GROUNDING PRACTICES IN CONSUMER AUDIO
As many of us have observed over the years grounding practices in consumer audio have for many of us been a frustrating experience. Compatibility/Synergy are consistently major issues, as well as safety, reducing ground loops, noise and interference. Designers are equally frustrated by this problem. Some of us wish it would go away and others attempt to deal with it as best as we know how. There does not appear to be much of any standardization that I, and many others are aware of. Many of us who work in the electronics industry outside of audio who have observed the posts from audiophiles, designers and dealers scratch our heads in near disbelief.
The following is not only an outline for reducing these problems, but a wake up call to this end of the industry. The following is for the application to consumer unbalanced audio, and could well be a possible outline for positive discussion and direction. It is also a way to use Earth Ground to our advantage instead of a problem.
Chassis Grounding:
The following is a technique used in instrumentation for low frequency applications.
The chassis will be earth grounded via the earth ground at the three-prong outlet. The audio signal ground contained in the chassis is not connected to the chassis ground. This will require the design to be electrically isolated from the chassis ground which is easily solved by using nylon stand offs to mount pc boards and isolated bulkhead RCA jacks. (I am going to break this rule later on but bear with me for the moment). The above forces the design to be star grounded at the return of the power supply, which is always good practice to reduce noise and ground loops. However, this does leave the present configuration susceptible to interference from the inputs. This interference can be reduced by the using a simple common mode ac line filter at the AC input and using either a well shielded coax or microphone cable at the line level input. This configuration also poses an additional problem due the fact that we now have two different grounds with two different potentials. In the past I have observed this problem when using a certain brand of rotary switch for a volume control. The rotary switch was not well isolated internally and had enough of a leakage current so that noise was developed when it was used. Moving to a different vendor with higher isolation devices corrected the problem. I have not observed any problem with standard switches for on off applications or anything similar.
I have applied this technique to basic audio chain of equipment consisting of an outboard DAC, line level pre amp, and power amps using the chassis and grounding design I outlined above. The transport that I presently use is a modified consumer device and is equipped with a two-prong plug. The system also has an FM tuner and an old pre amp that is used as a phono pre amp. Both of these devices are standard consumer issue with a two-prong plug. I have observed no compatibility issues with the older style units.
As I outlined earlier I am going to break this rule at one point. The line level pre amp now has a connection from the return of the line stage pre amp power supply to the chassis of the unit. I have now connected earth ground to the analog “center point of the system.” This did not cause a ground loop at all, and to be more precise, for CD playback the inherent ground loops that are typical for unbalanced circuitry simply disappeared. The FM tuner and the old pre amp appear to be unaffected by the center point earth ground. A welcome addition was that the rotary switch that had a leakage problem because of the two different potential levels described earlier no longer had the problem due to center point earth ground.
The use of the system center point earth ground for low frequency applications has been in the textbooks for at least 30 some odd years and has been applied to other low frequency applications. Applied at this level to a simple chain of audio playback equipment CD playback now has reduced hum and hiss to levels more akin to balanced design than unbalanced design. Playback of FM tuner and Phono pre amp remains unaffected.
One thing that has surprised me relates to the issue of low frequency applications. I was expecting to find problems with the digital portion of this playback chain. I have not found one to date but I think this needs to be investigated further when time allows.
A note to all of the tweakers who read this: I am not recommending any changes to existing designs; in fact I would discourage it.
To DIY folks: You may wish to rethink some of your present chassis/grounding schemes.
To the rest of the industry; this is a subject that not only deserves discussion but an active participation to reach acceptable standards.
The grounding system described above will not address the problem of toroidal transformers mechanically vibrating due to either DC on the AC lines or as I have observed on occasion, low frequency oscillations.
Dan Banquer
R.E. Designs
P.S. Some folks prefer the term safety ground vs. earth ground. For those folks with those preferences feel free to substitute the word safety whenever the term is used in conjunction with the phrase earth ground.
P.P.S. I would like to keep this thread as a discussion of grounding practices only.
As many of us have observed over the years grounding practices in consumer audio have for many of us been a frustrating experience. Compatibility/Synergy are consistently major issues, as well as safety, reducing ground loops, noise and interference. Designers are equally frustrated by this problem. Some of us wish it would go away and others attempt to deal with it as best as we know how. There does not appear to be much of any standardization that I, and many others are aware of. Many of us who work in the electronics industry outside of audio who have observed the posts from audiophiles, designers and dealers scratch our heads in near disbelief.
The following is not only an outline for reducing these problems, but a wake up call to this end of the industry. The following is for the application to consumer unbalanced audio, and could well be a possible outline for positive discussion and direction. It is also a way to use Earth Ground to our advantage instead of a problem.
Chassis Grounding:
The following is a technique used in instrumentation for low frequency applications.
The chassis will be earth grounded via the earth ground at the three-prong outlet. The audio signal ground contained in the chassis is not connected to the chassis ground. This will require the design to be electrically isolated from the chassis ground which is easily solved by using nylon stand offs to mount pc boards and isolated bulkhead RCA jacks. (I am going to break this rule later on but bear with me for the moment). The above forces the design to be star grounded at the return of the power supply, which is always good practice to reduce noise and ground loops. However, this does leave the present configuration susceptible to interference from the inputs. This interference can be reduced by the using a simple common mode ac line filter at the AC input and using either a well shielded coax or microphone cable at the line level input. This configuration also poses an additional problem due the fact that we now have two different grounds with two different potentials. In the past I have observed this problem when using a certain brand of rotary switch for a volume control. The rotary switch was not well isolated internally and had enough of a leakage current so that noise was developed when it was used. Moving to a different vendor with higher isolation devices corrected the problem. I have not observed any problem with standard switches for on off applications or anything similar.
I have applied this technique to basic audio chain of equipment consisting of an outboard DAC, line level pre amp, and power amps using the chassis and grounding design I outlined above. The transport that I presently use is a modified consumer device and is equipped with a two-prong plug. The system also has an FM tuner and an old pre amp that is used as a phono pre amp. Both of these devices are standard consumer issue with a two-prong plug. I have observed no compatibility issues with the older style units.
As I outlined earlier I am going to break this rule at one point. The line level pre amp now has a connection from the return of the line stage pre amp power supply to the chassis of the unit. I have now connected earth ground to the analog “center point of the system.” This did not cause a ground loop at all, and to be more precise, for CD playback the inherent ground loops that are typical for unbalanced circuitry simply disappeared. The FM tuner and the old pre amp appear to be unaffected by the center point earth ground. A welcome addition was that the rotary switch that had a leakage problem because of the two different potential levels described earlier no longer had the problem due to center point earth ground.
The use of the system center point earth ground for low frequency applications has been in the textbooks for at least 30 some odd years and has been applied to other low frequency applications. Applied at this level to a simple chain of audio playback equipment CD playback now has reduced hum and hiss to levels more akin to balanced design than unbalanced design. Playback of FM tuner and Phono pre amp remains unaffected.
One thing that has surprised me relates to the issue of low frequency applications. I was expecting to find problems with the digital portion of this playback chain. I have not found one to date but I think this needs to be investigated further when time allows.
A note to all of the tweakers who read this: I am not recommending any changes to existing designs; in fact I would discourage it.
To DIY folks: You may wish to rethink some of your present chassis/grounding schemes.
To the rest of the industry; this is a subject that not only deserves discussion but an active participation to reach acceptable standards.
The grounding system described above will not address the problem of toroidal transformers mechanically vibrating due to either DC on the AC lines or as I have observed on occasion, low frequency oscillations.
Dan Banquer
R.E. Designs
P.S. Some folks prefer the term safety ground vs. earth ground. For those folks with those preferences feel free to substitute the word safety whenever the term is used in conjunction with the phrase earth ground.
P.P.S. I would like to keep this thread as a discussion of grounding practices only.
Connecting safety ground ( earth ) to the analog signal ground will pose ground loop problems when the tuner has an ( earthed ) distribution cable input. At least this is the case in my country. When an antenna is used there won't be a problem but the use of an aerial is rare nowadays. Most people are connected to a cable system over here.
So in this country the general advice is to *not* connect safety ground to analog signal ground because it causes more trouble than it solves. For example earth/safety ground can be pollute your system. Since 99 % of the gear sold is double insulated as it is called safety grounding ( using "earth" ) is not needed. This gear does not have an earth/safety ground connection at all.
That is not the case when a 100 Ohm resistor is soldered from the center/star of the starground to the case possibly paralleled with a small ceramic cap for shunting RF to earth/safety ground. 100 Ohm is high enough for preventing a ground loop and low enough for solving the potential difference. All this of course when the case is connected to safety ground/earth as in the grounding scheme you describe !!
The grounding scheme you propose is a classic one and it is a good functioning scheme too. But the 100 Ohm resistor was included in this ground scheme as well in the old textbooks.
So in this country the general advice is to *not* connect safety ground to analog signal ground because it causes more trouble than it solves. For example earth/safety ground can be pollute your system. Since 99 % of the gear sold is double insulated as it is called safety grounding ( using "earth" ) is not needed. This gear does not have an earth/safety ground connection at all.
That is not the case when a 100 Ohm resistor is soldered from the center/star of the starground to the case possibly paralleled with a small ceramic cap for shunting RF to earth/safety ground. 100 Ohm is high enough for preventing a ground loop and low enough for solving the potential difference. All this of course when the case is connected to safety ground/earth as in the grounding scheme you describe !!
The grounding scheme you propose is a classic one and it is a good functioning scheme too. But the 100 Ohm resistor was included in this ground scheme as well in the old textbooks.
Grounding
Good point!
However, for those folks who don't have their systems grounded via the antenna earth ground this could well be more viable. What do you think of a switch on the back of the pre amp that allows the user to either earth ground the audio signal or not?
One possible drawback of the 100 ohm resistor is that if we have two units that use the same configuration we should get some form of groundloop. Some sort of standardization is what I really looking for here.
d.b.
Good point!
However, for those folks who don't have their systems grounded via the antenna earth ground this could well be more viable. What do you think of a switch on the back of the pre amp that allows the user to either earth ground the audio signal or not?
One possible drawback of the 100 ohm resistor is that if we have two units that use the same configuration we should get some form of groundloop. Some sort of standardization is what I really looking for here.
d.b.
The problem Jean-Paul brought up is a very common problem also
here in Sweden, where we often have the tuner connected to
a cable network. Since most residential houses do not have
earth in the wall outlets, this used not to be a problem. However,
since many people connect their PC to the stereo this becomes
problematic. PCs are expected to be plugged into three-prong
outlets, and so have these X and Y capacitors. Since we usually
do not have three-prong outlets, these caps will cause the "signal"
ground in the PC to follow the mains voltage at about half the
amplitude. The caps are small enough that you won't get any
dangerous currents when touching computer and being grounded
at the same time. However, when you connect the computer to
the stereo you get a spark, and you get hum problems. Actually,
you may still get hum problems even if you have a three-prong outlet since you will create a ground loop through the mains
earth and back through the antenna cable network.
There is a tweak to fix this that is often recommended, and I think
you can even buy such a device in Finland. The trick is
to solder capacitors in series with both the signal and the
ground lead of the antenna cable. Choosing caps on the order
of a few nF will give very little attenaution in the FM band, but
a very large impedance at mains frequency, and since there is
a cap also on the ground lead there is galvanic isolation. I don't
know if this is recommendable practice, but it is often recommended
to overcome the problem mentioned above. Is this a good solution,
or is there some serious problem with it?
here in Sweden, where we often have the tuner connected to
a cable network. Since most residential houses do not have
earth in the wall outlets, this used not to be a problem. However,
since many people connect their PC to the stereo this becomes
problematic. PCs are expected to be plugged into three-prong
outlets, and so have these X and Y capacitors. Since we usually
do not have three-prong outlets, these caps will cause the "signal"
ground in the PC to follow the mains voltage at about half the
amplitude. The caps are small enough that you won't get any
dangerous currents when touching computer and being grounded
at the same time. However, when you connect the computer to
the stereo you get a spark, and you get hum problems. Actually,
you may still get hum problems even if you have a three-prong outlet since you will create a ground loop through the mains
earth and back through the antenna cable network.
There is a tweak to fix this that is often recommended, and I think
you can even buy such a device in Finland. The trick is
to solder capacitors in series with both the signal and the
ground lead of the antenna cable. Choosing caps on the order
of a few nF will give very little attenaution in the FM band, but
a very large impedance at mains frequency, and since there is
a cap also on the ground lead there is galvanic isolation. I don't
know if this is recommendable practice, but it is often recommended
to overcome the problem mentioned above. Is this a good solution,
or is there some serious problem with it?
* Just to avoid confusion: a ground breaking switch ( compromising safety in a untolerable way ! ) as found on older tube equipment like tube guitar amps is not meant. *
To be honest I would avoid any connection of safety ground/earth to signal ground as it is polluted with noise and other nasty things that shouldn't be there in theory. In other words: big chance you'll connect noise to your gear that you wanted to solve by connecting those grounds.
I do believe stargrounds are an elegant and correct way of grounding signal grounds in audio equipment and I think you're right that some standard should be created. For instance the inputRCA's-ground should be connected together and then via one wire to the starground. The shields of the input cables should be connected too at the inputRCA's ground and *disconnected* at the PCB or receiving side. Mmm, when I read my own writing I think a drawing will say more than a 1000 words...
Grounding
I think the real problem is that a lot of the world is still using the two prong plug. Given the popularity of the home computer and the three prong plug that comes with it I am sometimes surprised by this. Then again.......
To Jean Paul: If the earth ground is giving you a lot of noise pollution I would guess that there are two possible issues. One is that the earth ground connection itself has a problem or there is a ground loop in the system. Is that a reasonable guess?
d.b.
I think the real problem is that a lot of the world is still using the two prong plug. Given the popularity of the home computer and the three prong plug that comes with it I am sometimes surprised by this. Then again.......
To Jean Paul: If the earth ground is giving you a lot of noise pollution I would guess that there are two possible issues. One is that the earth ground connection itself has a problem or there is a ground loop in the system. Is that a reasonable guess?
d.b.
The real problem is that a 2 prong device should be plugged in an "un-earthed" wall outlet and 3 prong devices should be plugged in an "earthed" walloutlet.
Devices that throw out RF have mains filters that should be earthed hence the 3 prong plug on a pc. The caps in those filters dividing the mains voltage in 2 on the unused earth prong. This is the case when the pc is plugged in a "unearthed" walloutlet, mainly a user error made possibly by the mainsplugs that fit on both unearthed and earthed systems. This can be classified as a design error too.
Dan, don't think safety ground is an ideal ground. It has resistance and it serves to lead the things we don't want to ground. The copper bars used for safety ground corrode and after some years groundquality will be worse.
Audio signal ground can be designed well and I don't see the point connecting it to an earth ground of ( in general ) unknown quality.
Devices that throw out RF have mains filters that should be earthed hence the 3 prong plug on a pc. The caps in those filters dividing the mains voltage in 2 on the unused earth prong. This is the case when the pc is plugged in a "unearthed" walloutlet, mainly a user error made possibly by the mainsplugs that fit on both unearthed and earthed systems. This can be classified as a design error too.
Dan, don't think safety ground is an ideal ground. It has resistance and it serves to lead the things we don't want to ground. The copper bars used for safety ground corrode and after some years groundquality will be worse.
Audio signal ground can be designed well and I don't see the point connecting it to an earth ground of ( in general ) unknown quality.
Grounding
I respectfully disagree. I'm pretty well convinced that when folks have a negative attitude toward earth ground it is due to it causing ground loops or the earth ground they are connected to has corroded or not been connected properly.
d.b.
I respectfully disagree. I'm pretty well convinced that when folks have a negative attitude toward earth ground it is due to it causing ground loops or the earth ground they are connected to has corroded or not been connected properly.
d.b.
Re: Grounding
OK but I haven't connected signal ground to my well earthed safety ground and I don't have hum or any other problems due to poor grounding/loops or other nasties. I just don't see what positive effect connecting both grounds can bring. The only safety ground connection I have is a mains filter that has a connection to earth/safety ground for return of the garbage it filters. That mains filter is connected to a 2 prong walloutlet ( un-earthed so to speak ) for connecting double insulated devices as 99% of the gear sold is. Of course the tuner has a connection to ground because of the earthed cable system.
So please explain why I should connect both grounds for what purpose. I use the grounding scheme as proposed by you without connection to earth already. Cases are connected with 100 Ohm to signal starground so that the case has a use as shielding without having influence on signalground-integrity.
Dan Banquer said:
OK but I haven't connected signal ground to my well earthed safety ground and I don't have hum or any other problems due to poor grounding/loops or other nasties. I just don't see what positive effect connecting both grounds can bring. The only safety ground connection I have is a mains filter that has a connection to earth/safety ground for return of the garbage it filters. That mains filter is connected to a 2 prong walloutlet ( un-earthed so to speak ) for connecting double insulated devices as 99% of the gear sold is. Of course the tuner has a connection to ground because of the earthed cable system.
So please explain why I should connect both grounds for what purpose. I use the grounding scheme as proposed by you without connection to earth already. Cases are connected with 100 Ohm to signal starground so that the case has a use as shielding without having influence on signalground-integrity.
jean-paul said:
I think user error is a strong term to use. PCs are sold for home
use, but most homes here don't have a three-prong outlet, except
in the kitchen and the bathroom, where at least the latter is not
suitable for a PC. Manufacturers are aware of this and yet they
sell us only this type of computers (apart from laptops). Strictly
seen, it is illegal to use PCs in most homes, so I would
say it is a manufacturer/sales error, not a user error. For the
user it is a catch 22. Of course, it is probably even more correct
to say that it is a system error not to have earth in the
mains outlets (at least all houses built during the past 10 years
are required to have three-prong outlets in all rooms).
Grounding
Hi Jean Paul;
It appears we are both having problems with the english language.
When I was talking about the switch I meant that the switch would only connect the audio signal ground to earth or not. The chassis would remain unaffected.
If one has the earth ground for the antenna then either one of two things need to happen. The antenna earth ground through the tuner is now the "centerpoint" earth ground or we use a line level transformer to isolate the tuner from the pre amp and earth ground the return of the pre amp power supply.
If you have two units with the resistor to earth ground using unbalanced connections you still will have a ground loop. The idea here is one and only one earthground point for the entire system.
Yes Jean Paul I have run a system without the signal ground connected to earth at all and it was pretty quiet. However at some point this will come back and bite you. If you are looking for an unbalanced system that is quiet and stays that way, without using line level transformers, then I think this is the only way to go.
Long live the two prong plug! The two prong plug is dead!
d.b.
Hi Jean Paul;
It appears we are both having problems with the english language.
When I was talking about the switch I meant that the switch would only connect the audio signal ground to earth or not. The chassis would remain unaffected.
If one has the earth ground for the antenna then either one of two things need to happen. The antenna earth ground through the tuner is now the "centerpoint" earth ground or we use a line level transformer to isolate the tuner from the pre amp and earth ground the return of the pre amp power supply.
If you have two units with the resistor to earth ground using unbalanced connections you still will have a ground loop. The idea here is one and only one earthground point for the entire system.
Yes Jean Paul I have run a system without the signal ground connected to earth at all and it was pretty quiet. However at some point this will come back and bite you. If you are looking for an unbalanced system that is quiet and stays that way, without using line level transformers, then I think this is the only way to go.
Long live the two prong plug! The two prong plug is dead!
d.b.
I do not care for ground loops either, however I do like have the chassis at the same potential as the supply. In communication systems, we often have some device running at the end of a wire on off -48vdc. These devices use small switches are not referenced to real earth ground. Therefore, you just can connect the ground of a scope to this device. In addition, anyone that has cable TV or Internet know what ground problems they can cause. Therefore, I always have of bunch of ground isolators-dubbed cheaters around it great when testing supplies..🙂
Attachments
Jean,
That what you get for for 58 cents here. It was nore than likely made in China. I have never seen one of higher quality.
That what you get for for 58 cents here. It was nore than likely made in China. I have never seen one of higher quality.
I don't mean your specific plug but the american type power plug in general. There is "play" ( don't know the correct english word ) when plugged in. That is not possible with the european or english power plugs which is more safe. The grounded/earthed 3 prong variant of the american plug has one advantage that the earth pin is longer than the others ( the ones I know to be american types have anyhow ) and with those the "play" is minimal. The longer earth pin is a clever design trick that enhances safety for reasons that are obvious.
Back to the subject please as Dan wanted to discuss only grounding techniques.
Back to the subject please as Dan wanted to discuss only grounding techniques.
- Status
- Not open for further replies.