Measuring the Progress of Kiln Drying

Although much progress has been made in the development of automatic systems for monitoring the drying of timber it is not yet clear whether these always represent a reliable substitute for the traditional method of monitoring which involves the periodic examination of withdrawable samples during the kilning operation. The withdrawable sample method is essentially the benchmark from which other methods have been developed and because it can be relied upon to give good quality drying under most circumstances, even with timber which has unpredictable drying properties, it has been presented in full in this guide. The broad principles governing the selection and positioning of withdrawable samples apply equally to the selection of samples selected for monitoring by other methods.

The particular advantage of the withdrawable sample method is that it enables an accurate overall assessment to be made of the condition of the timber at any point duing the drying operation. If the samples have been correctly selected and positioned in the load, the following assessments can be made:

• The extent of surface checking and distortion within the load can be assessed.
• The average moisture content for known pieces can be calculated and this can be used to estimate the variation in average moisture content in the load and also the average moisture content of the load.
• The steepness and variation in moisture gradient within the load can be assessed.
• The severity of any casehardening stresses can be evaluated.

As well as providing information which can be used to guide the drying correctly this kind of monitoring provides a useful check on the uniformity of drying conditions horizontally and vertically within a kiln load. At a more fundamental level, information obtained during drying trials with the withdrawable sample monitoring method is used in allocating new commercially available timbers to the most appropriate drying schedule.

A disadvantage in monitoring timber in this way is that it is a time consuming process; the samples have to be selected, incorporated into the load, withdrawn at intervals during drying, weighed, and cut to obtain test sections. In addition a small loss in value of the timber will be incurred when test sections are removed from each sample. However, in many circumstances, particularly when the drying properties are unpredictable or when little experience has been gained with a particular species, the advantages of the withdrawable sample method will generally outweigh its disadvantages. With experience it may be possible to reduce the extent of sample monitoring, especially with timber which has predictable drying properties. In such instances the introduction of automatic monitoring, based on the moisture content or weight of the load, may often give satisfactory results. Although electrical resistance moisture meters are useful for monitoring moisture contents below fibre saturation point, an alternative means of monitoring will normally be required at higher moisture contents to enable a kiln schedule to be followed.

• Selection of Samples

The samples should be selected to provide an indication of the likely range of drying rate and moisture content which will occur during kilning. There are several variables which have a major influence on drying including:

• The difference in drying rate between species
• The proportion of sapwood and heartwood within pieces of the same species
• The thickness of the timber
• The initial range of moisture content of the timber
• The way the timber has been converted (quarter sawn timber usually dries at a slower rate than plain-sawn)
• Differences in drying conditions across and along the kiln load


• Number of Samples

  The number of sample pieces required and their selection is governed by which and how many of the above variables apply. For example in a kiln load containing a single species, the number of samples which would be required to monitor drying accurately would only be about one half of the number required for a load containing two species with different drying characteristics.

  For a typical drying operation involving an average size kiln load, and a single species of timber converted through and through to a constant width, it is recommended that the number of sample pieces in the sample series (see Figure 9, below ) is a minimum of six. This sample series should consist of both plain-sawn and quarter-sawn samples and if possible it should include pieces which are representative of the wettest and dryest wood in the kiln load.

  The number of samples should be increased proportionately in larger kiln loads. However there is a practical and economic limit which may restrict the number that can be monitored. For example in large loads containing more than one species of more than one thickness, it may be impractical to have enough samples to provide all the necessary information about the progress of drying. Conversely it may be possible to include less than six samples when kilning timber which has a fairly uniform moisture content, and well known and predictable drying characteristics .

• Size of Sample Pieces

  The size of the sample pieces should be as typical as possible of the pieces making up the load. With long stock it may be impracticable to handle and weigh the full length material and it may be necessary to monitor shorter pieces (normally not less than 1500mm). Because moisture loss occurs more rapidly in the end zones of a piece, the pattern of moisture loss will differ quite markedly between shorter and longer pieces of the same species. This can result in inaccuracy in the determination of moisture content and when, in the drying of long stock, it is necessary to use short samples, these should be end sealed to prevent disproportionately large moisture loss.

  Samples should be of similar widths to those of the timber in the load, but in the case of very wide boards or planks they may for convenience be cut to a width of about 230mm.
   

• The Position of the Samples

Accurate estimates of the range of rate of drying and the range of moisture contents depend on the correct positioning of the samples within the load. Samples should be positioned so that they provide an indication of the drying rate and behaviour of the wood in the fastest and slowest drying regions of the load. For example, with overhead fan kilns, by placing the wettest of the quarter-sawn samples at the air outlet side of the load it is possible to estimate the slowest drying rate and therefore to judge whether adequate drying has been achieved. The wettest plain-sawn samples positioned at the air inlet side would dry the most rapidly and as drying degrade would be most pronounced in these pieces, they provide a means of assessing the risk of degrade within the load.

To avoid overdrying and excessive distortion, it is also important to be able to estimate the lowest moisture content which will be obtained under a particular drying schedule. An estimate of this value can be obtained from the drier of the plain-sawn samples if positioned at the air inlet side of the load.

Under frequent air flow reversal, which is normal practice, the fastest and slowest drying areas will vary depending on the configuration of the kiln. It should be noted that, whichever way the air is circulating, the slowest drying region is situated halfway along the path that the air follows through the load before it is reheated (see Figures 3 and Figure 4).

A system of numbering or lettering should be adopted so that each sample can be identified and re-positioned correctly after being withdrawn for weighing.


• Overhead and Some Side Fan Kilns

  With overhead and some side fan kilns (see design D1 Figure 4) the heated air only passes through the load once before it is reheated and recirculated. With these kilns the slowest drying position is in the centre of the load, and arranging the easy withdrawal of the samples from this position may present difficulties. However, the provision of these samples is essential for accurate monitoring, particularly with older kilns having wide loads and slower air circulation. Apart from initial weighing to determine their moisture content these centre samples need not be weighed until those on the inlet sides are approaching the desired moisture content. With overhead fan kilns the samples may be placed in the end of the load which is easily accessible from the doors.
  

• Side Fan Kilns

  With horizontal flow side fan kilns and some other side fan designs (see design D2 Figure 4) the air passes twice through the load before it is reheated. The slowest drying part, under frequent air flow reversal, will therefore be on the side of the load which is remote from the fan.

If more than six samples are required, their positioning will depend upon the composition of the load and on kiln design. If necessary, the advice of the kiln manufacturer should be obtained on the correct positioning of samples.

To ensure that the drying response of a sample is typical of the wood in a particular part of the load, samples must be incorporated in a way which does not interfere with local drying conditions. Additionally the method of incorporation must allow easy withdrawal of the sample for weighing. However difficulties of sample withdrawal may sometimes be unavoidable with timber which is prone to distort badly.

One method of accommodating withdrawable samples is to place sticks over them which have been notched out to about half the normal thickness. This tends to be time consuming and rather wasteful of sticks unless it is feasible to have samples of a standard width and position, in which case the notched sticks can be used repeatedly. An alternative arrangement, using suitable lengths of sticks of half the normal thickness, is shown in Figure 10, below. Both these methods involve opening the main kiln doors for the removal of the samples but with modern kilns the air conditions soon return to normal.

Samples can be accommodated in the sides of loads by cutting one or more sticks off short (Figure 10) leaving the sample free. Access to side samples is usually by the small side doors rather than by the main door and because the operator therefore has to enter the kiln, there may be practical difficulties in retrieving samples under certain kiln conditions.

Samples should always be positioned within the load and not mounted on short projections at the ends of loads, or between two separate loads, where abnormally fast drying may occur.

• Using the Samples to Assess Timber Condition

Information on average moisture content, moisture distribution and intensity of casehardening stress is obtained by examination of cross-sections cut from each sample (Figure 12). If the series of withdrawable samples has been correctly selected and positioned within the load, the combined results from all these cross-sections will enable a fairly accurate overall assessment to be made of the condition of the load.


• Estimation of the Average Moisture Content of a Sample

By first estimating and then monitoring the fall in average moisture content of the samples during drying, a given schedule can be followed accurately.

The average moisture content for a particular sample is estimated by first measuring, using the oven method the moisture content of one (or more) test sections cut as indicated above. If it is then assumed that this measured moisture content is typical of the remainder of the sample, the dry weight of this remainder can be estimated and changes in its average moisture content can be monitored throughout drying by weighing.

Suppose that the initial average moisture content of the sample was estimated to be 35%. If the sample weighed 12.40kg then its dry weight can be estimated:

Dry weight = (wet weight/moisture content)/100 + 1

=12.40/[(35/100) + 1]

=12.40/1.35

=9.18kg

This estimated dry weight remains a constant quantity as long as no further wood is cut from the sample.

If after a period of drying the actual weight of the sample has fallen to 11.72kg, its new average moisture content can be estimated as follows:

New average moisture content = [(Current weight/Dry weight) - 1] x 100

=[(11.72/9.18) - 1] x 100

=27.6%

This technique is based on the assumption that the test section(s) do provide an accurate estimate of the average moisture content of the sample from which they were cut. However, even in the ideal situation where a section is removed from each end of the sample (and the values obtained averaged) an appreciable error can occur. This is because moisture content will inevitably vary along the length of a piece and for this reason it is always advisable to redetermine the moisture content of a further batch of test sections towards the end of a kiln run.

With certain species, such as teak, there is a marked tendency to retain pockets of moisture along the length. In these cases the average moisture content is difficult to determine in the normal way and, if it is necessary to dry such timbers to a uniform moisture content, there will be a need to cut more test sections to obtain a suitably accurate estimate of average moisture content.


• Estimation of the Average Moisture Content of the Load

Having obtained the average moisture contents for the individual samples, it is then quite straight-forward to use this information to estimate the average moisture content of the load. Again, this can be estimated at any time during drying and it helps the kiln operator to judge the progress of drying. It is estimated simply, as follows:

Average moisture content of the load =	Sum of average moisture contents for all for all samples being used to monitor the load __________________________________
	Total number of samples  being used to monitor the load

 

For example, if at a particular time, the samples gave average moisture contents of 16, 14, 13, 12, 10 and 10%:

Average moisture content of the load

= (16+ 14+ 13+ 12+ 10+ 10)/6

= 75/6

=12.5%

Once below fibre saturation point, electrical resistance moisture meters can be used to augment the information obtained by oven drying.

 

• Assessment of Moisture Distribution and Casehardening Stresses
 

The importance of assessments of moisture distribution and casehardening is in evaluating the risk of drying degrade. Both assessments can only be made by cutting fresh test sections. It is often convenient to cut these at the same time as sections are taken for the average moisture distribution. Casehardening stresses should be assessed at least twice during a kiln run: first before drying is commenced (to ensure that the appropriate drying schedule will be suitable without modification) and again towards the end of the kiln run to ensure that drying is progressing correctly. More assessments may be necessary when drying difficult sizes of timber or species which have unpredictable drying qualities.

Moisture distribution is assessed by removal of a test section and by sub-dividing this into strips as shown in Figure 12, below. The strips are cut so that the inner and intermediate strips are representative of increasing depth within the original sample. Moisture distribution is then assessed by measuring by moisture content of each strip separately by the oven method. In parallel (Figure 12, below) a section can be removed for casehardening tests. Although these assessments are time-consuming and labour intensive, careful monitoring in the manner indicated will always be economic if it avoids extensive and unnecessary degrade within a load.

Figure 12 Re-cutting of kiln sample for testing moisture content, average and distribution: also casehardening

• Monitoring Small Dimension Stock

When dealing with short dimension stock the samples are selected and positioned as normal but a slightly different technique for assessing the condition of the timber has to be adopted. In such cases a test section is cut from the middle of each sample which may be, for example, a short furniture square. Immediately after removal of the test section, which is used for moisture content determination, the two freshly cut surfaces of the remaining parts of the sample are painted and fastened together. The estimated moisture content of the joined sample can then be monitored by weighing as drying proceeds.

This method cannot be used when even smaller items such as flooring blocks or brush backs are to be dried. Here it will be necessary to estimate the moisture content of the load by measuring the average moisture content of a few whole items using the oven method. The progression of drying can then be monitored roughly by weighing an additional series of samples at intervals. A moisture meter can be used to estimate moisture contents if samples are first allowed to cool. When the desired moisture content is apparently reached, further checks using the oven method and/or a moisture meter should be carried out on samples from various parts of the kiln load.  

Copyright © 1992-2004 Malaysian Timber Council
18th Floor Menara PGRM, 8 Jalan Pudu Ulu, Cheras 56100 Kuala Lumpur, Malaysia
Tel: +603-9281 1999 Fax: +603-9282 8999 Email: council@mtc.com.my

Last modified: January 24, 2002
Send mail to webmaster@mtc.com.my with questions or comments about this web site.