Fine grinding will decrease the weight per cubic foot, the fine cement not packing as closely as the coarser product. In "Cement for Users," by Mr. Henry Faija, the following results are given, showing the relation between fineness, weight, and specific gravity: (a), cement as delivered; (b), siftings that passed through sieve with 2,500 holes per sq. in.; (c), coarse, retained on above sieve; (d), cement all ground to pass above sieve; (e), coarse particles reground to pass above sieve.
Weight per Bushel.
Since the coarse particles of cement are practically inert, there is every reason to believe that finer grinding will increase the activity of a sample, since it will render some inert particles active. For the reason mentioned in § 81, however, it is difficult to show this difference in time of setting by actual tests.
Tests reported by Mr. David B. Butler1 showed that several Portland cements which took an initial set in 20 to 30 minutes and hard set in 45 to 120 minutes would, when reground to pass a sieve having 180 meshes per linear inch, begin to set in from 1 to 7 minutes and set hard in 5 to 15 minutes. These may be considered extreme results; the rise in temperature of these cements during setting was so great as to indicate they were not normal cements, and variations in consistency of the pastes may have influenced the time of setting.
Since the best burned clinker of Portland cement is the hardest, it follows that the unground particles would, if ground fine enough to become active, form the best portion of the cement. This is not, a priori, true of natural cements, because burning renders some varieties of cement rock softer at first, but when the burning is carried beyond a certain point they become harder again. The coarse particles in a natural cement may thus be either from underburned or overburned rock; hence it is possible that in some cases it might be better to leave the hardest particles in an unground state. Thus, while it has been generally accepted that fine grinding improves Portland in a twofold degree, — by bringing into action the best burned clinker, as well as by rendering a given weight of cement capable of coating a larger number of sand grains, — a similar conclusion concerning natural cement is not well established.
1 Proceedings Inst. C. E., 1898.
Tensile Strength, Pounds per Square Inch.
1 Part Standard Sand to 1 Cement.
2 Parts Standard Sand to 1 Cement.
3 Parts Standard Sand to 1 Cement.
4 Parts Standard Sand to 1 Cement.
6 1/2 mo.
2 3 4
268 283 278 392
538 473 538 592
224 230 307 368
381 350 433 538
207 245 292 271 21
354 433 469 344
291 426 406
202 302 305 274
48 65 61 110
156 212 240 205
49 78 65 90
Fineness of Cement, Per Cent. Passing Sieve Number.
1. Cement as received passed through No. 20
2. Cement as received passed through No. 100
3. Reground in mortar, not sifted ....
76.5 100.0 95.8
72.4 94.6 91.5
Cement; Natural, Brand Jn. No. 1. Passing No. 20 sieve.
" 2. Passing No. 100 sieve.
" 3. Reground before sifting.
" 4. Particles retained on No. 50 sieve, reground to pass No. 100 sieve, " 5. Particles retained on No. 50 sieve, reground to pass No. 50 sieve, but retained on No. 100 sieve. All briquets made by one molder and immersed in one tank. In general, each result is mean of five specimens.
92. Some tests bearing upon the value of fine grinding have already been given in Table 15. Samples 3 and 8 were reground with mortar and pestle before being sifted. If we compare the results given by sample 3 with those obtained with samples 1 and 2, not reground, it appears that the regrinding diminishes the strength in neat mortars but increases it in mortars containing three parts sand to one of cement. Regrinding appears to be no better, however, than sifting. Comparing sample 8 with samples 6 and 7, it is seen that regrinding Portland cement does not diminish the strength in neat mortars to the same extent as sifting does, and in sand mortars regrinding generally results in a greater increase in strength than sifting.
93. The results in Table 17 were obtained with another sample of natural cement and are of greater practical value as indicating the importance of fine grinding, since in these tests a sample is included obtained by regrinding the original cement without previous sifting. The conclusions concerning the cement retained on No. 50 sieve reground to pass No. 100, and the coarse particles alone retained between sieves 50 and 100, are practically the same as those drawn from Table 13.
As to the other three samples, the No. 20 sieve removed only a very few coarse particles, and that passing this sieve may be considered to represent the cement as received The No. 100 sieve removed about 24 per cent, by weight from the original cement, and the cement that was reground contained but about 4 per cent, of particles which would not have passed the No. 100 sieve. The third sample, reground cement, may be compared with the first to indicate the improvement obtained by finer grinding, and it may be compared with the second to determine the difference between removing the coarse particles by sifting and reducing them by finer grinding. In considering these results it will be best to neglect the two-year tests, since all of the samples failed at this age. A comparison of the results obtained with these three samples indicates that while the advantage of finer grinding is not apparent in neat tests, in sand mortars the value of finer grinding is more marked the larger proportion of sand used, so that with three or four parts sand, the strength with the fine samples is about 50 per cent, greater than with the cement as received. It also appears that the reground sample gains its strength more rapidly than the sifted sample, though at six months it seems to make little difference whether the coarse particles are removed by sifting or reduced by grinding.
The general conclusions to be drawn concerning fine grinding and sifting may be summarized as follows: According to the tests given, it appears that to remove the coarse particles from a sample of natural cement by sifting, or to reduce them by finer grinding, generally diminishes the strength obtained in tests of neat cement mortars. In one-to-one mortars, the strength of the finer samples is not much greater than when the coarse particles are present; but in mortars containing greater proportions of sand, the advantage obtained by eliminating the coarse particles is very marked in the case of natural cement, the strength given by the finer samples sometimes exceeding that of the original cement by more than 60 per cent. While the advantages of sifting and finer grinding are also important for Portland cements, there does not result such a large proportionate increase in strength.