THE RESTORATION OF OLD DISTRIBUTING SYSTEMS
“The Restoration of Old Distributing Systems” is a subject which is coming more forcibly each succeeding year to the attention of superintendents, managers and engineers having to do with waterworks operation We have buried these systems with no thought of their to-morrow until it is evident from all the symptoms that the doctor is needed. Some of these symptoms are poor fire streams, dirty water after fires, high cost of pumping, and poor efficiency in general operation of the plant. Economy is the goal sought by operators in all kinds of industries to-day, but economy can only obtain through proper efficiency in waterworks operation to a large extent depends upon the condition of these old distributing systems. How can these distributin systems be restored? Let us see the effect of sediment. scale and corrosion in the mains, and then look for the remedy. Water conduits which have been down for ten, twenty, twenty-five years or longer have in most cases lost a considerable portion of their efficiency as carriers. This is due to the growth of organisms or the deposit of sediment or scale of one kind or another. It does not take a very large growth or deposit to seriously affect this carrying capacity and this deterioration runs into operating cost very rapidly. as I shall attempt to show you in the brief discussion under the following heads:
First. The carrying capacity of dirty water mains compared with that of the same mams after cleaning, and with that of new mains of the same diameter and laid under similar conditions.
Second, -Economy of operation as affected by reduction of friction in pipe.
Third.—The useful life of cast iron pipe as affected by cleaning.
Fourth. Increase in investment due to cleaning.
First.—The carrying capacity of dirty water mains compared with that of the same mains after cleaning, and with that of new mains of the same diameter and laid under similar conditions.
A great many tests have been made to compare the discharge and carrying capacity of pipes before and after cleaning. These tests have been made in various ways. In some the flow has been measured by a 6 -inch meter, in some with a pitometer, in some cases the pitometer and meter have both been used to check against possible errors, and in others the flow has been measured by nozzle discharge. The frictional loss has been measured by pressure gages, and sometimes by mercury column, and the difference in levels between the testing points taken into consideration so that the exact frictional loss could be ascertained. The tests below will show some results which will be of interest and of value as bearing on this subject:
Referring to the tables it is worth while to mention particularly some few of these tests and explain the conditions under which they were made. Test No. 4 made in Brooklyn is a remarkable example of the effect of cleaning old pipes. The flow through 950 feet of 12-inch pipe discharging through an open end on to the street was measured before cleaning and found to be 1,235 gallons per minute; while after cleaning it was 3,751 gallons per minute. The frictional loss for the increased flow was 69 feer, compared with 98 feet before cleaning. Here was an actual increase in discharge of 204 per cent. Taking into account the frictional loss, the actual carrying capacity was increased 205 per cent. Another test worth noting was No. 12, made at Belle Plaine, Ia., where the discharge before cleaning was 248 gallons per minute, but the frictional loss per 1,000 feet was only 9.7 feet after cleaning compared with 39.8 feet per 1,000 before cleaning. Again looking at test No. 21, made at Lockport, N. Y., where the same conditions existed and the same pressure was maintained at the pumping station before and after cleaning, the increase in discharge was 135 per cent. Perhaps the most interesting experiment of all though is recorded as No. 10. this test having been made on a 6-inch pipe at Pertn Amboy, N. J., Here the incrustation which consisted of iron corrosion, did not average ¼-inch in thickness, and in the top of the pipe was not more than ⅛-inch in thickness, and in some places even less, but the actual increase in discharge was 32 per cent., while the increase in carrying capacity was 51 per cent. This goes to show that it is not the amount of volume taken up by tlie incrustation but the roughness of the surface which causes the trouble in these distributing mains. I would now like to call your attention to tests No. 6, No. 7 and No. 8. where no measurements were taken before cleaning, but a very careful test was made after cleaning, because one of the requirements in regard to the cleaning of these lines was that they should have a carrying capacity after cleaning within 5 per cent, of that of new pipe, of the same diameter. The carrying capacity in each case was figured by the Darcy formula as used on page 77 of Westons friction of water in pipes. In test No. ti the carrying capacity came within 3 per cent, of that of new pipe. In test No. 7 the carrying capacity was 4 per cent, below that of new pipe, while in test No. 8 the carrying capacity came within 1 per cent, of that of new pipe of the same diameter. In regard to restoring the carrymg capacity of mains it can lie safely guaranteed that tlie capacity after cleaning will he within o per cent, of that of new pipe, as figured by the Darcy formula, providing that the pipes were originally coated, which is in all probability the case with water mains laid since 1870.
Second.—Economy of operation as affected by reduction of friction in water mains.
There are so many different examples and demonstrations of reduction in operating expenses due to cleaning that it is hard to tell which are the most important to present. Among these I will call your attention to tests Xo. 28, Xo. 29 and Xo. 30. At Atlanta, Georgia, where a 30-inch main was cleaned the conditions existing were not uncommon or unknown in other plants. They had a large pump which was delivering water through this line at the rate of 11,880,000 gallons per day against a total head of 110, and they were losing water in their reservoir daily. By cleaning this line they were able to deliver 17,820,000 gallons per day against the same pressure, an increase of nearly 6,000,000 gallons. Here it can be readily seen that not only was there great economy affected in operation, but had this line not been cleaned there would have been the necessity of laying new pipe, or of getting new pumps. Again in test No. 29, made at Carbondale, Ill., where they had reached a maximum pressure against which they could pump and in fact were straining their pumps, before cleaning ll,500 feet of 8-inch pipe they were delivering 115 gallons per minute, against a total head of 140 lbs., while after cleaning practically the same amount of water was pumped ahead of only 48 lbs. This was a private line owned by the Illinois Central Railroad, and while it was cleaned nearly a year ago, the water supply foreman in charge of this plant tells me that they are saving over 60 per cent, on their fuel and in addition to this are saving $300 a year in repairs to the pumps, and in addition to affecting this great saving, they are pumping more water than they did before cleaning. In Centralia. I11., there was a very marked increase in efficiency, which perhaps could not be credited to saving in operating expenses, unless by taking into account what it would cost to maintain the increased pressure without cleaning, but at the same time the results deserve to be mentioned here. Cuts herewith given show records of recording pressure gauge at the City Hall before and after cleaning. The same pressure was maintained at the pumping station before and after cleaning, but the average pressure maintained in the city was raised from approximately 30 lbs. to an average of 60 lbs. I would also like to call your attention to the very marked example of saving as affected in Mt. Vernon, I11. In this city they were troubled with poor fire streams and with muddy water and also with high cost of pumping. They pumped by electrical power and before cleaning their force line and part of their distributing system they maintained at their city office 20 lbs. pressure during the heavy draught hours, but to do this they were compelled to have 60 lbs. at the pumping station. After cleaning with the same pressure at the pumping station, they had 45 lbs. to 50 lbs. at the city office, and they found they could cut down their average pressure at the pumping station 20 lbs. and still maintain a better pressure at the city office than they had carried heretofore. Before cleaning they used 41.5 kilowatts to do their pumping, compared to 30 kilowatts after cleaning, a reduction of 27.7 per cent, in power required to do their pumping. From this and many other examples which could be given, I believe that the economy of operation due to cleaning is fully proved.
* Paper read at Rochester (N. Y.) convention of Ameri can Waterworks Association by Burt B, Hodgman. C.E., General Superintendent National Water Main Cleaning Company.
Third.—The useful life of cast iron pipe as affected by cleaning.
Heretofore the life of cast iron water mains has been given as from 80 to 100 years. Some authorities have claimed greater life and some shorter, but where pipes are affected by growth on the inside their useful life is very much less and in some cases is not over 20 years, while the loss in iron is not great enough in most instances to make the actual life of the shell of the pipe anywhere near as short as 100 years. Mr. Nicholas S. Hill, Jr., in his paper before this association in 1907 gave an example of the amount of loss of iron due to corrosion inside of water mains at Washington, D. C., and showed that there was 8½ per cent, loss of iron by weight in do years. He further stated that this is not perhaps the total loss, as some of the remaining iron might be soft and useless. Also examination made upon the pipes which had been in the ground in Brooklyn for 60 years showed that the shell of the pipe was in very good condition. Some pittings were found on the interior and the corrosion was very bad, some of the tubercules being 2-inch in diameter and more, but the examination which was made by expert foundrymen, engineers, and by the writer showed that the shell of the pipe was in excellent condit on. In talking of the amount of iron which has been destroyed 1 have not mentioned the exterior of the pipe, but corrosion on the outside of water mains when laid under ordinary conditions is almost negligible when figuring the amount of iron destroyed. It is true that certain soils affect pipe badly and of course when laid in peat hogs or tn cinders they are liable to be pitted on the outside, but these are extraordinary conditions anti the actual shell of the pipe under ordinary conditions. I believe, is good for a great ueai longer life than has heretofore been considered There are pipes in this country to-day which have been in use for more than 100 years but the useful life of cast iron for water mains is often very short unless they are cleaned at the proper intervals. The length of these intervals depends entirely upon the local conditions for each different pipe, but from the experience thus far. I believe that cleaning will increase the useful life of cast iron pipe several fold.
Fourth.—Increase in investment due to cleaning water mains.
As has been shown heretofore the pipe which is badly corroded does not have anywhere near the capacity of new pipe of the same diameter, and in figuring what the capacity is we find that a badly corroded large pipe often does not have the capacity of a much smaller clean pipe. Some examples of this are the tests made in Brooklyn. The 6-inch pipe before cleaning had a capacity equal to that of a 3 1/2-inch new pipe. This pipe was uncoated as it was laid in the days when present methods of coating were not known, but after cleaning it had the capacity of a 5 1/2-inch clean coated pipe. Also in the case of the 8-inch line tested in Brooklyn, the test showed that before cleaning it had the carrying capacity of a l-inch new pipe, while after cleaning it had the capacity of a 7½-inch new coated pipe. Test No 4 on a 12-inch main in Brooklyn showed that before cleaning it had a capacity of a 7-inch new coated pipe, while after cleaning it was equal to a 10-inch new coated pipe. These results indicate what can be done for uncoated pipe. In the case of those which are coated take the test at Carbondale. I11., No. 29, where the pipe before cleaning bad a carrying capacity equal to that of a new 5-inch pipe, but after cleaning it had the full carrying capacity of an 8-inch pipe, and m almost every case where coated pipe is cleaned the full carrying capacity is restored. These facts certainly indicate that where old distributing mains develop such large frictional losses that their value is equal only to that of the smaller new pipe which would carry the same amount of water The depreciation would then be the difference In value between the cost of the original pipe and the cost of the smaller pipe. As has been shown, the carrying capacity can be restored by cleaning and. therefore, the value of the mains increased. I have tried to show you by the introduction of these statistics that the greatest deterioration which water distribuiug systems suffer is from corrosion and deposit in the pipe. It is a clearly demonstrated fact that where this trouble exists it can be removed by cleaning, operating expenses reduced, the eflicieney increased, the useful life of the pipe prolonged and the investment increased. The restoration of old distributing systems can lie very effectually secured by cleaning.
MR. DUNHAM : I would like to ask whether where a pipe is cleaned rusting immediately goes on much more rapidly than at first, so that the more they are cleaned, the more they have to be cleaned. Some experience of the author in regard to that feature I think will be of interest.
MR. HODGMAN : I can not give any definite answer to that question. The same question has been put by the author to several superintendents and engineers for whom pipe has been cleaned and they have stated that the incrustation did not grow any faster after cleaning than it did betore cleaning. I tested pipe in Little Rock, Ark., three years after it had been cleaned, and found it was carrying 155 per cent, more water than it had carried before it was originally cleaned, that it, it had a carrying capacity of 71 per cent, ot that of new pipe three years after cleaning, while before it was originally cleaned, it had only 29 per cent, or 30 per cent, of the carrying capacity of a new pipe. I do not believe, however, from observations that 1 have made, that under ordinary conditions the rusting takes places any more rapidly after cleaning mains than it does in new pipe.
MR. CHESTER: Wouldn’t there be more tuberculation if you removed a large percentage of the coating in the cleaning?
MR. HODGMAN : Undoubtedly there would be more tuberculation if a large proportion of the coating were removed in the cleaning, but from observations which I have made after cleaning, the coating has not been removed, although in some instances the original imperfection of the coating allowed the water to act directly upon the pipe and when the tuberculation was removed Ike poor coating came with it. I quote herewith from letter received from Chas. R. Bettes, chief engineer, Queens County Water Company, Far Rockaway, N. Y., which is similar to many others which have heen received covering the same lines:
“Replying to your favor of April 20th, 1909, asking our experience as to the result of cleaning pipe. We have had occasion to cut out sections of the 12-inch, 8-inch and 6 inch pipe after it was cleaned and in no case have we found the coating destroyed or injured. For the first few days after cleaning there may be some discoloration in different parts of the system, but this is easily remedied by flushing the mains. Except as above mentioned there has been no discoloration from cleaning. We have noticed that so far there is no deposit in pipe cleaned three years ago and are satisfied that the cleaning does not injuriously affect the quality of the water.”
MR. CHESTER: In that tabulation of results obtained by the cleaning of old water mains in various localities in the United States, do Nos. 6, 7 and 8 refer to Huntington, W. Va.? Have you made any tests or got any results as to any lists there that you have with you?
MR. HODGMAN : I have not.
MR. HODGKINS: The supposition is that the coating protects the pipe. I would like to know as to the age of pipe when cleaning becomes advisable. Do I understand that cleaning destroys whatever coating there might be there? In that case it might be the greatest error to clean too early. I should like to know the length of time that should elapse before cleaning the pipe.
MR. HODGMAN: In regard to the length of time before they should be cleaned, that differs for every different water, and even with the same kind of water with different castings of pipe. There are some stations around the Great Lakes where they have had very little incrustation after eighteen or twenty years; and I do not know of anywhere they have had very serious results short of that time. I know of other than lake waters, more particularly surface waters, that have affected pipe very seriously inside of six or eight years; so that it is hard to give any definite answer to that; but I can say very positively that where the coating is intact under the tuberculalion that cleaning does not injure it.
MR. HODGKINS: There are three things which it seems to me it would be interesting to know in connection with this paper; one is the length of time that the pipe in each one of these cases had been laid; second, an analysis of the water pumped or that flowed through the pipe; third, the source of supply, whether ground, water, surface water, or lake water, or what. Such data it seems to me would be of great value. I do not know that the author of the paper can give us that off-hand, but if he can supply it later in connection with the publication of the paper, it would be desirable.
MR HODGMAN : No doubt data can be given as to, the source of supply. As to this analysis I would refer Mr Hodgkins to the paper read by Mr. Nicholas S Hill, Jr., before this association in 1907 on the subject, “Tuberculation and the Flow of Water in Mains ” I doubt very much, however, if the age of the pipe can be given in a large number of the cases mentioned in the tables connected with this paper, because a great deal of the work was done for municipalities and several of them have no records to show when the pipe was laid.
MR CHESTER Municipalities never keep records of pipe laying, is that it?
MR, HODGEMAN : I have known some that did not
MM. MCMILLAN: We consider our records fairly good although we are municipal. We had a new connection to make to the water tower about two years ago. We had removed some pipe that bad ben in place there for thirty-five years We found that the original tar coming was still inside of the pipe; it was in such good condition that the pipe that was taken out was put back in place and considered as good as new. There was simply no corrosion in the pipe. It had not been cleaned by any mechanical means. The source of supply there is Lake Michigan. So that this paper would only suit certain localities as to the condition of the water, etc.
MR WORRELL: We have just finished cleaning about 11,000 fet of 16-inch conduit at Meridian, Miss, which was done bv the National Water Main Cleaning Company I was constantly on the work, because I was very much interested in getting results. My observation made very carefully and closely was that the coating of the pipe was not destroyed at all except by the little nodules of rust that adhered to the pipe itself; and the results obtained on tnat work showeu about 100 per cent, increase in the carrying capacity of the pipe. It is used as a conduit to bring water from the reservoir, and the reservoir is supplied by springs. By the way, that answers the gentleman’s questions as to the kinds of water, it works under a head of about 80 lbs., but unfortunately for the records we have not had a test since January 11, 1910. That pipe was then carrying slightly less than 46,000 gallons per hour. Of course some incrustation was had during that period. The last test was made on Sunday a week ago and showed that the pipe was delivering 92,000 gallons per hour.
MR. CHESTER: What was the condition of the head ?
MR. WORRELL: The records are not quite as good, but more favorable to the head carried in January, 1910, because we have more water now m the plant and have more water in the reservon in January than in May as a rule.
MR. CHESTER: It seems that you have less head at the other end.
MR. WORRELL : We made an allowance for that also.
MR. CHESTER: You have a possible variation of something like 30 or 40 ft. head?
MR. WORRELL: Not that much; it is possible to have about 28 ft. We did not find it necessary to use booster pumps after it was cleaned out. We shut down to booster pumps at a saving of $250 per month.
MR. CHESTER: What was the character of the scale you took out?
MR. WORRELL : Such as is usually found in that kind of pipe. We have not had it analyzed. It is in the hands of a chemist for analysis.
MR. CHESTER: You did not find your water was bad for boiler use?
MR. WORRELL: No sir, not very bad.
MR. DUNHAM : If it can be obtained I would like to have some idea as to the amount of attention that was given to these mains before it was decided to clean them; how often they were flushed, and how they were treated in general. That would be interesting, I think.
MR. HODGMAN: In reply to Mr. Dunham’s question regarding the attention given to pipes, I would say that from observations made regarding the efficiency of water mains which have been frequently flushed, they are no better than those that are unflushed except where there is only a collection of mud in the pipes. Flushing will undoubtedly remove most of the sediment except where it has been allowed to gather for a long time before any flushing is done. There is one plant with which the author is familiar which is owned by a private corporation where flushing is carried on with great regularity, but they are seriously troubled not only with dirty water, but with corrosion and it would seem that where a great deal of flushing is done that there is a tendency toward greater corrosion.
MR. SMITH (Atlanta): With reference to the cleaning of that 30-inch main in Atlanta, at the time that was cleaned I was not in charge of the department although connected with it. It is a force main from the river pumping station to the storage reservoir. It was found necessary a few years ago on account of the lack of capacity of the mains between the stations and the reservoir to put in another main. We paralleled this main with a 36-inch main from the river station to the reservoir. There was a constant flow through this main. I don’t know the condition of this 30inch main at the present time. I have made no examination of it since cleaning; but we found that this incrustation occurs in the pipes within the city as well as in this main that carries only raw river water. We cut into a 16-inch main the other day for the purpose of getting a connection and found it in as bad condition as the 30-inch main was that was cleaned by the National Water Main Cleaning Company. We found also that in some sections of the city in mains that have been in the ground as long as this 16-inch main that we cut into very recently, that there is practically no corrosion at all in the pipes. It is a condition that seems to exist in parts of the city, possibly in the larger mains covered by the flow of water; but there in certain sections of the city, certain pipes that we cut into that have been in the ground longer that are not in as bad condition as other mains that were laid at a later date.
MR. HODGMAN: I would like to ask Mr. Smith if it is not a fact that those pipes which are affected the worst are the ones which carry the most water, or in other words, under ordinary conditions have a higher velocity in them.
MR. W. Z. SMITH (Atlanta): In the case of the mains within the city I think that is the case; but the other main had a constant pumpage through it. The 30-inch main was carrying raw river water and had a constant flow through it, delivering over 11,660,000 gallons, which was increased according to the record to 17,000,000 gallons or something.
MR. WEHR: AS a further illustration that the character of the water is the leading element in the incrustation of the pipe, I might cite an instance in our plant where in 1907 we took up about 12,000 feet of 10-inch pipe and about 5,090 feet of 12-inch pipe which had been in service from the pumping station in the outlying district. That has been down for twenty-one years. We hauled that pipe ten miles across country and laid it in the ground again. Before laying it we thought it might be well to scrape it out. It felt smooth, but we ran the brushes through it and did not get out enough to pay for the time spent in cleaning, so we did not stop to complete the cleaning, we laid it back in the ground. I apprehend that we will not have to consult a pipe cleaning company during my administration at least.