I think we are close to a climax. People who think that the last 6ft of power cord makes a diff need to go outside and scream upwards to the stars.
Same with magic bricks.
Same with Tip-toe cone feet.
Same with the magic clocks.
Same with the hi-$$ solid maple what-nots from ... Maple Shade.
Same with snake oil.
Same with magic bricks.
Same with Tip-toe cone feet.
Same with the magic clocks.
Same with the hi-$$ solid maple what-nots from ... Maple Shade.
Same with snake oil.
Good to hear Frank. I'm about to build a new set of mono block power amps and I've run out of mains leads so i will either buy a couple of cheap off the shelf ones or make up a couple to the length I need. I have a couple of iec female plugs kicking around somewhere so will probably do the latter. It's roughly 5 years since I tried out a so called upgrade mains cord and as far as I could tell, it made no difference at all, but then I didn't expect it too so maybe I somehow prejudiced the outcome eh 
Oh dear... Who is talking about noise rejection of a power cord, not me. I responded to the thread title and mentioned a reason why power cords may make a difference.
And do not apportion statements to me that I have not made, it is the height of rudeness.No straw men thanks and I will give you the same courtesy.
The system ground currents will have a different magnitude and spectrum depending upon the electrical characteristics of the wires linking the component grounds together. Resistance is but one of the electrical parameters; I mentioned impedance, which is what should be talked about in AC circuits.
What I have said does not present a solution via power cords, merely highlights what is going on, that people like you don't appear to acknowledge or understand. I will be unsurprised if people say they hear differences from inserting different power cords, particularly in unbalanced systems. They are just tuning a circuit after all.
I am afraid that when there are ground current loops enclosed by system interconnects and power cords there will be noise pick up.
A slight diversion if I may: in fact people try all sorts of ways to reduce ground loops in their equipment, some of which render them unsafe, or non compliant with safety regulations. These include inserting series resistors on safety ground, using back to back diodes etc. Placing an inductor in series with safety ground would be a good way of filtering ground currents but then will violate the regulations I mentioned and might prevent breakers tripping when there are fault currents.
Last edited:
Not all gear in any of the studios I've been in has been balanced. And did you stick around in the studios you've designed when people bring in there own gear and start hooking it up? Do you sit thru there sessions? I doubt it. Try being an engineer in a few different studios, then tell me there are no grounding problems. And reread my last post, I agree that cords don't make a difference.
I'm not sure where to start. So, I color coded your erroneous understandings. (sorry, red is difficult to read)
The ground loop currents depend on the loop resistance of the ground, the loop inductance of the ground, and the transfer coupling between the source of the time varying magnetic fields and that of the ground loop. Single ended equipment, which is the majority of consumer equipment, is very susceptible to ground loop currents, and they are NOT designed to prevent creation of ground loop currents by their operation.
Where are you getting this??? First, the level of current is dependent both on the induction voltage AND the reactance of the path. Resistance is but one aspect of the path reactance.
A very misleading if not outright erroneous strawman argument. You should know better than that.
Another strawman argument, and entirely without merit. The hot and neutral can indeed be considered a balanced system, that is exactly what GFCI's use to measure hot to ground currents. But the argument here is NOT, and I repeat NOT a consequence of the balanced power, it is a consequence of the GROUNDING conductor within the power cords forming a conductive loop in concert with the chassis and interconnect cords.
You basically present the "miles and miles argument", mis-directing the argument to that of IR losses and supply holdup characteristics, which is not what the discussion is about.
Because they do not create an intimate grounding loop using consumer based high power equipment.
Again errors...First, look up Pin 1 problems. It's well established. The funny thing is, on another forum, there is a guy who states he is a pro soundman, works a large pro venue, and designs/builds and installs large systems for other customers...you know what he builds into the quote and the build?? A huge quantity of line level isolation transformers to break ground loops where he runs into problems. And you know why he needs to do that? Because the professional equipment he purchases, specifies, and uses is not immune to ground loop issues. So much for the pro world.
Then you too toss out the IR drop/supply holdup argument in a discussion about something entirely different. Stop the strawman arguments, I certainly know better.
Now, for the real question.
The line cord problem is for the most part, a problem caused by the design of the equipment. Given poorly designed (EMC) equipment, trying to design a line cord to compensate is going to be a completely random thing. As such, throwing silly high priced cords at the problem is in my opinion, a total waste of time and money.
Your line cord MUST meet code, it should be listed by an NRTL, and it must be sized for the ampacity requirements of the load.
I recommend you use only one line from the service panel if at all possible, to limit loop trapping on the grounding conductors in the branch romex. If two are necessary, have the electrician put the romex together such that the ground to ground distance is minimal. Near the panel, other loads can induce ground loop currents if their wires couple into the ground loop of your system. Within the load panel, have the electrician put the grounds against each other to reduce loop trapping within the panel.
I recommend that should you find you have a hum or noise problem, first try routing the cables to reduce the line cord to IC loop area. It sounds counter-intuitive, but you may find it solves hum and noise problems. Me, I've wrapped 125 foot unbalanced IC's around a simple home depot extension cord along it's 100 foot length, and resolved ALL hum and noise problems in a particular system.
I also recommend a multiport SPD in HT systems, that to help limit near stroke transients damage to equipment inputs. (While not part of the current discussion, I feel it's a great way to protect the equipment.)
jn
Last edited:
Changing a power cord will not change the sonic charactaistics of any system.
You are also incorrect on this too. I will repeat:
Current does not seek the path of least resistance to ground, it seeks "ALL" paths to ground. Apparently you do not understand basic electrical characteristics.
I toss this out so that my comments are not mistaken (quite like your lack of comprehension of what I wrote) to state that changing the power cord makes no difference in sound. This is not to be confused with using an undersized cord for the application where it starves the power supply and creates clipping or brown out from lack of current. Other than that, a wire is a wire and will make no changes to the sound.
I suggest that you re read before you comment.
Wow. No technical discussion, just a blanket "it doesn't happen".
If you wish, I can elaborate for you on ampere's law, faraday's law, and to understand loop coupling with shielded and unshielded wires, lenz's law. Also, while IEEE-1050 was retired about a decade or so ago, it still contains huge quantities of accurate information you clearly do not understand.
You crack me up. Clearly, you have no clue what I do for a living, so you feel you can simply say silly things and they'll be believable.
You obviously do not understand the topic. If you would like, I can point you to some primers on em theory, EMC theory and practice, heck even Whitlock has progressed in the last 10 years, presenting nowadays my engineering analysis' and discussions from a decade ago.
I did. I find your misunderstandings consistent with misunderstandings I found ten years ago.
I am surprised you continue to hold onto such misguided concepts.
jn
ps. I do not sell, make, or recommend any line cords. Ever. Just to code, and listed.
Allowing that possibility, a power entry filter will tend to prevent, or at least minimize that problem, too.
One must be careful with power entry filters as well. Some cause ground current as a consequence of the design. Never over 4 milliamps 60 cycle IIRC, but it may produce some. If you have a single ended amp or pre that is susceptible, the ground loop might induct some of that current's generated field based on the loop area and the aggressor power cord conductor geometry.
For an internally generated transient or signal, a power entry filter could actually aid in coupling between the transient aggressor and the chassis/line cord/IC loop. Poor layout of a power amp supply rails or power supply bridge to caps are typical agressors, and it's unknown of the entry filter would help or hinder the coupling. That would have to occur during the design phase, and they'd have to build a test setup to measure that. There is no industry standard test for either ground loop agressors or ground loop victims yet. Whitlock is closest to what is needed, but so far he only uses 60 hz half sine in an uncontrolled, low bandwidth method. Not the level of test I recommended back in 2004.
Whitlock gives industry presentations where he speaks of 60 hz currents in the field as high as 58 milliamps, so it's a real concern even in the pro world.
jn
Since you mention Whitlock, here's a paper of his that discusses many of the issues we've touched on here:
http://www.jensen-transformers.com/an/generic seminar.pdf
http://www.jensen-transformers.com/an/generic seminar.pdf
A simple question
Within a single ended stereo system, using two ic's to connect a source to an amp, where both components use a three prong plug...
How many of you think that the return current for the left channel low level signal goes back to the source by way of the left channel IC shield??
How many of you think that the return current for the left channel low level signal goes back to the source by splitting 50/50 between the left shield and the right shield, leaving the IC to IC loop formed susceptible to external magnetic field interception, thereby allowing induction??
And finally, how may think that the VAST bulk of the low level signal current in the bass through midrange regime will flow through the power cord grounding conductors from one chassis to the other because the path reactance is an order of magnitude lower (minimum) at bass and mid frequencies.. In the upper audio range, the IC's will split the return 50/50, and only when the frequency is sufficiently high (tens to hundreds of KHz up), will the lowest reactance path will be the individual shields.
This is rudimentary EMC understandings. The first pages of a reasonable EMC tutorial. to wit: understanding where the currents go, and how to control them.
jn
btw. I have asked these questions specifically for a very good reason. A coaxial wire does NOT shield from magnetic coupling if the current within the coax core is not the same as the negative current on the shield. In other words, in order for a coax to work properly, all the current within the core MUST return by the shield around it. This is NOT what happens in single ended systems where the chassis are bonded to ground for safety, or when two IC's are used interchassis. In fact, my avatar is the analytical result of the magnetic field created by a coax comprised of two shells of current carrying conductor.
Within a single ended stereo system, using two ic's to connect a source to an amp, where both components use a three prong plug...
How many of you think that the return current for the left channel low level signal goes back to the source by way of the left channel IC shield??
How many of you think that the return current for the left channel low level signal goes back to the source by splitting 50/50 between the left shield and the right shield, leaving the IC to IC loop formed susceptible to external magnetic field interception, thereby allowing induction??
And finally, how may think that the VAST bulk of the low level signal current in the bass through midrange regime will flow through the power cord grounding conductors from one chassis to the other because the path reactance is an order of magnitude lower (minimum) at bass and mid frequencies.. In the upper audio range, the IC's will split the return 50/50, and only when the frequency is sufficiently high (tens to hundreds of KHz up), will the lowest reactance path will be the individual shields.
This is rudimentary EMC understandings. The first pages of a reasonable EMC tutorial. to wit: understanding where the currents go, and how to control them.
jn
btw. I have asked these questions specifically for a very good reason. A coaxial wire does NOT shield from magnetic coupling if the current within the coax core is not the same as the negative current on the shield. In other words, in order for a coax to work properly, all the current within the core MUST return by the shield around it. This is NOT what happens in single ended systems where the chassis are bonded to ground for safety, or when two IC's are used interchassis. In fact, my avatar is the analytical result of the magnetic field created by a coax comprised of two shells of current carrying conductor.
Last edited:
Thanks. Note at the bottom of page 40 the scenario. While OT, that is one reason I recommend using a multiport SPD for, and one reason I recommend using one branch only for multiple pieces of ht equipment.
You linked to his 2005 presentation. He updated in 2012 IIRC, and I believe it's now over 200 pages. 223 for some reason strikes a bell.
My old comp died, so I am unable to provide linkage to it, nor am I able to access many of my diagrams, test setups, analysis', nuttin. I retrieved the HD from it, but it's SATA, and my portable HD enclosure doesn't work for that..sigh.
ps. Lots of my stuff was on another forum, but I will not link inter-forum.
jn
Last edited:
The mechanisms by which cables can have an effect was discussed by numerous members in another thread (cant remember which), with some excellent input from certain people who are much more knowledgeable in this field (jneutron springs to mind).
So cables can have an effect, but this effect is caused by bad engineering practice or by plain lack of understanding about what is going on.
One of the problems faced in audio is the range of frequencies of concern and their preference for different return paths, the low end wanting to follow the path of least resistance, as the frequency increases this changes with the higher frequencies moving towards the path of least inductance. This can be the cause of unthought-of loops where the thicker wires in power leads earth connections have a lower resistance than the return path through the IC connecting two pieces of equipment together, thus some of the signal return travels one path, some by another.
Again though if cables do change anything then there is an underlying problem with the equipment that needs rectifying.
Missed all this other discussion while typing up my reply and trying to work at the same time...
So cables can have an effect, but this effect is caused by bad engineering practice or by plain lack of understanding about what is going on.
One of the problems faced in audio is the range of frequencies of concern and their preference for different return paths, the low end wanting to follow the path of least resistance, as the frequency increases this changes with the higher frequencies moving towards the path of least inductance. This can be the cause of unthought-of loops where the thicker wires in power leads earth connections have a lower resistance than the return path through the IC connecting two pieces of equipment together, thus some of the signal return travels one path, some by another.
Again though if cables do change anything then there is an underlying problem with the equipment that needs rectifying.
Missed all this other discussion while typing up my reply and trying to work at the same time...
Last edited:
Thanks very much for spending the time and energy in sharing your expertise on this; I didn't feel like the fight nor do I have the expertise.
Rob.
On this side of the pond, the neutral is at the same potential as the ground conductor as it is connected at the load panel. If there's a line filter between hot/neutral and the internal chassis of a double insulated chassis, there is still a possibility of induction causing ground loop currents. But now it's only capacitive.
60 hz may not be so bad, but as the frequency climbs through the kHz region, capacitance plays more of a role.
I've actually had problems of this exact nature with a phono preamp/ttable combo. I ended up twisting the ttable line cord to about one twist per inch (25.4 mm for you length challenged guys on that side of the pond..
) And then twisting the phono IC's and phono ground wire together, then bundling them all together. Worked great.
I've also had times when I had absolutely no problems with the mix and match. Go figure..
jn
Electrons can apparently move in mysterious ways!
Over here we use a ring mains system where all outlets on the same ring can feed and return in both directions. I was told this allows for a higher load with smaller diameter cables and was brought in to save copper during the second world and we have been stuck with it ever since. Draw enough current and you create a massive loop antenna transmitting a fairly high energy signal at 50Hz.
All it takes is for one outlet to have a lower resistance connection at a screw terminal for any attached load to draw unequal amounts of power from the loop in different directions. Some houses noticeably hum and although I don't normally hear it, my missis says it affects her quite badly at times. She has epilepsy and is sensitive to things like too many mobile phones being switched on around her. She says they give her a headache.
Over here we use a ring mains system where all outlets on the same ring can feed and return in both directions. I was told this allows for a higher load with smaller diameter cables and was brought in to save copper during the second world and we have been stuck with it ever since. Draw enough current and you create a massive loop antenna transmitting a fairly high energy signal at 50Hz.
All it takes is for one outlet to have a lower resistance connection at a screw terminal for any attached load to draw unequal amounts of power from the loop in different directions. Some houses noticeably hum and although I don't normally hear it, my missis says it affects her quite badly at times. She has epilepsy and is sensitive to things like too many mobile phones being switched on around her. She says they give her a headache.
electricity does not “take the path of least resistance.” It takes all paths available—in inverse proportion to the impedance of the paths. Current flows through all available paths. The magnitude of the current flowing in each path depends on the voltage and impedance of each path. The lower the impedance of the path (assuming voltage remains constant), the greater the current. Conversely, the higher the impedance of the path (assuming voltage remains constant), the lower the current.
Picture two unequally-sized resistors in parallel. The current flowing through one resistor depends on the size of that resistor—not the one next to it. Assuming an infinite power supply, you could add 1,000 resistors in parallel and the current in that one resistor wouldn’t change. Because of this principle, such things as the integrated circuit are possible.
IEEE Std. 80 uses a value of 1000 ohms for the human body for touch voltage calculations. A 25-ohm ground rod in parallel with a 1000-ohm human will not make an installation any safer from electric shock. For example, if you touch a metal pole energized by a 120V line-to-case fault and there’s no effective fault current path, the touch voltage will be enough to kill you even if you bond the metal pole to a ground rod with a measured ground resistance of 25 ohms. The Figure helps illustrate the following:
1. A ground rod with a resistance of 25 ohms does not provide an effective fault current path. The pole will remain energized with dangerous touch voltage because the fault current will be only 4.8A (I = 120V/25 ohms). This is not enough to trip a 15A breaker.
2. Electrons take all available paths, and one of those paths is your 1000-ohm body.
3. OSHA and NFPA 70E say dangerous touch voltage is any value over 30V, and death from electric shock can occur from as little as 50 mA in a few seconds.
4. Touch voltage from an energized object is about 75% of the line-to-case voltage. So a 120V line-to-case fault results in a touch voltage of 90V. This can result in 90mA flowing though the human body indefinitely(I = 90V/1000 ohms), click here.
For many years, the street lighting and traffic signaling industries used ground rods without an effective fault current path, to ground metal parts of the electrical system. The thinking was that these installations were safe because “electricity takes the least resistive path and it bypasses high resistive paths.” People also thought the ground rod (earth) provided an effective fault current path to quickly remove dangerous touch voltage. Unfortunately such thinking resulted in practices that causes many deaths.
This thinking still exists. Worse yet, some equipment manufacturers state in their installation instructions that a 25-ohm ground rod without an equipment grounding conductor is an acceptable means of providing a safe installation.
Make sure poor grounding practices don’t lay you to rest. Yes, electricity does take low resistance paths, including the one of least resistance. But, it takes every other path available to it also. You can’t suspend Ohm’s Law and Kirchhoff’s Laws by driving 10 feet of copper-clad steel into dirt. To make an installation safe, ensure the touch voltage on metal parts never exceeds 30V for more than a few seconds. You can do this by bonding all metal parts to an effective fault current path in accordance with Article 250.
- Status
- Not open for further replies.