CHECK-SCALING 



117 



From these results a rule of thumb may be suggested as follows: Add one-third 

 to the per cent of sweep as expressed in terms of diameter of log to obtain the 

 per cent of cull; e.g., a log 16 feet long and 16 inches in diameter scales 159 board 

 feet. With a sweep of 4 inches or 25 per cent, deduct f X25=33| per cent or 

 53 board feet; scale, 106 board feet. With a sweep of 8 inches, deduct f X50 = 66f 

 per cent, or 106 board feet; 3cale 53 board feet. With a sweep of 2 inches no 

 deduction would be made, since this is merely the normal crook. 



Logs which crook in two or more planes must be culled far more heavily than 

 when the axis lies in a single plane. For a given per cent of crook the scale is 

 roughly proportional to the square of the per cent scaled by the deductions set 

 forth above; e.g., a log which scales 50 per cent or one-half if crooked in one plane 

 will, if crooked in two planes, scale (J) 2 or 25 per cent of its contents. 



94. Check-scaling. By check-scaling is meant the re-scaling of 

 selected logs or of a portion of a total run of logs, in order to determine 

 the relative accuracy of the original scale, check the methods used by 

 the scaler and detect and correct errors in these methods. A re-scale 

 requires the remeasurement of all of the logs. ' The necessity for a 

 re-scale is usually revealed by a check-scale. 



Where a number of scalers are employed, check scaling becomes 

 necessary in order to maintain uniformity in scaling practice. No 

 matter how carefully the standard of scaling practice is set forth in 

 printed instructions which cover not only the " scale " with respect 

 to diameters, length, taper and trimming allowance, but rules for deduc- 

 tions for defects, individual scalers tend to vary from this standard 

 through habit or carelessness and inexperienced men are slow to acquire 

 accuracy, especially in scaling defective logs. 



A check scale should be made by the most experienced man available as fre- 

 quently as possible, but usually at from three to six months' intervals. Where 

 logs are numbered, the original scale should show the deductions made from the 

 full scale of each log (§ 85). The check scale can be made at random on as many 

 logs as there is time for. The total scale for the logs checked is then compared 

 with the original scale of the identical logs, keeping separate the sound and the 

 defective logs. Using the check scale as 100 per cent, the per cent of error in 

 scaling is computed according to the following plan: 



Sound logs 



Scale by 



No. of logs 



Scale per cent 

 + or — 



Defective logs 



No. of logs 



Scale per cent 

 -f- or — 



Total 



No. of 

 logs 



Scale per cent 

 + or — 



James Smith 

 Check scale by 

 John Kipp 



The standard of accuracy in the U. S. Forest Service for check scaling requires 



