FIGURE 15.-PORTION OF WAEFFLER'S WRAPPING TABLE.

Dr. HERZFELD in his work on the Technical Testing of Yarns and Fabrics, gives the following table of variations above and below reputed count, but it does not appear that either these. allowances or any other are in general use, although some manufacturers insist on having their yarn delivered on the fine side of the reputed count.

One sometimes hears the expression "one count above or below will pass." The effect of a slight variation in count is clearly shown in the following example:

1000 Looms. 42" in reed; 68 picks per inch; reputed count 40's fillling at 36 cts.; actual average count 39.5; 180 pks. per minute; 60 hours per week; 50 full weeks per year; 80 % effective production.

Looms.

Pks. Mins. Hrs. Weeks. Pr Ct. Pks.

Ins. Cts. 1000 X 180 X 60 X 60 X 50 X 80 X 68 X 42 X 36 36 X 68 X 100 X 840 X 40 X 100

1000 X 180 X 60 X 60 X 50 X 80 X 68 X 42 X 36

36 X 68 X 100 X 840 X 39.5 X 100

$328,101 - $324,000

=

$4,101 Loss.

These may appear startling figures, but there does not appear to be any reason for doubting their accuracy, and they show how necessary it is for a manufacturer to keep a very strict eye on the count of the material as it comes into his mill.

Testing the Count from a Small Sample of Cloth.- When a small sample of cloth is submitted to a manufacturer, the count of materials has to be ascertained, and there are several methods of doing this. Perhaps one of the commonest plans adopted is by comparison with known counts of yarn, usually by twisting a certain number of threads from known count and sample together. In the hands of an expert this is probably sufficiently reliable for all practical purposes, but it requires some skill and experience before the results obtained in this manner may be relied upon. In addition, when this plan is adopted it is necessary to have a fairly wide range of counts and spun from various kinds of cotton, as the test is one of touch.

Now while there is no desire to cast any doubts on the general accuracy of this method, attention should be drawn to the fact, that count is a matter of weight for a given length, and not a question of thickness of material. Under these circumstances it would appear that the correct method is to weigh the yarns and calculate the count from the weight.

There are a number of specially designed balances for ascertaining the count of short lengths of yarn, all of which however may be classed in two groups, A, those which are based on the principle of length units per weight unit, and B, those in which a given length of yarn causes a pointer to indicate the count on a graduated scale; each system has advantages and disadvantages, but if a special balance has to be used, preference should be given to the latter type, because in most cases a greater length of material is tested, they allow of multiples and measures of the stated length being used, and they are nearly all provided with a standard weight by which they may be calibrated. The former type may be used with advantage when only very short lengths are available.

Figure 16 shows a drawing of a balance belonging to the first type, and is known as Staub's balance. It consists of a light wire beam, at each end of which there is a hook, the one on the left representing the unit of weight, and that on the right being used to carry the short lengths of yarn being tested. Small brass templates are supplied with the apparatus, each of which bears the same relation to the length unit that the left hand hook bears to the weight unit. The number of threads which balance the beam indicate the count.

Figure 17 illustrates a Swiss balance belonging to the second class. This apparatus can be used for several purposes. First, for testing the count of four yards of yarn; second, the count of forty yards of yarn; and third, the weight of one hundred square yards of cloth. Two templates are supplied with this balance, one of which is one-tenth of a yard square, and the other onehalf a yard long. The first is intended when only small pieces. of cloth are available, and for cutting cloth to get the weight, forty threads are cut to length of template when the count is shown on 4-yard scale, whilst the second is used when larger pieces are available, in which case eighty lengths of one-half yard each are cut and put on the hook, when the count is indicated on the 40-yard scale.

Whilst both these types of apparatus are useful, a wellconstructed chemical balance is, after all, the best and the most

reliable. It can be used for all kinds of fine weighing, and for both long and short lengths of yarn. All that is required is a small series of templates representing different weight units. For example, suppose one grain is taken as the weight unit, then the length unit will be (840 × 36) ÷ 7000 =4.32", and if a template is cut exactly to this size, the number of threads of this length which balance the grain weight will equal the count. Other weights and other sizes of templates will readily suggest themselves.

Method of Measuring Short Lengths of Yarn from Cloth.— Seeing that the material to be weighed is at most only a small quantity, it is of great importance that it should be carefully measured. It is well known that the bending of the threads. of warp and weft due to weaving varies in different fabrics, and it is also very clear that in order to get the right length of yarn for weighing, the bent yarn must be straightened out. From some experience in this work it has been found that the threads. can be cut to the proper length better out of the fabric than in the texture. The "buckle" should be removed by passing the threads between the finger and thumb, but without stretching.

This point has been specially mentioned because the makers of some of the small balances say that the cloth should be cut to the template and then the prescribed number taken out and weighed. Obviously this is not correct, as no account is taken of the threads being longer out of, than in the cloth.

Testing for Strength and Stretch to Breaking-Having dwelt at some length on matters relating to the count of yarn, attention is now directed to the various methods by which the tensile strength of a thread is determined.

The strength and elasticity of a thread are two of its most important features, as during the process of weaving these properties are often severely taxed, and if they fall below the necessary standard, much trouble is caused and the quality and product of the loom is often materially reduced.

There are various means employed in testing yarn for strength and elasticity. For example, men of much experience can