Some figures will show that the lamp is safe with the new gauzes .sx A lamp was filled with the most explosive mixture of firedamp and air and the mixture was fired by an electric spark .sx When the gauze had twenty wires to the inch the explosion did not pass through it .sx Nor did the explosion pass when the gauze had eighteen wires , and even a gauze with only sixteen wires was able to stop the explosion in every trial made .sx There is thus a good margin of safety in the new twenty-mesh gauzes .sx Other ways of improving the light of the flame lamp are described in Paper No .sx 40 of the Safety in Mines Research Board .sx Firedamp Explosions are imprisoned in the Flame Safety Lamp .sx It is clear , from what has been said , that the safety lamp is safe because , if its flame sets fire to firedamp inside the lamp , the flame of the explosion cannot get out of the lamp .sx The gauzes do not prevent an explosion inside the lamp , but they keep the explosion harmlessly imprisoned .sx We may now pass on to consider how electrical machinery has been made safe by the use of the same principle .sx FLAME-PROOF ELECTRICAL MACHINES .sx Electric Sparks and Firedamp .sx The smallest flame of any oil lamp is able to set fire to an explosive mixture of firedamp and air .sx Flames are the most certain means of igniting firedamp .sx Some electric sparks are also possible means of ignition of firedamp , and the danger from them has also been guarded against .sx Although all flames can ignite firedamp it is not possible for all electric sparks to do so .sx Some sparks are too feeble .sx For instance , when an ordinary comb is drawn through the hair on a dry day a crackling sound may sometimes be heard , because the comb has become electrified ( see Figs .sx 9 and 10) .sx Tiny sparks may be drawn from the comb if it is put close to the back of the hand ( see Fig. 11 ) , but these sparks are too weak to set fire to gas .sx They are so weak that they can only be seen in the dark .sx Stronger electric sparks may occur at bare signalling wires in the pit , but these also cannot ignite firedamp if their strength has been properly regulated .sx The sparks from an ordinary electric lighting or power supply are usually much stronger , and it may be taken for granted that all sparks from such currents of electricity will fire gas , even when they are not very bright .sx Hence special means are used to avoid the danger of starting explosions by these sparks .sx Gas-tight Casings .sx A sure way to prevent the sparks of an electric machine from igniting gas in the pit would be to put the machine in a gas-tight case , for the machines have not to be fed with air as safety lamps are fed .sx Unfortunately , it is almost impossible to make the casing of a machine gas-tight .sx The case has to be made so that it can be opened for inspection and repairs .sx This means that there must be joints between covers and casings and gas will find its way sooner or later through the best of joints , or through the bearings of spindles used for operating the machine from outside the case ( see Figs .sx 12 and 13) .sx In No .sx 2 of this series of books , called " Gas and Flame , " an experiment is described which proves that gas works its way at a great rate through the walls of a pot of unglazed earthenware .sx If it can thus pass through earthenware , it will find plenty of room to enter casings through joints and bearings .sx Explosion-proof Casings .sx As gas may sooner or later be in the air round a machine used at the face and , if so , will surely find its way inside the casing , and as sparking may occur in the casing at the moment when gas is inside it , the casing must be made in such a way that the flame of an explosion of gas cannot get out of it and start an explosion in the mine , as it would if the casing burst .sx There are two ways to do this .sx One is to make the casing as tight as possible and so strong that it will not break under the force of the strongest firedamp explosion inside .sx Another way is to use some device to weaken the force of the explosion and at the same time to imprison the flame , based on the principle of the gauze in flame safety lamps .sx Some results of research on gas explosions in casings and similar vessels will now be described .sx First .sx Research on the way in which flame spreads from an electric spark in an explosive mixture .sx Second .sx Measurement of the force of explosions in closed casings .sx It is necessary to know exactly what sort of conditions will produce the most violent explosion that can occur within the casing .sx Third .sx Methods of weakening the force of an explosion in a casing without allowing the flame to pass out and ignite explosive mixtures outside the casing .sx How Flame spreads from an Electric Spark .sx Before making any experiment , we might predict that , when an explosive mixture of gas and air is fired at some point by an electric spark , the flame of the explosion would spread from the spark in all directions at the same speed .sx The flame would grow as a soap-bubble grows when air is blown into it .sx Let us see how this prediction is borne out by experiments .sx Many snap-shot photographs of flames have been taken .sx One of them , Fig. 14 , shows a flame kindled by a spark in the centre of a glass sphere a hollow ball of glass when it has spread a very short distance from the spark .sx The next picture , Fig. 15 , is another snap-shot taken a moment later when the flame had grown bigger .sx Two more pictures , Figs .sx 16 and 17 , show the flame when it had grown still more .sx In all these pictures the flame appears as a circle , proving that it continued to move in all directions at the same speed .sx The flame is therefore shaped like a sphere .sx The next experiment was carried out in the same manner , but with one difference .sx Instead of taking separate snap-shots on four photographic plates , .sx one for each position of the flame , only one plate was used , and it was left in .sx the camera while four snapshots were taken rapidly one after the other .sx The result is that all four pictures are brought together on one plate ( Fig. 18) .sx The .sx flame grew like a soap-bubble until it had burnt through all the gas and reached .sx the walls of the glass sphere .sx Then it faded out , because all the gas was burnt .sx One conclusion from these experiments is that , when a gas explosion is .sx started at the centre of a sphere , the flame reaches the walls and dies out at the .sx same instant at every point , because it has moved at the same speed in all .sx directions .sx The next conclusion is important .sx Since all the hot gases produced by the flame are on the inside of it , the walls of the vessel cannot begin .sx to cool the gases until the flame has passed through the whole explosive mixture .sx Explosions in Vessels that are not Spheres .sx The conclusion that has just been drawn was reached by experiments in spheres .sx The results are different when explosions are made in vessels that are not spheres .sx For example , the next four snap-shot photographs ( Figs .sx 19 to 22 ) were taken of a flame while it was spreading in a glass tube .sx The flame started ( Fig. 19 ) with nearly the same shape as in the experiments in a sphere , hut it soon lengthened and divided into two parts ( Figs .sx 20 , 21 and 22 ) which moved separately to the ends of the tube .sx While this was happening , the hot gas near the middle of the tube was touching the walls of the tube and was therefore cooling .sx Fig. 23 shows the photographs of Figs .sx 19 to 22 all together on one plate .sx The force of an explosion is due to the heat of the flame , and increases as the flame spreads through the explosive mixture .sx Hence the greatest force must be expected when no heat is lost before all the gas is burnt .sx Therefore the explosion in the sphere would produce more force than the explosion in the tube .sx Two more photographs of flames will illustrate this conclusion .sx Fig. 24 is a number of photographs of a flame in a square box , all taken on the same plate , like the pictures in Fig. 18 .sx The flame was started by a spark in the centre of the box , but before all the gas in the corners was burnt the gas next to the sides of the box was cooling .sx Fig. 25 is another set of photographs of a flame in a sphere , but the flame was started at the bottom , and the gases were cooled at the sides before the flame died out at the top of the sphere .sx These experiments show that the greatest force of a gas explosion is to be expected when the gas is fired at the centre of a sphere .sx If the force of such an explosion is measured , we know the greatest force that a casing of any shape will have to withstand .sx How the Force of Explosions is measured .sx Fig. 26 is a photograph of the apparatus used to measure the force of explosions .sx Near the top of the picture is a thick bronze sphere , called for short a " bomb .sx " It is made in two halves , clamped together .sx The inside is a perfect sphere , and contains the mixture of firedamp and air for the experiments .sx To the right of the bomb is the electrical apparatus for making a spark inside the bomb , for igniting the gas .sx On the table below the bomb the apparatus for measuring the force of the explosion .sx It is connected with the bomb by a pipe which can just be seen in the picture .sx The results of the explosion experiments in the bomb are best understood when shown as " graphs .sx " These must first be explained .sx A record of the ups-and-downs of the football teams in a league can be made by cutting out the league table from the paper every Saturday night .sx The past progress of any one team can be seen by looking through these cuttings .sx That is slow work , however , and may be made simpler by a " graph .sx " A sheet of paper is ruled in lines across and up and down , like Fig. 27 , and the horizontal lines are numbered downward from 1 to 22 .sx The upright lines stand each for one week , starting with the left-hand side for the beginning of the football season .sx A " graph " has been made in Fig. 27 of the position of the Sheffield United team in the Football League all through the season 1928-1929 .sx On Saturday night , September 3rd , 1928 , they stood seventeenth in the league .sx The second Saturday they were thirteenth , and so on .sx By October 8th they had fallen next to the bottom place and remained very low until nearly Christmas .sx After that they began a steady rise which lifted them out of the last two and thus kept .sx them in the first division of the league .sx The whole story is told by the " graph , " except that the reason for the improvement at Christmas is not told .sx The progress of an explosion in the bomb is shown in the same way .sx The explosion is finished in less than a second , however , while the English football season lasts eight months .sx Whilst a record of the position of the football team is made at the end of each week , the record of the progress of the explosion has to be marked at the end of each one-hundredth part of a second .sx Some readers may think it nonsense to talk of one-hundredth of a second .sx Look at Fig. 28 .sx The man drawing the chalk line on the blackboard took just one second to draw it .sx