You are viewing an archived copy of this website captured Mon Nov 05 11:31:03 AEDT 2012

All Content © CRC for Forestry 2007

The Log - Issue two - Evaluation of log measurement errors as related to harvester measurement accuracy

Martin Strandgard, University of Melbourne

Introduction

Most mechanical harvesters can optimise the log mix cut from each tree by estimating its characteristics from initial length and diameter measurements, and predicting the log combination that maximises return while fulfilling market requirements. Optimisation requires accurate harvester measurements. Murphy (2003), in a study of 39 mechanical log-making systems, found on average 20 per cent of potential value was lost through measurement inaccuracy and poor selection of log combinations. Although there are no comparable Australian studies, observations suggest similar levels of loss are likely to be occurring.

Harvester measurement accuracy is assessed through comparison with manual log measurements by assuming manual measurements represent true log dimensions. However, until manual measurement errors are known, differences between manual and harvester measurements cannot be solely ascribed to harvester errors. Potential manual measurement error sources include instrument errors, poor measurement technique and flawed assumptions. The major manual measurement error source is believed to be the unstated assumption that logs are regular in shape. In reality, branch stubs, bumps, sweep and non-square log ends can effect log length measurements while out-of-round stem cross-sections can effect log diameter measurements.

Summary of results

Manual measurements of length and small end diameter were made on 200 radiata pine logs at two sites near Mt Gambier, South Australia. Length was measured on three sides of each log and diameter was measured with a diameter tape and single horizontal calliper measurements. At one site, measurements were repeated by two operators. Marked variability was found between log lengths on different log sides and between diameter measurement instruments on a substantial number of logs. This variability is believed to mainly result from:

  • Non-square log ends caused by stem taper and sweep tilting harvester heads (Figure 1)
  • Out-of-round logs caused mainly by stem eccentricity (Figure 2)
  • Figure 1

Variability in manual measurements can obscure errors in harvester measurements. Large log samples can “average out” the variability but at the cost of increased measurement time and reduced productivity, which is particularly important in harvester calibration. Recommendations for reducing time and cost by minimising log sample size include:

  • Excluding butt logs as they were found to be more variable then other logs
  • Selecting straight logs free from bumps or branch stubs
  • Selecting logs with a circular cross-section

For more information please contact Martin.

References

Murphy GE (2003) Mechanisation and value recovery: worldwide experiences. In 'Proceedings of the Woodfor Africa Forest Engineering Conference'. July 2002. pp. 23-32. (Pietermaritzburg: South Africa).