A Hohenheim type solar tunnel drier was redesigned, fabricated and installed at the Department of Farm Power and Machinery, Bangladesh Agricultural University, Mymensingh, Bangladesh. The drier basically consisted of a plastic sheet covered flat plate solar air heating collector, a drying tunnel unit, two dc fans and a 40 W photovoltaic module. The drier was 20 m long and 1.80 m wide. The solar collector unit was connected in series with the drying tunnel . The collector and drying chamber were made of plain metal sheets and wooden frames in a number of small sections and were joined together in series. These sections can be opened easily for transportation from one place to another. Glass wool was used between the two metal sheets at the bottom of the drier as an insulation material to reduce the heat loss from the bottom of drier. The collector was painted black to facilitate absorption of solar radiation. The drying area of the drier unit was same as that of the collector. Both the collector and the drying units were covered by 0.2 mm thick transparent UV stabilised plastic sheet. The plastic sheet was fixed on the collector side of the drier to the metal frame using U-type aluminium channel and rubber rope. At the drying unit one end of plastic sheet was fixed to a metal tube, which allows rolling Nomenclature db dry basis DR drying rate, kg kg1 h1 G solar flux density, W m2 M moisture content kg kg1 (dry basis) T temperature, C wb wet basis 86 M.A. Hossain, B.K. Bala / Solar Energy 81 (2007) 85–92 Author's personal copy of the plastic sheet up and down for loading and unloading of the drier. To prevent the entry of water into the drier during rain, both lateral ends of the plastic cover were fixed at 15 slope. A 40 W solar module was installed at the inlet of the solar collector as a power source to operate the fans, which supplied air over the product. The whole system was placed horizontally on tables made of iron angle frame 0.8 m above the ground floor for ease of loading and unloading of the products. Plastic net of 2 mm · 2 mm size was used as a tray and the tray was placed 75 mm above the floor of the drier. The drier was installed at a place free of shade, particularly for the period of 8:00 a.m. to 4:00 p.m. 3. Experimental procedure Three experimental runs in full load conditions were carried out for red chilli and another three for green chilli during the period of February–April in the years 1999 and 2000. The whole pods of red and green chillies were water blanched in hot water at about 85 C for 3–4 min. The blanched chilli was then spread on a bamboo mat to drain out excess water. After the draining out of excess water and cooling to ambient temperature, the chilli was weighed and spread out over the tray in a single layer in the solar tunnel drier. To compare the performance of the tunnel drier with that of conventional sun drying, three control samples of blanched chilli and another three samples of unblanched chilli were also placed on trays in a single layer beside the drier in the open sun. Drying was started after completion of the loading, usually at 9:00 a.m. and discontinued at 4:00 p.m. Weight loss of both the samples in the solar tunnel drier and the control samples in the open sun were measured during the drying period at 1 h interval with an electronic balance (Model CT1200-S, OHAUS Corporation, Florham, USA: accuracy ±.01 g). In the afternoon, after 4:00 p.m., the samples in the solar tunnel drier were kept in the drier and the control samples were kept in a room at ambient conditions. These control samples were again put out in the sun next morning usually at 9:00 a.m. Then both the solar and sun drying samples were subjected to dry under the same weather conditions. A K-type thermocouple (Chromel–Alumel) was used to measure the drying air temperature along the flow direction of the air inside the collector and drier at seven fixed points. A solar meter (Model 776, Dodge Products, Houston, Texas, USA: accuracy ±1.5%) was used to measure the solar radiation at the position of the PV module. Relative humidity and temperature of the ambient air were measured with a digital humidity/temperature meter (Model Fluke 51, John Fluke MFG, Co. Inc., USA: accuracy ±2.5%). Velocity of drying air was measured with a vane type anemometer (Model Taylor 3132, Taylor Instruments, Toronto, Canada: accuracy ±.01 m/s) at the outlet of the drier. The ambient temperature, ambient relative humidity, temperature at seven points in the drier, relative humidity at the inlet and outlet of the drier, air flow rate at the out let of the drier, solar radiation, generated voltage and flow of current of the PV module were recorded at 1 h intervals during the solar drying of chilli. The moisture content of the chilli sample was measured at the starting and end of each run of the experiment by drying the samples in an air ventilated oven at 105 C for 24 h. After completion of drying, the dried chilli was collected, cooled in a shade to the ambient temperature and then sealed it in the plastic bags. About 80 kg of fresh red and green chillies were dried to about 20 kg. 4. Colour measurement The hot chilli pods were destalked, sliced longitudinally into two halves and the seeds were removed from the placenta. The sliced pods were oven dried at 58–60 C for 2 days and ground using a grinding mill (Model 1/T-204, BU ¨ Hler, Goldon) equipped with a 1-mm screen. The chilli powder was sealed in plastic bags and stored at 20 C until processed. The colouring strength of chilli was determined by the internationally accepted EOA (Essential Oil Association of USA) method. The EOA method based on the absorbance of a 0.01% w/v solution of the extract in acetone at 458 nm and multiplied by 61,000, gives the EOA colour value (Verghese et al., 1992). The absorbance was measured using a UV spectrophotometer (Model UV- 1201, SHIMADZU, Japan). 5. Pungency test Four grams of hot chilli powder was extracted with acetone till a colourless acetone solution was obtained. The volume was then made up to 100 ml with acetone. The extract was kept 3 h at room temperature. After 3 h, 5 ml of acetone was taken in a beaker and heated on a water bath till fully dry. To this, 5 ml of 0.1 N NaOH solution Fig. 1. Solar tunnel drier: 1. air inlet; 2. fan; 3. solar module; 4. solar collector; 5. side metal frame; 6. outlet of the collector; 7. wooden support; 8. plastic net; 9. roof structure for supporting the plastic cover; 10. base structure for supporting the drier; 11. rolling bar; 12. outlet of the drying tunnel. of the plastic sheet up and down for loading and unloading of the drier. To prevent the entry of water into the drier during rain, both lateral ends of the plastic cover were fixed at 15 slope. A 40 W solar module was installed at the inlet of the solar collector as a power source to operate the fans, which supplied air over the product. The whole system was placed horizontally on tables made of iron angle frame 0.8 m above the ground floor for ease of loading and unloading of the products. Plastic net of 2 mm · 2 mm size was used as a tray and the tray was placed 75 mm above the floor of the drier. The drier was installed at a place free of shade, particularly for the period of 8:00 a.m. to 4:00 p.m. Three experimental runs in full load conditions were carried out for red chilli and another three for green chilli during the period of February–April in the years 1999 and 2000. The whole pods of red and green chillies were water blanched in hot water at about 85 C for 3–4 min. The blanched chilli was then spread on a bamboo mat to drain out excess water. After the draining out of excess water and cooling to ambient temperature, the chilli was weighed and spread out over the tray in a single layer in the solar tunnel drier. To compare the performance of the tunnel drier with that of conventional sun drying, three control samples of blanched chilli and another three samples of unblanched chilli were also placed on trays in a single layer beside the drier in the open sun. Drying was started after completion of the loading, usually at 9:00 a.m. and discontinued at 4:00 p.m. Weight loss of both the samples in the solar tunnel drier and the control samples in the open sun were measured during the drying period at 1 h interval with an electronic balance (Model CT1200-S, OHAUS Corporation, Florham, USA: accuracy ±.01 g). In the afternoon, after 4:00 p.m., the samples in the solar tunnel drier were kept in the drier and the control samples were kept in a room at ambient conditions. These control samples were again put out in the sun next morning usually at 9:00 a.m. Then both the solar and sun drying samples were subjected to dry under the same weather conditions. A K-type thermocouple (Chromel–Alumel) was used to measure the drying air temperature along the flow direction of the air inside the collector and drier at seven fixed points. A solar meter (Model 776, Dodge Products, Houston, Texas, USA: accuracy ±1.5%) was used to measure the solar radiation at the position of the PV module. Relative humidity and temperature of the ambient air were measured with a digital humidity/temperature meter (Model Fluke 51, John Fluke MFG, Co. Inc., USA: accuracy ±2.5%). Velocity of drying air was measured with a vane type anemometer (Model Taylor 3132, Taylor Instruments, Toronto, Canada: accuracy ±.01 m/s) at the outlet of the drier. The ambient temperature, ambient relative humidity, temperature at seven points in the drier, relative humidity at the inlet and outlet of the drier, air flow rate at the out let of the drier, solar radiation, generated voltage and flow of current of the PV module were recorded at 1 h intervals during the solar drying of chilli. The moisture content of the chilli sample was measured at the starting and end of each run of the experiment by drying the samples in an air ventilated oven at 105 C for 24 h. After completion of drying, the dried chilli was collected, cooled in a shade to the ambient temperature and then sealed it in the plastic bags. About 80 kg of fresh red and green chillies were dried to about 20 kg. 4. Colour measurement The hot chilli pods were destalked, sliced longitudinally into two halves and the seeds were removed from the placenta. The sliced pods were oven dried at 58–60 C for 2 days and ground using a grinding mill (Model 1/T-204, BU ¨ Hler, Goldon) equipped with a 1-mm screen. The chilli powder was sealed in plastic bags and stored at 20 C until processed. The colouring strength of chilli was determined by the internationally accepted EOA (Essential Oil Association of USA) method. The EOA method based on the absorbance of a 0.01% w/v solution of the extract in acetone at 458 nm and multiplied by 61,000, gives the EOA colour value (Verghese et al., 1992). The absorbance was measured using a UV spectrophotometer (Model UV- 1201, SHIMADZU, Japan). 5. Pungency test Four grams of hot chilli powder was extracted with ace tone till a colourless acetone solution was obtained. The volume was then made up to 100 ml with acetone. The extract was kept 3 h at room temperature. After 3 h, 5 ml of acetone was taken in a beaker and heated on a water bath till fully dry. To this, 5 ml of 0.1 N NaOH solution was added followed by 3 ml of 3% phosphomolybdic acid solution and was kept at room temperature for 1 h. Finally optical density values were measured at 650 nm with the help of a UV spectrophotometer. The value of optical density is considered to be the pungency index of hot chillies (Mangaraj et al., 2001). Different values of optical density were obtained for different chilli samples. The sample for which the optical density was higher is considered to contain more capsaicin or more pungent.The colour values and pungency indices of solar, improved sun dried and conventional sun dried chillies were statistically analysed using randomised block design (RBD). The data of colour values and pungency indices of red and green chillies obtained experimentally in the year 1999 and 2000 were analysed by analysis of variance using the software SPSS 9.0. The mean differences of colour values and pungency indices were graded by Duncan’s Multiple Range Test (DMRT).