I recently bought and built Elenco’s XP-720 power supply. The main rationale for choosing it was that it provides both positive and negative power rails that are typically needed for audio applications. That and the affordable price and compact size helped too.

However, I am concerned about how “clean” the voltage sources will be as this also impacts audio applications. The spec states a max ripple of 5 mV(rms), which is around 14 mV(pp), but I wanted to see if that was true so purchased some 10W resistors to measure the amount of ripple when under under load, as well as how hot the device gets under high current draw, and if there is significant voltage drop on the outputs.

Never done any testing like this before, and while it was kind of fun figuring it out, didn’t want to damage anything or start a fire so only tested the positive 1.25V to 15V output set at 12V and increased the current draw in small increments. Started with a single 10W 50Ω resistor which creates a 0.24A current and then added 4 more 50Ω resistors in parallel, one at a time, to get 0.24A increments in current.

You can see the setup here:

The Amprobe meter on the far right is reading the current coming out of the supply. The Radio Shack meter in the middle is measuring the voltage, and the oscilloscope is measuring any ripples in the voltage. The Fluke on the left is measuring the surface temperature of the power supply’s heat sink (that has the voltage regulators attached to it).

Used leads with clips to add and remove resistors to the circuit as shown in this image:

So here were the readings for the first resistor: 50Ω

232 mA @ 11.99V. Heat sink is slightly warm. No visible ripple, but noise @ 10mVpp.

Two 50Ωs in parallel: 25Ω

464 mA and the voltage has drop by 0.05V. Heat sink definitely hotting up. Kinda looks like ripple is starting to show. Noise still around 10mVpp.

Three in parallel: 16.66Ω

700 mA and voltage down 0.1V now. Heat sink is hot. Ripple starting to be visible but is still only 10mVpp.

Four: 12.5Ω

928mA and voltage drop is 0.17V. Ripple more pronounced but still only 10mVpp. The frequency is 120Hz.

Finally all five: 10Ω

1.148 A with total of 0.3V dropped. Heat sink hot but not bad at all. Ripple has more definition but still not greater than 10mVpp.

Definitely pleased with those numbers. Unless I’m doing something really wrong it looks like this power source is super clean.

Will do a more rigorous test soon with both rails pulling current at the same time.

11/25/21: Update

I knew there would be more to testing a power supply’s ripple and noise. Crikey!

Not sure sub 10mV accuracy is needed in my case but will definitely try to improve my methods in the next tests.