Issue 149 December 2002
Quad Bench Power Supply

Start • The Analog Core • The Zetex ZXCT1009 • An Ideal Isolator • MCU and User Interface • Firmware • Sources and PDF

THE ZETEX ZXCT1009

You can monitor the current drawn from the power supplies in two ways. Both methods involve inserting an accurate low-value resistor in series with the power supply output, and then measuring the voltage drop across that resistor. A measure of the current drawn then will be equal to the voltage drop/resistor value. If that resistor is placed in series with the negative output terminal of the power supply, the resulting voltage drop will be referenced to the power supply’s common terminal. This makes it easy to measure with an ADC (or DPM) that is powered by, and referenced to, the power supply’s common terminal.

The downside of this method is that whatever voltage is dropped across, this current sense resistor is lost (i.e., the load gets a little less voltage than the power supply thinks it is providing, and you see an inflated reading on the voltage meter).

Alternately, you can place the current-monitoring resistor in series with the positive output terminal of the power supply. Then, the voltage feedback network of the pass regulator can be wired to follow this resistor, eliminating the lost voltage problem that I described earlier.

This method, however, introduces the main problem associated with the measuring of a small current-sense voltage riding on a large common-mode voltage: the power supply voltage itself. You can minimize this problem by using a high-quality instrumentation amplifier and precision-matched resistors, but they are somewhat costly. This second approach is called high-side monitoring.

In his lab supply project, Robert devised a clever circuit to compensate for the lost voltage problem that plagued the first method I described. In my design, I chose to go with the second approach—high-side monitoring.

I came to this decision after discovering a clever IC made by Zetex called a high-side current monitor. The ZXCT1009 is a three-pin device in an SOT23 package that converts the voltage dropped across a high-side current sense resistor into a current. This current is sent through a resistor to the power supply’s common terminal, providing an easy-to-measure voltage proportional to the current draw.

The problems of measuring the low sense voltage riding on the high power supply common-mode voltage are addressed inside the ZXCT1009; therefore, you don’t have to worry too much about this. Because the device costs roughly $1, it certainly beats designing in an instrumentation amplifier to perform this task.

However, the ZXCT1009 isn’t a universal solution to the current-sensing problem. It requires an input voltage of 2.5 V or greater, so you can’t easily monitor current if you want to run your power supply at voltages less than this. The maximum input voltage it can withstand is 20 V without additional circuitry. Neither limitation was a deal breaker for me, so I incorporated one of these devices in each power supply. My biggest concern was holding the tiny device steady while I soldered it to the PCB!

You may want to consult the Zetex datasheet for more information, but the only other detail I’ll mention is that the device produces 10 mA for every 1 V dropped across the current sense resistor. I had 1-W, 1% 5-W resistors in my junk box, so that’s what I used for the current-sense resistors in all three supplies. This didn’t waste too much of the power supply’s voltage capability.

The lower-current floating supplies used a 220-W resistor to convert ZXCT1009’s output current into a voltage. The higher-current, ground-referenced supply has a fitted 100-W resistor, and the MCU’s software performs the math that’s necessary to convert the ADC’s output into the correct current reading on the meter.