There are few problems more interesting to the pattern-maker than the determination of the weight of the castings which his patterns will produce. Some years ago the author investigated this subject very carefully, both theoretically and experimentally, and published the results in the first volume of The Manufacturer and Builder, of which he was editor.

The relative weight of patterns and castings can, of course, always be determined most accurately by measuring the pattern and multiplying the number of cubic inches which it contains by the weight of a cubic inch of the metal of which the casting is to be made. The weight of a cubic inch of the various metals in common use may be found in any table of specific gravities. This plan should always be adopted in the case of very large castings, because it eliminates several important sources of error; and when proper allowance is made for shrinkage it gives results which are very near the truth. But many patterns, especially those of small size, are so irregular in shape that accurate measurement is tedious if not difficult. In such cases a tolerable approximation may be obtained by weighing the pattern, and comparing this weight with the weight of the same bulk of the material of which the casting is to be made. If the specific gravity of all samples of the same kind of wood were alike, and if the casting were always the same size as the pattern and of uniform specific gravity, this method would be perfectly accurate. But even with every drawback it gives tolerable approximations.

In the following table we give a series of multipliers which express the relative weight of patterns of different materials when cast in different metals. To find how much a casting from a given pattern will weigh proceed as follows: -

Weigh the pattern. Then in the first column find the material of which the pattern is made, and opposite this and under the material of which the casting is to be made will be found a number which when multiplied by the weight of the pattern will give the weight of the casting.

Example: A pattern made of St. Domingo mahogany weighs 8 lbs. 6 oz. How much will a casting of iron weigh?

 Weight of pattern ............................ ...............8 lbs, 6 oz. Multiplier for east-iron set opposite . . . . 10 St. Domingo mahogany, Weight of casting ........................... .............83 lbs. 12 oz.

## Table Of Multipliers For Finding The Weight Of Castings From Patterns

 Material of Pattern. Specific Gravity. Cast Iron. Gun Metal. Yellow Brass. Zinc. Mahogany, . . . 854 8 10 9.8 8 10.2 St. Domingo do. . 700 10 12 11.5 9.5 12.2 Maple,..... 700 10 12 4 12 9.8 12.5 Beech,..... 624 11 13.8 13.4 11 14 Cedar,..... 596 11.5 14.5 14. 11.4 14.7 Yellow Pine,. . . 541 13 16 15.5 12 6 16.2 White Pine, . . . 473 14 2 17.8 17 14.5 18

The causes of error are: shrinkage in the castings; weight of nails and screws in the pattern; variation in specific gravity of material of which pattern is made; variation of specific gravity of metal of which casting is made. Shrinkage is too large an element to be left out of consideration, and we have diminished our multipliers by a proper proportion to allow for it. In the construction of patterns an allowance is usually made for this contraction, either by calculation or by the use of a shrinkage rule as it is called, - that is, a rule on which 12 1/8 inches is called a foot, and divided accordingly. But in making allowance for shrinkage in casting, pattern-makers understand that different shapes will shrink differently. The standard table of allowance for shrinkage in use in the best shops of the country is as follows: -

For Loam Castings,.....1/12 inch per foot.

" Green Sand Castings, . . . 1/10 " " " " Dry " " . . . . 1/10 " " "

" Brass Castings, . . . . 3/16 " " " " Copper " . . . . 3/16 " " "

" Bismuth " . . . . 5/32 " " "

"Tin " . . . .1/4 " " "

" Zinc " . . . . 5/16 " " "

" Lead " . . . . 5/16 " " "

When cores are to be used a suitable allowance must be made for them, but this is in general most easily and accurately done by measuring the cubic contents of the hole left by the core and calculating the proper weight to be deducted.

A singular oversight occurs in one of our architects"' Pocket-books" in the treatment of this problem. The reader is directed to use a series of multipliers which are arranged for the metal of which the casting is to be made, utterly irrespective of the material of which the pattern is made!