1845.] 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



347 



trict of the sirepis of Loiulon. It may I).- oljocted, lliat Uie gas e?cBpins 

 in the soil becomes decomposed, and that tlurefore no smell is observed ; 

 but this h>potliesis is inadmissible, as in tliat case the hydrogen would he 

 formed inio water, and the ciiibon wouNl be deposited in the soil ; therefore 

 the niiantity, by weisht, annually left in the sod, from the supposed escape 

 of 30 per cent would be upwards of 3,000 tons. This is from the works 

 of one company only ; and if we consider the enormous additional quantity 

 that would be" produced from the other ten metropolitan companies, we 

 cnnnot avoid pronouncing it to be utterly impossible. 

 }Vuler Meters. 



The ordinary wet meter is necessarily placed in the basement, in order 

 to receive ilie water which, risiuR as vapour out of the meter, is condensed 

 in the (iitings. The meter is therefore almost always in a dark cellar or an 

 obicure corner, and in some instances fraudulent methods li^ve actually 

 been in existence without detection during the inspection by the company's 

 oflicers. 



It is well known that the water meter invented by Clegs, and improved 

 by Crosslev, Is subslaniially the same as that now generally in use. The 

 action of the \va!er gas-meter is generally understood. The gas is intro- 

 duced at a central opening of the measuring drum, which is sealed with 

 water; the pressure is exerted upon the surface of llie water, and the 

 diagonal divisions of the revolving wheel, which is the measuring chamber 

 of the water meter. The revolutions of the wheel, caused by the passing 

 of the gas, are shown by the index. Now the elftct of this meter can be 

 easily evaded by various methods. Tlie correctness of its measurement 

 depenils entirely upon the water in the meter being kept at the proper 

 height ; thus, for instance, if the case of the meter be tilled forward to an 

 angle of from .5° to 13° (according to its construction), and a proportion of 

 the water drawn oil", so as to nnseLiI the outlet of the measuring chamber, 

 the gas passes through it without affecting the index, and without being 

 registered at all. During the winter, the water contained in the wet meter 

 is sometimes (ruzm ; and then it is necessary, in order to enable the gas 

 lo pass through the fittings, that the meter sliould be removed. The bad 

 example of avoiding the measurement of the gas, until a fresh meter can 

 be substituted, is thus set by the company itself. The numlier of new 

 meters required on such occasions is so great, that a considerable period 

 often elapses before the frozen meter can be replaced. 



Dry Gas Meters. 

 . To meet these admitted evils, various kinds of dry meters have been at 

 different times invented ; but hitherto they have been open to such serious 

 objections as to prevent the general adoption of them in practice. The lirst 

 machine of any importance was that of the Dry Meter Company. The 

 material of wliich liie measuring chamber of that meter was formed was 

 leather, which has been found liable to several objections. If the meter is 

 .nred only occasionally, at intervals, the action of the gas upon the leather 

 produces coneiderable contraction and causes a registration of an increased 

 proportion against the consumer. On the other hand, when the meter is in 

 CODS ant use the leather is expanded, whereby more gas passes into con- 

 (uinptiun than is marked by the index. This of course operated unfairly 

 against the manufacturer of gas. These imperfections have therefore ren- 

 dered this meter obsolete. 



The only other dry meter of sulTicient importance to be mentioned, is 

 that constructed by Defiies. In this instrument, each of the tliree mea- 

 suring chambers of which it is composed is separated from the others by a 

 flexible partition formed of leather, partially defended from the chemical 

 action of the gas by metal plaies. This flexible partition is expanded by 

 the pressure of the gas, and in the alternate expansion and contraction it 

 forms a cone. Now as a cone is one-third part of a cube, one-third part 

 .of the surface only is available to the pressure of the gas. Independently 

 of the loss of power thus occasioned, a further loss arises from the sides of 

 the flexible partiliun being fixed, and the centre only being moveable, and 

 registering by its motion the gas consumed. I'urther, if where the leather 

 is attached to thf sides of the case there be a play of -^'h of an inch be- 

 tween the plates and the line of attachment of the leather, in the backward 

 and f(H-\v»rd motion, it passes through Jtli of an inch; thus, when by 

 Mse an I exposure to the atmosperic air, the leather has contracted only j'^nd 

 »f an inch, ilien in its motion it passes through this ^^ih part of an in'h, the 

 cfiect of whii h is, that the measuring chamber is uiniinished by this /jih 

 of an inch, over tlie uliole diaphragm. Now since it meafuies merely a 

 conical space, it must be evident that this loss of ^'h of an inch over its 

 surface very much lessens the measuring chamber. In cases where the 

 meter has been some time at work, it is stated tliat it has thus registered, 

 against the consumer, as much as from 9 to 11 per cent The reverse of 

 this occurs «hen the meter has been some time in use, without any admis- 

 sion of air, and then the manufacturer incurs a loss. Each fiexihie parti- 

 tion consists of four triangular divisions, each of which is protected by a 

 metal plate, and between each division and all round the outer rim of the 

 partition, where it is atlarlied to tlie case of tlie meter, there is necessarily 

 an uncovered surface of the leather, lo allow the partition lo move freely 

 backwards and forwards; this leather is consequently liable to be acted 

 ■ pon by the gas. This circumstance must be an objection to every meter 

 iu which tlie tiexible material forms part of the measuring diaphragm. 



C'roll and Uichards' meter avoids these objections, wliich have hitherto 

 prevented the general use of dry nieteis. The machine will be more readily 

 understood by imagining a steam engine measuring its steam, as it really 

 docs, in all cases. The steam enters the cylinder, from the boiler, on to 



the top of the piston, forcing it Ihroagh a certain space ; the supply is cut 

 off, and the action is reversed, the bulk of steam occupying the space 

 through which the piston moves, is thus measureil ; for presuming the pis- 

 ton to beof B given area, and the distance through which it moves at every 

 stroke, to be constant, it can readily be conceived how the actual quantity 

 of steam employed could be indicated or calculated. The meter in ques- 

 tion bears a strong resemblance to a double engine. It consists of a cylin- 

 der divided by a plate in the centre into two separate cylindrical depart- 

 ments, which are closed at the opposite ends by metal di*c> ; these metal 

 discs serve the purpose of pistons, ami they are kept in their places by a 

 kind of universal joint, adapted to each ; the space through which the discs 

 move, and consequently the means of measurement, is governeil by metal 

 arms and rods, which space, when once adjusted, cannot vary. To avoid 

 the friction attending a piston working in a cylinder, a band of leather is 

 attached, which acts as a hinge, and folds with the motion of the disc ; this 

 band is not instrumental, to any extent, in the measuring, S'l that if it were 

 to contract or expand, the registering of the meter would not be affected, 

 inasmuch as it would only decrease or increase the capacity of the hinge, 

 the disc still beMig at liberty lo move through the required space; the 

 leather is also distributed in such a manner, being curved, and bending 

 only in one direction, that it prevents any wrinkles or creases from forming, 

 and renders it, therefore, much more durable. The arrangements of the 

 valves and arms are somewhat dillerent to that of a steam engine, altliougU 

 similar in principle. 



Meters leith Three Chambers. 



Mr. Defries said, there were some points of Mr. CroU's paper, with 

 which he could not accord, although he perfectly agreed in the statement 

 of the general deficiencies of the wet meters, and the facilities they afl'orded 

 for fraud. Being aware of the objections against all meters with only two 

 companmenls, as being liable to canse oscillations of the lights, he had, in 

 the construction of his meter, adopted three chambers, in order that its 

 action, like that of a three throw pump, might be continuous. In practice 

 this was found to he the case, and at the Thames Tunnel, the House of 

 Lords, and in many private establishments, where very large meters made 

 by him were used, and their measurements were tested daily, none of the 

 contraction or expansion of the leather hinges, or any alteration in the size 

 of the chambers, hail occurred. The leather used was prepared expressly 

 for the purpose, and he believed, that the theoretical objections, both to the 

 use of leather hinges, and to the form of the chambers, were not well- 

 founded ; at all events, no ill effects had been found to result from either 

 ill a period of seven years, during which tune upwards of ten thousand 

 meters had been made. He contended it must be evident, from the form 

 and the arrangement of the chambers of his meter, that it would work cor- 

 rectly under a low pressure ; indeed at less, he thought, than if the disc 

 moved bodily forward ; but that point could only be ascertained by actual 

 comparative experiments. 



Mr. J. Farf.y said, the meter wilh three chambers appeared at first sight 

 most likely to keep up an equal How of gas, and he did not think the di- 

 mensions of the chambers would be subject to alter so as materially to 

 all'ect the capacity. The meters with two chambers were somewhat on the 

 principle of the diaphragm pump, patented by Benjamin Manin nearly a 

 century ago, but which did not succeed in water. The same kind of pump 

 was more extensively used for the " Carcel" lamps, and in oil it was very 

 durable ; but I'e was of opinion, that when working in dry gas, the leather 

 diaphragm would crack, unless it was prepared in a peculiar manner tu 

 resist this tendency. 



Leakage by the Porosity of the Pipes. 



Captain W. S. Moobsom said, that a point of much interest connected 

 wilh the subject was the leakage of the gas through ihe metal pipes. He 

 understood the same efl'ect had been observed on the atmospheric railway, 

 where the leakage through that portion of the main which was composed 

 of close pipes was as great in proportion as in that part with the continuous 

 valve. 



Mr. J. T. CoorF.R said, there could not he any doubt of the porosity of 

 the ordinary metal pi pes, so that the process of "endosmose and exosmose" 

 occurred to a great extent, particularly wilh soft iron pipes. This subject 

 had been discussed at length last Session. If harder iron, of a greater 

 density, were used, there would be less porosity. He was astonished to 

 hear the statement relative to the leakage of the air through the nielal of 

 the pipes in the main of the atmospheric railway; but with respect to the 

 pipes in the streets, it was not surprising that carburetled hydiog.n "as, 

 which was very volatile, shoiiM traverse the pores of soft iron. The method 

 of exhibiting the " endosmose and exosraiise" process, by means of jars 

 covered with a sheet of India rubber and filled with gases of various den- 

 sity, was well known to chemical students, 



Mr. Farey believed, that if the gas pipes were made from belter mate- 

 rials they would not be so poroiisas to be in any way prejudicial ; but now 

 for the sake i>f a low prici-, they were cast from any sort of coarse bad iron, 

 and they could scarcely be expected lo be sound. Much also depended 

 upon the method of casting them. If they were cast in moulds placed 

 vertically, with fountain jets, tlie thickness of the pipes would be more 

 uniform, and the metal would be less liable to be spongy. 



Mr. LowK said one source of loss arose from the destruction of the 

 wrought iron service pipes, by oxydation under ground. It hud been as- 

 serted, that in opening the ground in the streets, it was not uncommon to 

 find, that the whole of the metal of the service pipe, was gone, and that Ike 



