Print full article

Testing the Test: Water absorption with RILEM tubes

All images courtesy Building Diagnostics Inc.
All images courtesy Building Diagnostics Inc.

by Adrian Gerard Saldanha and Doris E. Eichburg
Water repellents are often applied to exterior walls to reduce absorption. A common question from building owners is, “When does the repellent need to be re-applied?” Since water repellents are clear, visual inspections are not useful to evaluate their durability. A simple field test frequently used to measure the effectiveness of water repellents is the RILEM tube. But is this widely used test reliable?1

The RILEM tube test was adopted in the United States in the early 1980s by the water repellent manufacturing industry to assess water absorption properties of walls and other substrates, with or without treatment. Simply put, the test measures the quantity of water absorbed by a particular substrate over a given time through an uptake tube. The uptake tube is sealed to the substrate with putty (Figure 1).

RILEM was an acronym for the original French name of a European organization founded in 1947 to promote scientific co-operation in the area of construction materials and structures. A RILEM technical committee developed “Test No. II.4 Water Absorption Under Low Pressure (Pipe Method)”—now commonly known as the RILEM tube test—and explained its uses in a 1980 report.2

RILEM uptake tube attached to substrate with putty, graduated from 0 to 5 ml.
RILEM uptake tube attached to substrate with putty, graduated from 0 to 5 ml.

The hydrostatic head developed by the column of water in the tube can be correlated with wind-driven rain of a certain speed. This relationship is not linear; pressure is proportional to the square of velocity, so the wind speed per 1 ml (0.04 oz) of water at the bottom of the tube is about twice that at the top of the tube.

The tubes described by RILEM in its 1980 report were shorter (98 mm [3.9 in.] from the center of the round cylinder to the top reading mark, labeled ‘0’), and were graduated from 0 to 4 cc (about 0.14 oz). The tubes readily available today are longer (about 120 mm [4.72 in.]) and graduated from 0 to 5 ml (about 0.17 oz). The 0-ml line (top of tube) correlates to 158 km/h (98 mph) and the 5-ml mark (near the bottom of the tube) correlates to 72 km/h (45 mph). Tubes have also been produced with graduations in wind speed.

Testing the test
Although widely used, there is currently no industry standard on how to perform a RILEM test. ASTM International has reported plans to develop an uptake tube test method for water repellents, but that standard is years away.3

The authors found manufacturers use different time intervals for their readings, different amounts and types of putty to attach the tubes, and different initial water fill levels. Therefore, RILEM tests were performed in a laboratory setting to assess whether those differences matter—essentially, ‘testing the test.’ The parameters included:

  • varying the initial water level;
  • constant pressure (refilling the tube during testing) versus diminishing pressure;
  • tube design—horizontal, vertical, short, and long;
  • altering the water/specimen contact surface area; and
  • different putties.

The first problem was to find a suitable substrate for repeated testing; the volume of water absorbed in a fixed time is a material property. Concrete masonry units (CMUs) were found to be too porous before treatment (and, interestingly, even after treatment with one repellent intended for CMUs). Clay paver tiles were found to be ideal specimens, absorbing water slowly but distinctly, with measurable differences before and after treatment.

Short RILEM tube, graduated in wind speed.
Short RILEM tube, graduated in wind speed.

Untreated clay tiles took an average of eight minutes to absorb 5 ml of water. To study the absorption of water over time, and to evaluate whether conclusions depended on the duration of the tests, readings were taken every minute for graphical presentation. Distilled water was used for all tests. Various repellent types were tested. (For more on this topic, see “Water Repellants”).

Varying the water level in the tubes
The original test recommendations described by RILEM permit flexibility in deciding the water level in the uptake tube based on likely rain exposure conditions and durability requirements of the substrate. This is reflected in manufacturers’ literature that recommends testing with only 3 ml (about 0.1 oz) of water in the tube for certain repellents on extremely porous substrates.

Short tubes are available for this purpose, and have essentially the same geometry as the regular tubes (Figure 2); during testing, there was no difference between the fully filled short tubes and partially filled regular tubes. However, the authors confirmed fully filled tubes drained into the clay tiles faster than the partially filled ones because the applied water pressure is greater. This can lead to erroneous interpretations about absorption when the test time is short (Figure 3).

Constant head pressure
A more severe test method is to refill the tube at regular intervals as water is absorbed by the substrate, measuring the water added. Since the higher pressure is maintained (rather than diminishing as the water level falls), more absorption would be expected to occur in the same time. This method is not commonly used in the field, but it represents one of the myriad ways this test has been modified by different users.

Horizontal tubes
‘Horizontal’ tubes (actually mounted vertically) are convenient for testing horizontal surfaces, like concrete slabs. The opening of the uptake tube is co-axial with the vertical column of water. From the authors’ testing, no significant difference was observed in the readings obtained from the two tube orientations on the same sample (Figure 4).

As shown by this graph, fully-fi lled uptake tubes drain quicker than partially fi lled tubes because of the higher hydrostatic pressure.
As shown by this graph, fully-filled uptake tubes drain quicker than partially filled tubes because of the higher hydrostatic pressure.
The absorption rate with horizontal and vertical tubes is essentially the same.
The absorption rate with horizontal and vertical tubes is essentially the same.










Altering the contact area
Assuming water travels through the substrate at the same velocity in all directions, the shape of the water absorbed in the substrate is initially cylindrical, and gradually approaches hemispherical.4 The geometrical boundary condition of the infiltrating water is complex and variable, and can contribute to inaccurate results.5

The contact area between the tube and surface, through which water flows into the test substrate, is bordered by putty. Larger openings lead to a higher rate of absorption as compared to smaller openings that cause a lower rate of absorption;6 therefore, putty that accidentally fills a portion of the contact area will appear to reduce the absorption. These results were confirmed by testing (Figure 5). The contact area can be recorded after the test by taking an imprint of the putty using an ink pad.7

Testing procedure recommendations
Small changes in attaching the uptake tube or performing the readings may lead to different results. The authors have the following recommendations regarding the testing.

A smaller contact area leads to lower rate of absorption.
A smaller contact area leads to lower rate of absorption.

Attaching the tubes
The supplied putty was used to attach tubes to the substrate during testing, as well as a butyl adhesive. Both adhesives worked, with similar difficulty in achieving a seal. Water leaking from the putty, instead of absorbing into the substrate, can dramatically change results (Figure 6). Therefore, it is important to watch for leaks during testing, and repeat the test if water is lost.

Both adhesives left oily stains on the substrates, the putty more than the butyl. The stains are not only unsightly, but also affect the substrate’s water repellency, and influence future repeated testing at the same location.8 It is recommended to perform an initial test in an inconspicuous area whenever possible.

Testing before and after a repellent is applied should be performed at the same locations for comparison. The same amount of putty (contact area) should be used for each test and care should be taken to ensure the opening is not covered by putty.

Filling the tubes
Air bubbles can affect results, and should be avoided. The authors found having the tip of a plastic squeeze bottle in physical contact with the back of the tube while filling it to the zero mark is the easiest way to prevent air bubbles. If air bubbles are present, the test should be repeated.

A timer should be started after the water is filled to the zero mark. The absorption that occurs before then is not usually taken into account, but it may affect the accuracy of measurements under five minutes.9

RILEM recommends readings at five-minute, 10-minute, 15-minute, 30-minute, and one-hour intervals, but also mentions in-situ measurements can be limited to the first three readings. Several repellent manufacturers require final readings at 20 minutes. In the authors’ testing, the results at 20 minutes were indicative of the results at an hour, so the longer test time does not appear to be necessary.

Even a slow drip from the putty may change results, so the test should be monitored for leaks.
Even a slow drip from the putty may change results, so the test should be monitored for leaks.

A case study on durability
To evaluate a repellent’s durability, it is essential to measure its effectiveness over time. Laboratory testing of cores removed from a treated building façade would be the most definitive method, but taking cores at intervals is expensive and disruptive. Water beading on a surface is often considered as a measure of repellent effectiveness, but it does not correlate with actual absorption (except immediately after treatment)—water can still absorb into a substrate that exhibits surface beading, and vice versa.

Field testing with RILEM tubes is a practical approach to confirming durability, and can be reliable when the variables discussed earlier are fully considered, as illustrated in a case study. A multi-story commercial office building constructed in 1989 has limestone panels cladding the lower three floors. The inferior quality limestone suffered significant erosion, discoloration, and organic growth due to high moisture absorption.

A water repellent had been applied to the limestone during original construction, but it had degraded. By the time of the study in 2001, cleaning the façade had become increasingly difficult. As the limestone eroded so easily, research and testing were required to find the least aggressive method of removing the existing sealer, dirt, and organic growth.

Several mockups were performed using various chemical and abrasive cleaning methods. The chemical cleaning was burdensome and required harsh acids. Sanding the panels was the most effective and least expensive method. Any stains that remained after sanding were removed with a biological stain remover. Following the cleaning, a clear, water-based siloxane water repellent was applied; the manufacturer provided a 10-year guarantee in 2003.

RILEM tests were performed after cleaning the limestone but before applying the water repellent. The limestone absorbed the water in the tubes within 30 minutes at all locations; three out of the five test tubes absorbed the water within 15 minutes. RILEM tests were then performed at the same locations after the water repellent was applied and cured; no water was absorbed after one hour. RILEM tests continue to be performed at regular intervals at the same locations; the last tests in September 2012 still showed zero absorption (Figure 7). The repellent will be re-applied once the limestone measurably absorbs water, indicating it is losing effectiveness.

RILEM tests on limestone cladding, measuring performance of the water repellent at the same locations.
RILEM tests on limestone cladding, measuring performance of the water repellent at the same locations.

Despite the range of variables that can affect results, RILEM tubes can be used reliably to test the resistance to water absorption of a substrate over time, which is the key to determining the durability of a water repellent. Locations should be selected that can be tested at regular intervals over the repellent’s service life. Results from successive tests can then be compared to indicate whether treatments are still effective or need re-application.

It is important to perform the testing in a consistent manner to get reliable results. Small changes in the testing locations, application of the RILEM tubes, or reading intervals may lead to differing results. Although there are numerous publications about RILEM tests, the authors have found no current published industry standard. Standardization should improve the uniform interpretation of results and increase the test’s use.

1 The authors gratefully acknowledge the continuing support and leadership of David W. Fowler, PhD, PE—the faculty advisor for the research being performed at The Durability Lab, a testing center at the University of Texas at Austin. The authors also thank Fran Gale, director of the Architectural Conservation Laboratory at the University of Texas School of Architecture, for her technical guidance. (back to top)
2 See the 1980 report, “Tentative Recommendations: Recommended Tests to Measure the Deterioration of Stone and to Assess the Effectiveness of Treatment Methods,” by RILEM Technical Commission 25-PEM. (back to top)
3 This can be found in ASTM WK27132, New Practice for Effectiveness of Field-applied Water Repellent Treatments Using a Water Uptake Tube, by ASTM Committee D01.47 on Concrete, Stone, and Masonry Treatments. (back to top)
4 See the Roel Hendrickx article, “Using the Karsten Tube to Estimate Water Transport Parameters of Porous Building Materials,” published in the 2012 Materials and Structures Journal (Springer Netherlands). (back to top)
5 See “Non-Destructive Determination of the Penetration Depth of Impregnation Materials” by Gerd Pleyers and H. Rainer Sasse, as part of ASTM STP 1355-1999, The Use of and Need for Preservation Standards in Architectural Conservation. (back to top)
6 See “Measurement Techniques” by E. De Witte, et al., which appeared in Evaluation of the Performance of Surface Treatments for the Conservation of Historic Brick Masonry Research (report No. 7, published in 1996 by the European Commission). (back to top)
7 See “Consolidants Lab Tests, Evaluating Performance & Durability” by Alice Custance-Baker of the Scottish Lime Center Trust ( (back to top)
8 See “Comparison of Non-destructive Techniques for Analysis of the Water-absorbing Behavior of Stone” by Delphine Vandevoorde, et al., delivered at the 12th International Congress on the Deterioration and Conservation of Stone (2012). (back to top)
9 See “Test Methods for the Evaluation of the In-situ Performance of Water Repellent Treatments” by Rob. P. J. van Hees, et al. (Delft University of Technology, 1995). (back to top)

Adrian Gerard Saldanha is completing his master’s degree in construction engineering and project management at the University of Texas at Austin (UT). He is a graduate research assistant in The Durability Lab—a testing center at UT established by Building Diagnostics Inc.—to study the durability of building components, identifying factors causing premature failure. He can be contacted at

Doris Eichburg is a principal with Building Diagnostics Inc., specializing in the investigation of problems with existing buildings, designing remedies for those problems, and resolving disputes that arise from them. She also participates in the company’s research group, The Durability Lab. Eichburg can be reached by e-mail at

Leave a Comment

2 comments on “Testing the Test: Water absorption with RILEM tubes”

  1. Plz sed me the procedure for Water absorption coffiicient and water capillary absorption ML @ 24 hours sepratelly
    ant ISI stendrad of Wall Putty .
    Surendra Jain

Leave a Comment


Your email address will not be published. Required fields are marked *