.452323 Variation and, allowances FOREST PRODUCTS LABORATORY t , FOREST SERVICE U. S.. DEPARTMENT OF AGRICULTURE SMALL SAWMILL IMPROVEMEN T PRACTICAL POINTERS TO FIELD AGENCIE S WASTE FROM VARIATION IN SAAING PRECISIO N A major loss sustained by small mills results from inaccurately . cut lumber .. The product brings •less per thot and boarrd feet and encounter s stiffer sales resistance than that from the more accurately cutting band mills, A more obscure loss is the excessive manufacturing waste that results from inaccurate cutting . Studies* by the Forest Products Laboratory indicate that th e portabl e.,mill operator, in sawing for thickness, cuts only about 20 percent " of the boards within 1/32 inch of the thickn :ese he sets for . The rem inin g SO percent may vary in thic .hieas from as much as £;/32 inch on the thin side to as much as 5/32 inch on the thick slide . (Fig . 1, upper pile .) Conse quently the operator must set to cu t . most boards too thick so that there , will not be too many that are excessively thin. Each 1/32 inch thus adde d to avoid thin boards reduces the ie aliblo cut exaetly a.-; if the saw keri:' were increased 1/32 inch . This wnstc can be f-aini :ivied by keeping th e equipment in good condition . The main causes for inaccurately cut lai ber arc ; (1) Fault y condition of the saw, such as uneven filing; of saw teeth, excessive or uneve n swage, dull teeth on one side, unequal tension ; ( ."'.) orn bearings' in mandril , carriage wheels, and particularly in the set-works ; (3) moor installation o f carriage and saw, chips between log and headbleck or on track ; (4) careles s setting, iaadeq-ua.te manipulation of dogs, miscalculation rosulting in the last board cut from each log being either undersized or oversized ; (5) frozen timber, unusual• stresees in wood . These will not be elaborated upo n now . By adjusting the saw, carriage, and track for the most accurat e work possible, sales advantage due to more accurately cut lciabcr is obtained . To get the second advantages an increased yield, set to cut all bohrd s thinner . Note that two stews are essential : (1) Adjusting to improve -cuttin g accuracy and (2) setting for thinner lau ber . The number of thin reject ; wil l not be increased because the effect of trtiing up eT .nment is to bunch board s closer to the intended. 1/32--inch group and thus to decrease the number o f extremely thick and thin beards . (Fig . 1, effect of ier:?roved cutting accuracy shows in middle pile . If 2S/32 inch is taken as the thinnest rough dry thickness that will 7aroduee the softwood standard 25/32 S2S, the setework s can be moved either by 1/ ) 2einrh intervals or, where the intervals arc l/ l inch, by placing a 1/ .6-inch leaf in the back stop so that the entire cut ' is 3/32 inch thinner . Lower pile . The consequent increase ' in yield i s '4-l/2 percent . ) *E . M . Davi s aF F . L. RS59-2 I 4 Maintained at Madison, Wisconsin in cooperation with the University of Wisconsi n °See outline in Small Sawmill Improvement Working Plan, March 1980, for explanation of indexing system proposed The ideal setting_; noes not attempt to a,-a lif,- every beard hut rather to set for thinner boards until a point is reached whore the gai n from ineroasin yields is baianced by the loss from thin rejecter : . Thi s bal. nce is reached when a 1-1/2 percent increase in rejects results fro .._ setting 1/32 inch thirxicr, hence continue to set thinner by 1/52-inc h intervals until the number of thin 'ejects increases by more than . 1-1/ 2 percent over the previous setting . The ideal adjustment is I/52-inc h thicker than this setting . 2:hi rejects for each setting can he 'de' termined by measx.in, t o the nearest 1/32 inch at the thinnest place, exclusive of the terminal foot , a hundred freshly sated, nominally -/4-inch boards . Th e . nunber of iuoaid s less than the thickness r e quired to meet standard . specifications + ;7:as an allowance for shrinkage in drying is the total percent of thin _'ejects . Consequently when the rejects from setting 1/32 inch thinner increase .two or more boards per hundred set for the next thicker 1/32 inch . SOP' i'â–ºOODS An allowance of 1/32 inch mast be made for shrinkage of inc h softwood lumber from a green to the equivalent of a thoroughly air-dry condition . Moreover, most softwood grading rules admit 2.0 percent 2S/32 inch rough dry lumber ; hence, the 23/32--inc h- green lumber 'group is th e thinnest to qualify . A mill cuttin, ; lumber° as in the upper pile in Figur e 1 should therefore set 1/32 inch thinner, since the 1-1/2 percent increased _ yield is but pertly offset by the 1 percent increase rejeects . nor a nil l grouping its cut as in the lower ,file in Figure 1, a 1/32 inch thinner set , although gaining the 1-1/2 percent increased yield, results in an increas e of 3 percent in rejects . HARDWOOD S The set-up for hardwood differs in that 2/32 inch must b e allowed for shrinkage (3/32 for beech and hic, _ory) and the grading rifle bas e is 32/32 inch for rough dry lumber so tha t . 34/32 inch is the thinnest gree n board to qualify . A mill grouping its cut as in the upper pile in Figure 1 for hardwoods should set four places thicker, so that the 2 percent show n in the 30/32-inch group would be in the 3+/32-inch group . , Contributed by C . J . Telford ' , Small Mill Specialist , Forest Products Laboratory , October 4, 1932 R899-2 T.? - .S7' Lti.7.Fc..;s7tY Before adjustin g for accurate cuttin g i/ ;2. r-r :. V/ AL {/ After adjustin g „/ i 1 Adjusted and set t o optimum thicknes s F./ -- I '- - 9 3 31 ,/;" -t/ 5637 3 g 39 40 41 Thickness (Thirty-seconds of an inch ) 2g 2 0 - Figure 1 .--Example of variation in thickness of each lumber as cut at portable ' mill s la) Forest Prbucts Laboratory, Madison ;Wis . B.89.c.,-2 boards of inc h LMORATOR Y