Controlling the kiln: uniformity, energy and quality routine

The best schedule fails if the charge is uneven or the control data are poor. This lesson ties the course together.

Uniformity: the spread matters as much as the average

A lot can leave the kiln with the right average and still contain wet and overdried boards. EN 14298:2017 (Europe), according to its scope text, assesses a lot of dried sawn timber by the target, the lot average and the variation between pieces, with an optional case-hardening specification. Its sampling plan and tolerances were not verifiable. Spread comes from sapwood share, thickness variation (about 4% drying-time change per 1/32 inch in 1 inch red oak, FPL), airflow and sensors.

What a UBC study found (Rahimi et al. 2023, Canada, western hemlock, small laboratory kiln, 42 pieces per run, target 12%).

• The control schedule gave a mean final MC of 10.87% with the lowest variation; Schedule I gave a mean of 15.59% with the highest.
• Initial weight was the most informative input; predicting a single board's MC did not generalise (R-squared 44.81% on test data), but classification reached 69.05% test accuracy.
• Classes in this study: below 10% over-dried, 10 to under 19% acceptable, 19% or more under-dried. These are the study's thresholds, not a standard.

Worked example. Eight boards finish at 9.5, 12.0, 18.9, 19.0, 10.0, 21.0, 8.0 and 14.0%. With the study's thresholds: over-dried 9.5 and 8.0; under-dried 19.0 and 21.0; acceptable 12.0, 18.9, 10.0 and 14.0, i.e. 4 of 8, 50%. The mean (about 14%) hides the spread.

The conclusions are cautious: collect data on both board properties and kiln settings, think in classes rather than exact numbers, and validate any model on your own species and kiln. A Swedish doctoral thesis (Vikberg 2015) on moisture estimation, airflow and heat recovery could only be read as an abstract that the portal could not re-verify, so none of its figures are used.

Energy

Garrahan (2011, Canada, citing a 2009 study) gives total softwood lumber energy of 1,514 MJ/m3, of which drying is 1,000 MJ/m3, about two-thirds. Evaporation takes 60 to 70% of softwood drying energy; fan electricity is 5 to 10% of drying energy but up to 20% of cost. A theoretical benchmark is about 1.6 kWh per kg of water removed; for oak, Campean and Lazarescu report 2.7 to 4.0 kWh per kg (conventional drying).

Worked example. Removing 2,500 kg of water from a charge needs about 2,500 x 1.6 = 4,000 kWh at the benchmark, but 2,500 x 2.7 = 6,750 to 2,500 x 4.0 = 10,000 kWh at the reported oak values (different species and boundaries, so not directly comparable).

Unit conversion: 1 kWh = 3.6 MJ. A Hungarian plant (Europe) measured thermal energy for 28 mm oak of about 1089 MJ/m3 in summer and 2312 MJ/m3 in winter, i.e. 1089 / 3.6 = 302 kWh/m3 and 2312 / 3.6 = 642 kWh/m3. Measures named in the sources include insulation, heat recovery, fan drives, uniform presorted charges and controlled air drying.

A quality-control routine

1. Define target and tolerance before the run (service class 1 = 12% or less, class 2 = 20% or less; North American tables: 8% interior, 12% exterior, framing at most 19%).
2. Sort by species, thickness and sapwood share.
3. Control with representative samples in the slowest and fastest zones.
4. Run the stages deliberately: low temperature early, humidity down after one-third of green MC is lost, temperature up only when the core is safe.
5. Equalise and condition, and check with prong and shell-and-core tests.
6. Assess the lot at unloading by mean and spread.
7. Log curves and results and keep them for claims and schedule corrections.

Limits of this course

This course was drafted by an AI from the portal's fact-checked articles and no other source. Figures are regional: most numbers are North American (USDA Forest Products Laboratory and Wood Handbook) and several European items come from single studies. Check every figure against the relevant standard (for example EN 14298, whose sampling plan was not available), your supplier's specification and your contract before using it. European species shrinkage tables, commercial European schedules, sensor accuracy and kiln capital costs were not available and are not covered.

Key points

• Judge a charge by its average and its spread; sort before drying to reduce spread.
• Initial weight and logged kiln settings are cheap inputs for a classification approach, validated locally.
• Drying is about two-thirds of softwood lumber energy in the Canadian figures; convert units carefully (1 kWh = 3.6 MJ).
• Follow a written routine: define, sort, control, run, equalise and condition, assess, log.
• Treat this course as a starting point and verify figures against standards and contracts.