31 



that the cable is frequently eaten off at the top of the pipe by the electric current 

 passing from the cable to the pipe, thereby putting the system out of order. Occa- 

 sionally, however, their inspector overlooks a gas-pipe grounding and accepts a 

 risk on a building so rodded. Since the 1912 report of the Protected Mutual they 

 iave lost their first rodded building by lightning, and curiously enough it was one 

 with the gas-pipe groundings. On examination it was found that the cable was 

 nearly disconnected, only two or three of the fine copper wires remaining intact. 

 Whether the injury was done in closing the gas-pipe too tightly on the cable or 

 by the current eating the wires off was not determined. It would seem that as 

 yet this method of making groundings is open to some question and it would 

 appear wiser to locate the ground-rods where there is no manure, then there is no 

 need for the gas-pipe. 



A simple way to make a ground-connection is illustrated in figures 37, 38 and 

 39. In figure 37 note the drill 10 feet long. It is a piece of half-inch iron rod, 

 with point swedged to make it about five-eights or three-quarters of an inch 



Fig. 38.— Drill down 10 feet. 



Fig. 39. — ^Ground rod down 10 feet. 



in diameter. An eye for a hand-hold is formed on the upper end of drill. A hole 

 large enough for a pail of water is dug in the ground. The hole is filled with 

 water, and the drill is placed in the centre of the hole and gradually worked down- 

 ward. More water may be poured in if required. When the drill is down full 

 depth (see Figure 38) it is withdrawn and the cable carefully slipped down into, 

 the hole (see Figure 39). The grounding shown in these figures was made while 

 rodding a silo, and took probably twenty minutes after the hole was dug. 



It is sometimes advised to dig a large, deep hole and bury a ground-plate to 

 which the ground-rods are attached. This does not appear to be necessary in 

 general farm conditions. In Michigan where the efficiency is 09.9 per cent, the 

 ground-connections are made in a manner similar to that Just described and shown 

 in the photographs. If efficiency without ground-plates is 99.9 per cent, there seems 

 to be no urgent need for them. Where the rock is near the surface and the soil 

 becomes dry to the bottom during the summer a ground-plate would be of some 

 value, especially if imbedded in coke, for coke, besides being a conductor, holds 

 moisture well, but better results would probably be secured by sinking the ground- 

 rod into a well or a good crevice in the rock, or a stream of water if one should be 

 near. Sometimes in shallow soil a large flat stone is laid over the ground-rod " to 



