Continued from - Costing of Pipelines - 1
22. Piping Cost Estimation
Piping always represents a sizable part of the total installed cost of any process plant. It may run to 20 percent or even more of the entire cost of the plant, including land and buildings, and may total from around 10 percent to as high as two-thirds of the cost of the equipment. But piping is usually complicated and often consists of a myriad of small, rather inexpensive parts. Its estimation can therefore be a headache to the estimator.
It makes his problem no easier that he is often asked for a reasonable estimate long before the final plans exist and before the project engineers have any idea how much piping will actually be needed. In this case, he must not only be an estimator, but he must be able quickly to convert the preliminary flow sheet into a fair approximation of an actual design to know what his estimate will involve. For this purpose the ability to make quick and rather accurate free-hand isometric sketches is of inestimable value.
However, the estimation of the piping need not be arduous if the estimator will make use of the N system developed by the another during the last several years. This system, first disclosed in Chemical Engineering (R.A. Dickson, Chem. Eng., November, 1947, pp 121-123) has now been carried much further and includes most kinds of pipes with which the chemical engineer is likely to be concerned.
The N System of Calculating Piping Costs. This new system of estimating
piping cost is quicker than the piece-by-piece, and more accurate than the “finagling-factor,” system, the two methods ordinarily used by estimators.
The first requires picking of each length of pipe with its fittings and valves
and pricing the material and labour costs in detail, then adding them all up for the total cost. The second consists in taking a percentage of the total cost of a project as the cost of the piping. This percentage the “finagling factor,” is supposed to be around 40 percent.
* The author wishes to express his appreciation to the firms listed and to
may others who prefer to remain anonymous, but who helped by furnishing cost data on which the N-factor tables are based. Specific acknowledgement is made to Armstrong Cork Co., Inc.; Mundet Cork Corp.; Johns Manville, Inc.; Cooper Alloy Foundry Co.; Taylor Forge and Pipe Works; Andrews Knapp Construction Co.; Knapp Mills Inc.; The Saran Lined Pipe Co.; The Rie-Wil Co.; and The Duriron Co.
See, for example, Table 10-33 of the N factors. The costs of different sizes of steel Pipe, butt-welded, schedule 40, black, with fittings and valves as specified, will be to each other approximately as the N factors. As the table shows, a string of 2-in nominal diameter pipe will cost 1.84 times as much as the same string in 1 in.
Take, for example, a string of such pipe of 2-in nominal diameter:
|Item||Unit Cost||Total Cost|
|500’ pipe||$ 0.23||$ 115.00|
The same string in 1-in nominal diameter:
|Item||Unit Cost||Total Cost|
|500’ pipe||$ 0.13||$ 65.00|
Hence, 208/119 = 1.84, the N factor for 2-in. pipe in terms of 1 in. pipe as unity.
Not only has the N-factor System been tested thoroughly, but it also shows up satisfactorily over a period of time. Several checks of this fact have been made. For example, table 10-39 was selected at random, and comparable strings were figured for 4- and 12-in. pipe, using first the costs of 1935, and then July, 1949, costs. Despite the fact that the costs of each string had risen approximately 83 per cent from 1935 to 1949, the N factor for 4-in. pipe for 1935 was 0.249, and for 1949; 0.248.
How the N-factor System Is Used: To use the tables, first calculate the cost of the reference sizes of the strings of pipe in question. Then use the N factors to get the cost of the same string in the required size.
The index of N-factor tables includes 66 materials and weights of piping,
counting some insulated, some bare. If the N system is adopted, this means that prices and costs of installation of these materials for the reference sizes only need be kept on file, instead of material and labour cost of every size of every material and wall thickness in common use. (Also as later explained, if less accurate estimating suffices, data given here can be “factored” to the date of use by using one of the available cost variation indexes.).
The tables cover only the cost of putting the strings of pipe together. The costs of burying or supporting pipelines are easily calculated. Moreover, many times the proposed lines lie on the ground or are supported on existing structures. These conditions are far too variable to include in any tables of costs.
An example of use: Assure that several strings of pipe of various sizes have been taken from the flow diagram of a small project involving some additional piping.
The specifications of pipe, fittings and valves are:
Pipe: steel, schedule 40, butt-welded, black, bare
Fittings: forged steel, screwed. 2,000# cold-water pressure
Valves: bronze, screwed ends, 200 # cold-water pressure
Assume that the pipe is to be installed on existing support.
First consult the index of N-factor tables. The specifications of the pipe, fittings and valves in Table 10-33 correspond to those required. The reference size, i.e., the size for which N = 1.00, is 1 in.
Next, calculate the cost of hypothetical string of 1-in. pipe of the required
specifications. (Or take the cost from Table 10-33, giving the cost of the
reference string in July, 1949. Then “factor” this cost to date of use.)
As in Table 10-33, the installed costs may be found to be:
Per foot of pipe .............................. $ 0.13
Per fitting ....................................... 4.50
Per valve ......................................... 11.00
Calculate all the strings taken from the flow diagram as 1-in. pipe, then multiply by the N factor for the actual size:
For comparison, the total cost of piping calculated by the conventional pieceby - piece system would be S764, as shown in the following check:
What Data Are Needed: The N-factor tables were derived by actual calculation of the installed costs of typical strings of pipe in all the sizes mentioned in each table, and in all the different weight and materials of construction listed in the 66 tables. For uniformity, all have been calculated on the basis of July, 1949, materials and labour costs in the New York area. For each specification the costs so calculated for the several sizes have been compared with the cost of one reference size which is taken as 1.00. The comparative costs of the other sizes of the same material are then listed as multipliers of the reference size. In Table 10-1, for example, the sizes range from ½ to 12 in. The reference size is 3 in, and the N factors (i.e. relative costs compared with that
of the reference size range from 0.320 to 6.730).
Hence, the N factors are all based on the reference size (or in a few cases, two reference sizes) for each piping specification, and to figure any string of pipe, it is necessary only to figure the cost for the reference size in that particular specification. Once the cost of the reference size is known, the costs for any other size or for several sizes can be determined immediately by multiplying by the proper N factor. This means that it is necessary to keep up-to-date the cost data only for the reference size (or sizes) in each specification, since the N factors remain constant through an extremely wide range of material and labour cost variations.
Estimating without Current Data: This means also that it is unnecessary to keep up-to-date data even on the costs for the reference strings, if a somewhat lower degree of estimating accuracy can be tolerated than is possible by having up-to-date figures. For this purpose, each of the tables includes as a second part (or in some cases, as a second and third part) the detailed calculations for the reference size or sizes for these materials are based on July, 1949 costs. As long as labour and material costs do not change much from July, 1949, these figures can be used directly. But since such costs are not likely to be stated in the future, it is possible to “factor” them to the date of use by the judicious use of one or more of the available cost indexes, in comparison to the index for July, 1949.
Cost Indexes: Many cost indexes are available monthly in the Survey of Current Business, published by the U.S. Department of Commerce. Among these are several building cost indexes for various kinds of labour and for a number of basic materials. The Survey, as well as the magazine Engineering News-Record, also publishes the ENR Construction Cost Index, which is widely used by estimators in dealing with plant and equipment cost variations, although it is intended to cover only heavy construction costs. Some organisations have successfully applied their own modifications to the ENR Index in using it to “factor” plant and equipment costs. The magazine Chemical Engineering regularly publishes the comparative equipment cost indexes for process industries compiled quarterly by the evaluation engineering firm of Marshall and Stevens, together with this firm’s 1947 industry average, covering equipment costs in a wide range of industrial and commercial activities.
None of these indexes applies directly to piping costs and must therefore be used with care. However, experience has indicated that such “factoring” can be sufficiently accurate at least for pre-construction cost estimating. The only indexes that actually deal with piping cost, as far as the author knows, are not so readily available as those mentioned. These are the Handy Indexes of Public Utility Costs, put out for the estate of William W. Handy by Whitman, Requard and Associates, of Baltimore, and Benjamin L. Smith and Associates, of Albany. This compilation is issued every six months and lists several kinds of piping including gas mains and power plant piping.
Table for N System
Pipe: aluminium, Aleoa alloy 618-T6, standard weight, bare, 1/2 “ to 12”
Fittings: aluminium, welding
Valves: ½” and 1" aluminium, screwed, gate; 1 ½” to 12" aluminium, flanged, gate, 125#
Reference size: 3#