13-foot beam will be supported at those points which will cause a minimum bending moment. An interior view of the moist room for storing beams, cylinders, cubes, and bond test pieces is shown in Pl. XIII. A 13-foot beam, supported by the trolleys, is shown at the right. The cylinders are piled three high with three cubes at the top. The beams are piled six high and the piles are prevented from toppling over by the wooden racks shown at the left. Each beam rests on two small timbers placed at the points that give the smallest bending moment. The register numbers of the test pieces may be seen. Either steam or water may be sprayed into the moist room from the ordinary type of fire sprinklers near the ceiling. At the present time water is sprinkled in fine spray over the test pieces at regular eight-hour intervals. The waterproofing of this room for the purpose of making it steam tight was done without cost to the laboratories under the supervision of a representative of the Hydrex Felt and Engineering Company, through the courtesy of the manager, Mr. Edward De Knight. Both the molding and storage rooms have self-registering thermometers giving a seven days' continuous record, and the moist room is equipped with a hygroscope for determining the percentage of moisture in the air. Each day the beams that are to be tested are raised by means of the chain hoists, moved out into the molding room, and placed upon two trestles or upon the floor, whence they are taken to the testing room. BEAM TESTING. APPARATUS. The testing room is equipped with one 3-screw 100,000-pound testing machine, one 4-screw 200,000-pound testing machine, and one 3-screw 200,000-pound testing machine, as shown in Pl. XIV. A 600,000-pound machine for testing columns and large beams is being installed in this room. The machine ordinarily used for-testing beams is shown in the center. The beams are raised and carried to the testing machine by means of a steel carriage shown in Pl. XVI, B. The carriage is pushed along the floor end first over the extension arm of the testing machine and one end of the beam is lowered to a dolly and rolled into place on the testing table. The beams are tested on a 12-foot span. The load is applied to the beams at the third points; that is, 4 feet from the ends and 4 feet apart. The apparatus through which the load is transmitted to the beams at two points consists of a steel box girder 4 feet long by 6 inches deep, built of two 6-inch channels and a -inch by 8-inch cover plate, top and bottom. 15767-Bull. 329-08- -5 The curved inner surface of the castings are somewhat more than semicylindrical, so that the rollers, while permitted to bear on the beam and turn freely with any motion of its upper surface, will remain in place when the steel girder is raised above the beam. The load is transmitted from the machine head to the steel girder through a spherical bearing block (a, Pl. XV, B) from the girder to the beam through two cylindrical rollers (b), which bear upon steel blocks (c) bedded with plaster of Paris upon the top of the beam. The upper surfaces of these blocks are cylindrical surfaces whose axes are parallel to the length of the beam. PLAIN-CONCRETE BEAMS. The testing machine is balanced with the poise at zero before the beam is placed in the machine; when the beam has been placed on the supports, its weight is read on the scale beam and is recorded. After the weight of the beam has been found, two sets of deformeters are placed on the beam, one having a gage length of 291 inches, with contact points 10 inches apart vertically, and the other a gage length of 24 inches, with contact points 5 inches apart vertically. A plain-concrete beam in place in the testing machine, with its deformeters adjusted ready for testing, is shown in Pl. XV, A. The supports for the beams are shown at d, d. These blocks have cylindrical tops and have a slight outward motion, so that they may move outward to accommodate the lengthening at the bottom of the beam. In several preliminary tests it was found that plain beams four weeks old failed under a total load but a few hundred pounds greater than their own weight, and therefore, if tested by the usual method of supporting them at their ends and applying two equal loads at the third points, but a few readings from which to interpret the test were obtained. On this account it was thought advisable to get a reading of the micrometers when the total deformations throughout the gage lengths are zero. A table was prepared showing the necessary amounts by which the reactions must be decreased for beams of different weights in order to make the total deformations throughout the gage lengths equal to zero. The uplift to decrease the reactions is applied at the third points of the beam by means of the stirrups marked a in Pl. XV, A. The head of the machine is raised so that the stirrups pull up on the beam until the support afforded the beam by the stirrups is sufficient to reduce the reactions at the ends of the 12-foot span to the value taken from the table. The micrometers are then read and the head of the machine is lowered until the beam rests freely on the 12-foot span, when the stirrups are removed and the micrometers on the deformeters are again read. The load is then applied through the box girder and readings of the micrometers |