Articles & News Stories

A Study on the Influence of the Suction Arrangement on the Performance of Twin Screw Compressors

Description
A Study on the Influence of the Suction Arrangement on the Performance of Twin Screw Compressors
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/267595921 A Study on the Influence of the Suction Arrangement on the Performance of Twin Screw Compressors Conference Paper  · November 2013 DOI: 10.1115/IMECE2013-62391 CITATIONS 0 READS 203 5 authors , including: Some of the authors of this publication are also working on these related projects: Low Temperature and Cryogenic Refrigeration, Editors: Kakaç, Sadik, Avelino, M.R., Smirnov, H.F. (Eds.), 2003   View projectNextORC: Fundamental studies on organic Rankine cycle expanders   View projectAhmed KovacevicCity, University of London 162   PUBLICATIONS   893   CITATIONS   SEE PROFILE All content following this page was uploaded by Ahmed Kovacevic on 31 March 2015. The user has requested enhancement of the downloaded file.   1 Copyright © 2013 by ASME Proceedings of the ASME 2013 International Mechanical Engineering Congress and Exposition IMECE2013 November 15-21, 2013, San Diego, California, USA IMECE2013-62391 A STUDY ON THE INFLUENCE OF THE SUCTION ARRANGEMENT ON THE PERFORMANCE OF TWIN SCREW COMPRESSORS Maria Pascu Howden Compressors Ltd Glasgow, UK Manoj Heiyanthuduwage Howden Compressors Ltd Glasgow, UK Sebastien Mounoury Howden Compressors Ltd Glasgow, UK Graeme Cook Howden Compressors Ltd Glasgow, UK Ahmed Kovacevic City University London, UK ABSTRACT Screw compressors are complex flow systems, but operate upon simple considerations: they are positive displacement machines consisting of meshing rotors contained in a casing to form a working chamber, whose volume depends only on the angle of rotation. Their performance is highly affected by leakages, which is dependent on various clearances and the  pressure differences across these clearances. Nowadays, the manufacturing and profiling techniques have matured so much, that rotors of even the most complex shapes can be manufactured to tolerances in the order of few microns, resulting in high efficiencies. With manufacturing tolerances this tight, there is only small amount of improvement expected from further exploration of this venue, and a rather different direction for analysis may be more rewarding, i.e. other components of the screw compressor, like the suction and discharge areas. While the available literature includes several references on improvements of the compressor performance  based on the analysis of the discharge port and discharge chamber, the investigation of the suction arrangement and inlet  port remains fairly unexplored. This is the area of concern for the present paper, where the influence of the port shape and suction arrangement on the overall compressor performance is investigated. Various suction models were investigated for a standard screw compressor by means of CFD, which allowed in-depth analyses and flow visualizations, confirmed by the experimental investigation carried out on the actual compressor. Keywords: screw compressors, optimum suction, CFD INTRODUCTION Although the basic operation of twin screw compressors is well known and the analytical methods for their performance  prediction are well established, only few attempts of investigating the flow in screw compressors by means of CFD can be identified in the available literature. Nevertheless, there are many advantages in considering CFD as integrated part of the design and optimization process of screw compressors (SC). This is mostly because CFD complements the experimental and analytical efforts by providing an alternative cost-effective mean of simulating real fluid flows and substantially reduces lead times and costs of designs and production compared with an experimental based approach, Tiu and Liu [1]. Probably the most noticeable efforts in the field of numerical analysis of SC were made by Kovacevic et. al. [2] and [3], where in addition to establishing a mesh procedure specific to such flow machines, the author also explains adequate boundary calculations to encourage good convergence and minimal numerical errors. Similar efforts were made by Sauls and Branch [4], where the commercial code ANSYS-CFX was used for the detailed analysis of a refrigeration SC designed for use with R134a in air- and water-cooled chillers. Also benefiting from the mesh technique documented in [2], Steinmann [5] reported results from the modeling of a helical-lobed pump and a SC using ANSYS-CFX. While the available literature includes several references on improvements of the compressor performance  based on the analysis of the discharge port and discharge   2 Copyright © 2013 by ASME chamber, Mujic et. al. [6], Huagen et. al. [7] and Pascu et. al. [8], the investigation of the suction arrangement and inlet port remains fairly unexplored. This is the area of concern for the present paper, where the influence of the port shape and suction arrangement on the overall compressor performance is investigated. EXPERIMENTAL BACKGROUND While its theoretical background has never really been the focus of research in the available literature, the shape of the suction port in a twin screw compressor is often the subject of experimental investigations. The general belief is that by opening the gas admission through a radial port at the suction will have a positive effect on the compressor performance, as more of the rotor area will be exposed to the working gas. In order to determine whether or not the inlet conditions  bear an effect on the overall compressor performance, two suction scenarios were investigated for the same compressor, characterized by equal rotor diameters 165 mm, L/D=1.45 and “N” rotor profile with 4/6 lobes: one with axial port at the compression chamber entry (referred to as srcinal), the second including the same axial port, as well as a radial port machined-off from the inner casing wall (modified), as shown in Figure 1.  Figure 1 Compressor models: left  –   srcinal compressor; right  –   modified compressor When analyzing the experimental results shown in Figure 2,  particularly the flow, a reduction of volume flow in the modified compressor compared to the srcinal design can be observed, especially at the higher pressure ratios. The maximum reduction of the flow is around 2 - 3% at the pressure ratio of 20 and reduces as the pressure ratio goes down. At lower pressure ratios, below 5, both compressors seem to provide roughly the same flow. There are two main flow mechanisms which govern the net flow going through a screw compressor: the flow induced into the suction space of the rotors and the gas leaks back from the high  pressure regions to the suction space. The amount of gas induced depends on the volume, pressure and temperature of the rotor cavities opened to the suction gas. Any pressure drop within the suction chamber and the suction ports would result in lower flow and increase the power. On the other hand, the amount of gas leaks back to the suction area depends on the size of various clearances and the pressure difference across these gaps. By examining the flow results it is clear that the amount of gas induced into the suction cavities remains the same for both designs  –   this is because the flows are somewhat identical for the low pressure ratios. This could be due to fact that the srcinal suction port arrangement had very little pressure drop within the suction chamber and the suction port, and the improvement to the pressure drop by opening up the radial port did not give a significant flow benefit in terms of the induced gas flow. However, the incorporation of radial port seems to increase the leakages. This is clear from the flow decrease at the high  pressure ratios. When analyzing the power measurements results, the two designs seem to absorb very similar amount of power, even though the modified casing seems to absorb very small amount of extra power for the higher pressure ratios  -  this could  possibly be due to the influence of the leakage gas. 0204060801001201400 5 10 15 20 25    P   o   w   e   r    [    k   W    ] PR [-] ORI_test MOD_TEST   0.000.050.100.150.200 5 10 15 20 25    M   a   s   s   F    l   o   w    [    k   g    /   s   e   c    ] PR [-] ORI_testMOD_TEST   Figure 2 Experimental results In conclusion, contrary to the initial assumption, this compressor has not benefited in any way from the introduction of the radial suction port, as test results have revealed no improvement in the compressor performance at smaller pressure ratios and a slight deterioration at higher pressure ratios, in terms of both power and flow.   3 Copyright © 2013 by ASME NUMERICAL MODELS In order to fully understand the flow mechanism in the suction area of the compressors and determine the reasons which influenced adversely the performance when opening a radial port into the compression chamber, a numerical investigation was carried out. Numerical mesh The critical sub-domains in this setup are the two rotors as they contain the working chamber as well as the clearances and leakage paths (radial, axial, interlobe and blow-hole area). Generating the grids for these domains is by far the most challenging part of the entire meshing procedure, as both micro- and macro- scales elements have to be solved (the interlobe clearances are in the order of several microns, whilst the rest of the rotor body measures over 1000 mm). In this case, a technique dedicated to screw compressor rotors was employed, as described by Kovacević [3], which is included in SCORGgg (Screw COmpressor Rotor Geometry grid generator). This  procedure is fully explained in several publications included in the reference list and therefore, will not be repeated here. A simplified representation of the rotors mesh (cross-sectional view) is presented in Figure 3. This technique resulted in 143,000 nodes for the male and approximately 140,000 for the female, of structured mesh.   The numerical model includes three more domains: the inlet casing, the main casing, which in turn includes the suction into the compressor and the discharge casing, including the discharge  port. ANSYS ICEM v14.5 was used for the mesh generation process and special mesh refinement techniques were employed for sensitive flow areas, i.e. the interfaces with the rotors). Figure 4 Representative mesh for the working chamber (modified compressor) The overall mesh statistics typically used for these compressor simulations are:    Main casing approx 82K nodes    Discharge casing approx 86K    Inlet approx 176K nodes An adaptive meshing technique is utilized to capture all the changes which occur within the working chamber during the compression process. The number of time changes required by the rotors mesh is 120 for the full rotation of the male rotor, with the number of nodes kept constant across the timesteps. Boundary conditions The numerical model includes the stationary domain with the major casing components (inlet, main and discharge) and the rotating domain, depicted by the two rotors, see Figure 5. Various interfaces were applied to each of these domains to ensure the flow transition between the different domains. All these interfaces were considered to be General Grid Interfaces (GGI). The srcinal compressor model includes one interface  between the rotors and casing (called axial port  , placed between the inlet casing the two rotors). The modified compressor includes an additional interface to the rotors  –   the radial port   machined in the main casing. Figure 5 Numerical model Both the suction and the discharge were simulated by pressure  boundary conditions. The pressure boundary conditions are similar to the inlet or outlet boundaries, firstly because they couple pressure and velocity directly and secondly because for all equations, apart from the momentum equation, the boundary  properties are calculated from the velocity. This procedure may   Figure 3 Mesh over rotors cross-section before 3D interpolation
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks