Hello all,

I have 2 almost similar PCBs for voltage and current measurement. I have used 230VAC from the outlet to test them both.
The circuit in both boards is similar which is just a voltage divider to reduce the input voltage (see attached image).


but I got some noise in one of them even though I am applying the same conditions as seen below.





I suspected that noise could be either voltage divider resistors noise (the Johnson noise), or ground/power rail noise because I am using a DC-DC converter module to power the OPAMP (+15/-15V), and these power modules are known to have higher noise.

What could be the reason that this noise got coupled to the signal? and how can I get rid of it?

Thank you,

 

Mains wiring picks up LOTS of noise and this can vary widely with time. Did you compare measurements of both boards at the same time?
The two capacitors in series across the two 300k resistors will pass unwanted high frquency noise. Why do you need them?

 

Mains wiring picks up LOTS of noise and this can vary widely with time. Did you compare measurements of both boards at the same time?
The two capacitors in series across the two 300k resistors will pass unwanted high frquency noise. Why do you need them?

Yes, I have tested both boards at the same time.
as for the capacitors I have no idea why they are used but both boards have them so if they introduce high-frequency noise then i should've got the same noisy results!
maybe something else is causing the noise?

 

At 1 GHz bandwidth and/or 100 MSa/s, you could be picking up Russian propaganda radio, Voyager I satcom, or every radio station within a 50 mile radius.

But seriously, at that bandwidth, millimeter differences in wiring layout will appear in the signal.

A 1 uF capacitor has three inductors associated with it, one for each lead and one fir its internal construction. These increase the capacitor's series impedance at high frequencies, decreasing its filtering effectiveness.

The ground clip on the scope probe is a loop antenna.

Others will chime in on noise sources and control techniques.

almost similar

Photo of the circuit construction - ?

ak

 

as for the capacitors I have no idea why they are used

Usually, the capacitors are the main series impedance in attenuating the amplitude of the mains signal. For example, at 50 Hz a 1 uF cap has an impedance of approx. 3.2K ohms. But your caps are so small, their impedance at 50 Hz is 318 Gohm. That's a lot. This means that the 2 x 300K resistors are doing all of the attenuator work, and the caps create a highpass filter with a corner frequency of 53 kHz. That is low enough to pass conducted line noise from every switching power device in the area - computers, cell phone chargers, LED bulbs, microwave ovens, TVs, game consoles, etc.

Offline power circuits, even one as low-power as yours, are inherently dangerous, often because the behavior of the capacitors is misunderstood. In your case, the caps serve no useful function, and in fact make the circuit worse - unless the intent is to capture high-frequency noise. If the circuit is not your own design, can you post a link to its source?

ak

 

“or ground/power rail noise because I am using a DC-DC converter module to power the OPAMP (+15/-15V),”

I just completed a project which used a 5V to a +/-15V converter and had similar noise problems. I solved them by modifying the common-mode filter as shown in the data sheet. I don’t have the schematic available right now, give me some time and I will post the solution that worked for me.

 

At 1 GHz bandwidth and/or 100 MSa/s, you could be picking up Russian propaganda radio, Voyager I satcom, or every radio station within a 50 mile radius.
But seriously, at that bandwidth, millimeter differences in wiring layout will appear in the signal.

The same bandwidth is used for both boards, so why doesn't the other on doesn't pick up as much noise?

a millimeter differences in wiring layout will appear in the signal.

OK, this might be it!!.
Although the 2 boards have similar parts, schematic, and size, they differ in layout a little bit.

a photo of the input section of the board is attached (I am measuring at blue circle).

 

“or ground/power rail noise because I am using a DC-DC converter module to power the OPAMP (+15/-15V),”

I just completed a project which used a 5V to a +/-15V converter and had similar noise problems. I solved them by modifying the common-mode filter as shown in the data sheet. I don’t have the schematic available right now, give me some time and I will post the solution that worked for me.

I suspected that the dc-dc converter module might be the problem because they did state in the datasheet that it has 75mV ripple & noise.

 

I suspected that the dc-dc converter module might be the problem because they did state in the datasheet that it has 75mV ripple & noise.

That is noise on its output voltage, measured under very specific conditions. Your results absolutely will vary, and for the worse.

There is no DC/DC module on your schematic.

ak

 

That is noise on its output voltage, measured under very specific conditions. Your results absolutely will vary, and for the worse.

There is no DC/DC module on your schematic.

ak

Yeah, it isn't shown in the schematic..
a 15Vto+/-15V converter module used to isolate the power rails so the opamp is powered by +/-15V and has different ground than the "digital part".
I might use an external power supply to power this part of the circuit to see if anything is changed!