Measuring AC power line for DC offset?

Measuring AC power line for DC offset?
While I have a real, real nice DMM, I'm not at all sure as how to go about measuring for DC offset

You could roll a simple LPF with say a few hundred K of R and several uF of C...

Do you expect to find DC ?

JR

I don't expect to, but audiophiles seem to find it all over the place.

Speedskater said:

I don't expect to, but audiophiles seem to find it all over the place.

Click to expand...

That's because they don't know how to use basic test equipment.

It the ac mains had DC content every transformer on that line would dissipate extra heat, and saturate sooner.  Not a good thing.

JR

JohnRoberts said:

It the ac mains had DC content every transformer on that line would dissipate extra heat, and saturate sooner.  Not a good thing.

JR

Click to expand...

Also eliminating it for the next step.

If you use a normal DMM with a NMRR of well, unspecified most of the time, but if a nice one has 60dB we could expect 40dB or 50dB in a good day for a cheap one, 1V could be expected to be seen at a 220V mains for a DC measurement. Adding a filter as JR said may be useful.

JS

Well if they hear it (mechanical transformer hum) and I suspect that a few really do, it could be caused by a DC offset from 2nd harmonic distortion.
This Plitron paper thinks that other causes are more likely.

"MEASURING ACOUSTIC NOISE EMITTED BY POWER TRANSFORMERS"
Menno van der Veen
Corresponding Author, [email protected]
Francisco de Leon, Brian Gladstone, Valeriu Tatu
Plitron Manufacturing Inc., [email protected]

http://www.idc-online.com/technical_references/pdfs/electrical_engineering/MEASURING_ACOUSTIC_NOISE_EMITTED_BY_POWER.pdf

How would a second harmonic introduce DC if it's a sine function with no DC? You would be summing two signals with no DC and no DC as result. It shift half cycle up and half down, any harmonic distortion will introduce DC...

JS

joaquins said:

How would a second harmonic introduce DC if it's a sine function with no DC? You would be summing two signals with no DC and no DC as result. It shift half cycle up and half down, any harmonic distortion will introduce DC...

Click to expand...

Mains transformers usually operate ju..ust under saturation.  Saturation results in higher current on the peaks.  You can see this by looking at the Magnetising current.

If there are even harmonics on the mains, one cycle will have a higher peak than the  other.  A higher peak on one cycle means the Magnetising current has a bigger peak on one side .. ie a DC current is introduced .. even if the Voltage has no DC component.

ricardo said:

joaquins said:

How would a second harmonic introduce DC if it's a sine function with no DC? You would be summing two signals with no DC and no DC as result. It shift half cycle up and half down, any harmonic distortion will introduce DC...

Click to expand...

Mains transformers usually operate ju..ust under saturation.  Saturation results in higher current on the peaks.  You can see this by looking at the Magnetising current.

If there are even harmonics on the mains, one cycle will have a higher peak than the  other.  A higher peak on one cycle means the Magnetising current has a bigger peak on one side .. ie a DC current is introduced .. even if the Voltage has no DC component.

Click to expand...

Well, that's fair. How big could 2nd harmonic be, I've never measured more than 0.'something'% which I guess won't do much DC current shift, which will also depend on the phase relation of that 2nd harmonic with fundamental, I just don't know, I'm asking.

JS

joaquins said:

How big could 2nd harmonic be, I've never measured more than 0.'something'% which I guess won't do much DC current shift, which will also depend on the phase relation of that 2nd harmonic with fundamental.

Click to expand...

Have a look at the AC voltage waveform.

If you can see ANY distortion, you probably have saturation somewhere.  There's usually enough impedance in your house wiring to see the effect.  Maybe your country & house has supa dupa Mains but  I expect to see 'visible' (maybe 1%) THD in the UK & Oz.

Also look at the Magnetizing (no load) current which WILL be distorted.

I don't have adequate equipment now, at the university last week THD was under 1% most of which odd harmonics, even was really low. Magnetization distortion will be symmetric, which couldn't make 'out of phase' even harmonics, it only would generate DC current if the lines already has them.

I don't know the magnitudes, but seems strange that DC could leak all the way up and down the lines, it could be present at some point but with each transformer in the way will disappear and with each resistance will decouple it. Maybe what I see as too little may be not so little and affect something, anyone did the LPF and measurement that JR said?

JS

joaquins said:

Magnetization distortion will be symmetric, which couldn't make 'out of phase' even harmonics, it only would generate DC current if the lines already has them.

Click to expand...

Any even harms. will be MUCH more evident in the Magnetisation current.

Have you got any noisy transformers?  No noise, probably no problem.

well, I have one bigish 220V/110V under my desk now and I doesn't hear it... I did hear one on the street some years ago (old and small to be in the distribution line), thanks for everyone that's no longer there. But a transformer will ring without DC if it's loose, and this I mention was ooold and probably loose.

JS

Rather than my poor attempts at explanations I'll let Rod Elliott have a go at it.

"Blocking Mains DC Offset"
A varying DC offset on the AC mains is no longer uncommon. There are many ways that a DC offset can be created, with most being totally outside the control of those who have to try to eliminate it, or put up with the mechanical noise created in toroidal transformers.
By Rod Elliott (ESP)

http://sound.westhost.com/articles/xfmr-dc.htm

http://groupdiy.com/index.php?topic=54315

JDB.

Speedskater said:

A varying DC offset on the AC mains is no longer uncommon.

Click to expand...

If it's varying, it isn't DC. 

-a

Andy Peters said:

Speedskater said:

A varying DC offset on the AC mains is no longer uncommon.

Click to expand...

If it's varying, it isn't DC. 

-a

Click to expand...

If it is varying between different same polarity DC voltages, it isn't quite AC either.  A significant DC component will bias magnetics away from zero and affect them negatively.

Even battery voltage can vary, I would accept a less strict definition for DC especially in this case.

JR

Really I think that it's a DC offset thing. Like when you use a editing program to examine an audio file and it calculates the average offset. An old fashioned 100 Watt light bulb and a in series power diode should generate some DC offset.

I knew that I had recently read something on the subject, but it took me awhile to look in the most obvious place.
From the Plitron paper that I referenced in reply #6:

We researched which mains conditions generate noise inside power transformers and found three major
causes:
1: The mains 'sinusoid' is not symmetrical. This is identical to a DC-voltage on the mains. A very nice example is:
the single phase rectifier used in hair dryers at lower power by means of one series diode. The combination of this
rectifier with the actual load resistance and the resistance of the mains wires create an effective DC-voltage on the
mains. See Figure 13 for an explanation of this effect of hair dryers.
(It is very easy to test a particular transformer for noise under this "DC on mains" condition. Connect a 100 or 200
W light bulb in series with a proper diode to the same mains socket of the transformer under test and check for
transformer noise by listening).
Our measurements indicate that only a few mV-DC can be large enough to bring a transformer into its noisy region.
We did many tests to determine the amount of DC-voltage on the mains by means of the light bulb and other loads
in series with a diode and by observing the DC content on the mains due to asymmetrical loading elsewhere in our
or other buildings connected to the same distribution transformer. We found a DC-component smaller than 100 mV
for 120 V at 60 Hz mains. However, by experience we know that bad conditions today surely will be worse
tomorrow. We therefore propose to take an extra margin and to use 250 mV-DC as our standard condition for
"adverse" mains. (For 230 V at 50 Hz mains, the standard adverse mains condition equals the same 250 mV-DC).