A solar cell characteristics plotter

0 downloads 0 Views 392KB Size Report
Feb 24, 1981 - (vi) Operates with any standard XY recorder with 0-10 V input. 3 Functional description of circuit. Figure 2 shows a block schematic diagram of ...
J. Phys. E: Sci. Instrum., Vol. 14, 1981. Printed in Great Britain

A solar cell characteristics plotter

Y W Lam Department of Electronics, Chinese University of Hong Kong, Shatin, NT, Hong Kong Received 24 February 1981 Abstract This paper describes a solar-cell characteristics plotter with its associated sample holder. The plotter has facilities for both manual and auto-plotting with built-in current limit against damage of the solar cell under test. Either current or power output can be plotted as a function of terminal voltage and there is provision for marking the maximum power point on the curve. The sample holder allows the temperature of the solar cell to be varied over a range from approximately - 188 to 250°C and permits investigation of other properties of materials used in the making of the solar cell, e.g., the antireflection layer. If temperature variation is not required the holder can be water cooled to prevent overheating of the solar cell during test. The plotter can also be used for the investigation of other semiconductor devices, and is ideal for use in small research and development laboratories.

while the former is only suitable for terrestrial simulation (AMI). A reference cell is used to set the level of illumination. The reference cell is calibrated at a central testing authority; hence, essentially all cells are tested under the same controlled illumination level ERDA (1977). It is essential that the reference cell is made from the same material as the actual cell under test so that they have similar spectral response. It may be mentioned, in passing, that the spectrum of a tungsten filament lamp is a function of the temperature. Thus it is advisable to choose a temperature that gives the best simulation of the sun’s terrestrial spectrum and then adjust the distance of the cell to the source until the correct level of illumination as determined by the reference cell is obtained. As regards the spectral response, unless the spectrum of the source is accurately known, it is more reliable for the measurement to be carried out by reference again to the calibrated reference cell. Moreover, since the response of cells to light is not linear with intensity variations more accurate measurement may be obtained by biasing the cell with white light of the appropriate spectrum and then measuring the incremental response to a small superimposed alternating component of monochromatic light. In this paper we describe a solar-cell characteristics plotter and its associated sample holder for current-voltage and power-voltage measurements. 2 Principle of measurement and facilities required For accurate measurement of current and voltage the fourpoint probe technique Valdes (1954) is used, i.e. separate pairs of contacts are provided for current and voltage measurement. This is important as resistance due to the connecting wires and contacts could amount to 1 L2 if special care was not taken, and at a current of 100 mA the voltage measurement would have an error of 0.1 V. The IV characteristics would thus be distorted in the same way as if the cell had a series resistance of 1 L2.

tight

!1gnt ::

1 Introduction A solar cell is generally characterised by its open-circuit voltage Vac, short-circuit current Zsc, fill-factor FF, and spectral response SR. To obtain these parameters, measurements are usually made in two steps. The first step is to measure the current-voltage characteristics of the solar cell at a specified temperature both in the dark and under illumination. The second step is to measure the short-circuit current of the cell per unit power content of incident monochromatic light as a function of the wavelength of the light. In order that meaningful measurements can be made the l e d of illumination must be accurately specified. One way to do this is to carry out the measurements at a specified time, say noon, in a day when there is a clear sky. Even then the measurement cannot be very accurate as there is a seasonal variation of the sun’s position in the sky and various gases in the atmosphere, particularly water vapour, will affect the precise spectral content of the sunlight. The approach now used by many laboratories is to use a simulated source which approximates the spectrum of sunlight. This can be a halogen-filled tungsten-filament projector lamp with dichroic reflectors which reflect only uv and visible light, or it can be a xenon light with a proper filter to simulate the airmass condition required. The latter provides a closer simulation of the sunlight spectrum under any airmass condition required as this is merely determined by the filter used

0022-3735/81/111302+04 $01.50 0 1981 The Institute of Physics

II

t

I/

/

- I

I

I

Figure 1 Principle of measurement.

Figure 1 shows the basic principle of IV measurement. In essence when the cell is under illumination, a photocurrent is generated which passes through the external load. A voltage is thus developed across this load which biases the cell in the forward direction. If the load RL= CO, the voltage developed is the open-circuit voltage. If the load RL= 0, the current measured is the short-circuit current. By the compensation theorem the load may be replaced by an EMF of the same magnitude and direction as the voltage developed across it. Thus the measurement may be carried out by actually applying a voltage across the cell under illumination and plotting the current in the circuit as a function of this voltage. VOcis then obtained from the plot at 1=0 and Iscat V=O.

A solar cell Characteristicsplofter X amplifier

ter

P Maximum c u r r e n t s e t t i n g 13 ranges1

0

Break Gate 1

Sweep genera tor

S+op sweep

~

Zero cdrrent setting

0

Point marking

detector marking duration genera tor

Figure 2 Block schematic of plotter.

The following facilities are considered essential. (i) IV plotting: Manual : - 7 V (presetable) to 1 V (wider sweep if desired for other types of solar cell). Current in three ranges: 0-2 mA, 0-20 mA 0-200 mA, (higher current ranges if desired for cells of larger area). Auto : -1 V (presetable) to a voltage at which I= 0. Current ranges as for Manual. (ii) Current limiter at maximum of range to prevent damage to cell under test. (iii) Adjustable sweep speed for Auto. (iv) Power-voltage plotting (Manual and Auto). (v) Maximum power output point indication on IV curve. (vi) Operates with any standard XY recorder with 0-10 V input. 3 Functional description of circuit Figure 2 shows a block schematic diagram of the solar-cell characteristics plotter. Both manual and auto modes are possible. In the manual mode the voltage applied to the solar cell may be varied from about - 7 V to about 1 V. The negative limit may be adjusted to a higher value if required, but - 7 V is considered sufficient for the dark ZV measurement of most solar cells, and is more than sufficient for the illuminated IVmeasurement of all solar cells. In the automode, the rate of voltage sweep is adjustable by a simple potentio-

+

meter (not shown in figure 2). The automode is used only for illuminated ZV plotting and as such the range of voltage sweep is set between - 1 V to 1 V. In addition a circuit is incorporated which terminates the voltage sweep when the current output from the solar cell reverses its direction, i.e., when it goes through zero. To prevent overload damage to the cell (e.g., as the result of accidental application of a high forward bias), a second circuit is incorporated which cuts off the bias to the cell when its current exceeds the top limit of the current range chosen (i.e. at 2mA, 20mA, or 200mA.) The two voltage terminals from the solar cell are connected to the X-amplifier whose output feeds the X input of the X Y recorder. The current output goes to the input of a currentvoltage converter whose output feeds the Y input of the X Y recorder. There is provision for the insertion of an ammeter to calibrate the Y axis of the recorder; similarly an appropriate voltmeter may be connected to either the input or the output of the X-amplifier to calibrate the X axis of the recorder. The outputs from the X amplifier and the V-Z converter are fed to a multiplier the output of which may be connected to the Y input of the X Y recorder if it is required to plot the power, i.e. the product IV, instead of the current, as a function of voltage. The multiplier is followed by a peak detector which operates by comparison of the voltage at the output

+

1303

Y WLam of the multiplier with that which occurred a short moment ago. After a maximum is detected, the following controls will come into action for a small fraction of a second; the actual duration may be adjusted to suit the speed of the X Y recorder. (i) In the manual mode gate 1 will hold the line voltage so that further manual variation of applied voltage will not alter the bias. (ii) In the auto mode the sweep will be stopped and the bias to the cell will remain constant. (iii) A small voltage will be applied to the X input of the XY recorder so that the pen will be effectively drawn back a little. (iv) The Y input of the XY recorder will be momentarily shorted to ground. The above actions will cause a kink in the IV curve at the maximum power point. The complete circuit of the plotter including the multiplier and its associated peak detector and controls is available on application to the author. 4 Sample holder and temperature adjustment The solar cell test holder consists of a copper platform of dimensions (in mm) 70 x 80 x 6, as shown in figure 3. Tunnels of 4 mm in diameter are drilled inside the platform as indicated. The entrances to some of these tunnels are then sealed off so that when cold air is pumped in at the inlet it travels through these tunnels and comes out at the outlet. The cold air comes from a heat exchange coil which is made of copper tubing, as shown diagrammatically in figure 4. The coil is immersed in a cold bath of ice or liquid nitrogen, depending on the temperature required, and the other end of the coil is connected to a nitrogen cylinder. The temperature of the plat-

7

A l l dimensions in mm 8 0 ____I

Copper plat form Tunnels inside Capper platf Air outlet

Handle-

\\

I

6

-7i r---

in

Lic;uid nitrogen

Copper heat exchcnge c o i l

1-1

Cewar

Figure 4 Heat exchanger and cold bath.

form can be regulated by varying the speed of the air flow. With a nitrogen bath, the temperature of the platform can be lowered to about - 1 8 8 T , provided that the platform is enclosed in an insulating box made of, say, polystyrene. This facility is useful for studying the characteristics of solar cells under low-temperature conditions. If the only facility required is to cool the solar cell so that it does not overheat during prolonged test, water cooling can be used instead. Flat heater pads (not shown in the diagram) are mounted at the bottom of the platform the temperature of which can be raised to about 250'C. The solar cell is held down to the platform by suction through five suction holes of 2 mm in diameter. Small samples such as 2 0 m m x 2 0 m m will only cover the three main suction holes, and the other two holes can be covered by putting a piece of soft plastic or similar material over them. For larger cells, all of the five suction holes will be utilised. The probes are fixed in an insulating probe holder which is pivoted to two mounts, one on each side of the holder. A small spring (not shown in diagram) is attached to the probe holder so that a certain amount of pressure is exerted on the solar cell contact when the probes are lowered. In addition a small cam (not shown in diagram) is fixed to the probe holder in such a way as to allow the holder and its probes to remain in an up position when lifted from the solar cell. The probes are made of phosphor bronze wires of about 1 mm diameter. Both the probes and the platform are nickel plated to minimise contact resistance due to oxidation. For higher current measurements, thicker probes with a larger contact area should be used. 5 Display Any X Y recorder with a high input impedance can be utilised for the display of the ZV and PV characteristics. For fast bias-sweep rates in the auto mode an oscilloscope can be used for the display. An ammeter inserted is used to calibrate the Y channel and a voltmeter connected across the input or output terminals of the X amplifier may be used to calibrate the X channel of the display unit. Figure 5(a) shows a typical dark IV curve in the manual mode of operation, figure 5(b) shows an illuminated IV curve under AMI conditions in the auto mode of operation, and figure 5(c) shows the P-V curve of a commercial pn junction solar cell of 20 mm x 20 mn in dimensions.

',

O @ @

Figure 3 Solar cell sample holder.

1304

Air o L t

6 Accuracy and concluding remarks The accuracy of the ZV plotter depends on the linearity of the amplifiers, particularly the ZV converter. Typically 3 % accuracy is easily achieved. For PV plotting, accuracy depends

120 100

-a +

1

curve provided that the peak power point can be automatically detected. A small error is inherent in the identification of the peak power point. This comes about because action can only be taken until the signal has actually passed its maximum. This amounts to an error of about 2 % in the position of the peak, and the problem can be overcome by employing a delay circuit, i.e. the signal fed to the peak detector is taken before the delay circuit while the actual power signal to the Y channel of display is taken after the delay

[n)

L

80-

0

I

20 -

,

1

1

I

I

[

circuit current directly. In our laboratory a halogen-filled tungsten-filament lamp is used which is fed from a highly stabilisedpower supply. Some lominutes after switch on should be allowed for the temperature and thus the resistance of the lamp to stabilise before actual measurements are carried out. If necessary, this problem of illumination variation as a result of resistance change of the lamp can be overcome by a feedback circuit which takes its control from the short-circuit current of a small solar cell mounted permanently to the test platform, thereby acting as an illumination monitor. In conclusion it may be remarked that the plotter described above allows rapid measurement of the solar cell characteristics both in the dark and under illumination. It is assembled from essentially standard components, provides graphical output, and, if required, may also be used to plot ZV curves of other semiconductor devices. The plotter is basically intended for the measurement of small solar cells in their prototype and is ideal for use in small research and development laboratories.

'

ibl

40 a

-

I

E

i60L

3 U 0

8 80-

-1

100-

12C"

0

I

Terminnl voltnge

01

oz

0.3

Acknowledgment The author wishes to thank K N Fu and D W Chan for their assistance in the construction of the plotter.

[v) 04

0.5 I

,

0.6 !

References Report ERDA/NASA1022-77/16 1979 Valdes L B 1954 Resistivity measurements on germanium for transistors Proc. IRE 42 420

+ 2

n L

a

3 a 0

40 -

Figure 5 (a) ZV (dark) characteristics of a pn-junction solar cell; (b),ZV (illuminated) characteristics of the same cell; (c) PV characteristics of the same cell.

1305