MEASURING MOISTURE CONTENT IN WOOD WITH BUILT IN PROBES 20+ YEARS EXPERIENCE Measuring moisture content in wood E.BRANDT and M. HJORSLEV HANSEN Danish Building Research Institute, Hoersholm, Denmark Durability of Building Materials and Components 8. (1999) Edited by M.A. Lacasse and D.J. Vanier. Institute for Research in Construction, Ottawa ON, K1A 0R6, Canada, pp. 669-679. National Research Council Canada 1999 Abstract In Denmark probes made with electrodes placed in a small plywood disc or a small piece of beech wood have been used for long-term measurement of moisture content in wooden constructions, especially roofs. The probes are embedded in the wood structure where the measurements are to be made and are equipped with wires in order to make remote measurements. This makes it possible to measure moisture content in closed cavities etc, where it would be difficult or impossible to use handheld moisture meters. The method is described in NORDTEST method 420 Building materials, wood: Moisture content. The paper describes the experiences gathered during the many years of use. A large amount of constructions have been equipped with this type of sensors giving experience about the usefulness and the long-term stability of the sensors. Besides, different investigations have taken place regarding, for instance, the influence of the shape and dimensions of the probes, the materials used for electrodes and the way of connecting the wires to the electrodes. For some years it has been feasible to perform continuous measurements with a datalogger. An example of results achieved in this way is given. The precision, the advantages and drawbacks experienced during the use etc. is discussed. Comparisons are made to what can be achieved by placing electrodes directly in the wood to be measured. Keywords: Durability, humidity, moisture content, moisture sensors, test method.
1 Introduction In Denmark the same principle as the one used in the hand-held moisture meters was applied during extensive roof surveys in the early 1970's, when the so-called moisture measuring discs were developed. The moisture measuring discs have since then been used quite intensively both for research work and for practical purposes where there was a wish of monitoring the moisture content of wooden constructions. The use of the moisture measuring discs are described e.g. in (Prebensen and Brandt 1984 and PHØNIX Posten, 1984) whereas the moisture measuring dowels are described in (Munch-Petersen, 1990). Finally the method has been described in detail in (Nordtest Method NT Build 420, 1993). The experience with the method is good but in measurements carried out at the Danish Building Research Institute, (SBI), it proved difficult to obtain concurrent results with moisture probes manufactured in different ways. Consequently a series of investigations were carried out to reveal the influence of different factors on the measurements. These results are examined and commented upon in the following. Results from moisture measuring disc and moisture measuring dowel as well as for electrodes placed directly in the wood to be measured are described. 2 The advantage of built-in probes The advantage of built in probes is that they are inserted permanently which means they are able to monitor the moisture conditions over a period of time. In this way it is possible to monitor variations in moisture content e.g. due to seasonal variations or changes in use of the construction. Another asset is that it is possible to detect any abnormal changes in the behaviour of a construction, e.g. it is possible to follow the moisture content in new innovative constructions to ensure that the moisture content remains on the expected (low) level. Finally, it is an advantage that the electrodes can be built in right from the beginning so that measurements can be carried out in inaccessible places, for instance in cavities. For this purpose it is only necessary to wire the sensors to a place convenient to measurement and registration. The advantage of using an intermediary link in the form of moisture measuring disc or moisture measuring dowel is that it assures that measuring is taking place in relatively homogeneous material with well known properties and that a direct relation exists between the measured resistance and the moisture content. Practical experience also shows that more stable measurements are obtained. The disadvantage of using probes is that it is necessary to calibrate and that, consequently, the price is relatively high.
3 Description and use of probes The moisture measuring disc consists of a 12 mm plywood disc with a diameter of 50 mm. Originally the plywood was of the same type as the one in which the moisture content was to be measured. Two electrodes are inserted in the plywood. Besides, the disc is equipped with a thermocouple, which is used for temperature registration as well as for temperature compensation of the resistance. Before use the moisture measuring disc's are calibrated by being placed in turn under different, controlled humidity conditions. In this way it is possible to find a correlation between the resistance and the moisture content of the plywood. At calibration, probes differing from the rest are identified and discarded. At a later stage moisture measuring dowel's were developed. These are in principle similar to the discs only differing by being made of beech wood and in somewhat smaller dimensions. Fig. 1: Plywood moisture measuring disc 50 mm Ø and beech wood dowel 10 mm Ø 3.1 The use of moisture measuring disc and moisture measuring dowel Moisture measuring discs and moisture measuring dowel have been available for quite a few years and the method has now been made into a NORDTEST method: NT BUILD 420 "Building materials; Wood; Moisture content ". After calibration the probes are embedded in the wood or plywood under investigation. After moisture equilibrium with the surroundings has been achieved the resistance between the electrodes is measured and thereby, indirectly, it is possible to find the moisture content in the probes and the surrounding wooden material. Measuring by hand is usually done by means of a log(mohm)-meter but the electrodes may also be hooked up to an ordinary hand held moisture meter in which case they function as an extension of the legs/pins of the meter, se Fig. 2.
Fig. 2: Hand-held moisture meters and log(mohm)-meter Until a few years ago it was rather cumbersome and expensive to collect data with a datalogger as the loggers were unsuitable for the purpose. There are, however, now loggers on the market which function well with moisture measuring disc and moisture measuring dowel. Suitable loggers should have a great input resistance because of the large resistance between electrodes in wood. Besides it shall use AC or alternating DC in order to avoid polarisation of the fluid in the pores of the wood. At temperatures other than 20ΕC, correction must be made. For example a deviation of - 8ΕC (i.e. measurement at 12ΕC) means a moisture reading which is approximately one per cent too low. The temperature correction is described in NT- Build 420. 4 Design parameters Several tests have been performed as regards the influence of the mounting method of the wires to the electrodes as well as the choice of electrode material and the size and shape of the probes. The results of these tests are described below. 4.1 Mounting of wire To find the influence of the electrode material a test was performed where brass nails, copper nails, stripped wire and Ø 1.5 mm cable wire were used as electrodes. The mounting of the wires were made by mechanical contact to the electrodes and a soldered solution respectively. The test results showed that all readings followed very closely the actual moisture content of the wood which was followed also by weighing the specimens during the tests, see Fig. 3. No significant difference was found between the various electrode materials.
Moisture content (weight-%) 30 20 10 0 RH = 50% RH = 90% RH = 80% 0 50 100 150 200 250 300 Time (days) Type A Type A1 Type B Type C Type D (Soldered) Determined gravimetrically Fig. 3: Moisture content in various moisture measuring dowels subjected to varying relative humidity as a function of time As regards the effect of the mounting of the wires to the electrodes it was found that soldering of the wires instead of mechanical fastening gives less resistance between the electrodes. Apparently soldering of the wire to electrodes already fixed in the wood causes some permanent changes in the resistance between the electrodes. This is probably due to heating of the wood during the soldering. Consequently, different calibration curves shall be used depending on the fixing of the wires. Calibration of probes should therefore not take place until after the wires have been fixed to the electrodes so that the effect of the fixing process is taken into account in the calibration. 4.2 Types of electrodes For hand-held moisture meters it is well known that electrodes may either be insulated or bare (non-insulated), see Fig. 4. For the insulated electrodes only the tip of the electrode is bare and the resistance is consequently measured between the two tips of the electrodes. If the electrodes are plugged into the wood it is, therefore, the moisture content of the wood that is decisive for the result of a measurement - as it should be. Contrary for bare electrodes, the resistance can be measured between any points of the electrodes. The resistance registered is the lowest between any two points of the electrodes. If, for instance, condensation exists on a surface, it is the resistance in the condensation-layer that is the lowest and therefore the condensation becomes decisive for the result of the measurement. The result of the measurement is therefore unlikely to be representative of the moisture content inside the wood. For use in probes, ready-made electrodes in the form of surface coated (insulated) nails are available.
If the probes are intended for use in places where so high humidity levels are expected, that there is a risk of corrosion, electrodes of corrosion resistant materials should be used. This can, for instance be done by drilling a narrow hole equal to the diameter of a thin plastic hose. A piece of hose is inserted into the hole and a nail, for example of stainless steel, is led through the hose and hammered into the wood so only the tip penetrates the hose. The hose functions as an insulator against the surroundings. For hand held meters, bare electrodes are the most common ones on the market due to price. As regards fixed electrodes, the price difference between the use of bare and insulated electrodes is so minimal that there is no reason to use bare electrodes. Recently the Danish Building Research Institute has started to use gold plated precision manufactured contacts, meant for use in connectors, as electrodes in moisture measuring discs. The wires are fastened to the contacts by means of a crimptool. 4.3 Size and shape of probe With a look to the difference in size of dowels and discs it seemed worthwhile to test the influence of shape and size of the probes, as this could open up for the use of probes tailor-made to specific purposes. From previous tests it is well known that distance between the electrodes has only a minor influence on the resistance (correct within wide limits). Therefore experiments have been made regarding the influence of the design of probes using different sizes, shapes and electrode distances and designs. The results showed that design is of no influence. The probe can, according to need, be made in the size and shape suitable for the task in question. Fig. 4: Measurement of moisture content (resistance) with bare and insulated electrodes respectively. With bare electrodes the lowest resistance is measured. In case of surface condensation, the condensation is decisive for the result and not the moisture content inside the wood 5 Experience from the use of moisture measuring probes Long-term experience with regard to discs and dowels for moisture measurement has been favorable. The long-term stability has been examined through measurements made on 20+ year old discs in a roof construction. These
measurements show that the long-term stability is good. A more inexpensive solution is to place electrodes directly in the wooden construction to be examined. The mounting is simple and inexpensive and if the only purpose is to observe changes in the moisture content and not the absolute moisture content, electrodes plugged directly into the wood in the construction could prove beneficial. By use of electrodes plugged directly into the wood it has proved practical to apply ordinary terminal blocks for measuring. Measurement can be carried out either directly on a coupler of conduits attached to the electrodes, or wires can be led from the electrodes to a suitable measuring point where they end in a terminal block. If the two outer links in a three-light terminal block are used the pins of an ordinary handheld moisture meter can be used for a direct measurement of the moisture content in the desired depth, see Fig. 5. Fig. 5: Measurement of moisture content in various depths using a hand-held moisture meter and three-light terminal blocks 6 Examples of usage of built-in electrodes for moisture measurement Over the years built-in electrodes have been applied both for laboratory tests and in connection with long-term surveillance of renovation solutions or new types of construction. As an example of usage for laboratory purposes can be mentioned usage in connection with surveys of roof constructions and windows. 6.1 Moisture conditions in unventilated roofs In connection with tests regarding moisture conditions in unventilated roof tile underlays in the SBI moisture test house, moisture was measured in the upper and lower surface of rafters as well as in counter battens. For this purpose a large amount of dowels have been used which were built into the elements before erection. The wires were hooked up to a datalogger inside the house. A regular registration of the moisture conditions in the various roof elements was performed every 6 hours for 2 years. This was done in order to monitor the moisture content in the wooden
construction thereby making it possible to judge whether there was a risk of degradation or not. 6.2 Accelerated ageing of windows When performing accelerated ageing tests on wooden windows, insulated electrodes were inserted directly into the frames in order to register the moisture accumulation and the drying-out, all depending on the position in the window. In these cases the electrodes were inserted where the greatest risk of moisture pick-up was likely to occur, for example 5 mm from the outer surface of the frame or 5 mm below the edge of the window fillet/groove, see sketch on Fig. 6. The electrodes were inserted from the backside of the window (the inner surface of the window) and the wires were hooked up to a measuring panel of terminal blocks fixed to a place convenient for manual measurement. Measurement took place with a hand-held moisture meter once a week. Reading of approximately 30 measurements could be done in less than 30 minutes. Fig. 6: Position of electrodes in window Fig. 7: Insulation of electrodes frames during accelerated ageing test with plastic hose 6.3 Tests concerning absorption and drying-out In the case of pilot projects or damage investigations, there is often a need for measuring the moisture content in certain depths to ascertain the moisture distribution in the construction. In order to illustrate these possibilities, some tests with a series of insulated electrodes embedded in different depths were made. The electrodes are executed as described earlier with a drilled hole into which a thin plastic hose is inserted. Through this hose, are inserted electrodes of brass or copper so only the tip protrudes, see Fig. 7. Identical tests with fir sapwood, fir hearth-wood and spruce were carried out. The electrodes are placed in heights varying from 5 mm to 40 mm above the bottom of the wood specimen. As regards the first pair of electrodes the y are placed approximately 15 mm from the end of the wood specimen, and as to the last pair of
electrodes approximately 120 mm from the opposite end. The electrodes are placed in rows, for fir two rows of sapwood and two of hearth-wood respectively. Half of the measuring points are doubled so that some absorption of the moisture through the end specimen can take place at one end of the wood specimen while at the other end a wetting/spraying primarily takes place through the side wood, see Fig. 8. The wood specimens were placed with the bottoms emerged 3-5 mm in a water bath for approximately 30 days. The moisture content is measured regularly by connecting the electrodes to an ordinary hand-held moisture meter by means of a terminal block. Close to the end of the specimen a rather quick wetting took place especially in the three pairs of electrodes in the deepest position (closest to the end grain). Also the remaining pairs of electrodes showed a faster wetting due to the influence of the endgrain wood (absorption). Very distinct differences in the wetting and drying-out cycles for the various qualities can be observed as the sapwood is wetted much faster and more thoroughly than hearth-wood which again is wetted to a higher degree than spruce. The test clearly shows that insulated electrodes are a suitable way of determining the moisture content in different depths. Subsequently approximately 4 weeks of drying-out took place by means of heat lamps placed below the wood pieces (so that the moist bottom was heated up). From Fig. 9 it can be observed how the heating causes an instant rise in the moisture content because the temperature drives the moisture into the wood by diffusion. This corresponds to conditions under which a moist window frame is exposed to sunlight Side view Seen from end-grain Height (mm) 50 40 30 20 10 0 Row 1 Row 2 Row 3 Row 4 Row 5 Water level Sapwood Line 3 Line 1 Line 2 Line 4 Heartwood 0 20 40 60 80 100 Distance from end-grain (mm) Fig. 8: Position of electrodes in wood specimen during wetting and drying experiment where the heating will drive the moisture into the wood. The temperature and thereby the drive pressure will increase on dark surfaces. This fact is confirmed by investigations concerning moisture measurement by means of electrodes in window frames and beams in the SBI moisture test house. For the highest-lying pairs of electrodes a certain delay can be observed both during wetting and during the following drying-out. This corresponds to the general experience that moisture accumulation and drying-out processes are dependent on the dimensions of the wooden pieces. Drying-out by diffusion through the wood is a slow process and is even slower when there are only small variations in the vapour pressure
50 Moisture content (weight-%) 40 30 20 10 0 Heating start 0 500 1000 1500 Time (hours) Fig. 9: Moisture content in wood specimen during water uptake and subsequent heating 7 Usage in practice Usage in practice is mainly to be found in roof constructions. Normally dowels or discs are installed in connection with erection/renovation and wires for measurement purposes are led to a convenient measuring point. When it is a question of manual reading it will, as a rule, only be possible to make one or at best a few readings annually. As far as possible the measurements should be made at the same time of the year each time so that the influence of the seasonal variation in the moisture conditions does not become too great. When assessing the measurement results, consideration should be given to how the moisture variations could behave outside the moment of measuring. In components exposed to the environment (roofs, outer walls, etc.) it would normally be most suitable to make measurements during the period of February- March (in the northern hemisphere) at which time the highest moisture content is to be found in the constructions. 8 Conclusion Based on experience moisture measurement in wood by means of discs, dowels or built-in electrodes must be considered as a good and reliable method. With insulated electrodes (nails), it is possible to apply the method for measuring in certain depths, for instance when assessing the moisture distribution. Measurement can be made by means of an ordinary contact plug meter and a 3-light terminal block. Experience shows that the long-term stability of built-in sensors/electrodes is good. Furthermore it is important to remember the following before a moisture measuring is initiated:
Knowledge of building physics is a prerequisite when interpreting the results exceeding ordinary moisture control. The moisture conditions in a construction may vary considerably from one area to another and consequently several measurements are necessary to give a clear picture of the conditions. The moisture conditions may vary considerably over the year and an assessment of the results should take this fact into consideration. As with hand-held moisture meters measurements should be made along the grain. It should be avoided to measure near cracks or nails or in impregnated wood. 9 References Munch-Petersen,J. (1990), Fugtmåledybler (in Danish). Report 198, Institutet for Husbygning, Technical University of Denmark. Nordtest Method, NT Build 420 (1993), Building Materials, Wood: Moisture Content. PHØNIX Posten (1982), Fugtmåling i konstruktioner (in Danish) Prebensen, K and Brandt, J. (1984), Fugtmålinger i træ (in Danish). Report, Teknologirådet, Publication 473, COWIconsult.