Monument Future

Conclusions

The presented survey methodology allows an evaluation of the masonry quality of an existing building by means of surveys and punctual inspections 286(1st level of analysis) and by means of non-destructive diagnostic investigations (2nd level of analysis). This procedure helps to recognize and correctly define the type of masonry for a correct structural evaluation but also, afterwards, to choose the most appropriate and compatible repair technique.

When dealing with historic masonry buidlings, the characterization of the single components of a stone masonry, without analyzing also the structure of the masonry as a whole, does not provide sufficient and representative data for its mechanical characterization. The structural efficiency of the masonry itself depends directly on the quality of construction, and in particular on those “rules of art” developed locally in an empirical way in order to optimize the quality of the masonry on the basis of available resources in terms of raw materials, labour, costs and processing times, etc., to ensure a monolithic behaviour to the masonry elements, an essential characteristic for a good response to horizontal actions or in case of eccentricity of loads. Future research will concern the inclusion of more specific types of masonry such as the herringbone of Romanesque architecture and the dry-stone masonry.

References

Barbaro D. 1567. I dieci libri dell’architettura di M. Vitruvio.

Binda L., Saisi A., Tiraboschi, C. 2000. Investigation procedures for the diagnosis of historic masonries, Constr. Build. Mater. 14(4), 199–233.

Binda L., Cardani G., Penazzi D., Saisi A. 2003, Performance of some repair and strengthening techniques applied to historical stone masonries is seismic areas, ICPCM a New Era of Building, Cairo, Egypt, Vol. 2, 1195–1204.

Binda L., Cardani G., Saisi A. 2005. Application of a multidisciplinary investigation to study the vulnerability of Castelluccio (Umbria). In: 9th Int. Conf. on Structural Studies, Repairs and Maintenance of Heritage Architecture. Malta, 22–24/06/2005, WIT press, 311–322.

Binda L., Cardani G., Saisi A., Valluzzi M. R., Munari M. & Modena C. 2007. Multilevel approach to the vulnerability analysis of historic buildings in seismic areas, Part 1. Restoration of Buildings and Monuments, 13(6), 413–426.

Binda L., Saisi A, 2010. Typological classification and investigation methodologies of historic stone masonry construction. In: Iscarsah Symp. Mostar 09, Assessment and Strengthening of historical stone masonry structures subjected to seismic action, 12/07/2009, Mostar, 21–42.

Binda L., Cardani L. 2011. Lo stato di conservazione e la qualità muraria. In: L. Milano et al. (eds.). L’università e la ricerca per l’Abruzzo. Il patrimonio culturale dopo il terremoto del 6 aprile 2009, L’Aquila: Textus s. r. l., 423–426.

Binda L., Cardani G. 2015. Seismic Vulnerability of Historic Centres. Handbook of Research on Seismic Assessment and Rehabilitation of Historic Structures, edited by Asteris, P. G. and Plevris, V, 01/2015, IGI Global, chapter 1, 1–29.

Borri A., Corradi M., Castori G. and De Maria A. 2015. A method for the analysis and classification of historic masonry, Bulletin of Earthquake Engineering 13 (9).

Cardani G., Binda L. 2015. Guidelines for the evaluation of the load-bearing masonry quality in built Heritage. In: Built Heritage: Monitoring Conservation Management, Edited by Toniolo L. and Boriani M. and Guidi G., 01/2015, Springer International Publishing, 127–139.

Eurocode 6. 2005. Design of masonry structures – Part 1-1: General rules for buildings. ENV 1996-1-1.

Giuffrè A. 1993. Sicurezza e conservazione dei centri storici: il caso di Ortigia, Bari, Italy, Laterza Ed.

IMIT. 2018. D. M. 17/01/2018, Technical Standards for Construction and their applicative documents: Circular n. 7 21/01/2019, Italian Minstry of Infrastructure and Transportation, Rome, Italy.

RELUIS-DPC Project. 2009. The state of earthquake engineering research in Italy, Manfredi G. and Dolce M. editors, Napoli, Italy.

Rondelet J. B. 1802. Traité theorique et pratique de l’art de bâtir, Paris, France.

287

KARSTEN TUBE PENETRATION TEST – NEW FINDINGS AND NEW EVALUATION METHOD

Peter Kozub

IN: SIEGESMUND, S. & MIDDENDORF, B. (EDS.): MONUMENT FUTURE: DECAY AND CONSERVATION OF STONE.

– PROCEEDINGS OF THE 14TH INTERNATIONAL CONGRESS ON THE DETERIORATION AND CONSERVATION OF STONE –

VOLUME I AND VOLUME II. MITTELDEUTSCHER VERLAG 2020.

Technische Hochschule Koeln, University of Applied Sciences, Cologne Institute for Conservation Sciences, Ubierring 40, 50678 Köln, Germany

Abstract

Water penetration tester according to Karsten is commonly used in conservation as a simple rapid test method with regard to water penetration into porous materials such as concrete, stone or plaster.

The decisive factor for the assessment is the water absorption coefficient (w-value). It describes how much water a defined area of a material is absorbed by capillary or absorptive forces in a certain period of time.

The water intake by means of the Karsten test tube represents a three-dimensional suction process. Until recently, it was assumed that the fully moistened area ideally has the shape of a central cylinder, which is surrounded by a quarter torus. The cylinder cross-section was accordingly corrected with the evaluation programs according to Wendler or Niemeyer. However, the mathematical expression of the wetted volume in 3D simulations of moisture transport for geometries with axial symmetry and the direct observations on the rock sample plates revealed a completely different water penetration behavior. The geometric shape of the moistening accordingly corresponds to the half-ellipsoid. This finding led to the fact that the mathematical calculation models of w-value from this measurement would have to be revised.

The new evaluation program “KARSTEN 2.3 ellipsoid” considers these new findings.

Supplemented by a curve fitting function and theoretical visualization of the course of the water absorption coefficient at depth, this new possibility allows an optimized evaluation of this measurement even for materials whose water absorption capacity is unknown.

Keywords: Karsten tube, penetration test, water absorption coefficient (w-value), new evaluation method

Introduction

Knowledge of the parameters of moisture transport is of great importance in stone conservation. In most cases, these can be determined with great accuracy under laboratory conditions. However, their determination requires the taking of samples, which is often unacceptable, at least for small and valuable objects. The determination of these properties directly on the object is more difficult. One of the most important methods to determine the water absorption coefficient is the measurement with so-called Karsten’s test tubes.

The water penetration test according to Karsten has been frequently used since its development in 1960 (Karsten 1960) as a simple, non-destructive rapid test method for the penetration of water into porous materials such as concrete, stone or plaster.

288The decisive parameter for the evaluation is the water absorption coefficient (w-value) (in earlier publications also called A-value e. g. Künzel 1971):

It describes how much water a defined area of a material absorbs in a certain period of time by capillary or absorptive forces (Schwarz/Künzel/Gösele 1971).

Significance of w-value

The importance of water absorption coefficient in stone conservation is clearly evident from many studies. Information obtained by measurement about the behaviour of the material towards water at different stages of weathering and at different stages of conservation work helps both to determine the weathering processes and to check the conservation measures carried out. This non-destructive measuring method can be applied directly on the object without any problems. Since no changes are to be expected by the application, the measurement of the same spot can be carried out several times. Simple handling and quick application allow a larger number of measurements to be carried out in a dense grid (Snethlage/Pfanner 2013). This allows both the scanning of large areas and increases and specifies the accuracy of the values at the same time. The resulting w-value is taken into account in many calculations. The water absorption coefficient is often used, for example, in building materials technology to classify the absorbency of materials in order to better assess the damage caused by water, for example, moisture penetration, frost damage or mould and algae infestation.

The following classification of the water absorption coefficient is common:

absorbing: w > 2 kg/m² × h^0,5 water-resistant: w ≤ 2 kg/m² × h^0,5 water-repellent: w ≤ 0.5 kg/m² × h^0,5 water-impermeable: w ≤ 0.001 kg/m² × h^0,5

The w-value is also used to determine the required minimum penetration depth for the strengthening agent by determining the depth position of the maximum of the mean moisture distribution, which in a first approximation depends on this value (Snethlage/Pfanner 2013):

w = 0.1 … 0.5 kg/m² × h^0,5: s = 1.0 cm w = 0.5 … 3.0 kg/m² × h^0,5: s = 3.0 cm w > 3.0 kg/m² × h^0.5: s = 6.0 cm

Also in the description of the drying behaviour of the materials by means of the so-called “Künzel number”, a product of w-value and s^d-value, it is also necessary to determine the water absorption coefficients of the materials involved (Künzel 1969, 1976):

sd × w ≤ 0.1 kg/m h^0,5 w ≤ 0.5 kg/m² h^0,5 sd ≤ 2 m

The requirements for hydrophobicity also explicitly refer to the minimum penetration depth of hydrophobing agents, which can be estimated with the help of the previously determined w-value. In addition, reference is made to the application for testing the hydrophobic effect. Here, the w-value after hydrophobing should be less than 0.1 kg/m h^0,5 (Snethlage/Pfanner 2013).

A summary description of the individual restoration steps, documented with limit values and correlations with other properties, requirements and quality testing proposals can be found in (Snethlage/Wendler 1995; Snethlage/Pfanner 2013).

Measurement, evaluation and calculation

The inventor of the water tester Rudolf Karsten suggests a simple application and evaluation of the measurement as “water intrusion”. “At regular intervals, the amount of water infiltrated is read off.” (Karsten 1960). In later publications, he refers to this quantity as “water ingress” and describes the measurement in more detail. The values are expressed in “ml of water per minute [or in] ml 289of water per minute and cm² […] by dividing the mean measured values by the size of the test area (usually 3 cm²)” (Karsten 1997). In this way, no water absorption coefficient is calculated.

The procedure for the measurement is not exactly defined. However, the type of application suggested by Karsten is adopted by numerous authors and suppliers of this test device in different variants. Although there are some standards which should regulate the measurement, the “confusion of standards in the determination of moisture-related material parameters” was already complained about by Fitz and Krus in their article (Fitz/Krus 2004). As already mentioned, there are numerous different standards which describe the determination of water absorption coefficients on sample materials (e. g. DIN EN 1015-18; DIN EN 1925; DIN EN 772-11; DIN EN ISO 15148; DIN EN ISO 15801). Equation (1) is used in most of them.

In a few standards describing the measurement with water penetrator according to Karsten (RILEM (2006); DIN EN 16302), a similar measuring principle is proposed as by the inventor himself. None of these standards suggests a calculation of w-value from this measurement. Nevertheless, some authors calculate the water absorption coefficient from this measurement using equation (1).

Obligation of conversion

In many investigations one is forced to compare the values determined with Karsten with the w-values from the literature, which have been determined according to DIN on sample materials under laboratory conditions. Many w-values found in the literature and which are used for comparison, whether for financial or conservation reasons or for material technical reasons, have been determined in this way. The classifications and calculations already mentioned are also based on the w-values determined according to DIN.

However, direct comparison of these values with the values from the Karsten measurement is not possible without further ado. With the values determined according to DIN, the sample geometry (drill cores) and the procedure (wetting of the standing surface) are clearly defined. The water thus spreads out in two dimensions in the sample (water front as a surface). Thus, the above-mentioned equation (1) applies here. The water absorption by means of the Karsten test tube represents a three-dimensional suction process, as the lateral areas of the measuring point also absorb water (Wendler/Snethlage 1989).

Therefore it is necessary to calculate the w-value differently.

There are some ideas to avoid this problem. E. g. in which the edge area of the moisture penetration is taken out of the calculation of the geometry by a double switched tube (Pleyers 1999).

Previous conversion models

There are though some attempts to solve the problem purely mathematically. Until recently, it was assumed that in the ideal case the moisture soaking body would have the shape of a central cylinder surrounded by a quarter torus. Wendler and Snethlage first described the geometry of the moisture penetration area (Wendler/Snethlage 1989) (Fig. 1).

Figure 1: Schematic illustration of the rock volume penetrated by the Karsten tube after Wendler / Snethlage 1989 (slightly modified).

To date, two programs are available for the evaluation of the measured data – the Calcarow program from Wendler und Pfefferkorn and the evaluation algorithm from Niemeyer. Both are based on the Microsoft Excel spreadsheet. The evaluation of the measured data with both programs is described in detail in (D’ham/Meinhardt/Niemeyer 2011). The calculation process is also very similar for both programs. Since other parameters of moisture transport are also mathematically linked with the w-value, including the water penetration coefficient 290(B-value) and the water absorption capacity (WAK), the conversion can be performed for a different moisture dispersion geometry. Both programs are based on the assumption that the moisture penetration area in the Karsten measurement has the shape of a central cylinder surrounded by a quarter torus.

New findings and new approaches

However, the mathematical representations of the moisture penetration area in 3D simulations (Hendrickx 2013) and the direct observations on the stone sample slabs revealed a completely different water penetration behaviour. The geometrical shape of the moisture penetration corresponds to the half-ellipsoid (Fig. 2).

This finding led to the fact that the mathematical calculation models of w-value from this measurement would also have to be revised.

The method of calculation has been retained. First, the volume of the moisture penetration area is calculated in the form of half-(rotational) ellipsoid (V_HE):

This results from the conversion to variable x:

and the volume of the centric region of the half-(rotational) ellipsoid (V_B):

The new evaluation program KARSTEN 2.3 ellipsoid takes the new findings into account. The calculation program is also based on the Microsoft Excel spreadsheet to facilitate access and use of the software.

In the last step, similar to the previous programs for this area, the equation (1) is used for the calculation of w-value (Fig. 3).

Apart from the adaptation of the geometry of the moisture penetration area, the so-called factor-y is also introduced. Depending on the pore structure of the examined material or pore sizes in different areas of the material, the capillary water transport in different directions can take place at different speeds. The variable y describes the lateral expansion of the half-ellipsoid, i. e. the length of the half-axis perpendicular to the axis of rotation (a) compared to the half-axis parallel to the axis of rotation (c), while the volume of 291the moisture penetration area (V_B) remains constant.

In contrast to the previous programmes, the water absorption capacity (WAK) does not need to be estimated. In most cases, this value is difficult to determine or unknown. The evaluation algorithm of the program assumes that the measuring curve represents the course of the water absorption with sufficient accuracy. From the measured values, the two decisive variables – WAK and variable y – are adjusted step by step. The adaptation is carried out by means of coefficient of determination R².

Figure 2: The geometrical shape of flattened half-rotationellipsoid as result from the rotation of a half-ellipse about its z-axis.

Figure 3: Schematic comparison of the models of water absorption according to Wendler/Snethlage 1989 and as ellipsoid. The differences in the area of the cylinder are marked red and at the sides violet.

The optimized values, which represent the closest possible agreement with the measured curve, are then automatically generated and displayed as a proposal. This objectifies the conversion process (the same results are generated for the same measured values independently of the user) and enables evaluation even for completely unknown materials. However, both variables are still freely selectable.

In addition, a different representation of the values is selected for the program. Most measured materials show an inhomogeneous course of the values in different depths or ranges. This is also partly represented by the course of the measurement curve and is a valuable information for the evaluation of the measurement results. Accordingly, apart from the arithmetic mean of all sections of the w-value and the B-value, the program also displays their course at the calculated depth in a histogram (Fig. 4).

Figure 4: The program displays the course of the curves for w-value and B-value at the calculated depth as a histogram. Below is an example from the measurement of the untreated Leistadt sandstone. The light red line marks the arithmetic mean value.

Additionally, the program offers the possibility to print the data (raw data, notes, calculated values and the associated diagrams) in the form of a protocol or to save them separately as a pdf file.

Conclusion

The measurement with so-called Karsten’s test tubes to determine the water absorption coefficient has been established in the restoration community for a long time. This simple and non-destructive measuring method provides reliable and important information about the capillary properties of the materials when used correctly. However, 292until recently, many factors that significantly influence this measurement were still not clarified. The calculation of the water absorption coefficient from this measurement was based on an incorrect geometric model of the moisture penetration range. Also the different lateral spreading of the water in the material structures was not considered. Previous presentation of the value only as an arithmetic mean value of all sections led to a lack of understanding of this quantity and to application problems with inhomogeneous and especially weathered materials.

The new free open source evaluation program KARSTEN 2.3 ellipsoid, which is offered here, takes the new findings into account, facilitates the application and optimizes the evaluation of this measurement even for materials whose water absorption capacity is unknown.