A flow-through type solar vacuum tube (model JJR-T-2140) was used as a receiver. It comprises a metallic absorber tube made of Stainless Steel (SS 304/316) bounded in a high-strength Borosilicate glass envelope, as shown in Figure 2. Antireflective coating on the glass envelope is provided to reduce optical reflection losses and enhance the transmission of sunlight through the tube. The annulus space between the glass envelope and absorber tube is evacuated (3 × 10⁻⁴ Pa at 200°C) to avoid convection losses.

Aluminum Nitride selective coating is also applied at the absorber tube to improve absorption of solar energy and minimize infrared losses. In addition, special seals provided at the ends of the tube help maintain vacuum at elevated temperatures while expansion bellows accommodate uneven thermal expansion of metal and glass tubes. The detailed specifications of the ETR are given in Table 1.

A CPC is a non-imaging concentrating device that consists of two parabolic segments at its ends (Vu and Shin, 2016). The new double parabolic concentrator is made by joining two parabolas. The inner portions of parent parabolas are removed from the common point, making a cusp in the middle, as shown in Figure 3A. The outer portions are then tilted inward by 30° (which also defines the acceptance half-angle of the resulting CPC) to form a combination of two parabolic sections, as shown in Figure 3B.

The receiver tube is placed below the common focus point so that the axes of both parabolic segments are tangent to the inner metallic absorber tube, as illustrated in Figure 4A. A small V-shaped cavity is also formed at the cusp point to lower the receiver’s position further. After reflecting from the CPC reflectors, the incident solar radiations are directed to a region below the common focal point where the receiver is placed.

As the upper part of reflectors become almost parallel to the common axis of CPC and thus contributes very little to the concentration of solar rays. This orientational issue provides the opportunity to remove upper portions, thereby saving material and manufacturing costs at the expense of a slight decrease in the collector’s aperture width and concentration ratio, as illustrated in Figure 4B. Furthermore, as a result of truncation, the view field of the receiver is also increased, allowing the direct solar radiations to reach the receiver for more time, even outside the acceptance angle of the CPC. In this way, the overall daily operating time of the collector is also increased. In the present study, truncation level was selected to be 50% which resulted in about 7.5% reduction in the aperture width and the concentration ratio, while reflectors height was reduced to half of the original height.

The detailed specifications of the developed CPC reflector are summarized in Table 2. In addition, details of the ray tracing analysis and optical performance of the designed CPC troughs can be seen in (Akhter et al., 2017).

The 50% truncated CPC, which emerged as the suitable choice, were selected for manufacturing and experimental testing. The CPC collector module comprised a newly designed three double parabolic troughs paired with evacuated tube receivers, which were named East (E), Central (C), and West (W) troughs for comparison of output temperatures when placed in the north-south direction. The optical axes of side troughs were tilted outward by an angle of 15°, as shown in Figure 5, to increase the overall effective time of operation and smooth out the power output by maximizing solar energy collection in the forenoon as well as afternoon. The north-south arrangement of this configuration allows more than one trough to be active during the effective time of operation of the module. The central trough captures maximum radiations around noon. The outer troughs intercept more radiations in the forenoon and afternoon sessions when the incidence angle of solar radiations falls within the acceptance half angles of the outer troughs.

The concentrating-reflecting surface of the CPC troughs was manufactured using highly reflecting sheets. The ETRs were installed in the focus line of the CPC trough. The experimental test rig consisting of three evacuated tube receivers paired with modified CPCs was installed to investigate the thermal performance of single-phase thermosyphon in the medium temperature range. The header pipe, connecting pipes, and storage tank were carefully insulated to curtail thermal losses. The temperature at the inlet and outlet of all tubes and the header pipe and storage tank were continuously monitored and recorded by installing resistance temperature detectors. A one-way valve was also installed at the inlet side to prevent the possibility of flow reversal during the night-time or low insolation period. The schematics diagram of the experimental test rig is illustrated in Figure 6A, while the CAD model is shown in Figure 6B.

The complete module of the CPC collector was fixed on an adjustable base frame and the measuring instruments were installed at designated locations, as demonstrated in Figure 7. All thermocouples and temperature sensors have been calibrated before installation in the test rig. Since the developed CPC collector was aimed to supply solar thermal energy in the medium temperature range, Therminol-55 oil was employed as a heat transfer fluid instead of water to avert phase-change inside the evacuated tube receivers at higher temperatures. The heat transfer oil was filled into the CPC, and the test rig was exposed to solar radiation for experimental testing. Temperatures at selected locations were continuously monitored and recorded at 60 s intervals. The data was stored by a multichannel data logger for postprocessing and analysis of the experimental results.

The fabricated system was installed at the Solar Thermal Advanced Research Center of the Universiti Teknologi PETRONAS, Malaysia (4.386°N, 100.979°E) for experimental testing under outdoor tropical weather conditions. The real-time weather data, including direct and global solar irradiance, and ambient temperature, were recorded during the experimental testing. The direct solar irradiance was measured using a pyrheliometer (EKO MS-56, EKO Instruments Co., Ltd. 1-21-8 Hatagaya, Shibuya-ku. Tokyo, 151-0072 Japan) installed on a dual-axis tracking platform (STR-21G). The global solar irradiance was measured by employing a first-class pyranometer (EKO MS-402, EKO Instruments Co., Ltd., 1-21-8 Hatagaya, Shibuya-ku. Tokyo, Japan) mounted on the mainframe in the plane of the aperture of the CPC collector. Finally, the temperatures at designated places were measured by resistance temperature detectors (PT-100) attached to a multichannel data logger (Graphtec GL-820, Totsuka-ku, Yokohama 244-8503, Japan). Details of the main measured parameters and the specifications of the related measuring instruments are given in Table 3. In addition, the energy absorbed by the CPC collector was determined, and the performance of the CPC collector was evaluated to determine the feasibility of a single-phase thermosyphon in the CPC collector in a tropical climate.

Since the outlet temperature was intended to be in the medium range, Therminol-55 oil was used as the working medium due to its high boiling temperature. The oil was filled in the storage tank of the collector until it reached a level slightly lower than the top of each absorber tube. The Therminol-55 oil level in the closed-loop was ensured by installed a level gauge in the oil storage tank. When the CPC collector was subjected to solar radiation, the heat collected by the metallic absorber tube was transferred to the heat transfer oil inside the absorber tube, causing it to expand. Due to a change in density caused by solar heat absorption, the oil started overflowing from the top of the absorber tubes into the header and onto the storage tank. In the meantime, the colder fluid was pushed down in the return line due to its relative density difference and gravity effect, leading to natural circulation called “solar thermosyphon”. The hot fluid flowing into the storage tank raised the temperature inside the tank. The heated thermal oil, in turn, reached the inlet of the receiver tube, forming a closed-loop system. Thus, the thermal oil temperature was continuously increased unless heat was extracted from the storage tank by circulating water in the secondary loop.

The overall thermal efficiency of the single-phase closed loop thermosyphon system is defined as the ratio of the heat gained by the primary fluid to the incident solar radiance on the aperture area of the solar collector, given by the following equation (Kalogirou, 2014); η = Q u G t A c , (1) Q _u is the useful energy gain by heat transfer fluid, G _t is the total incident solar irradiance on the aperture of CPC, and A _c is the aperture area of the solar collector. The effective or total solar irradiance for a CPC is given by following equation (Kalogirou, 2014). G t = G b + G d C , (2) where G _b is direct or beam irradiance on aperture, G _d is diffuse solar irradiance and C is concentration ratio of the CPC.

In thermosyphon, flow is induced by a pressure difference created by density change which depends on the temperature gradient in the loop. For steady-state conditions, the total driving pressure head (due to buoyancy) is balanced by the frictional pressure head (determined based on Darcy’s equation). Assuming one-dimensional, steady and laminar flow, with linear temperature distribution along the tube, the momentum equation of the loop could be manipulated as under (Recalde et al., 2015): Δ ρ g H = 2 f L   m 2 ρ d A 2 , (3) where, ρ is the density (kg/m³); g is the gravity (m/s²); H is the vertical distance between the heat source and the heat sink (m); f is the coefficient of flow resistance; L total length of thermosyphon loop (m), d is tube diameter (m), A is the cross-sectional area of the tube (m²) and ṁ is the mass flow rate through the tube (kg/s). For small values of ΔT, the change in density between hot and cold fluids can be determined using Eq. 3 (Recalde et al., 2015). Δ ρ = ρ ¯ β Δ T (4) where β is the normalized thermal expansion coefficient of the fluid. By substituting the density difference in the momentum balance, Eq. 3, the equation for mass flow rate can be written as below: m = ( ρ ¯ ρ β g A 2 d 2 ) ( H Δ T f L ) (5)

Useful heat gained by the working fluid is determined by Eq. 6. Q u = m C p Δ T (6) where C _p is the specific heat capacity; ṁ is the mass flow rate; ΔT is the temperature difference of working fluid between the inlet and the outlet. Substituting the value of ṁ from Eq. 5, the energy equation for one dimensional single-phase flow can be written as below: Q u = B ( cos β t f ) 1 / 2   ( Δ T ) 3 / 2 (7) where β _t is the tilt angle of the collector measured from a horizontal plane and the parameter B is given by the following equation: B = C p ( ρ ¯ ρ β g d   A 2 2 ) 1 / 2 (8)

The accuracy of the experimental results is determined by estimating the uncertainty of the measuring parameters using the B-type method (Moffat, 1988; Chen et al., 2015). In the present experimental investigation, the major errors are related to solar irradiance measurements and temperatures. The errors in the collector area and specific heat capacity of the heat transfer fluid are assumed to be negligible for the determination of total uncertainty. The overall uncertainty of the system is estimated using the following equation. U η = ( ∂ η ∂ G u G ) 2 + ( ∂ η ∂ Δ T u Δ T ) 2 (9) where u _G and u _ΔT are the possible errors in the measurement of solar irradiance and temperature, respectively.

In the present experimental study, the errors in the measurement of solar irradiance and temperature (together with datalogger) were found to be 1.7 and 1.5%, respectively. Thus, the maximum uncertainty in the experimentally determined thermal efficiency was 5.34%.

The variation in the temperatures at selected collector locations is plotted in Figures 8, 9 for two different testing days, day 1 and day 2. The effect of outward tilting of side troughs can be seen in the rising temperature trends for a south-facing collector. The truncation of top portions of the reflectors further increases the overall operating time by enabling the receivers to intercept direct radiations beyond the acceptance angles. The incidence angle of solar radiation affects the energy absorbed by the collector, as the oil temperature in the eastern tube started rising earlier in the morning.

On both measurement days, the temperature in the central tube increased and reached the highest value near noon. Following this trend, the oil temperature in the western tube should have taken the lead in the afternoon. However, it remained slightly lower than the central tube due to a large decrease in solar radiation in the afternoon, typical in tropical regions. The average outlet temperature also dropped in the afternoon due to a large reduction in solar radiation. As a result, it became less than tank temperature, which could cause reverse flow in the evacuated tube receivers. However, flow reversal was avoided by employing a non-return valve in the return line. The maximum temperature achieved by the collector at the outlet was in the range of 115–120°C.

The maximum value of average and storage temperature is shifted from the middle of the day towards the afternoon due to the working fluid’s heat capacity and thermal inertia. As the thermal oil in the ETR reached up to a temperature of 90°C even on a day with low solar radiation, the data collected supported the hypothesis that the CPC collector can function effectively and deliver sustainable solar thermal energy at a medium temperature range under the tropical climate conditions.

The model presented by Kalogirou et al. (2013) for thermosyphon in a non-concentrating solar collector was modified to take into account the effect of concentration of solar radiation by including the aperture area of the developed CPC collector. The characteristic curve for thermal efficiency was determined using the experimental data collected during the selected days. Figure 10 shows the overall efficiency of the developed CPC solar collector operating in thermosyphon mode plotted against reduced mean fluid temperature (T _m ^* ) given by following equation: T m ∗ = T m − T a m b G t A c (10)

It can be observed that the zero-loss thermal efficiency of the developed collector, which represents the system optical efficiency, reached up to 70% during the test period. On the other hand, the thermal efficiency of the developed CPC solar collector slumped at higher inlet temperatures and low solar irradiance. The systems operating in thermosyphon mode are typical behaviour as the natural circulation of working fluid is quite low in these systems compared to the active mode (forced circulation).

The developed CPC solar collector was also tested using water as a working fluid. Figure 11 shows the experimental results of weather parameters and temperatures at designated locations of the collector recorded on day 3. The weather was mostly sunny on the selected measurement days with an average solar irradiance of about 800 ± 20 W/m² except for sometimes intermittent clouds, as illustrated in Figure 11A. The average outlet temperature of the water as working fluid reached the saturation temperature of water at atmospheric pressure at about noon. The maximum temperature in the tank reached 90°C, as an average of all testing days. The effect of tilting the side troughs is also evident in the case of water as a working fluid, as shown in Figure 11B. The outlet temperature of the eastern tube started rising earlier than the central tube, followed by the western receiver tube. Some fluctuations in outlet temperature of the western tube were observed after reaching the saturation limit of water which might be attributed to some reverse flow or boiling of the water.

The experimental results, using water as working fluid, demonstrate that phase change can occur in an evacuated tube receiver paired with CPC if used in thermosyphon mode at standard atmospheric pressure. This indicates that the loop should be pressurized if the water is used as the working fluid in a single-phase or some other heat transfer fluid with a higher boiling point (such as thermal oil) is suitable for a closed loop thermosyphon system operating in the medium temperature range.

The experimental thermal efficiency of the ETR paired with modified CPC is also compared with the thermal efficiency estimated by using the model proposed by (Kalogirou et al., 2013) following EN12975 standard for testing solar collectors. Figure 12 presents the comparison of both efficiencies under tropical climate conditions. It is observed that the results are closely matching at lower inlet temperatures and higher solar irradiance. However, significant improvement in the experimentally determined thermal efficiency is achieved at higher inlet temperatures and low solar irradiance. This improvement in thermal efficiency is attributed to relatively higher differential temperatures between the inlet and outlet of the receiver, which is achieved due to the concentration of solar radiation by CPC reflectors.