Buzzing noise off mains, silent on UPS battery?

[rowertnk]

Thanks all for your help. I have to put my dunce cap on The fluctuating readings were actually due to the multimeter's pins being reversed when testing. I know that in theory this has no effect when testing AC power but somehow it does with my multimeter. Maybe it's because it's an el cheapo one.

Anyway I'm usually getting readings around 238V stable in any socket in the house. Sometimes it goes down to 226V/230V stable.. btw I just remembered that my area doesn't even have its own substation and we're sharing with another area.

Neutral to ground is 3V or 4V, Live to neutral is 238V and Live to ground is exactly (Live to neutral) minus (neutral to ground).

I had a technician come in who thinks that the noise is due to a lower incoming voltage or crappy incoming mains.


Tried plugging in my old Enermax and guess what there's noise too, just that it's much much softer. It's already many years old and it's still working along with all the other components so I guess whatever causing the noise is not harming my PC. But it's still darn annoying


Here's a recording of the sound for your enjoyment, you may hear some fan and HDD noise too.
http://www.mediafire.com/?xgjkg9u3k8jlnyb

I wonder whether this is 50Hz mains hum.



Found this thread which is similar what I'm experiencing.

http://www.jonnyguru.com/forums/show...highlight=buzz


Buzzing at back of PSU, near power cable. Intermittent, happens at certain times during the day. Happens nearly always when hot water shower is on, but can happen randomly too. Happens with incoming AC power regardless whether the PC or PSU switch is on.

Happens with 3 PSUs. Silverstone loudest, Compaq loud, Enermax barely audible.


Do I have a serious grounding problem, or is it pretty harmless?

Should I change my PSU in the hope that it will fix the noise issue? My only available choices are a Coolermaster GX 550w/Silent Pro M 500W/Xigmatek NRP 600W. I really don't know if they are any good.

I could probably survive with the Silverstone if it's not being damaged or causing any damage...and hopefully be able to retain some sanity after a few years


Thanks guys, I don't know what I'd do without your forum

[TheSlakey]

Audacity reckons it's 10KHz, which is a really common fluorescent ballast operating speed. If it's not bad grounding, I'm officially amazed.

Grounding problems can be a real safety issue. You'd need to call out an electrician to fix it, or get your land lord/building management/parents/whatever to do it.

[rowertnk]

Thanks TheSlakey, I'll look into the ground issue.


The issue is getting weirder though, need you gurus help.

The buzzing started again just now. I decided to see if something connected to the UPS was causing the issue.

Connected to UPS:
-CPU
-Modem
-LCD monitor with separate adaptor block (like the laptop kind)

Unplugged the monitor's adaptor. Perfect heavenly silence.

Tried a few times with the same result. The PSU is totally 100% silent without the monitor's adaptor.

Do you think the adaptor is dying although there's no noticeable symptoms? The adaptor is silent!

Could the adaptor in this room be affecting the PC in the other room? Should I get the adaptor replaced (not cheap!) and expect the noise issue to be resolved, or is the adaptor in itself not the cause of the problem?


TIA

[McSteel]

Originally Posted by rowertnk

Thanks TheSlakey, I'll look into the ground issue.


The issue is getting weirder though, need you gurus help.

The buzzing started again just now. I decided to see if something connected to the UPS was causing the issue.

Connected to UPS:
-CPU
-Modem
-LCD monitor with separate adaptor block (like the laptop kind)

Unplugged the monitor's adaptor. Perfect heavenly silence.

Tried a few times with the same result. The PSU is totally 100% silent without the monitor's adaptor.

Do you think the adaptor is dying although there's no noticeable symptoms? The adaptor is silent!

Could the adaptor in this room be affecting the PC in the other room? Should I get the adaptor replaced (not cheap!) and expect the noise issue to be resolved, or is the adaptor in itself not the cause of the problem?


TIA

There's probably a parasite resonance between the monitor's power brick's AC-DC converter and the UPS. All DC power supplies introduce electrical noise into the AC circuit they're connected to, and all of them introduce some degree of harmonic distortion to the original input AC sinewave. I suppose that the PC PSU picks this electrical noise up and amplifies it, because it's within the frequency response of it's transistors and coils, and it's doubtful there's any filtering in it meant to clean up noise at those frequencies, since they normally shouldn't be there in the first place. This should be a non-issue, as far the performance or the longevity of your equipment is concerned, but I can see it being fairly annoying.

Is the noise there when the monitor is connected directly to mains, and not trough the UPS? If there isn't any, just connect it that way, and if you ever really need to work off the UPS's battery (power loss), simply connect the monitor to it at that time. Shouldn't be too often, and then it would only take a few seconds, so I see no problem in doing it that way.

## EDIT ##

I just remembered that I used to have a Scientific Atlanta CMTS a while ago, which had it's external power brick squeal quite annoyingly. I remember noticing that the squealing was there only after the coax power connector was pulled from the modem. It would start off barely noticeable, only to increase in amplitude over a couple of seconds to being quite easy to hear. The P'brick would continue to squeal until it was disconnected from the mains, upon which the sound would slowly fade, and would not be there after reinserting the mains power connector. Not sure what to make of it, just thought I'd mention it...

[rafal_iB_PL]

There's probably a parasite resonance between the monitor's power brick's AC-DC converter and the UPS. All DC power supplies introduce electrical noise into the AC circuit they're connected to, and all of them introduce some degree of harmonic distortion to the original input AC sinewave. I suppose that the PC PSU picks this electrical noise up and amplifies it, because it's within the frequency response of it's transistors and coils, and it's doubtful there's any filtering in it meant to clean up noise at those frequencies, since they normally shouldn't be there in the first place. This should be a non-issue, as far the performance or the longevity of your equipment is concerned, but I can see it being fairly annoying.

Where's transient filter in all this?

[McSteel]

Originally Posted by rafal_iB_PL

Where's transient filter in all this?

You mean this?

Ever wonder what's the impedance on those caps @ 10 kHz?

[rafal_iB_PL]

Ever wonder what's the impedance on those caps @ 10 kHz?

And what's impedance of those inductors @10 kHz?

[McSteel]

Originally Posted by rafal_iB_PL

And what's impedance of those inductors @10 kHz?

Depending on how high the parasite capacitance of the coil(s) used is, at that frequency, attenuation of outside signals would be less important, as the real source of the EMI could be the coils themselves. Now if the power brick uses el cheapo coils, not wound to make their capacitance as low as possible, that would offer a nice explanation for the observed behavior.

I'm really stretching this, aren't I?

[TheSlakey]

The impedance of the caps is low and the impedance of the common mode choke is high at 10kHz. That's pretty much the whole idea of the transient filter.

You're not even seeing anything new with the monitor adapter, it's just the same problem repeating itself. You've attached another device that introduces differential noise to the line, and the Y-caps in your PC power supply's transient filter can't shunt the noise to ground like they're supposed to, so your chokes sing again. The monitor's power brick is doing fine, it just doesn't have a line filter.