Content Map

 

Introduction

DOI 10.1615/hedhme.a.000457

4.8 COSTING OF HEAT EXCHANGERS
4.8.1 Introduction

A. Background

Perhaps the most important step in the application of heat exchangers to industrial processes is that which occurs at the process design stage. Here, the process designer makes a selection (which often turns out to be the final selection) of the type of heat exchanger which is to be used. In practice, many types can be eliminated at this stage on the grounds of unsuitability for the process; thus, for example, the process pressure may be greater than the maximum operating pressure of some exchanger types. Similarly, the toxicity of the process fluids may be such as to eliminate exchangers where the possibility of leakage occurs (for example through the gaskets of plate-and-frame heat exchangers). However, in most situations, there are several feasible types of heat exchanger available for a given application. Often, the selection is made on the basis of previous practice; thus, if shell-and-tube exchangers have always been used for a particular process application, then a safe option is to use them in any new process plant for the same application. It would be more logical to make a choice on the grounds of capital (and perhaps also operating) cost, but this depends on having sufficiently accurate costing methods at the process design stage. The object of this Section of HEDH is to provide a platform for the presentation of such methods.

Since perhaps the majority of heat exchangers are custom-built (or at least built by integrating mass produced parts) for each application, accurate costing is quite difficult to achieve and is usually done using proprietary computer codes developed and held by the manufacturers. These codes depend on estimation of materials, labour, overheads and shipping costs which are specific to a given manufacturer. However, the dictates of competition generally lead to costs from different manufacturers which are surprisingly similar. Here, one should make a clear distinction between cost and price, the latter being the result of a market judgement by the manufacturer’s sales team. If rapid delivery is required, the price may be much higher than the cost. If, on the other hand, the manufacturer wishes to maintain loading on his plant, they may actually sometimes quote a price which is lower than the cost as a short term measure (often called “buying the job” in the industry). Clearly, such a marketing policy could not be pursued in the longer term! Thus, it is assumed here that, on average over time, the price will be a reasonable reflection of the cost.

There are a number of techniques for carrying out rapid costing of heat exchangers. These include:

  1. Using thermal design data, carry out a costing approximating to the full costing methodology. This could include, for instance, correction factors allowing adjustment of costs for a given design relative to a standard design. This is essentially the methodology given in Section 458, Section 459 and Section 460 for shell-and-tube, air-cooled and plate and frame exchangers respectively.

  2. Base the cost on a predicted heat exchanger area. Here, the prediction is achieved normally by using estimated overall heat transfer coefficient (U) values and determining the surface area A from the relationship.

    \[\label{eq1} A=\dfrac{\dot{Q}}{U\varDelta T_m} \tag{1}\]

    where is the total rate of heat transfer (W) and ΔTm the mean temperature difference.

  3. Base the cost on a predicted heat exchanger volume. Here, the volume (V) can be determined from the following relationship:

    \[\label{eq2} V=\dfrac{\dot{Q}}{B\varDelta T_m} \tag{2}\]

    where B is a volumetric overall heat transfer coefficient (W/m3K).

  4. Base the estimate of cost on cost (C) per unit  /ΔTm. This approach avoids problems in definition of heat exchanger area and allows direct comparison of various heat exchangers for a given duty. The method was originally devised by Hewitt, Guy and Marsland (Hewitt et al., 1982) but has been extensively adopted in later work by the Engineering Sciences Data Unit (ESDU, 1986; ESDU, 1994; ESDU, 1995; ESDU, 1997; ESDU, 1994).

... You need a subscriptionOpen in a new tab. to view the full text of the article. If you already have the subscription, please login here