|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
208.58.2.83
Of course it *might* be against code, but I'm just askin'...
Follow Ups:
Mine is almost like that of maabx. Starting at the breaker is a continous 12ga twisted pair with a ground wrapped around the twisted pair in the opposite direction at half the twist rate (twisted pair at 6 turns per foot and ground wire at 3 turns per foot). The entire length is in flexible plastic conduit, I'm down in a hole away from town so not much RFI here. I built strain relief into the box near the amp using nylon pieces and nylon ties. Going out of the box is a pigtail in armored cable with a female IEC connector. The whole cable is continuous from the breaker to the IEC connector. I talked with several electricians including a friend that works for Bonneville power and all of them said this was defensible under the code.I use this with a PS Audio HCA-2 which had a some transformer noise and played little gurgles and squeaks all by itself before I made this change.
x
Why the last few feet of power cable should make a difference? I'm not asking whether it does make a difference; rather, I'm asking for an explanation.In capacitors, we can talk about DF and charging dielectrics and stuff like that--there is reasonable explanation available.
Why would a few feet of high-priced fancy OFC or whatever in series with a bunch of regular 40-year-old copper wire (usually Romex or worse) make a difference?
And while I don't know for sure, I doubt that Clark's idea violates code. Our air conditioners are directly wired, but not our stoves and dryers.
There is a specific mechanism that I have been tracking down that is similar to what Al mentions.Many components that have traditional linerar power supplies (not switchers) generate a fair amount of high frequency noise in their power transformer. When the diodes in the PS switch off they generate a pulse of noise, this by itself can be dealt with fairly easily, BUT this noise frequently excites resonances in the power transformer which is much greater than diode noise itself. The transformers literally ring for quite some time after each 120Hz waveform. My checking of a bunch of components shows this noise to be in the 100-300KHz range for most components. This is high enough to sail right through most power transformers but low enough to be affected only slightly by most AC line RF filters.
Components can even be excited by each other. I had one component that was not generating any of this on its own, but when plugged into a strip with another component that was ringing, its own resonances were excited, winding up with two different ringing frequencies in the box and on the power cords. With all my usual boxes plugged in I had four separate ringing frequencies throughout the system!
What does this have to do with power cords? They radiate all this stuff! It can get picked up by your ICs, speaker cables, components themselves etc. My tests show that different power cords can have radically different amounts of radiated fields from this source. The cords can also effect the resonance itself by changing the electrical parameters the transformer sees. Different cords can change the frequency of the resonance in some components.
I did some experiments with a spectrum analyzer and a short piece of wire connected to a probe and traced out the field around different parts of the system, plugs, cable, fuse, transformer etc. The results were kind of interesting. The transformer itself actually radiates little energy, the power cord itself ratiates more, but the amount varies significantly with cord. Cords with twisted wires radiate less, shielded twisted much less. The greatest radiation is at the fuse. This makes sense because this is where the wires separate. In the cord the fields from the two wires tend to cancel, there is no cancellation around the fuse where the wires separate. The plug is alao a place where the wires separate a little more than usual which also causes a significant increse in radiation.
So the upshot is that cords do make a big difference in the amount of radiated fields. Fuse holders and fuse wireing also make a huge difference. I didn't do any tests about different fuse types or materials, but I could easily see how this could make a difference. Anything which keeps the two AC wires close to each other will decrease the radiated filds.
BTW in my system I attacked the problem at the source, I dampped all my power transformers so they don't ring anymore. Jim Haggerman has a nice paper on this at his site, I followed his procedure to measure the resonant frequency and impedance of each transformer and compute a damping network for each. This made amazing improvements in the system sound. Unfortunately the measurements are not easy, I happened to have the right test equipment to do this.
As Al mentioned Higher frequency RF issues can also be an important part of power cord behavior, especially with all the digital stuff floating around in our environment, but I found that the transformer generated ringing is far more of a problem, at least in my system.
Thanks for the interesting post.Yes, conventional linear supplies with capacitor-input filters generate RF noise, even if they use fast-recovery diodes. Schottky rectifiers should prevent this, but I don't have experience with them.
Power cords have to be flexible and resist abrasion. These requirements affect their base-band (power and audio frequency) performance. Their geometries also influence base-band noise rejection. RF resonance control is a third aspect of power cord performance. It is not widely understood, but can make the difference between acceptable and unacceptable audio performance.The AC outlet and plug present impedance changes to RF noise relative to the house wire and power cord impedances. RF energy is reflected by impedance changes ("mismatches").
This makes the power cord, the house wire cable from the outlet to the breaker panel, and even the feed from the utility transformer secondary to the breaker panel, act as separate RF resonators.
Resonance in low-loss objects can take modest inputs and create strong outputs. Consider organ pipes as a familiar example.
Power cords and house wire cable segments are low-loss objects, unless you obtain versions that have RF loss mechanisms built in. Oyaide makes a carbon fiber-jacketed house wire cable that will suppress RF resonances, but it is very expensive. Some after-market power cords are designed to suppress RF resonances, or manage to do so as a consequence of some other design choice. The carbon fiber outlet plates made by Furutech and Oyaide, and the carbon fiber plug bodies on some Oyaide models, also help reduce the RF resonances. Some folks use ferrite beads or clamps, but I find the audio performance of the ferrites to be unacceptable.
Why the last few feet of power cable should make a difference? I'm not asking whether it does make a difference; rather, I'm asking for an explanation.Not sure about other countries, but in the U.S., the voltage at the utility pole outside the house is 10kV (at the utility pole transformer primary). So there's a voltage stepdown by a factor of 10k/120 or 83.33 before going into the house. The impedance is stepped down by the square of that, so it's stepped down by a factor of 6944. So you end up with a really low impedance at the secondary of the stepdown transformer.
> > Why the last few feet of power cable should make a difference?Why should the water filter at the end of a pipe make any difference?
æNormal is just a setting on my dryer.
_____________________________
out the the hose connected to it? While this is a bit of the reverse of how we tend to think of the last few feet, it is still valid. Resistance, capacitance, inductance all may vary, which affects the final product. Heavens, why would it matter what value fuse you put in the power line at the end if you believe the last few feet "don't matter"?
Are you suggesting that the last few feet of power cable somehow acts as a filter? Please explain.
That plastic Romex is the worst. I use a very heavy gauge solid core teflon coated wire which is noticeably better. (Just wiring it to a good AC outlet right now).
a
...have a look at the book "The Complete Guide To Home Wiring", put out by Black and Decker tools. There's lots of great stuff there that they don't teach in EE school. They have some really good illustrations of different wiring configurations, including the dedicated circuit that maabx mentioned. Their example is a dedicated outlet for a microwave oven in a kitchen.When I get my house, I'm going to have a subpanel for the listening room. The idea is to run a high current line from the main service (main breaker panel) to the subpanel. Each outlet in the listening room will be a so-called "isolated ground" outlet, wired to the subpanel. That name is somewhat of a misnomer. It just means that each outlet has an individual ground wire back to the panel, without any daisy-chaining of grounds. So there will be a kind of "star grounding" arrangement, with the star being subpanel ground. The book I mentioned gives instructions on how to install and wire up a subpanel.
Regarding electrical code, I scrounged a PDF of it, but when I tried to read it my eyes glazed over. It's like trying to read a legal document.
Pray tell, where did you get the PDF document??
Pray tell, where did you get the PDF document??I'd get in trouble if I mentioned it :). If you'd like a copy, feel free to shoot me an email.
First, local laws can be more stringent that the NEC, so even if it's not against NEC code, it may be against local codes.I believe that hardwiring a consumer piece that's not designed for hardwiring will require some sort of junction box to which you can attach the IEC pig-tail. Said box will have to be accessable, so you can't bury it behind drywall. And any cable coing in or going out of the box will have to be strain-relieved and maybe protected. Don't be suprised if the local inspector throws a fit.
BTW - household wiring is not taught in EE school (by which I assume you mean an college or university with an engineering curriculum), it's taught in trade school (or Vo-Tech if you will.)
And yes, the NEC is essentially a legal document, written to protect people from lawsuits and insurance fraud.
which can either be adopted by the jurisdiction or not...in part or in whole...the NEC has no authority in and of itself...it's all home rule local code on the state and local level.So, the converse of that may also be true...it may be against NEC, but allowable under local code...
And just to add, NEC is bare minimum.
My Balanced Power CPC has a removeable IEC power cord. My dedicated circuit is terminated with a female IEC connector.
x
This post is made possible by the generous support of people like you and our sponsors: