3. Experiments
3.1. LED solar simulator
Four color LED are used, namely, red-R at 632 nm (LTL2F3VEKNT), green-G at 525 nm, TOL-50bUGdCTa-M4), blue-B at 468 nm (TOL-50aUBdCEa-ETB6) and white-W (LTW-2S3D7).
Five LED arrays, each at 227.5 mm 227.5 mm, are constructed and serve as light sources of simulators, namely (Fig. 3),
(a) Four single-color LED arrays. Each array consisting of 1024 LEDs has 64 parallel strings, each string has 16 LEDs in series with a current limiting resistor.
(b) One combined RGB array. It has 1024 LEDs. The array has 128 parallel strings, each string has 8 LEDs. Percentages of irradiance from red, green and blue LEDs to total irradiance are 35%,
33% and 32%, respectively.
Each LED array is placed 3 mm above a glass diffuser which is 30 mm over the test plane.
LED heat sinks and forced air cooling (between heat sinks and a glass diffuser) provide heat removal and maintain a constant temperature heat sink temperature of 25 C.
On the test plane, a solar cell and calibrated photo-detector (13DAS003) are placed.
A pulse power supply for an array is constructed from a high voltage DC supply (Xantrex XDC20-600) and a pulsing circuit.
Amplitudes of pulse signals can be continuously varied from 0 to 150 V with a pulse width of 10 ms and a period of 1 s.
Study is made on power from pulsed operation (at different amplitudes, pulse widths and pulse periods) and compared with continuous operation of LEDs.
This is to obtain the maximum power that LEDs can withstand before being damaged due to temperature rise.
For pulse operation, we find that 2-3 times of LED rated voltages and 8-10 times of rated currents are feasible with forced air-cooling.