The prediction of long-time viscoelastic creep from short-time data

Experimental evidence is produced to corroborate the theory that creep of wood can be divided into two parts: a ‘primary’ or short-term part, and a ‘secondary’ or long-term part. The primary part was completed within one to two days. The secondary part followed a straight line on a plot of creep against logarithm of time, thus simplifying long-term creep prediction from short-term data. In addition to the carefully-controlled experiment described here, the analysis was applied to eight-year creep data published by Gressel, who also supplied numerical values. Alternative prediction methods by Hunt were based on a ‘normalising procedure’. Most of the predictions for the eight-year tests came within a few percent of the measured value, and were generally conservative. It was not possible to confirm whether the secondary creep is likely to follow the straight line indefinitely, or whether the rate decreases eventually towards a viscoelastic creep limit.