PRACTICAL WORKINGS OF DLTLCTOR-MLTERS*
At the request of Secretary Diven., I will endeavor to give some of my experiences with Detector-meters. The secretary, in calling upon me for this paper doubtless had in mind the writer’s membership on the private fire-service committee of this association, and of the New England Waterworks association, both of which have given active and serious attention to the problem of the metering of private fire service pipes. My connection with the tests and experiments of large meters, particularly fire service meters, made by the Lowell waterworks department, of which I am superintendent, also probably influenced Mr. Diven in this matter. In J905 w’e equiped our Centralville pumping station with what is believed to be the most complete and at the same time permanent apparatus for the testing of large meters ever installed. It is true, the same general arrangements and apparatus used by Mr. E. V. French at Salem, Mass., in 1897; by Mr. F. C. Kimball, at Knoxville, Term., in 1903, and by Mr. F. H. Crandall, at Burlington, Vt., in 1904, were followed by the Lowell department; but all of these earlier installations were of a temporary character, whereas our apparatus was permanently erected in the engineroom of the pumping station. Each and every one of these testing plants was installed for the purpose of determining which, if any, of the commercial meters then in use were adapted for metering tire-services, and, also, to encourage the production of new devices by furnishing adequate means for their testing. The result of Mr. French’s investigations will be found in the Journal of the New England Waterworks association of December, 1897, under “Loss of pressure caused by meters in factory fire-supplies;’’ Mr. Kimball’s will be found in transactions of the same association of December, 1903, under “Some 6-in. meter tests and how they were made,” and Mr. Crandalt’s are imbodied in the “Report of the committee on private fire protection,” presented to the New England Waterworks association, February to, 1904. The Lowell investigations form the subject of a paper by Mr. W. F. Sullivan, formerly assistant engineer at Lowell, now engineer and superintendent of the Pennichuck waterworks, of Nashua, N. H., entitled “Tests of large meters and fire-service devices.” read on March 8, J905, before the New England Waterworks association. The writer, because of bis membership on the committees before mentioned, was interested in all of these experiments in a general way and in a more especial manner, jointly with the Lowell water board, for the reason that our city, being full of manufacturing establishments and provided with some 125 or more private fire-service connections, is and must necessarily be one of those which feel the drain upon its water supply on account of misuse of private fire-service privileges. We had for many years been suspicious that water was used improperly and in large quantities; but the matter was finally and very forcibly brought to our attention in 1902, when, as a result of shutting off fifteen fire-services, the daily consumption of water fell off 275,000 gals. This circumstance finally led the board to adopt a set of rules governing private fire-services. The first of these rules provides that “all water for fire-service pipes shall be metered, meter to be furnished and set by city at the owner’s expense.” This brought about the question as to the kind of meters for the city to furnish. Naturally, the meters then in the market received the first attention. To determine their respective merits the various meter manufacturers were each requested to forward a 6-in. meter for test. The meters received were at first all set on a 6-in. main in the basement of our repair shop and observed for accuracy of registration, especially on small flows. This means of testing, of course, proved incomplete, inasmuch as it did not allow any proper way of ascertaining the capacity or delivery, so we then put in the testing plant at our Centralville pumping station which we have already referred to. In this work we had the assistance of Mr. J. A. Tilden (member of this association and general manager of the Hersey Manufacturing company), who had previously been actively interested with Messrs. French, Kimball and Crandall in the several tests above mentioned. Mr. Tilden designed and furnished a good deal of the apparatus, together with a number of experimental devices and, finally, the one which is now known to the waterworks people throughout the country as the Hersey Detector-meter, and, as this is the only meter of its type now on the market, it is proper to say that this paper will he directed towards the practical workings of this particular device in the city of Lowell. If you consider a Detectormeter on a fire-service as a sealed automatic valve with a metered bypass round the valve, you will go a good way towards comprehending how the Detector-Meter works out in actual practice. You have simply to consider that, in lieu of the ordinary sealed valve, with seal to be broken, the valve to be opened and re-sealed, all by hand, you have, not only a device in which these three operations are performed automatically, but, also, one through which you are informed by means of a little time-meter attached to the Detector (automatic valve) how long the valve has been opened—a thing you do not have in the manually operated valve and bypass. I speak of the manually operated valve with metered bypass and make comparisons with it, because it is a device which is accepted and used even in carefully managed water departments, though, in my opinion, its use should be limited to very exceptional conditions. We certainly do not need the objection of the underwriters to convince us that a manually operated valve is a serious menace to the efficient service of our works, owing to the fact that the man who should operate the valve is very likely to be absent or to lose his head when an emergency arises. The question is asked: What do you do, if you find from the little time meter that water has been used through the Detector? I answer: “What would you do when using the manual device, if you found the valve-seal broken?” You would send for the owner of the premises and ask him why, would you not ? and the chances are that you would receive a more or less satisfactory answer. His statement would be believed or not according to circumstances, according to the man’s reputation, according to whether or not the story seemed plausible, and you would pass over the circumstance or request settlement or even go so far as to enforce settlement as the circumstances of the case seemed to warrant. For example, if the man represented that he had just put in a new fire-cistern capable of holding 100.000 gal., and he wished to fill it and fill it quickly by turning in several hydrant streams, and if the man’s reputation and the circumstances bore out the statement, you would charge him for 100,000 gal. of water, re-seal his valve, and everybody would be satisfied. If, on the other hand, some of your people had caught him misusing the water, and he was inclined to be rather independent about it, when taken to task, you would undertake to see that he settled and settled right, too, even going so far as to discontinue the service. Now all this is not materially different when you use Detector-Meters except that you have the advantage with the Detector-Meter that you know between the meter readings, not only if the Detector, which corresponds to the manual valve of the old system, has been opened; but you know about how long. This is a tremendous leverage on your water-taker, especially if you are so unfortunate as to be up against one whom you feel you have reason to suspect. When he tells you that the water was drawn only for a few minutes—for example, to keep a boiier normally fed from the manufacturing service from blowing up. but which for the time being might have been shut off, and you know from the reading of the little time-meter that the valve has been open hours, instead of minutes, and you explain to him how you know, you have him at a great disadvantage, you have him at your mercy, and the thing does not happen again. Now we come to the real point of value in actual experience. The honest but careless water-taker and the dishonest water taker are the two who get into trouble right away when a DetectorMeter is installed. The honest but careless man does not say anything to his manager or to others in immediate charge about not using water in excess of what can be taken care of by the bypass meter, and. when he is called to stand on the water superintendent’s carpet, he is surprised; but he proceeds to investigate and forthwith reports. In nine cases out of ten this is his last offence. The Detector is never again found ooen. except in cases of that emergency when it should open—that is, for fire. The dishonest water-taker undertakes righ awav to beat the machine: but the Detector catches him. and. when lie is called upon to explain, he is made to appear to such disadvantage that he, too. is settled for all time. It is now four years since Detector-Meters were first put into use by us, and our experience is that, after the first month or so the Detector is not opened, except in case of fire. The water-taker learns early and learns well that the Detector is not a thing to monkey with, that the opening of a Detector is a more serious thing than opening a sealed valve—and this, because of the time-element which is present in the Detector, which is not present in the manually sealed valve. The question as to how much water does or can pass the Detector, when open, is of small consideration to the water department. We say this, because we have proved in actual practice that it does not open; therefore, if it is a fact that, when once a fire-service settles down into normal condition, the Detector remains closed, there is, of course, no water passing it and nothing to estimate. It has often been contended that the system of sealing valves with bypass meters, inside fire hydrants and automatic sprinkler-valves furnishes adequate protection against the misuse of water from fire-services ; but such a system does not prevent or disclose the presence of concealed connections with the fire-service, and the Detector-Meter does. I regard this type of meter as the biggest watersaver and biggest money-earner for a water department in the line of a water meter yet produced. The domestic meter does a retail business; the Detector-Meter does a wholesale business. Especially is this true where the private fire-service pipe runs a long distance under ground, having branches for hydrants and other connections with the possibility of imperfect and leaky joints, split pipe and other defects, which you who have considerable experience know is not improbable nor uncommon. All of which the Detector-Meter, if used, will absolutely discover, while sealed valves would be of no avail. I think it will be generally admitted that the loss of water through supplies designed for fire protection exclusively is very large, often larger than from all other sources combined—and this, in spite of all rules adopted by waterworks officials or by underwriters prohibiting the use of water for other than fire purposes. No matter how stringent such rules may be, they are without avail as long as the water is free or without charge (to measure the water and render the bill for it is the only effective cure for this evil). I think that quite often this condition exists from the fact that the underwriters or managers of property to which an unmetered fire-service has been laid, overlook its existence after a while, so that leaks and misuse of the water go unreported. Experience and observation teach us, I think, that, even under the closest supervision and the best of intent, there is always some water used and quite often a very large quantity. I have in mind one instance where Hersey Detectors were installed on a mill property where the mill agent was honestly of the opinion that no use or waste of water was taking place, and, when the meter proved there was a considerable consumption, it took several days before it was discovered where the water was going, and today, after two years’ time, they have not succeeded in stopping the waste entirely. Many times, without the knowledge of the mill agent or superintendent, a subordinate needing water under pressure for some purpose, through ignorance of the wrong he is doing, will make a connection to a convenient water pipe; and sometimes this is done, not through ignorance, but because they think it is smart. If this goes undiscovered, they will repeat the offence, probably telling someone else employed in a like capacity in some other mill how easy it is. Fire hydrants in mill yards are also very handy to draw from for various purposes. Thus the abuse grows, until the only remedy—a meter— is applied. The Detector type of meter seems to offer a common ground on which water departments and underwriters can stand, particularly as applied to those services which are intended for fire protection only. Water depart ments and underwriters have been and are seriously at variance with each other with regard to combined services—that is, services which are used both for manufacturing and for fire-protection purposes. Water departments insist, and of right should insist that all such services shall be fully metered—that is, metered for large and small flows; and underwriters have contended, and rightfully, too, that the presence of the ordinary types of commercial meters on such services have been a serious menace to fire protection. A step in the right direction was taken when water departments and underwriters agreed that those services intended strictly for fire-purposes should be separated from manufacturing services. Water departments expected that the. system of sealing just referred to would furnish adequate protection; but in this they have been greatly deceived, as all waterworks superintendents who have had experience can testify. Underwriter* very decidedly object to the use of the same service for both fire and manufacturing purposes—and this, for the reason that the manufacturing use is liable to run up to an amount which would seriously cripple the service in case of fire. For example, supposing there was a 6-in. pipe leading to a mill capable of delivering 1,000 gal. per minute, and supposing the manufacturing use at times to run as high as 500 gal. per minute, a fire starts, and, of course, in the excitement of the moment, no one thinks of shutting off the manufacturing use. There are, therefore, but 500 gal. per minute available for extinguishing the fire, one-half of the supply being unavailable. Further than this, most “protected mills” have what is called a secondary supply consisting of a tank and a fire-pump of -—say, 750 to 1,000 gal. per minute capacity. The instructions are to start this pump immediately there is a fire—the pump and local cistern in that case taking the place of the regular supply. Now, if there is a draught on the pump of 500 gal. per minute for manufacturing use, it is obvious that one-half or more of the value of the pump is destroyed. Here is where the Detector-Meter comes in. The underwriters declare that too to 150 gal. per minute is the utmost draught that can be allowed on a 6-in. fire-service for purposes other than fire-extinguishing, and that, even in permitting this much, they are allowing the fire-service to be weakened 10 to 15 per cent. They insist that the draught shall not be so great as to open the Detector, and they are in a position to know by the reading of the little time-meter on the Detector whether or not this rule has been violated. The Detector-Meter gives them a chance to say to the insured that they can use just so much water and no more, and that, if this amount is exceeded, the water will be shut off. It will be seen by this that the moral effect of the use of the Detector-Meter is enormous. The consumer not only has to be very careful about using any water, but he has to be extremely careful that the draught does not reach to any material amount. Thus, in practice, it works out that the bypass meter measures all the water used, and that the Detector rarely, very rarely opens. Here then is the common ground between water departments and underwriters referred to. The water department places the Detector-Meter and impresses upon the consumer that only in case of fire shall the Detector be allowed to be opened, that, otherwise, they would be taking water in quantities not contemplated in a fire protection system. The underwriter insists that the Detector must not be allowed to be opened, because that would represent a flow of water which would seriously cripple the fire-service in case it were called on for the extinguishing of fires.
•Paper read at the twenty-eighth annual convention of the American Water Works association; Washington, D. C.. May. 1908.
Secretary Diven: I would say that Mr. Thomas has given us just exactly what we wanted. I hoped that there would be a lot of waterworks superintendents here who would add their personal experiences to what Mr. Thomas has already said. In the Detector-Meter we have something on which we and the insurance people can agree, about the only thing in heaven or on earth I think that we have agreed upon so far. What I particularly wanted to know was, what the Detector-Meter was doing from a practical waterworks standpoint? Upon that point I think Mr. Thomas has satisfied our minds quite clearly. I am sorry that there are not a whole lot of people here who have used it and who could add their experience. Mr. Kimball: I can say that a company that I am interested in has about twenty of them in service. I think that is proof enough that we consider them satisfactory, and that they are doing the work, without my going into details. They are being used very largely, I know, not only satisfactorily to waterworks people, but to the insurance people without any lfs or ands, or increase of rates, or anything else. In fact, in that town, when a manufacturer or storekeeper applies now to the insurance underwriters for a sprinkler-risk, as they call it, they are told that, as a part of their equipment, they will have to do certain things for the water company, which include a meter of this description. In reply to a member, Mr. Kimball said: We have two methods there, one in which the owner can pay for the meter, and another in which he pays a rental which is equivalent to about 12 per cent, of the cost of the meter, which we consider covers interest, depreciation, etc. They can take their choice, so far as we are concerned; it is as broad as it is long. Mr. Twiggs said, he had installed some seven of these meters and had found them a “perfect success.” “The first point that attracted my attention to the Detector meters (he added) was the possibility of putting a meter on a fire-line that would more accurately and correctly register small flows. We could not get that in the larger meters, whereas, with the Detector-Meter, I have some lines of 10-in., and I can so connect them with a bypass that 1 can register practically all the small flows. That was the first thing that attracted my attention to these meters. While we have had all kinds of controversies in regard to them, they have about won out, and with these seven meters that we installed we decreased our daily consumption 750,000 gal. an hour and increased our revenues about $9,000 a year.”