How to measure DC offset on the mains voltage?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
DISCLAIMER: Please be advised that the following content in this thread is revolving around mains voltages which can be lethal, read the content in the link as follow before proceeding.
>>> https://www.diyaudio.com/forums/tubes-valves/announcements.html <<<

****************************************************************************************************



I come to think of the DC offset on the mains voltage and that a relatively small DC offset can easily make transformers upset and make louder humming, so I thought this sounds like a trivial task to find out with perhaps ones multimeter and so started to search the internet for an answer on how to do it, but it seems to be not that trivial after all.

I tried to measure the DC voltage with my multimeter set on 1000V but it didn't show anything, maybe either due to no DC offset present and/or my old multimeter isn't sensitive enough, and the next lower step on the multimeter is 200V which I don't dare to try as we have 230VAC mains voltage. While searching the internet I came across some text mentioning some multimeters can measure true RMS VAC with a DC offset, but it appears to be just a sum of both DC and AC (True RMS ac + dc =SQRT( ac^2 + dc^2)), so, how can one in the easiest pragmatical way determine (measure) the DC offset value on the VAC mains?


Tip 2: Measuring an AC Signal with a DC Offset | Keysight (formerly Agilent’s Electronic Measurement)
AC Measurements Using the Oscilloscope and Multimeter by Mr. David Fritz - PDF Free Download
Mains DC and Transformers
 
Last edited:
So I did draw a simple and quick mock-up circuit (using an online circuit editor Circuit Diagram Web Editor) which could aid in measuring DC offset on the mains using a multimeter, the multimeter is connected in DC mode and measures any voltage offset over the capacitor C.

R and C values are chosen so that the R is of sufficiently high value so that the major part of the voltage falls over R while near to zero voltage fall over C (that is when there is no DC offset on the mains) at 50-60 Hz, C should be a bipolar electrolyte capacitor, or two back-to-back polarized electrolyte capacitors, the values R = 100k, C = 1000uF in the attached circuit were only arbitrarily chosen for now.
One thing to consider with regards to the R and C values is that they have to accommodate for electrolytes leakage current.

Any opinion whether this could be a suitable circuit to use measuring DC offset on AC mains voltage ,and will I get a sufficiently good read out of the DC value? :)
 

Attachments

  • Measuring DC offset on mains voltage.png
    Measuring DC offset on mains voltage.png
    5.1 KB · Views: 678
A two-pole, symetrical filter would be safer and more effective:

attachment.php


Note that if your multimeter has a low internal resistance, you should take the 350K of the filter into account.

The filter leaves 100mV peak of 50Hz on its output, which should be no problem for any type of multimeter
 

Attachments

  • DCmains.png
    DCmains.png
    69.2 KB · Views: 3,056
What makes you suspect a DC offset on your mains? As the mains current to your home comes through various transformers, it would seem there is not a lot of opportunity for DC.

Why not contact your utility and ask if there is ever DC offset on their lines. In fact, if there were, I would wager they would come out and test for it.
 
In 2020 most of the DC is actually asymmetry in the AC power waveform caused by SMPS power supplies resulting in a DC offset. While old hair dryers and tea pots cause a real DC current difference. But hair dryers don't operate for long periods of time.
So is the problem 24 by or just for short periods? Short period problems are easier to correct.
 
Safety note on above examples- whatever else you do, add a large-value discharge resistor across the output side - say 1Megaohm - just to ensure once you unplug there is no residual HV charge on the filter cap(s) that might sting

(yes I know there shouldn't be, given it is a lowpass filter: but I'd not want anyone stung.)

Note - this means you'd need to scale the resultant dc measurement. Example: 1M added across the output of Elvee's suggestion means, multiply the DC mV you measure by (150+150+22K+22K+1M)/ 1M = 1.34 x

AND DIYaudio has some strong policies on direct-to-Mains Line experimentation - for good reason: it's NOT inherently-safe, and there is essentially NO LIMIT to RISK.

So - sooner or later - I expect a Mod will lock this thread.
 
Last edited:
A two-pole, symetrical filter would be safer and more effective:

Note that if your multimeter has a low internal resistance, you should take the 350K of the filter into account.

The filter leaves 100mV peak of 50Hz on its output, which should be no problem for any type of multimeter

Thanks Elvee, that is a better circuit with regards to safety having high value resistor on both side, and good point regarding multimeter internal resistance.

If you dont have a scope,

Thanks, just an important warning regarding using oscilloscopes and its probes connected to mains voltages for anyone reading this thread, the ground of the probes BNC connectors and the whole chassis of the oscilloscope is connected to the mains connectors ground pin as well, by hooking up the ground clip of the probe to the mains will cause the fuses to blow and in worse case if the ground and fuses wouldn't work can make the whole oscilloscope chassis conducting lethal mains voltage.
A much better alternative is a fully isolated differential probe.

Safety note on above examples- whatever else you do, add a large-value discharge resistor across the output side - say 1Megaohm - just to ensure once you unplug there is no residual HV charge on the filter cap(s) that might sting

(yes I know there shouldn't be, given it is a lowpass filter: but I'd not want anyone stung.)

Note - this means you'd need to scale the resultant dc measurement. Example: 1M added across the output of Elvee's suggestion means, multiply the DC mV you measure by (150+150+22K+22K+1M)/ 1M = 1.34 x

AND DIYaudio has some strong policies on direct-to-Mains Line experimentation - for good reason: it's NOT inherently-safe, and there is essentially NO LIMIT to RISK.

Thanks for your input, a discharge resistor is a good practice, and note taken with regards to mains voltages and safety, I have added a disclaimer in my first post.

Cheers Michael
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.