Need to filter out noise (probably from power line)

I was measuring current from a device. I saw some EM interference in my oscilloscope. For example, one frequency is about 100kHz (an obvious sinusoid wave with 10us period on top of my signal; and there are also some higher frequencies). I was thinking may be it was from the power line (maybe some high power equipment near my lab generated such noise and the noise was coupled into the power line). I want to get rid of those noises. I noticed that some people were using RF chokes. Is it a good option? Or do you have any better suggestions? Thanks!

hi,

in order to get the accurate reading i would suggest you use inductor it will help a lot in this kind of application.

Pity you do not mention the amplitudes. As you possibly use a typical oscilloscope with 10 mV/div/. one of the problems is that some probes are not screened well enough, and on the 1 MOhm input impedance it is easy to see induced interference.

Measuring a current by a scope often involves "floating" resistors; this means the ground of the scope is not identical with the ground of DUT. Ground lines are often long and this is a good path for the interference.

Testing sensitive circuits requires corresponding precautions like screened enclosures, use of linear power supplies or batteries (also screened in a metal enclosure), etc.

Try to connect a resistor in your scope probe tip, and search around the tested circuit. 100 kHz is often generated by switching power supplies, dimmers and other devices connected to power lines.

Connect next a suitable coil ( from a small audio transformer, without its core)- you can sense a lot of nice stuff from around AC and DC power lines.

Thanks for your reply. Do you mean connecting a coil between the device and the oscilloscope? Will that affect the signal I am measuring?

By the way, can you please point me to some materials that explain why people use a "floating" resistor to measure current? I actually hear people talking about it, but don't know exactly why.

oscilloscopes often have ground currents floating on their chassis, so it is hard to tell if the signal you are seeing is real or not. Try looking with a different oscilloscope or a DC blocked spectrum analyzer. Try floating the chassis of the scope from earth ground. Try running your unit off of battery power, etc.

The coil I mentioned is a "search" coil, sensing magnetic field around lines and components. The coil is connected between ground and "live" pin of the scope probe.

The word "floating" means that to measure current in a circuit, the oscilloscope, voltmeter or another device is connected across that circuit, between points A and B, none of which can be connected to scope ground. If you are lucky, one of points A or B can be "AC grounded", so you can connect a suitable capacitor to scope ground. The capacitor prevents a DC short to ground and the other point is used to see what happens - if there is a resistor between points A and B, then the voltage is proportional to a current in the resistor.