- Location
- Bremerton, Washington
- Occupation
- Master Electrician
See art 647. Not allowed in dwelling units, but there are cord and plug units you can use
Improving ground impedance
- Thread starter JonM4
- Start date
- Status
zbang
Senior Member
- Location
- Roughly 5346 miles from Earls Court
Please start with reading Bill Whitlock's work for grounding, transformers, and interfacing (https://www.prosoundweb.com/tag/bill-whitlock/ and others).
grich
Senior Member
- Location
- MP89.5, Mason City Subdivision
- Occupation
- Broadcast Engineer
Furman makes some of these cord-and-plug balanced-power devices. They ain't cheap.
drcampbell
Senior Member
- Location
- The Motor City, Michigan USA
- Occupation
- Registered Professional Engineer
Any connection to the mathematician/computer scientist Alan T. Furman?
grich
Senior Member
- Location
- MP89.5, Mason City Subdivision
- Occupation
- Broadcast Engineer
No clue. I didn't see a reference to him on their website.
Russs57
Senior Member
- Location
- Miami, Florida, USA
- Occupation
- Maintenance Engineer
"All of the connections I have between the sources and the preamp, and to the power amps, are balanced/differential."
If that is factually correct you are well on your way. Make I ask model/brand of sources/preamp/power amp?
In general, balanced means the outputs from one device, to the inputs of another device, have equal impedance to ground. Differential is more about the circuits inside the active devices (preamp/amp). This is a very simplified answer.
This link isn't too bad www.hairballaudio.com/blog/resources/diy-resources/balanced-and-differential
Reasonable discussions here about somethings that claim to be balanced aren't really what we think we are paying for https://www.audiosciencereview.com/...d-balanced-unbalanced-and-all-that-jazz.1352/
Best of luck with your hobby/obsession/addiction.
If that is factually correct you are well on your way. Make I ask model/brand of sources/preamp/power amp?
In general, balanced means the outputs from one device, to the inputs of another device, have equal impedance to ground. Differential is more about the circuits inside the active devices (preamp/amp). This is a very simplified answer.
This link isn't too bad www.hairballaudio.com/blog/resources/diy-resources/balanced-and-differential
Reasonable discussions here about somethings that claim to be balanced aren't really what we think we are paying for https://www.audiosciencereview.com/...d-balanced-unbalanced-and-all-that-jazz.1352/
Best of luck with your hobby/obsession/addiction.
Thanks for the links, I will definitely check them out.
The streamer I am using is an NAD M50.2. DAC is PS Audio, preamp is Aesthetix, and the power amps are Vandersteen mono blocks. I do run the streamer and DAC through a Furman but the pre and power amps are plugged direct into wall.
Strathead
Senior Member
- Location
- Ocala, Florida, USA
- Occupation
- Electrician/Estimator/Project Manager/Superintendent
I agree, and my understanding is that it really has to do with noise on the ground. A true audiophile would just use and isolation transformer for the amplifier and equipment power and make sure nothing else is plugged in to it. Isolated grounds may also be a solution.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
In the context of signalling, including analog audio, balanced does not refer to matched source and sink impedances. And except for power circuitry that is not particularly important. Balanced means that the two signal leads of a differential pair make equal and opposite excursions from signal ground at all times.
Differential means that the signal is the difference in voltage between two signal leads.
If the signals over a differential connection are in fact also balanced noise immunity is greatly increased.
A differential input can accept a single ended (unbalanced) signal simply by connecting one side of the differential input to signal ground. But that loses most of the advantages of balanced differential circuits.
A single ended circuit cannot in general accept a differential source unless that source is also balanced, so that the signal from one lead to ground has the same waveform except for amplitode as the differential signal.
In audio environments the conversion between differential and single ended can be made with active circuitry or throiugh a simple balanced to unbalanced transformer. The transformers useful for a mike level signal are much smaller than the transformers used for a line level signal. The latter must be larger to avoid core saturation at higher signal levels. The inline balanced to unbalanced (and low to high impedance) transformers for use on mike cables are not a versatile as the larger ones in a Direct Insert (DI) box.
- Location
- Illinois
- Occupation
- retired electrician
In a dwelling unit with a dedicated branch circuit, NM cable, and non-metallic boxes, an isolated ground does not do anything. (assuming that it actually does something in other cases, which I tend to doubt)
ptonsparky
Tom
- Occupation
- EC - retired
It does affect my income, of which I say "Thank you very much!"
Russs57
Senior Member
- Location
- Miami, Florida, USA
- Occupation
- Maintenance Engineer
"balanced does not refer to matched source and sink impedance"
I can see how my comment could be interpreted to mean that. Thank you for pointing it out. What I was trying to say is the sink sees the same impedance to ground on both source wires. Not that source and sink having matching output/input impedance. That would be a very bad idea for audio
Some equipment accomplishes this by inserting a resistor between ground and a non-signal containing wire. So sure, it is balanced per definition but it doesn't have a mirror image signal on both wires. Part of the games played in the audio market.
I think we can have agreement that the ideal situation is two signal carrying wires, of opposite polarity and of equal impedance to ground. Star quad cable has good "noise" immunity and need not be expensive. I'm a fan of balanced power too, having worked with isolated power systems in hospitals for decades. I understand it may not be code compliant in a home setting. Doesn't mean it can't be safer/better.
I can see how my comment could be interpreted to mean that. Thank you for pointing it out. What I was trying to say is the sink sees the same impedance to ground on both source wires. Not that source and sink having matching output/input impedance. That would be a very bad idea for audio
Some equipment accomplishes this by inserting a resistor between ground and a non-signal containing wire. So sure, it is balanced per definition but it doesn't have a mirror image signal on both wires. Part of the games played in the audio market.
I think we can have agreement that the ideal situation is two signal carrying wires, of opposite polarity and of equal impedance to ground. Star quad cable has good "noise" immunity and need not be expensive. I'm a fan of balanced power too, having worked with isolated power systems in hospitals for decades. I understand it may not be code compliant in a home setting. Doesn't mean it can't be safer/better.
zbang
Senior Member
- Location
- Roughly 5346 miles from Earls Court
Not quite.... just differential signaling does not make for a true balanced line. Nor does a reference to a signal ground. (And none of that considers where you measure that equal excursion.)
Try this from The Penguin Dictionary of Electronics
Strathead
Senior Member
- Location
- Ocala, Florida, USA
- Occupation
- Electrician/Estimator/Project Manager/Superintendent
IF there is noise on the ground, then an isolated ground to the audio circuits should keep that noise off of the audio circuit.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
I think our only disagreement here is that you are describing a balanced signal path, which does not tell you anything about the actual signal that is applied to it. I was describing a balanced signal, which is best carried over a balanced path, but can still be balanced regardless of the path.
To get the best noise immunity when processing a balanced signal you need a differential input.
However, I was too limited in my definition of a balanced signal. The key is that at all times the sum of the two signal voltages is constant (but not necessarily zero). And if you want to be picky, the point in the circuit where the signal is determined to be balanced should be at the source. (Although you could combine a somewhat unbalanced source with just the right unbalanced path to produce a balanced signal at the reciever,)
For those who wonder why a balanced signal source rather than just a differential signal source is useful, one benefit is that for distances from the transmission line which are large compared to the conductor spacing the electric and magnetic fields associated with the balanced signal fall off very rapidly, reducing the likelihood of inducing noise into other, not necessarily differential, signal circuits.
Last edited:
Jared Foster
Member
- Location
- Bakersfield, Ca
- Occupation
- Instrumentation Tech
Are you certain that audio quality is being affected by grounding because you have already implemented adequate acoustic remedies? (Wall panels, monitor stands, sound clouds, sound sweeps, etc) I will assume that you have.
I seriously doubt that extra ground rods and such will make any difference. The primary source of audio corruption from electrical is from ground loops and electrical interference, not from the earth. Review all your analog connections to determine if the shielding is connected at both ends to a device that is also grounded at the wall outlet. A ground loop can circulate fairly large amounts of current, with little electrical interference, as it is essentially a short circuit. That current can then alter the reference plane on any impedance pads in use, and subsequently distort the audio. Loop isolators are relatively cheap (especially compared to audio equipment). Consider them as cheap insurance.
I seriously doubt that extra ground rods and such will make any difference. The primary source of audio corruption from electrical is from ground loops and electrical interference, not from the earth. Review all your analog connections to determine if the shielding is connected at both ends to a device that is also grounded at the wall outlet. A ground loop can circulate fairly large amounts of current, with little electrical interference, as it is essentially a short circuit. That current can then alter the reference plane on any impedance pads in use, and subsequently distort the audio. Loop isolators are relatively cheap (especially compared to audio equipment). Consider them as cheap insurance.
- Location
- Illinois
- Occupation
- retired electrician
The so called isolated ground that you speak of, connects to the exact same location as the EGC in NM cable when the circuit originates at the service equipment. How could a second grounding conductor connected at the same location accomplish anything? Note that I am assuming a dedicated circuit for the audio equipment with that NM only serving the audio equipment and I specified the use of non-metallic boxes for the circuit.
smalltime
Member
- Location
- Roanoke, VA
A "better ground" doesn't have anything to do with reducing noise. Hopefully this will help.
This is from Grounding and Shielding in Facilities, Ralph Morrison and Warren H. Lewis, pages 168-169:
"Let's separate fact from fiction as we review Circuit Theory in contrast to Sump Theory. Circuits are closed loop systems and noise doesn't seek to go to ground (earth).
There is a school of thought that suggests that a noisy system can be improved by finding a better or quieter ground. This concept of good, better, or best ground has no technical basis and is not supported by the presently understood laws of physics. It is true that it is possible to experiment with grounds and sometimes there can be an improvement and sometimes things can get worse. These experiments are not the engineering approach that should be relied on. If all the grounding stories were compared it is obvious that they would tend to disprove one another.
The search for the perfect ground falls into the main category of the 'Holy Grail' and a subcategory called the 'sump theory of electronics.' In this theory noise in the form of current is skillfully directed by the designer to a point where it is poured into an electrical sump (usually earth) where it permanently disappears. The connection at the sump is called the 'quiet' or 'noisy' ground depending on the designer's perception of the day.
This theory is not taught (we hope) in any reputable school where the prevailing circuit theory concepts require all currents to flow in loops. If the current is poured into the ground here it must come out of the ground over there. The return path must either be metallic or through some capacitance or both. In the 'sump' view the sump never fills up. The designer believes that only certain selected spots can work well. This is analogous to the idea that sewage will not run up hill.
Some designers mix circuit theory and sump theory. The flow into the sump is equivalent to a shower head directed at the sump. The flow returns from the sump to the shower head. Since many shower heads may be pointing at one sump the designer cannot truly control the effect.
The search for this perfect ground leads to some marvelous rationalizations. Terms such as 'quiet,' 'noisy,' 'computer grade,' 'isolated,' 'dedicated,' 'insulated,' 'single point,' 'separate,' and 'signal' appear in the literature to classify grounds without any definition. The rationalization continues by assuming the other party just does not understand grounding and believes that it is a black magic art form and not an engineering discipline."
And correct, a "balanced circuit" isn't about matching the impedance of the output with the impedance of the input.
From Noise Reduction Techniques in Electronic Systems, 2nd Edition by Henry Ott:
"A balanced circuit is a two-conductor circuit in which both conductors and all circuits connected to them have the same impedance with respect to ground and all other conductors.”
Your good signal could be differential on both conductors or you could have a good signal on only one of the two conductors. If you use the differential signal you'll gain 6 dB. Regardless, if you have a balance of impedance and the conductors are twisted around one another, common-mode noise will be rejected at the differential input. How well it does that is called the CMRR for Common-Mode Rejection Ratio.
I personally find it fascinating that balanced circuits were developed during the transition from telegraph to telephone where voice required a higher signal-to-noise than the dots and dashes used for telegraph.
From A History of Engineering and Science in the Bell System:
"In the meantime, other problems were arising from adherence to telegraph practices. It had long been the custom in telegraphy to use a single wire with a ground return, not only to save the cost of the return wire but also because the ground return, on long circuits, offered a lower resistance path than a wire. Such a circuit is coupled conductively, inductively, and capacitively to other nearby electrical circuits and picks up unwanted noise from them. The interference of such circuits with the telegraph signal was not serious, but with telephony even small interfering signals were annoying and the higher frequencies employed increased the inductive and capacitive coupling to the point where they were intolerable. It was soon learned that a full wire (metallic) circuit would reduce the interference from external sources, particularly those at some distance, since the induced currents in the two wires would be roughly of the same magnitude and tend to cancel each other out.
It is generally agreed that the first major demonstration of a 2-wire metallic telephone circuit took place on January 10, 1881, on a commercial line between Boston and Providence. This demonstration was carried out by J.J. Carty, who is therefore credited with the introduction of the metallic circuit into commercial use. However, because of the expense of the additional wire and the need for new switchboards, several years elapsed before conversion to metallic circuits became general.
While the metallic circuit greatly reduced the interference from sources at some distance from the telephone pair, it did not completely solve the problem of interference or 'crosstalk' from a closely adjacent telephone pair on the same crossarm. The reason was that the disturbing wires were not both at the same distance from the two disturbed wires and consequently the induced currents did not completely cancel.
In 1886, the General Manager of the American Bell Telephone Company made the following statement:
We found that between metallic circuits with wire connected straight through, there was a very considerable amount of crosstalk, not quite as much, perhaps, as there would have been on grounded circuits, but not materially less. We have entirely removed that by securing a balance between circuits.
This so-called balance was achieved by a system of transposing the wires of each pair in such a manner that each side of the circuit was balanced not only with respect to ground, but also relative to the wires of other pairs on the line."
Transposing the wires = twisted pair and a balance of impedance. in 1886. Once again, the ancients steal our inventions.
This is from Grounding and Shielding in Facilities, Ralph Morrison and Warren H. Lewis, pages 168-169:
"Let's separate fact from fiction as we review Circuit Theory in contrast to Sump Theory. Circuits are closed loop systems and noise doesn't seek to go to ground (earth).
There is a school of thought that suggests that a noisy system can be improved by finding a better or quieter ground. This concept of good, better, or best ground has no technical basis and is not supported by the presently understood laws of physics. It is true that it is possible to experiment with grounds and sometimes there can be an improvement and sometimes things can get worse. These experiments are not the engineering approach that should be relied on. If all the grounding stories were compared it is obvious that they would tend to disprove one another.
The search for the perfect ground falls into the main category of the 'Holy Grail' and a subcategory called the 'sump theory of electronics.' In this theory noise in the form of current is skillfully directed by the designer to a point where it is poured into an electrical sump (usually earth) where it permanently disappears. The connection at the sump is called the 'quiet' or 'noisy' ground depending on the designer's perception of the day.
This theory is not taught (we hope) in any reputable school where the prevailing circuit theory concepts require all currents to flow in loops. If the current is poured into the ground here it must come out of the ground over there. The return path must either be metallic or through some capacitance or both. In the 'sump' view the sump never fills up. The designer believes that only certain selected spots can work well. This is analogous to the idea that sewage will not run up hill.
Some designers mix circuit theory and sump theory. The flow into the sump is equivalent to a shower head directed at the sump. The flow returns from the sump to the shower head. Since many shower heads may be pointing at one sump the designer cannot truly control the effect.
The search for this perfect ground leads to some marvelous rationalizations. Terms such as 'quiet,' 'noisy,' 'computer grade,' 'isolated,' 'dedicated,' 'insulated,' 'single point,' 'separate,' and 'signal' appear in the literature to classify grounds without any definition. The rationalization continues by assuming the other party just does not understand grounding and believes that it is a black magic art form and not an engineering discipline."
And correct, a "balanced circuit" isn't about matching the impedance of the output with the impedance of the input.
From Noise Reduction Techniques in Electronic Systems, 2nd Edition by Henry Ott:
"A balanced circuit is a two-conductor circuit in which both conductors and all circuits connected to them have the same impedance with respect to ground and all other conductors.”
Your good signal could be differential on both conductors or you could have a good signal on only one of the two conductors. If you use the differential signal you'll gain 6 dB. Regardless, if you have a balance of impedance and the conductors are twisted around one another, common-mode noise will be rejected at the differential input. How well it does that is called the CMRR for Common-Mode Rejection Ratio.
I personally find it fascinating that balanced circuits were developed during the transition from telegraph to telephone where voice required a higher signal-to-noise than the dots and dashes used for telegraph.
From A History of Engineering and Science in the Bell System:
"In the meantime, other problems were arising from adherence to telegraph practices. It had long been the custom in telegraphy to use a single wire with a ground return, not only to save the cost of the return wire but also because the ground return, on long circuits, offered a lower resistance path than a wire. Such a circuit is coupled conductively, inductively, and capacitively to other nearby electrical circuits and picks up unwanted noise from them. The interference of such circuits with the telegraph signal was not serious, but with telephony even small interfering signals were annoying and the higher frequencies employed increased the inductive and capacitive coupling to the point where they were intolerable. It was soon learned that a full wire (metallic) circuit would reduce the interference from external sources, particularly those at some distance, since the induced currents in the two wires would be roughly of the same magnitude and tend to cancel each other out.
It is generally agreed that the first major demonstration of a 2-wire metallic telephone circuit took place on January 10, 1881, on a commercial line between Boston and Providence. This demonstration was carried out by J.J. Carty, who is therefore credited with the introduction of the metallic circuit into commercial use. However, because of the expense of the additional wire and the need for new switchboards, several years elapsed before conversion to metallic circuits became general.
While the metallic circuit greatly reduced the interference from sources at some distance from the telephone pair, it did not completely solve the problem of interference or 'crosstalk' from a closely adjacent telephone pair on the same crossarm. The reason was that the disturbing wires were not both at the same distance from the two disturbed wires and consequently the induced currents did not completely cancel.
In 1886, the General Manager of the American Bell Telephone Company made the following statement:
We found that between metallic circuits with wire connected straight through, there was a very considerable amount of crosstalk, not quite as much, perhaps, as there would have been on grounded circuits, but not materially less. We have entirely removed that by securing a balance between circuits.
This so-called balance was achieved by a system of transposing the wires of each pair in such a manner that each side of the circuit was balanced not only with respect to ground, but also relative to the wires of other pairs on the line."
Transposing the wires = twisted pair and a balance of impedance. in 1886. Once again, the ancients steal our inventions.
- Status