Electronic "Watt-Tester" Power Transducer turns any Voltmeter into a Precision Wattmeter Brad O'Mara c.1992 Brad O'Mara The benefits of knowing more about what really runs our electrical world Ä watts of power Ä go largely unknown because of the difficulty in measuring them. Volts and amps are easy, but watts are new territory for many. You can't troubleshoot or search out your system's 'lost watts' if you can't measure them. Knowing actual output power can also help you negotiate a better deal when purchasing used 'bargain' priced modules. Low cost becomes high cost if power is lacking, regardless how well Isc or Voc check out. With the WattTester you'll never bring home a worn out 'sundog' module again. Power Meters Don't (Usually) Come Cheap For under $50 you can construct a portable 'power transducer' to instantly turn your trusty voltmeter into a precision power meter which works like industrial models priced upwards of a $1,000 or more. You can dial in any full scale (FS) reading you want by changing a few resistor values. From 1W for you micro-power designers, to more than 10KW for those with the largest array or inverter on the block, measuring power is fun when all you need is a current shunt and a voltmeter. Now any low cost digital multimeter (DMM) can measure watts with a precision of 0.10W per millivolt (W/mV) at an accuracy of 2 %. All of this from 3 smart chips and a 9 volt battery ! Why Watts Haven't Been Much Fun (Until Now) Measuring power is a complicated affair. Both voltage (V) and current (I) must be measured, recorded (or memorized), and then manually multiplied to compute wattage. Should sunlight, battery or load conditions fluctuate rapidly this becomes tedious if not impossible to do. Using two separate meters simultaneously helps, but the inherent need to manually multiply the I and V readings prevents getting an intuitive feel for what's going on at the moment of concern. Taking measurements, recording columns of data and then punching a calculator or setting up a computer spreadsheet to later calculate watts is too much work for most of us ... and with the WattTester, unnecessary. Theory of Operation Ä Or, How It Works Figure 1 shows at a glance how the WattTester works. It uses a Harris ICL8013 Analog Multiplier chip to multiply two 'signal-level' voltages (see Fig. 2) proportional to I and V. A gain stage amplifies the millivolt signal from a current shunt for the '8013's Xin input. Attenuation factor A reduces V for the Yin input. Scale factor K=0.1 is built into the '8013 because it's output is limited to ó+10V. Amplifier gain G and A are set by the readout conversion factor and the desired FS. Highest accuracy occurs if Xin and Yin see +10V at FS. For ease of use and to avoid skewing the multiplier's calculations (by weighting either voltage or current unevenly) A and G are set so Vout is a power of 10 of the measured power at FS. The schematic in Figure 3 shows how a true +15V split-voltage supply is obtained from a single 9V battery using the Linear Technology LT1026 Voltage Converter. It is a special device which eliminates need for an inductor or complicated switching circuitry by using an internal high- frequency oscillator to drive C1-C4 in a 'charge-pump' fashion. The Maxim MAX480 is a micropower precision Op Amp also powered off the +15V. Two quadrant '8013 operation that allows bidirectional (+) current flow without the need to reverse the shunt connections. The voltage input Vv+ must always be connected to a positive voltage relative to COMMON unless 4 quadrant operation (thus dual-polarity voltages) is also desired. C5 must then be a non-polarized type. The 480's key spec is input offset voltage,Vos. Typically 25 micro-volts, 80 times less than the 2 mV of a general purpose Op Amp, it's needed because Vos is amplified by gain G into a large error voltage. It is trimmed out by R9 but if Vos is too large to begin with its temperature coefficient will cause the error to quickly reappear and drift badly with temperature. The 480's 14uA bias current is also low enough for the '1026 to power the entire circuit from the 9V battery. Attenuation and Gain Selection Table 1 gives values for G and A. They depend upon the shunt resistance, maximum V and I, and FS. For example, with a 0.001ê shunt and a 1KW FS reading (i.e., @ P=1KW, '8013 Vout=10V, thus conversion factor=1,000W/10,000mV=0.1W/mV) the gain needs to be 1000. Potentiometer R5 varies gain approximately 100 to 1000. Before adjusting it first trim potentiometer R9 for Op Amp Vo=0V for 0V input by shorting together the I inputs, COMMON and Vi+. Then hook them to the shunt and adjust R5 so Vout=10V when maximum I is flowing. How To Use Your 'WattTester' Power Transducer Your WattTester is very versatile. You are free to choose any FS and V and I maximums you desire. The examples in Table 1 are based on a 1:1 ratio between measured power and the actual integers displayed on your voltmeter's LCD readout. This simplifies readings because you only need shift decimal places to read watts directly. Circuit variables can be changed, but be careful to make note of your conversion factor if you choose other than 1:1 lest you forget it while measuring power away from home. It would be embarrassing if your neighbor's brand new 50 Wp module tests 5 watts in full sun, or 500 watts in the shade ! Figures 4-6 show how to make various power measurements. Remember that the voltage across the transducer's current sense inputs, COMMON and Vi+, must always be low level, on the order of a few hundred millivolts or less. Diodes D1,2 are cross-connected across the inputs to offer a degree of protection should they become disconnected from the shunt. Adding a resistor ò 10Kê between R8 and the Vi+ input will increase protection but at the expense of requiring more gain G. Lastly, bear in mind the transducer's output voltage is bipolar. Negative voltages can mean positive power or vice versa. What's positive and negative power is only a matter of definition. You can choose simply by reversing your voltmeter leads. A negative reading occurs in Fig. 4 because the transducer sees a positive Voltage across Vv+ and COMMON but a negative voltage across Vi+ and COMMON. A positive reading occurs when the battery is discharging in Fig. 5 because both inputs see a positive voltage. Polarity then reverses when charging. The PV array power readings in Fig. 6 are negative unless you switch the leads to your voltmeter. However you hook things up, though, always remember that it's watts that do the work, or as Mr. Sun sez; " Volts are fine and Amps a perk, but ... it's Watts of mine that DO your work! " Access Author: Brad O'Mara, Outside Power Company, Midwest Lab Facility, 7477 Lakeshore Drive, Spirit Lake, IA 51360 ù 712-336-5045 ù Fax 712 - 336-5046 Pacific Northwest Address, 130 E. Main St. , Suite 325, Medford, OR 97501