Experimental Investigations of Heat Transfer in a Channel with Ribs and Obstacle

Abstract

This thesis presents experimental investigations of local heat transfer in a rectangular channel (aspect ratio AR = 4) with continuous transverse ribs as well as an obstacle. Liquid crystal thermography (LCT) is employed to obtain the temperature fields in the heat transfer experiments. The liquid crystal applied here has a red start at 35°C and a bandwidth of 1°C. The work consists of (1) periodic ribs, (2) obstacle (3) integration of ribs with an obstacle.

(1) The effect of (a) rib-pitch-to rib height ratio (p/e), and (b) rib height-to-channel hydraulic diameter ratio (e/Dh) on the local heat transfer in the first inter-rib regions (i.e., where thermal field is not yet periodically fully developed) are investigated. The Reynolds number, based on the channel hydraulic diameter, has the values 57000, 89000, and 127000. Concerning (a) with e/Dh = 0.078, the particular emphasis is on large p/e ratios, i.e., 10, 20 and 30, in the first inter-rib regions. It is conjectured that the main flow for all studied p/e ratios separates at the edge of the first rib and reattaches in the first inter-rib region at a distance of about 9-10e from the upstream rib edge, where a local maximum of the Nu number occurs. This distance is larger than that typically occurring in the periodically fully developed region. This may negatively affect the heat transfer in the entrance region if the p/e ratio is less than 10. Concerning (b), e/Dh = 0.039 and 0.078 have been selected for p/e = 10 and 20. The significant effect of the small blockage ratio on the local Nu number in the first inter-rib regions is found remarkable compared to the high blockage ratio. This effect is explained by the small thickness of the boundary layer in the developing region and it is conjectured that the core flow is strongly disturbed by the presence of the rib with high blockage ratio.

(2) The end-wall heat transfer around a single obstacle is studied. The obstacle has a rectangular cross-section and blocks the whole height of the channel. The Reynolds number, based on the spanwise width of the obstacle, has the values 35600, 55600, and 79400. The appearance of a double peak of locally high heat transfer in the upstream junction indicates the existence of more than one vortex. As the Reynolds number increased, this double peak merged to a single peak. The Reynolds number effect is expected to affect the strength of the vortices demonstrated by higher local heat transfer from Reynolds number 35600 to 79400. The size of wake region with low heat transfer is shrinking by increasing the Reynolds numbers.

(3) The aim is to control the heat transfer around an obstacle with the aid of a rib. The rib is positioned in (a) upstream region, and (b) downstream region of the obstacle. The Reynolds number varies between 35600 and 55600. In this study, the spacing between the rib and the obstacle is of primary importance. Concerning (a), the spacing S is normalized by the spanwise width of the obstacle and had values 1.25d and 0.625d. The effect of e/Dh is also of concern. The e/Dh varies between 0.039 and 0.078. It is found that the local heat transfer especially in the upstream region was strongly affected by the S/d and e/Dh. In the downstream region, the local heat transfer was more affected by the Reynolds number than S/d and e/Dh. Concerning (b) the spacing between the rib and the obstacle had the value 1.25d. It is shown that the local heat transfer in the upstream region of the obstacle remained unaffected by the presence of the rib for the considered rib heights and Reynolds numbers. In the downstream region of the obstacle, the heat transfer pattern was substantially modified by the presence of the rib for two considered rib height. In the upstream region of the rib, the small rib height presents higher local heat transfer than the large rib height for the considered Reynolds numbers. A larger region of high transfer in the downstream area of the rib reflects a stronger effect on the reattachment process caused by the large rib height.

Abstract (Swedish)

Popular Abstract in English

This thesis concerns experiments of heat transfer in a rectangular channel with ribs and an obstacle. Liquid crystal thermography (LCT) is used to record the temperature fields in the experiments. LCT is based on measurement of the color response of a heated surface. When illuminated with a white light, the liquid crystals react to temperature changes by changing color, i.e., from red at the low end to blue at the high end of the active temperature range. The liquid crystal used has a bandwidth of 1°C. The research work consists of three cases, namely several periodic ribs, a single obstacle, and interaction of ribs with an obstacle, respectively.

The results with periodic ribs placed on one wall expand the current knowledge frontier for two geometrical parameters, i.e., the rib spacing to rib height ratio and the rib height-to-channel hydraulic diameter ratio, for cases where the thermal field is not periodically fully developed. Three different flow velocities were considered. Particular emphasis is on large ratios of the rib spacing to rib height in the first inter-rib regions. It is conjectured that the main flow for all cases separates at the edge of the first rib and reattaches in the first inter-rib region at a distance of about ten times the rib height from the upstream rib edge, where a local maximum of the heat transfer coefficient is found. The effect of the rib height-to-channel hydraulic diameter ratio is investigated by two rib heights in combination with two rib spacings. The significant effect of the small rib height on the local heat transfer in the first inter-rib region is found remarkable compared to the bigger rib height. This effect is explained by the small thickness of the boundary layer in the developing region and it is conjectured that the core flow is strongly disturbed by the presence of high ribs.

The second part presents local end-wall heat transfer distributions in the upstream as well as downstream regions of an obstacle. The obstacle has a rectangular cross-section and blocks the whole height of the channel. Three flow velocities are considered. A double peak of locally high heat transfer occurs in the upstream wall-obstacle junction and indicates the existence of a complex vortex system. A primary peak upstream of the junction is explained by a vortex formed at the front corner.

The last part aims to control the heat transfer effect of the obstacle by using ribs. A rib is positioned upstream or downstream of the obstacle. Two flow velocities are considered. The spacing between the ribs and the obstacle is a primary parameter. In the upstream region, two spacings and two rib heights were considered. For the small spacing the result is contradictory to that of the large spacing. It is found that the local heat transfer especially in the upstream region is strongly affected by the spacing and rib height. In the downstream region, the local heat transfer is more affected by the flow velocity than by the rib height and spacing.

As the rib is in the downstream region, a single spacing prevailed but two rib heights were tested. No impact on the local heat transfer in the upstream region was found. In the downstream region of the obstacle, the local heat transfer was significantly affected by the presence of the rib and its height.

The results of this thesis will be of significance for understanding and handling heat transfer issues in gas turbines and heat exchangers.

• author: Ghorbani-Tari, Zahra
• supervisor: Bengt Sundén
• opponent: Professor Murray, Darina, Trinity College Dublin, Ireland
• organization: Heat Transfer
• publishing date: 2014
• type: Thesis
• publication status: published
• subject: Energy Engineering
• keywords: Liquid crystal thermography, ribs, entrance region, control the heat transfer around an obstacle by rib
• defense location: Lecture hall M:E, M-building, Ole Römers väg 1, Lund University Faculty of Engineering
• defense date: 2014-03-28 10:15:00
• language: English
• LU publication?: yes
• id: 4dfbc423-d4d8-4c18-849c-38a3e877a435 (old id 4350434)
• date added to LUP: 2016-04-04 09:10:56
• date last changed: 2018-11-21 20:51:20