Hydro project engineering costs9 August 2010
One of the first questions owners ask is ‘what will be the cost of engineering services’. This paper discusses the effect of documented/non-documented designs on project engineering costs, based on detailed man-hour data obtained from 17 hydro projects ranging in size from the 6.4MW Maggoty development in Jamaica, to the 2304MW La Grande 3 development in Quebec. By J L Gordon, P.Eng
Over the past 58 years, the author has been fortunate to work on both large and small hydro developments, and managed to keep detailed statistics on project engineering services. The author found that there was a vast difference in the project documentation required by a private utility, as compared with that demanded by a public utility. Also, there was a significant difference in how the design and construction work was undertaken for small as compared with large hydro developments. After many years, a distinct pattern in man-hours was discerned, as will be discussed in this paper.
Project engineering man-hours
After completion of a feasibility report, one of the first questions asked by the hydro project owner; is how much will the detailed engineering work cost? This is very difficult for the consultant to estimate with any degree of accuracy. It was found to depend on several factors:
• How much documentation of the work will be required by the owner?
• How will equipment bids be analyzed and what justification will be required on contractor selection?
• How will orders for the equipment be placed?
• Will the owner question all invoices for the engineering services?
The answers to these questions will directly affect the man-hours required to undertake the design and project management, resulting in two types of engineering services offered to owners - (1) a documented design service, and (2) a non-documented design service.
A documented design would entail the production of –
• A ‘Design Criteria report’: A document showing the criteria to be used in design of each structure, forwarded to the client for perusal by the client’s Review Board, prior to starting detailed designs.
• A ‘Design Transmittal’: A document showing how each structure was designed, for permanent retention in the client’s project files, to be used whenever there are repairs or modifications to the structure.
• A ‘Contract Award document’: A report on how the contractor was selected, to include all specifications and correspondence with the selected contractor prior to award.
• A ‘Project Completion Report’: This document would include copies (on CD) of all project drawings and photographs, and copies of all progress reports issued during construction.
• A ‘Project Operating Manual’.
Such detailed engineering services would normally be required for all major hydro developments, by large utilities and for international bank-financed projects.
A non-documented design is one where a client or contractor is willing to entrust the engineering services to a consultant, without requiring any documentation on the design or construction, and the only documentation provided would be: a copy of all project drawings; an operating manual.
This standard of work would be equivalent to the design work undertaken by a consultant for the general contractor on a design-build project, or that required for a small hydro development of less than about 50MW capacity.
From 1952 until 1990, the author worked for Montreal Engineering (Monenco), a consultant who provided design, construction and operating services to many associated hydro utilities. These utilities did not require any detailed project documentation; hence the author was able to accumulate data for non-documented engineering services, where most of the contracts for construction, turbine and generator were negotiated with reliable known contractors and manufacturers. Bids for other equipment were solicited from at most, three manufacturers. Specifications were concise. For example, all the technical specifications for civil works and mechanical equipment at the 356MW Brazeau project were contained in a 2.5cm three-ring binder. About 1960, Monenco began providing services to ‘outside clients’ who required more formalized engineering with full documentation of the work, and required all contracts to be open for bidding by pre-qualified contractors. As a result, specifications became more detailed and contractual conditions more complex. This provided an opportunity to obtain data on documented designs. The results are shown in Table 1, where the data was accumulated over 32 years between 1954 and 1986 for 8 non-documented projects and for nine documented projects.
Analyzing the data, the author found that there is a factor of about 2.5 between a ‘documented’ as compared with a ‘non-documented’ design and managed project, as shown in Figure 1. There are two distinct lines on the figure. The lower line (B) represents the man-hours for a ‘non-documented’ design, and the upper line (A) represents a documented design. Both lines can be expressed by the following equation-
Man-hours = k (MW/h0.3)0.54 (1)
Where: h = rated net turbine head in meters; MW = installed capacity in megawatts; k = a factor with a value = 8,300 for a non-documented design, and = 21,000 for a documented design.
In other words, a documented project requires over 2.5 times the effort expended on a non-documented project. This is very important, and a factor which must be taken into account by any owner contemplating awarding a contract for engineering services.
There are some projects on the chart which do not fit the data. These can be explained by:
• 2 – Snare Falls. The dam site on this project was changed at the last minute, when the owner decided that the project capacity should be increased by about 50%. This was accomplished by moving the dam site downstream to the next set of rapids, after all contracts had been issued for tender. This required a complete re-design of the project layout.
• 4 – Rattling Brook. This project proceeded with only a pre-feasibility study, requiring extensive additional studies on dam and spillway locations.
• 10 – Dadin Kowa. This project was terminated when the client ran out of money, and still requires completion of the powerhouse and installation of spillway gates.
• 13 – Cat Arm. During the design engineering work, the owner decided to add about six utility engineers to the consultant’s electrical design team to learn how to design the automatic controls, with the intent to retrofit the design onto older powerplants. Also, there were significant changes to the project concept when detailed design work commenced, which included use of a U-shaped weir spillway instead of a gated spillway, impulse units instead of Francis units, and elimination of the surge tank.
• 7 – Bighorn. This project was undertaken after a change in the management at the utility. The new management required all contracts to be tendered instead of being negotiated, adding substantially to the weight of both the technical and contractual conditions in specifications, and requiring more formal documentation of the work.
It is remarkable how consistent the data appears to be. All of the projects were undertaken by Monenco, with the exception of Wreck Cove, engineered by SNC, with Monenco acting as the owner’s consultant, and LG3, engineered by SNC in association with Cartier Engineering, a subsidiary of Monenco. Also, two of the projects, at Maskeliya Oya and Bayano were designed from project offices in Colombo and Panama, staffed with a few Canadian engineers, with all the other engineers and draftsmen being provided by the utility client.
The chart is based on data from projects undertaken before the introduction of computerized drafting, (CAD) which one would expect to reduce drafting time. The author has tried to determine whether CAD has had a significant effect on man-hours, to no avail, since consultants are naturally reluctant to divulge such data. However, the author is of the opinion that CAD and other similar programs have instead increased drafting time, since now it is very easy to produce more drawings, and to produce three-dimensional images of powerplant interiors and even individual concrete pours, both of the latter requiring significant additional input data. The added engineering simplifies construction work, but increases engineering man-hours.
Also, the work was undertaken before personal computers became available. Again, one would expect that computers would reduce the engineering effort. However, based on observing the extent and detail of recent specifications for large developments, the author is of the opinion that computers have indeed facilitated, but have also increased engineering man-hours.
Hydro specifications for documented and non-documented projects
This raises the issue of the effort required to produce specifications and contract documents. A perfect illustration of the difference is shown in Figure 2, where the large 3-ring binder contains the technical specifications only for the civil works on a medium-sized documented 200MW hydro project, and the small document on the left contains the civil specifications, the geological report, contractual conditions and the environmental guidelines for the civil works contract at a small low-head non-documented 600kW hydro development. The right-hand document was produced by a large consultant in Vancouver, and the other document by a small consultant in Halifax.
The author is of the opinion that it is very difficult for a large hydro consultant to produce a simple short specification. Large consultants have the ability to retain specialists in many disciplines, and consequently have few ‘generalists’ able to work on a variety of structures and equipment. For example, the mechanical department in a large consultant’s office could have an engineer specializing in powerhouse cranes, another in gantry and gate hoists, another in powerhouse pumps, compressors and piping, and another in turbines. All would be required to contribute towards an equipment specification, and the result would be a large and expensive document requiring considerable co-ordination between the specialists. On the other hand, a small consultant would have perhaps only one mechanical engineer, and this person would be required to produce the specifications for all the mechanical equipment. The resulting document would be far more concise than that produced by the large consultant.
Documented project design and management
To illustrate the difference in project management between a documented and non-documented work, the data for the large documented Jebba hydro project is presented. Jebba is located on the Niger River in Nigeria, and includes a large ship lock. It was commissioned in 1984. It has an installed capacity of 560MW at 27.6m head from 6 vertical axis Kaplan turbines. The completed project can be clearly viewed on Google Earth at 9-08-20N, 4-47-27E.
Major project quantities at Jebba included –
• Earth excavation: 1,230,000m3.
• Rock excavation: 3,270,000m3.
• Earth fill: 2,120,000m3.
• Rock fill and rip-rap: 2,930,000m3.
• Spillway, lock, weir concrete: 241,000m3.
• Powerhouse concrete: 245,000m3.
Total cost of the project was over $1B in 1980 US$. Time spent on engineering and project management was 386,000 man-hours, of which 27% (104,000mh) was on project management and construction supervision. The project was fully documented, which required the production of both “design criteria” documents and “design transmittal reports” for each of the nine major structures.
Based on detailed statistics maintained throughout project execution, the project management group supervised the production and distribution of –
• 23 contract documents, with a total of 2478 pages of technical specifications, plus contractual conditions.
• 364 tender drawings.
• 1406 construction drawings.
• 7335 manufacturer’s drawings received for review by the design engineers.
• 131 construction reports, with a total of 12,569 pages.
The production of drawings had to be kept ahead of the contractors’ requirements, and peaked at 512 approved for construction drawings issued in the month of February 1980. Of course, it was not possible to engineer such a large number of drawings in one month; they were produced over the previous year when design staff peaked at 72 individuals. The general civil work contract was signed at the end of January 1978, and in March seven copies of 198 drawings were packed in a large box and sent by air freight to the contractors head office, knowing that it was essential to keep ahead of the contractors requirements for detailed construction drawings.
Cost control was maintained by an accounting team at site, and was complicated by having to use four currencies – US dollars, Italian Lira, Japanese Yen and German Marks. During construction, it was difficult to determine the current total cost due to the constant change in the relative value of the currencies. The project was completed in 1984, after 6 years of construction. Management staff averaged nine persons over this period, and peaked at 13.
Non-documented project design and management
A non-documented design will not require such a large project management team as at Jebba. Another factor to be taken into account is the recent use of water-to-wire equipment contracts, wherein the contractor undertakes a major proportion of the powerhouse electrical and mechanical design work. This means that the consultant can often dispense with the services of electrical and mechanical engineers, further reducing the cost of project execution. Another factor is the introduction of high speed internet services over the last decade. This has allowed many engineers and technicians to work from home, working within informal groups to provide design services for small hydro projects. There are several such groups, where a hydrologist, civil engineer, geotechnical engineer and a CAD draftsman produce all drawings, and all working from home.
An early demonstration of this concept was provided at Ragged Chute, where the addition of a small 6.3MW, 13m head development at an existing dam in Canada was undertaken in 1992. The work was supervised by the owner (a retired contractor) from a rented nearby cottage, a civil engineer and draftsman with no hydro experience were engaged to undertake powerhouse and intake design, another to design the penstock, a senior hydro consultant provided advice, and costs were audited by a hydro engineer appointed by the bank providing the financing. A geotechnical engineer was engaged to provide advice on slope stability during excavation work. Ragged Chute is located at 47-16-35N, 79-40-19W, and is clearly visible on Google Earth. Total staff involved were; 3 engineers (owner + civil + draftsman) for about 18 months, plus another 4 engineers very much part-time. There was a water-to-wire contract for the equipment, and a general contractor undertook the construction and also designed the plant plumbing, heating, lighting and ventilation systems. There was no documentation other than an operating manual provided by the W/W contractor, and a set of drawings.
The first decision facing a hydro owner; is how much documentation of the project execution work will be necessary? For large projects, documentation is required. There is a grey zone between about 25MW and 100MW where documentation is optional. However, for small hydro work, of less than about 50MW, documentation should be kept to a bare minimum. It is just too costly and adds considerable expense to the project.
Recently, the author has seen specifications issued by hydro utilities for engineering services. They include extensive requirements for very detailed project documentation, even for small hydro projects, and the author is of the opinion that such specifications have been issued without understanding the effect on engineering costs. It is hoped that this paper will help to clarify such issues.
Jim Gordon is an independent hydropower consultant, and can be reached at [email protected]