How to filter noises from mains power
- Thread starter gnd
- Start date
Help Support GroupDIY Audio Forum:
mr coffee
Well-known member
I beat the hum loop issue with my mixer feeding into a computer A/D sound card by using a ground adapter on the mixer power cord (to lift the ground) and plugged that into an extension cord plugged into the same AC socket that my computer was plugged into.
I also tried running the computer power cord via an extension and plugged it into the same socket as the mixer (near the mixer instead of near the computer) using the adapter ground lift trick on the mixer - didn't work the same and I haven't got a clue why.
I live in the USA so this may not work for you in Europe where mains are different. Or you may have a different problem. Here in the states the electricians sometime reverse ground and neutral and there a little neon light gizmos to test for that (and open ground).
Good luck.
Keep us posted.
I also tried running the computer power cord via an extension and plugged it into the same socket as the mixer (near the mixer instead of near the computer) using the adapter ground lift trick on the mixer - didn't work the same and I haven't got a clue why.
I live in the USA so this may not work for you in Europe where mains are different. Or you may have a different problem. Here in the states the electricians sometime reverse ground and neutral and there a little neon light gizmos to test for that (and open ground).
Good luck.
Keep us posted.
SonsOfThunder
Well-known member
>>>using a ground adapter on the mixer power cord (to lift the ground)
And if the mixer is not grounded somewhere, its a shock hazard!
And if the mixer is not grounded somewhere, its a shock hazard!
mr coffee
Well-known member
Hi Charlie,
Thanks for adding the caution about shock hazards.
I was presuming too much knowledge based upon where gnd was posting (here at Prodigy professional forum) and the general fund of knowledge of such matters that his posts suggested he possessed. But someone else might find this with a search engine later that knew considerably less about what they were doing and put themselves at risk.
No argument. :thumb: You're absolutely right. And I'm sure UL wouldn't approve, either.
I remember the days before 3-prong outlets. Most Guitar amps had "ground switches" that coupled the chassis to either side of the AC supply with a 600 volt capacitor (and/or lifted it entirely). And yeah, I remember hearing on the news in the 70's that some musician actually got electrocuted grabbing a mic - presumably while holding "grounded" guitar strings when the setup was wrong and, I presume, the cap (or something else) failed. Of course, the "grounded" mic was part of the circuit, too, so it isn't a guarantee of safety, either, newbies.
Caution and tip to any newbies about shock hazards:
Old geezers like me still grab the guitar strings and just barely touch the back of our hand to ANY mic before we use it. If there's a problem, you get a bit of a shock and your bicep contracts and yanks your hand away from the mic. Simple technique and you get to live to tell about it and avoid ending up in the Darwin awards... works touching accident victims if the power lines are down, too. FWIW
gnd,
If you are trying to run a high end pro studio, yeah, you really do need to get the ground circuits completely straightened out (if that's the problem - it's hard to say for sure). But since you are recording with a sound card (albeit a high end one) in a computer, I'm guessing this is more mid-range at the present time. Sounds like you are recording tracks NOW and are having problems isolating the problem and could really use a quick fix. That's why I mentioned it. Ground loops are a major pain in causing hum and injecting noises into audio, and can be a real pain to troubleshoot.
Don't get hurt lifting grounds, but it can be a useful trouble-shooting tool and it won't kill ya if you use good common sense and remember the chassis is potentially hot. Those 3-prong to 2-prong adapters can be a help troubleshooting and possibly getting the track out that you need to do this afternoon. Many of us lived through the years before the third "ground" prong was added and lived to tell about it.
I think you are on the right track doing what you can inside the computer case to isolate the audio circuits from the RFI that permeates the inside of computer cases. You might also review, if you haven't already, the capacity of your power supply and whether power management is spinning your disks up and down causing internal glitches or if possibly cycling cooling fans were causing the glitches that might be getting into the audio through your computer's power supply connections. Computer supplies are notorious for not meeting their specs if they are heavily loaded. The glitching CAN propagate as "clicks" though the supply lines.
You mentioned AC line filters... RFI filters on the AC inputs may help. Or they may worsen things if they are feeding EMI into your "clean" AC ground circuit. That "clean ground line" can also be a conduit that can feed EMI\RFI from the computer or whatever into your mixer via it's ground. Many computer supplies sold here in the states have some of this filtering built in and it can route garbage into your pristine separate ground circuit and from there into low level circuitry like in your mic mixer (or sound card input circuitry if any is outboard and powered).
Happy hunting.
Hope these ideas help.
Thanks for adding the caution about shock hazards. I was presuming too much knowledge based upon where gnd was posting (here at Prodigy professional forum) and the general fund of knowledge of such matters that his posts suggested he possessed. But someone else might find this with a search engine later that knew considerably less about what they were doing and put themselves at risk.
No argument. :thumb: You're absolutely right. And I'm sure UL wouldn't approve, either.
I remember the days before 3-prong outlets. Most Guitar amps had "ground switches" that coupled the chassis to either side of the AC supply with a 600 volt capacitor (and/or lifted it entirely). And yeah, I remember hearing on the news in the 70's that some musician actually got electrocuted grabbing a mic - presumably while holding "grounded" guitar strings when the setup was wrong and, I presume, the cap (or something else) failed. Of course, the "grounded" mic was part of the circuit, too, so it isn't a guarantee of safety, either, newbies.
Caution and tip to any newbies about shock hazards:
Old geezers like me still grab the guitar strings and just barely touch the back of our hand to ANY mic before we use it. If there's a problem, you get a bit of a shock and your bicep contracts and yanks your hand away from the mic. Simple technique and you get to live to tell about it and avoid ending up in the Darwin awards... works touching accident victims if the power lines are down, too. FWIW
gnd,
If you are trying to run a high end pro studio, yeah, you really do need to get the ground circuits completely straightened out (if that's the problem - it's hard to say for sure). But since you are recording with a sound card (albeit a high end one) in a computer, I'm guessing this is more mid-range at the present time. Sounds like you are recording tracks NOW and are having problems isolating the problem and could really use a quick fix. That's why I mentioned it. Ground loops are a major pain in causing hum and injecting noises into audio, and can be a real pain to troubleshoot.
Don't get hurt lifting grounds, but it can be a useful trouble-shooting tool and it won't kill ya if you use good common sense and remember the chassis is potentially hot. Those 3-prong to 2-prong adapters can be a help troubleshooting and possibly getting the track out that you need to do this afternoon. Many of us lived through the years before the third "ground" prong was added and lived to tell about it.
I think you are on the right track doing what you can inside the computer case to isolate the audio circuits from the RFI that permeates the inside of computer cases. You might also review, if you haven't already, the capacity of your power supply and whether power management is spinning your disks up and down causing internal glitches or if possibly cycling cooling fans were causing the glitches that might be getting into the audio through your computer's power supply connections. Computer supplies are notorious for not meeting their specs if they are heavily loaded. The glitching CAN propagate as "clicks" though the supply lines.
You mentioned AC line filters... RFI filters on the AC inputs may help. Or they may worsen things if they are feeding EMI into your "clean" AC ground circuit. That "clean ground line" can also be a conduit that can feed EMI\RFI from the computer or whatever into your mixer via it's ground. Many computer supplies sold here in the states have some of this filtering built in and it can route garbage into your pristine separate ground circuit and from there into low level circuitry like in your mic mixer (or sound card input circuitry if any is outboard and powered).
Happy hunting.
Hope these ideas help.
Hey, Mr. Coffee.
My mixer and computer were pluged into same phase, surely, but not into same extension cord. I plugged them now into same extension cord, as close as it goes, and I'm almost convinced that hum is a bit better now.
Anyway, hum is not heard really, it is not loud as ground loop, nowhere close.... It is a tiny 10dB bump at 50 and 150 Hz on my analyzer, with peaks at -130dB. Probably it is due to mixer construction. It is not disturbing at all, you cannot hear it even by far, it only drives me crazy to see it on analyzer, and I want to remove it for good. And I will! :thumb:
I cut ground loops on cables, so no shock hazard.
Yea, my studio is more low end than high end... :grin: Lots of computers though, and I will find some solution for them.
In a few days I'm getting isolation transformer, and I will feed my mixer with balanced power then. That should clean up mains a bit too. If I insert RFI filtering in balanced AC power, it cannot disturb ground, because pulses cancel themselves relative to ground. At least in theory. Depends on CMMR of transformer. My will be bifilar wound, so it should work better.
Will post my findings when I install iso trafo.
...
My mixer and computer were pluged into same phase, surely, but not into same extension cord. I plugged them now into same extension cord, as close as it goes, and I'm almost convinced that hum is a bit better now.
Anyway, hum is not heard really, it is not loud as ground loop, nowhere close.... It is a tiny 10dB bump at 50 and 150 Hz on my analyzer, with peaks at -130dB. Probably it is due to mixer construction. It is not disturbing at all, you cannot hear it even by far, it only drives me crazy to see it on analyzer, and I want to remove it for good. And I will! :thumb:
I cut ground loops on cables, so no shock hazard.
Yea, my studio is more low end than high end... :grin: Lots of computers though, and I will find some solution for them.
In a few days I'm getting isolation transformer, and I will feed my mixer with balanced power then. That should clean up mains a bit too. If I insert RFI filtering in balanced AC power, it cannot disturb ground, because pulses cancel themselves relative to ground. At least in theory. Depends on CMMR of transformer. My will be bifilar wound, so it should work better.
Will post my findings when I install iso trafo.
...
julianzizko
Active member
AFAIK Industrial isolation transformers won't cure your the dirty main problems the industrial ones are designed for safety reasons, and the ones I had in my studio hummed really loudly, so much so I stopped using them.
Look here, this may be of some use to you: http://www.hometheaterhifi.com/volume_10_3/feature-article-isolation-transformer-8-2003.html
Look here, this may be of some use to you: http://www.hometheaterhifi.com/volume_10_3/feature-article-isolation-transformer-8-2003.html
Yes, this is exactly the kind I ordered, also with static shield between pri/sec. Only mine goes a step further, and has bifilar secondary, to improve hi frequency cmmr. And my toroid is 300W, and costs 40euro, shipping included.
I tried to order bigger toroid, but there were some winding machine limitations. You need bigger machine for bigger toroids with hi voltage bifilar secondary, so I was told.
Anyway, I should be getting it today or tommorow, and will post my findings.
gnd
mr coffee
Well-known member
gnd,
Glad you're running it down.
Maybe one of the geniuses around here can explain why plugging them into the same outlet matters. I can sure tell a difference in low end home studio apps.
I'm not talking hum thats better than 130db down though. 130db down for hum sounds like about as good as it will get without big bucks fixes. You could get that much hum from the length of you analyzer probe ground lead.
Have you gotten rid of the clicks yet?
Glad you're running it down.
Maybe one of the geniuses around here can explain why plugging them into the same outlet matters. I can sure tell a difference in low end home studio apps.
I'm not talking hum thats better than 130db down though. 130db down for hum sounds like about as good as it will get without big bucks fixes. You could get that much hum from the length of you analyzer probe ground lead.
Have you gotten rid of the clicks yet?
MrCoffe.
AFAIK plugging physically close makes ground potentials more equal, and thus less difference, less ground currents, less noise. Similar to internal star grounding.
My -130dB hum is not from analyzer probes, becasue I observe it at recording computer with software analyzer. It is there in recording card inputs, and gets recorded into master take.
Clicks are still there, I'm still waiting for iso toroid to arrive, maybe it will come tomorrow... :roll: Those clicks appear only on mixer. Master rec computer has switching PS, and clicks probably don't pass through. But mixer's linear PS seems to be more sensitive to clicks. I hope that iso toroid with balanced mains will sort that out. If not, I can still insert RFI filter after iso toroid, and it will work fine on balanced mains, without polluting ground.
gnd
AFAIK plugging physically close makes ground potentials more equal, and thus less difference, less ground currents, less noise. Similar to internal star grounding.
My -130dB hum is not from analyzer probes, becasue I observe it at recording computer with software analyzer. It is there in recording card inputs, and gets recorded into master take.
Clicks are still there, I'm still waiting for iso toroid to arrive, maybe it will come tomorrow... :roll: Those clicks appear only on mixer. Master rec computer has switching PS, and clicks probably don't pass through. But mixer's linear PS seems to be more sensitive to clicks. I hope that iso toroid with balanced mains will sort that out. If not, I can still insert RFI filter after iso toroid, and it will work fine on balanced mains, without polluting ground.
gnd
mr coffee
Well-known member
Hey gnd,
Did your isolation tranny do anything for the clicks?
How about the line filter?
Does the mixer still have clicks without anything plugged into the inputs (no mics, no line ins, no inserts)?
130 db down for a soundcard a/d sounds pretty good to me. I'm surprised the noise floor isn't higher than that.
Keep us posted!
Did your isolation tranny do anything for the clicks?
How about the line filter?
Does the mixer still have clicks without anything plugged into the inputs (no mics, no line ins, no inserts)?
130 db down for a soundcard a/d sounds pretty good to me. I'm surprised the noise floor isn't higher than that.
Keep us posted!
[quote author="mr coffee"]Hey gnd,
Did your isolation tranny do anything for the clicks?
How about the line filter?
Does the mixer still have clicks without anything plugged into the inputs (no mics, no line ins, no inserts)?
130 db down for a soundcard a/d sounds pretty good to me. I'm surprised the noise floor isn't higher than that.
Keep us posted![/quote]
Hi.
I just received tranny few hours ago, and am putting it in box (old PC PS), trying to figure a best way to fit it in... :? No line filter yet. Maybe I'll finish tomorrow, need to make bifilar 0.1mH inductors still.
Pls note, that -130dB is not level, as read by peak meter. Peak meter reads noise at -90dB. But looking on analyzer, that -90dB is a flat noise line at -140dB. And hum is sticking out from these -140dB a bit, with peaks at -130dB (50Hz and 150Hz). So it is not like my soundcard has 130dB SNR. No, SNR of soundcard is some 110dB, and connecting mixer raises noise to -90dB. Just to make it clear. :grin:
I'm using WaveLab as analyzer, on my master recording comp. I can only recommend that. Putting it in realtime mode, you can fire up all kinds of analyzers on it, including level meter, bit meter and phase meter and several more. I have it turned on all the time on my recording computer, so it is like big$$$buck hitech LCD analyzer some have standalone besides their console. Totally cool. But this one is better, looks more fancy with all the colors, and you can tweak it as needed, from rearanging meters on display to changing ballistics of peak/Vu meter and so on. Nice stuff, this WaveLab. And of course, you can even record into it. :grin: And burn master cd afterwards. :thumb:
gnd
Did your isolation tranny do anything for the clicks?
How about the line filter?
Does the mixer still have clicks without anything plugged into the inputs (no mics, no line ins, no inserts)?
130 db down for a soundcard a/d sounds pretty good to me. I'm surprised the noise floor isn't higher than that.
Keep us posted![/quote]
Hi.
I just received tranny few hours ago, and am putting it in box (old PC PS), trying to figure a best way to fit it in... :? No line filter yet. Maybe I'll finish tomorrow, need to make bifilar 0.1mH inductors still.
Pls note, that -130dB is not level, as read by peak meter. Peak meter reads noise at -90dB. But looking on analyzer, that -90dB is a flat noise line at -140dB. And hum is sticking out from these -140dB a bit, with peaks at -130dB (50Hz and 150Hz). So it is not like my soundcard has 130dB SNR. No, SNR of soundcard is some 110dB, and connecting mixer raises noise to -90dB. Just to make it clear. :grin:
I'm using WaveLab as analyzer, on my master recording comp. I can only recommend that. Putting it in realtime mode, you can fire up all kinds of analyzers on it, including level meter, bit meter and phase meter and several more. I have it turned on all the time on my recording computer, so it is like big$$$buck hitech LCD analyzer some have standalone besides their console. Totally cool. But this one is better, looks more fancy with all the colors, and you can tweak it as needed, from rearanging meters on display to changing ballistics of peak/Vu meter and so on. Nice stuff, this WaveLab. And of course, you can even record into it. :grin: And burn master cd afterwards. :thumb:
gnd
mr coffee
Well-known member
Hi gnd,
So if I get you right, the noise floor with the mixer hooked up and no inputs is -90 db and the hum and 3rd harmonic peaks you see on the analyser module are 10 db above that or -80db? That's not bad for a home studio. You MAY be able to get better than that, maybe not.
I take it that you are disconnecting the shields at one end on your cables and leaving the grounded AC cords alone to provide the return path. You might still want to try lifting the ground prong of the AC cords and connect the shield of cables at one end only but make the connection from output of one unit to input of the next (i.e., make cables with two internal conductors for an unbalanced line, one for ground, one for signal, with the cable shield connected to ground at one end only). This sometimes helps with hum, sometimes not. I know the books say to do it different, but if you aren't using transformers at both ends and balanced lines, hum is always a bit of a bugger when connecting audio equipment with unbalanced lines. Especially when the signal levels are -10db instead of +4 db as in many sound card line inputs. Sometimes it makes a difference (and defies the common logic to connect the shield at the input end) to have the shielded end plugged in one way or the other for lowest hum.
Let us all know what the transformer and filters do for the clicks and hum.
Best of luck with it.
:sam:
So if I get you right, the noise floor with the mixer hooked up and no inputs is -90 db and the hum and 3rd harmonic peaks you see on the analyser module are 10 db above that or -80db? That's not bad for a home studio. You MAY be able to get better than that, maybe not.
I take it that you are disconnecting the shields at one end on your cables and leaving the grounded AC cords alone to provide the return path. You might still want to try lifting the ground prong of the AC cords and connect the shield of cables at one end only but make the connection from output of one unit to input of the next (i.e., make cables with two internal conductors for an unbalanced line, one for ground, one for signal, with the cable shield connected to ground at one end only). This sometimes helps with hum, sometimes not. I know the books say to do it different, but if you aren't using transformers at both ends and balanced lines, hum is always a bit of a bugger when connecting audio equipment with unbalanced lines. Especially when the signal levels are -10db instead of +4 db as in many sound card line inputs. Sometimes it makes a difference (and defies the common logic to connect the shield at the input end) to have the shielded end plugged in one way or the other for lowest hum.
Let us all know what the transformer and filters do for the clicks and hum.
Best of luck with it.
:sam:
[quote author="mr coffee"]So if I get you right, the noise floor with the mixer hooked up and no inputs is -90 db and the hum and 3rd harmonic peaks you see on the analyser module are 10 db above that or -80db? [/quote]
No, not at -80dB.... Noise and hum are at -90dB. But removing hum out of noise will not change SNR much. I don't know how to put it properly, but it has something to do with power spectrum of signal. Single frequency (like 50Hz hum) can have high amplitude, but small power compared to whole spectrum. Noise without hum and noise with hum can measure almost the same on peak meter. Adding a single frequency 10dB above others in spectrum doesn't show at all on peak meter.
I don't know how to put this technically, maybe somebody else can explain it. Best is just to fire up WaveLab analyzer and try it for yourself.
Regarding ground loops I deal with them with cabling, in all different ways.
Today more on my new iso transformer. Will hook it up during the day.
gnd
No, not at -80dB.... Noise and hum are at -90dB. But removing hum out of noise will not change SNR much. I don't know how to put it properly, but it has something to do with power spectrum of signal. Single frequency (like 50Hz hum) can have high amplitude, but small power compared to whole spectrum. Noise without hum and noise with hum can measure almost the same on peak meter. Adding a single frequency 10dB above others in spectrum doesn't show at all on peak meter.
I don't know how to put this technically, maybe somebody else can explain it. Best is just to fire up WaveLab analyzer and try it for yourself.
Regarding ground loops I deal with them with cabling, in all different ways.
Today more on my new iso transformer. Will hook it up during the day.
gnd
OK, I plugged iso transformer in.
I made it for two basic purposes:
1. reducing noise
2. lowering voltage of existing transformer, which is too high.
Broadband noise is the same with iso transformer. Hum is slightly reduced (I'd say :? , I think, whatever...). Clicks and pops are well reduced in level, hi freq content of clicks is reduced, and they will not be noticable in the mix anymore, even if they appear. I installed RF filter between both transformers, in balanced mains section - I cannot say what it does. :?
So, was it worthed?
- For reducing voltage to 180V, instead of 230, and thus reducing PS voltage to 22V (from 29V), it works perfectly. :wink:
- For reducing broadband noise levels in audio, well.... Not working at all in my case. Equitech advertises its balanced mains designs to reduce noise by 20dB or something. Well, good for them. My balanced, bifilar toroid does nothing to noise. Probably I should do some filtering on ground wire. Or put entire studio on balanced power, which I will not do, btw.
- For reducing clicks and pops, there is vast improvement. Something is still passing through, but possibly something comes through ground wire too. It is much much less annoying now, and you cannot hear it in the mix.
Specially wound transformer cost me 50euro, while normally wound would cost me 40euro. I paid 10euro more for a good feeling, that I did my best to reduce clicks to minimum. It was an experience, and I learned new stuff. In my case it was worthed, although I expected more. In your case, I don't know. :grin: :thumb:
gnd
I made it for two basic purposes:
1. reducing noise
2. lowering voltage of existing transformer, which is too high.
Broadband noise is the same with iso transformer. Hum is slightly reduced (I'd say :? , I think, whatever...). Clicks and pops are well reduced in level, hi freq content of clicks is reduced, and they will not be noticable in the mix anymore, even if they appear. I installed RF filter between both transformers, in balanced mains section - I cannot say what it does. :?
So, was it worthed?
- For reducing voltage to 180V, instead of 230, and thus reducing PS voltage to 22V (from 29V), it works perfectly. :wink:
- For reducing broadband noise levels in audio, well.... Not working at all in my case. Equitech advertises its balanced mains designs to reduce noise by 20dB or something. Well, good for them. My balanced, bifilar toroid does nothing to noise. Probably I should do some filtering on ground wire. Or put entire studio on balanced power, which I will not do, btw.
- For reducing clicks and pops, there is vast improvement. Something is still passing through, but possibly something comes through ground wire too. It is much much less annoying now, and you cannot hear it in the mix.
Specially wound transformer cost me 50euro, while normally wound would cost me 40euro. I paid 10euro more for a good feeling, that I did my best to reduce clicks to minimum. It was an experience, and I learned new stuff. In my case it was worthed, although I expected more. In your case, I don't know. :grin: :thumb:
gnd
CJ
Well-known member
GND, this is $$, but kind of cool.
You just plug it into the wall, and it records all the garbage that happens on your line for a week, so that you can look at it and see what kind of power you have.
http://www.omnicontrols.com/detail.aspx?ID=2191
Maybe go in on one with a couple of other owners.
You won't need it all your life.
You just plug it into the wall, and it records all the garbage that happens on your line for a week, so that you can look at it and see what kind of power you have.
http://www.omnicontrols.com/detail.aspx?ID=2191
Maybe go in on one with a couple of other owners.
You won't need it all your life.
