig. 17) are conducted, toa certain extent as a continuation and application ofthe works in East Saxony. In this building sensors inthe outside wall of two bathrooms (north and eastside) and in one living room were installed.Additional to temperatures, heat fluxes, and relativehumidities within the construction, moisturecontents are measured in a number of woodenbeams. Hitherto, the comparisons of themeasurements with the simulations show goodcorrespondance.

Fig. 17. Test building framework house near Hanover, viewfrom north-east

CONCLUSIONS

The presented building physical measurementsdemonstrate the function of several inside insulationsystems at a number of different outside wallconstructions. The application of capillary activeinside insulation materials proves advantageous forthe drying process of potential built-in moisture aswell as for the limitation of the condensation amountduring winter. The U-value of the building wallscould approximately be halved in the presentedcases without the necessity of vapour barriers.

All measurements were reproduced bysimulations with good correspondence using the in-situ measured hourly values of inside and outsideclimate.

REFERENCES

Grunewald, J. 1999. Mass and Energy Transport in PorousMedia – Thermodynamical Identification of Driving Potentials.Int. Zeitschrift für Bauinstandsetzen und Denkmalpflege, 5.Jahrgang, Aedificatio Publishers, Heft 2, pp. 113-142,.Häupl, P.; Grunewald, J.; Fechner, H. 1999. Moisturebehaviour of a "Gründerzeit”-house by means of a capillaryactive inside insulation. Proceedings of the "Building Physicsin the Nordic Countries 1999, pp. 225-232, GöteborgHäupl, P.,Stopp, H., Petzold, H. 2002. Thermische Sanierungvon Fachwerkwänden mit kapillaraktiver Innendämmung. In:Fachwerksanierung nach WTA, Band. 2, Fraunhofer IRBVerlagFechner, H., Häupl, P.; Martin, R., Neue, J., Petzold, H. 1998.Thermische Sanierung von Fachwerkbauten mittelsInnendämmung - ein Vergleich zwischen Messungen undnumerischer Simulation, Bauklimatische Hefte, Heft 5, TUDresdenPlagge, R.; Grunewald, J.; Häupl, P. 1999. Simultaneousdetermination of water retention characteristic and moistureconductivity using instantaneous profile techniques, Procee-dings of the "10. Bauklimatisches Symposium", pp. 433-444,TU Dresden