How to measure temporary drop in voltage with meter

Question

What's the easiest way to measure the period of time between the "peaks" when a 5V DC signal goes from high to low to high?

I'm trying to debug a power-loss problem, and one of the solutions I'm considering is decoupling the power input by adding a capacitor across the VCC and Gnd pins, but I need to know how long the power drop lasts in order to choose an appropriate capacitance.

I have a Fluke multimeter and a DS202 DSO, but I'm not familiar with using either for this type of measurement, nor can I find anything in their manuals that explicitly tell me how to make this type of measurement.

Answer 1

Normally you'd use an oscilloscope with a suitable bandwidth. For example, an inexpensive 50MHz digital scope would reproduce a 1usec drop fairly well. If that DS thing has a 1MHz bandwidth, then you should be fine if the drop is > 50usec.

Most digital scopes allow you to see what happened just before the trigger, so you should look at your manual to see how to do that. That allows you to trigger a single sweep from the drop in voltage and see the entire event displayed.

Answer 2

Your DSO is the best tool for this job.

Spend some time reading the instruction manual for your scope. Learn how to set up the trigger stage to NOT auto-run and set the trigger settings for: Falling Edge, DC-coupled, Trigger voltage just below normal DC voltage.

You should have no problem seeing the low-voltage transient when it occurs. Adjust the horizontal timebase so that you can see the leading edge of the transient as well as how fast the voltage recovers back to it's normal value.

Learning to use your scope is a very important step in your learning. With practice, you will find it to be probably the most useful instrument on your bench. It certainly is on my bench: I use the scope all the time for making even routine measurements on whatever I'm working on. My multimeter gets used when I need very accurate voltage readings or if I am measuring resistance values.

Answer 3

Contrary to the other opinions, I found my DSO is not able to capture the signal between peaks. It has a trigger function, and I could capture a single drop curve or a single rise curve, but I could find no way to do both, much less measure the precise time between them.

I ended up coding an Arduino to read the input on an ADC pin and latch on the drop and rise event and output the time over serial. That also gave me a much more accurate reading than estimating the value from a blurry oscilloscope graph.