(94b) Optimization of Shell & Tube Heat Exchangers Operating in Heat Recovery Services with Tube Side Heat Transfer Enhancement

Optimization of heat exchangers for efficient heat recovery in thermal processes is an important aspect to reach net zero goals. To achieve high heat recoveries, a so-called temperature cross, where the cold fluid temperature intersects the hot fluid temperature, is encountered in these units. For shell & tube exchanger design, this results in a challenging set of conditions to work with.
Multi-pass shells in series configurations are common in situations with a small temperature approach, and in many cases are used in order to compensate for the temperature cross between the process streams. In such designs, the equipment and piping costs are high. Alternatively, a two pass F shell design is often considered, however here thermal and physical leakage across the longitudinal baffle and damaged sealing strips during bundle removal are major drawbacks. For these reasons F-type exchangers are frequently disallowed in end-user specifications. In general, the most cost-effective option is a single-pass E shell exchanger. However, the single pass E shell design is often not pursued, due to the following:

• Low tube side heat transfer due to low flow velocity in single pass design.
• Low tube side flow velocities often shift the flow regime from turbulent in multi-pass design into transitional or even laminar in single tube pass design.
• Operational issues in start-up and turn down conditions when operating near transitional flow regime.
• Bundle flow maldistribution often caused by very low frictional tube side pressure drop.
• Possibility of temperature stratifications (in-tube temperature pinch).

All these shortcomings can be overcome by using hiTRAN® Thermal Systems tube side heat transfer enhancement technology. Measured tube side heat transfer in laminar, transitional and turbulent flow are presented. The measurements in this region were conducted with different Grashof and Richardson numbers with and without tube side heat transfer enhancement. Results show that possible operational issues in the transitional flow region can be eliminated entirely. Over the entire operation range there is a linear response in terms of heat transfer with flow velocity (Reynolds number), as opposed to the response observed with empty tubes, thus eliminating the risk of sudden changes in the transitional flow region.

In addition, the much-improved tube side heat transfer of up to 16 times in laminar and 4 times in turbulent flow leads to smaller units with large capital cost savings.

The avoidance of possible in tube temperature stratification in low flow velocity single pass designs, when using hiTRAN^® technology is explained in detail. In in-tube temperature pinch situations, large areas of effective heat transfer area is lost and leads to underperforming heat exchangers. This situation is difficult to predict and is not well captured with standard heat transfer correlations since it depends very much on complex mixed convection flow behaviour. In general, software design packages do not account for this situation and often require the user to make a judgement based on a simple raised warning message.
CFD flow videos in tandem with experimental measurements demonstrate the significant difference between plain empty tube and hiTRAN® enhanced flow in those situations. It is demonstrated that the tube side flow pattern is changed from stratified mixed convection flow to homogeneously forced convection flow thus greatly improving the predictability in terms of heat transfer and pressure drop.

As indicated, single pass S&T designs often lead to low flow velocities, with very low associated frictional pressure drop. This in turn can lead to uneven tube side flow distribution, especially in large bundles. Increased frictional pressure drop, generated by the use of hiTRAN® inserts, can lead to improved fluid distribution in the bundle due to the increased frictional pressure drop. By applying Computational Fluid Dynamics to real industry cases, the level of improvement is demonstrated.

The findings are rounded up by an industrial case study comparing multi-pass shells in series, two pass F shell and hiTRAN® enhanced single pass designs. The comparison calculations were performed using HTRI Xchanger Suite® and CALGAVIN’s hiTRAN® plug in. In addition, HTRI Xchanger Optimizer® was used to evaluate reliably the cost implications of the different design.

Furthermore, in 2-phase heat transfer applications, hiTRAN® Thermal Systems is used to optimize industrial heat exchangers. This will be demonstrated with a comprehensive case study, where a large feed/effluent exchanger (Texas tower) is retrofitted on the tube side with hiTRAN® elements. The goal was two fold, an increased product throughput without the need to replace the heat exchanger and large energy savings with correspondingly large CO2 reduction. Paraxylol is evaporated and superheated to around 327°C on the tube side before entering the fired heater for meeting reaction temperature in the reactor. After leaving the reactor the superheated vapour is then condensed on the shell side of the feed effluent exchanger transferring energy to the tube side.
A detailed analysis of the tube side flow and boiling regimes was undertaken using Aspen and HTRI design software. The results of the thermal and hydraulic results were then used to design a bespoke hiTRAN® insert geometry. In this case, two distinctive zones are present, the 2-phase region with good heat transfer and relatively high incremental pressure drop. Here an insert with low packing density was selected. In the subsequent superheated region, the heat transfer dropped (which is expected for gas flows) and the associated incremental pressure drop was lower. For that reason, a higher packed insert with higher enhancement characteristics were selected.

After in-situ installation during a planned refinery shut down, the measured performance increase was very much in-line with the predictions of the heat transfer and hydraulic model. The throughput was increased by around 18% and the heat recovery increased by around 3 MW. Since installation this equates to around 80,000 t of CO2 savings.