Return to search

A critical evaluation of the design of removable cover-plate header boxes for air-cooled heat exchangers

Large air-cooled heat exchangers (ACHEs) are most popularly implemented in the petrochemical and power industries at arid locations. They operate on a simple concept of convective heat transfer, whereby air in the surrounding atmosphere is caused to flow across a tube bundle, which in turn transports a process fluid. The distribution and direction of the process fluid flow may furthermore be guided via a set of appropriately located header boxes, which essentially consist of a collection of welded flat plates and nozzle attachments. Perforations on one of the faces of these boxes serve as an interface to the tube bundle. The overall design and construction of an ACHE is commonly regulated by an American Petroleum Institute (API) standard, which is required to be used in conjunction with acceptable design codes. In spite of this, the design of certain header box configurations remains of prominent concern. It is the focus of the present study to investigate the approach adopted for a header box variant labelled as the removable cover type. In this configuration, one of the plates used to construct the header box is fastened and sealed by a collection of bolted joints and a gasket, allowing it to be removed. One appropriate design code for the header box equipment is the ASME (American Society of Mechanical Engineers) boiler and pressure vessel code. However, it provides no specific approach pertaining to the removable cover design. Instead it has been commonplace in industry for a number of aspects from this code to be synthesized, together with a collection of assumptions surrounding the header box behaviour, into an all encompassing design by rule approach. In this approach, the header box behaviour is accepted as being planar, whilst circumstances such as nozzle attachments and associated loading would suggest that a more comprehensive approach should be undertaken. The aim of the present study is therefore to critically evaluate the current practice, and establish its adequacy. To do so, a detailed three-dimensional finite element model (FEM) of an example header box design is developed. Subsequent comparisons with the stress distribution predicted via current practice show that the existing analytical model gives inaccurate and, in cases, overly conservative results. A new analytical approach developed from rigid frame theory is demonstrated to provide improved correlation with FEM. The linear elastic design by analysis approach, presented in the ASME code, is also utilised as a method for establishing design adequacy. Results obtained via design by analysis incorporating the finite element method are shown to be less conservative than those arising from design by rule methods. The design by analysis approach is also used to conduct a more detailed investigation of nozzle placement and external loading. In general, the effect of including a nozzle did not result in a significant increase in side plate stress, with failure more likely to occur within the nozzle wall. / Dissertation (MEng)--University of Pretoria, 2011. / Mechanical and Aeronautical Engineering / unrestricted

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:up/oai:repository.up.ac.za:2263/28133
Date22 September 2011
CreatorsPrinsloo, Lionel
ContributorsDr S Kok, Dr H Inglis, prinsloo.lionel@gmail.com
Source SetsSouth African National ETD Portal
Detected LanguageEnglish
TypeDissertation
Rights© 2011, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.

Page generated in 0.0021 seconds