The Structural Conservation of Panel Paintings

moisture content cannot be predicted, expected dimensional change can-

not be calculated with precision. The theoretical dimensional change for a

given piece of wood can be calculated using the following formula:

where: DD 5 dimensional change, in linear units; Di 5 initial dimension, in

linear units; MCi 5 initial moisture content, in percent; MCf 5 final mois-

ture content, in percent; FSP 5 fiber saturation point, in percent (if not

known for species, use 30%); and S 5 published value for shrinkage, in

percent (Sl, Sr,or St).

In calculating dimensional change for pieces of wood with inter-

mediate or variable growth-ring placement, a modified shrinkage percent-

age would have to be estimated by rough interpolation between the radial

and tangential values. It should be noted that because shrinkage takes

place only below the FSP, neither MCinor MCfcan be greater than the FSP.

Positive values of dimensional change indicate shrinkage; negative values

indicate swelling.

As an example, suppose a painting panel is assembled by the edge-

gluing offlat-sawn boards that have been identified as poplar (Populusspp.);

the finished panel measures 76 cm in width. The panel is placed in a build-

ing where records have shown a seasonal variation from a high of60% RH

in the summer to a low of 25% during the winter heating season. What

dimensional changes in width can be expected?

From the oscillating curve of Figure 7, one can assume EMC

extremes of a high moisture content (MCi) of 10.9%, a low moisture

content (MCf) of 5.4%, and an FSP of 30%. From Table 1, Stfor poplar is

given as 8.5%.

The estimated change in the width of the panel from its summer

to its winter condition is calculated as

The panel would be assumed capable of shrinking by approxi-

mately 1.25 cm. It is important to realize that this calculation would pre-

dict the behavior of normal wood free to move, whereas a painting panel

may be subject to restraint by its frame and cradling or mounting hard-

ware and by the applied layers of gesso and paint.

Careful evaluation of the formula presented above leads to some

important general conclusions. It is apparent that the overall dimensional

change, DD,is directly influenced by the magnitude of each of the three

factors Di, S,and DMC(i.e., MCi 2 MCf), which should be considered sepa-

DD 5

2 30% 1 10.9%

76 cm (10.9% 2 5.4%)

30%

8.5%

5

2 0.30 1 0.109

76 cm (0.109 2 0.054)

0.3 0

0.085

5 76 cm (0.055) 5 1.25cm

3.338

16 Hoadley