How about a manufacturer that installs captive 16 gauge two conductor power cords on all his equipment.
http://www.avahifi.com/
Power cords - do they really make a difference, other than potentially reduce noise?
- Thread starter ack
- Start date
How about a manufacturer that installs captive 16 gauge two conductor power cords on all his equipment.
http://www.avahifi.com/
Lack of UL certification
BTW many high-end manufacturers don't submit their products for UL testing. Ahem...
Supposedly Los Angeles County, among many jurisdictions, prohibits sale of non-UL devices. Furthermore, insurance policies may exclude non-UL approved gear from coverage.
BTW many high-end manufacturers don't submit their products for UL testing. Ahem...
Supposedly Los Angeles County, among many jurisdictions, prohibits sale of non-UL devices. Furthermore, insurance policies may exclude non-UL approved gear from coverage.
I have gotten far more mileage out of redesigning the power supplies, and rewiring the amplifiers in just about every case I've worked on.
Years ago, I had a Phase-Linear D-500 that was sensitive to buzz from light dimmers that use SCRs. One year, I tore the D-500 apart and rewired it. The manufacturer had not a clue about proper wiring practice. They had run all of the wires in one bundle, around the chassis. AC line, DC power, control wiring, and input audio wires--all in one bundle. I separated that, shortened all the leads and made direct runs. I even modified one of the PCBs, because there was a long trace that wasn't shielded on an audio input, and it was picking up hum from a nearby transformer. A strip of FR4 copper PCB, glued over that and connected to ground cured that. I also checked for audio across the power supply caps. Ended up putting 4.4 uF poly caps across the main electrolytics to shunt >1Kc signals that were backfeeding. Several things resulted: the buzzes and other strange noises from the line were gone. When the amp was driven to clipping, it no longer would produce a burst of ultrasonic oscillation as it comes out of clipping--the amp could clip cleanly for the first time in its life. It sounded better after those mods than it sounded when new.
My latest escapades into chasing noise came when I produced The Ultimate Fireworks Video on Blu-ray disc. It has an opening sequence with the national anthem being heard on a car stereo that was over 650' away on the other side of the airport, peaking at -85dB. Most Blu-ray players can't even separate this from the hiss and noise of their DACs. I had to get the Oppo BDP-83 to hear it without a ton of hiss. But every hum, ground loop, power supply rectifier switching pulse eddy currents, etc, were audible with such high gain playback conditions. I set to work on my Carver C4000 preamp, modifying the power supply, adjusting the ground return path to minimize switching pulses on the ground side, and installed snubber caps across the rectifiers. Then I went through the entire preamp and installed DC bypass caps on all the PCBs. I also put noise filtering in the power supply. The result is greatly reduced hum and noise. Now all the rest of the hum and noise come from the external components and ext processor chain.
The point of all this? It's not the power cord. It's poor design inside the equipment that it supplies power to. A good power supply design will filter out all the trash and noise that's coming down the line. And lord knows we have a lot of trash coming down our power lines, being in the middle of the aperture of a 2-bay ERI rototiller at 499' HAG driven by a Harris FM25K running full tilt for 50,000 watts ERP on 95.1MHz. When I built the house, I installed shielding in the walls and ceilings to keep the direct stuff out. The field is so strong that in our driveway, personal electronics that use microprocessors will stop responding to user input. That's right, a video camera will freeze up if you power it on in our yard, driveway, etc. We can hear rock music along with dialtone on our phones, too. And no doubt, it is in the power lines, too. So everything I modify has both RF filtering and AF filtering in the power supplies after I'm done. Hardly any audio manufacturer bothers with line filtering. Maybe an MOV transient protector, but no effort to screen out noise. It's really quite surprising. But worthwhile if you listen to high dynamic range content.
Years ago, I had a Phase-Linear D-500 that was sensitive to buzz from light dimmers that use SCRs. One year, I tore the D-500 apart and rewired it. The manufacturer had not a clue about proper wiring practice. They had run all of the wires in one bundle, around the chassis. AC line, DC power, control wiring, and input audio wires--all in one bundle. I separated that, shortened all the leads and made direct runs. I even modified one of the PCBs, because there was a long trace that wasn't shielded on an audio input, and it was picking up hum from a nearby transformer. A strip of FR4 copper PCB, glued over that and connected to ground cured that. I also checked for audio across the power supply caps. Ended up putting 4.4 uF poly caps across the main electrolytics to shunt >1Kc signals that were backfeeding. Several things resulted: the buzzes and other strange noises from the line were gone. When the amp was driven to clipping, it no longer would produce a burst of ultrasonic oscillation as it comes out of clipping--the amp could clip cleanly for the first time in its life. It sounded better after those mods than it sounded when new.
My latest escapades into chasing noise came when I produced The Ultimate Fireworks Video on Blu-ray disc. It has an opening sequence with the national anthem being heard on a car stereo that was over 650' away on the other side of the airport, peaking at -85dB. Most Blu-ray players can't even separate this from the hiss and noise of their DACs. I had to get the Oppo BDP-83 to hear it without a ton of hiss. But every hum, ground loop, power supply rectifier switching pulse eddy currents, etc, were audible with such high gain playback conditions. I set to work on my Carver C4000 preamp, modifying the power supply, adjusting the ground return path to minimize switching pulses on the ground side, and installed snubber caps across the rectifiers. Then I went through the entire preamp and installed DC bypass caps on all the PCBs. I also put noise filtering in the power supply. The result is greatly reduced hum and noise. Now all the rest of the hum and noise come from the external components and ext processor chain.
The point of all this? It's not the power cord. It's poor design inside the equipment that it supplies power to. A good power supply design will filter out all the trash and noise that's coming down the line. And lord knows we have a lot of trash coming down our power lines, being in the middle of the aperture of a 2-bay ERI rototiller at 499' HAG driven by a Harris FM25K running full tilt for 50,000 watts ERP on 95.1MHz. When I built the house, I installed shielding in the walls and ceilings to keep the direct stuff out. The field is so strong that in our driveway, personal electronics that use microprocessors will stop responding to user input. That's right, a video camera will freeze up if you power it on in our yard, driveway, etc. We can hear rock music along with dialtone on our phones, too. And no doubt, it is in the power lines, too. So everything I modify has both RF filtering and AF filtering in the power supplies after I'm done. Hardly any audio manufacturer bothers with line filtering. Maybe an MOV transient protector, but no effort to screen out noise. It's really quite surprising. But worthwhile if you listen to high dynamic range content.
Mark-another case of shoulda, woulda, coulda. Show me the perfect power supply? As in any electronics design, certain compromises have to be made whether it be for cost, space, sonic tradeoffs, etc. One of the biggest issues compromises in lower priced gear is in the size of the power supply and as a result, the equipment's dynamics. I've heard that time and time again in gear from manufacturers when comparing their statement product with their trickle down designed models.
If there was one perfect power supply, how come every manufacturer does theirs differently? Even in the rarified air of high-end audio.
Myles, there is a great deal of ignorance in the industry when it comes to RFI proofing. Almost no one addresses it. As a designer of commercial broadcast equipment, I am acutely aware of the problems, and with audio gear, when people are talking about the problem as some sort of mystery, for me it is obvious at a glance.
In higher end gear, there is no excuse not to provide even basic filtering at the input. Only the $99 stereo receivers from RatShack would stoop this low, in my opinion, but no... even the higher end manufacturers are completely ignorant of RFI protection.
As for dynamics, this is a matter for power amplifiers. Preamps with normally-designed and regulated supplies are limited by the op-amps themselves. Having a properly bypassed DC rail for each op-amp is the most important thing a designer can do, besides making sure the voltage regulators can supply enough current for the entire preamp. I've never run across one that could not. Noise in the rails is not as bad as I once thought, because op-amps have common mode rejection ratio and power supply rejection ratio exceeding 80dB. So as long as the power supply is stable and the DC lines are clean so no feedback/oscillation results, the power supply is then functioning properly and no further benefit would be had from overbuilding, other than possibly extending service life.
Power amps are a different animal. Rarely is the power supply regulated for the final pass transistors, so much of the momentary headroom comes from the store of energy in the capacitors. However, depending on the stiffness of the power transformer and rectifier, there may not be much voltage sag, so increasing capacitors on such an amp may offer little to no benefit. If there's a 55lb iron core transformer for a 1,000W amplifier, chances are it can deliver the current with less than 2% sag. However, many big amps with smaller transformers will sag 30% under load, long term. These can benefit with larger capacitors to store energy to extend the duration of voltage output before sag settles in.
In higher end gear, there is no excuse not to provide even basic filtering at the input. Only the $99 stereo receivers from RatShack would stoop this low, in my opinion, but no... even the higher end manufacturers are completely ignorant of RFI protection.
As for dynamics, this is a matter for power amplifiers. Preamps with normally-designed and regulated supplies are limited by the op-amps themselves. Having a properly bypassed DC rail for each op-amp is the most important thing a designer can do, besides making sure the voltage regulators can supply enough current for the entire preamp. I've never run across one that could not. Noise in the rails is not as bad as I once thought, because op-amps have common mode rejection ratio and power supply rejection ratio exceeding 80dB. So as long as the power supply is stable and the DC lines are clean so no feedback/oscillation results, the power supply is then functioning properly and no further benefit would be had from overbuilding, other than possibly extending service life.
Power amps are a different animal. Rarely is the power supply regulated for the final pass transistors, so much of the momentary headroom comes from the store of energy in the capacitors. However, depending on the stiffness of the power transformer and rectifier, there may not be much voltage sag, so increasing capacitors on such an amp may offer little to no benefit. If there's a 55lb iron core transformer for a 1,000W amplifier, chances are it can deliver the current with less than 2% sag. However, many big amps with smaller transformers will sag 30% under load, long term. These can benefit with larger capacitors to store energy to extend the duration of voltage output before sag settles in.
Mark, so as to bring this interesting thread back OT, given your obvious vast experience in examining and sometimes modifying the guts of a variety of gear in what way can/does a power cord influence sound and, if it does, how would that difference manifest itself?
Well Mark I don't agree any of your contentions.
Maybe in your arena the designers don't pay attention to RFI but not high-end designers I speak with. Pretty much all the high-end designers I speak with are esp cognizent of RFI. Many however feel that the cure is worse than the disease eg. most filtering efforts degrades the sound.
And sorry but power supplies do make a difference in preamps. Go back and read the POOGE articles in AA published years ago. I'm curious though. What high-end preamps have you actually heard in your system to base your contentions on??? And I'm talking about a real high-end audio system, not a boom box. I'm not talking about a 20 yo crappy Phase Linear solid-state designs that were about as equally one step above some of the crappy later Crown amplifiers.
What I would like to see is a comparative FR test done straight out of an amplifiers output terminals between a factory provided power cord and any highly regarded power cord using pulses of full range white and pink noise at 1w, and at maximum rated output as well as some points in between.
Then I'd like to see the same test done between amps that manufacturers say benefit from after market cords and manufacturers of amps that say theirs do not.
Then I'd like to see the same test done between amps that manufacturers say benefit from after market cords and manufacturers of amps that say theirs do not.
Last edited:
But Jack
That would be ... measuring. For many that would not suffice to explain the differences we ..err... hear ...?
No amount of objective proofs would convince a person who believes. Even when confronted with the hammer of DBT they have an explanation, not that their senses can be fooled but that such test is stressing shut down their discerning abilities .. admitting inone way that psychology plays a role in perception but only in the negative way in the absence of stress then the ears is superior to any measuring device.
Maybe I will get out and just watch this one unfold , maybe not. I do know however how humbling it can be to realize how much our ears in fact all our senses are unreliable.
OK so let's ask the question, what do you expect to see from these tests and what do these measurements add to how we perceive the sound of a cord or cable? I might point out that some pretty brilliant people have worked on the issue of measurements vs sound and walked away baffled. After all if FR told the story, then we'd all be reading Hi-Fidelity today. I suspect that one needs much more sophisticated measurements to show any sort of differences between cords. After all, in many cases we're talking subtle differences. And also, what measurements would correlate (it's certainly not a cause-effect relationship) when something degrades the sound?
Oh I hear differences Frantz. I really don't care if you believe me or not. What bothers me personally (only on an intellectual level) is that I can't explain why. When the music starts playing, none of this matters. As far as expectations and the emperors clothes and the mega watt bass pig goes, I've been disappointed more often than I've been impressed. I've rejected way more than I've taken in.
To my knowledge no noise-pulse tests have ever been done. It's always continuous pure tones/waves. To my knowledge there aren't any measuring tools that will give us the electrical equivalent of an acoustical waterfall plot. If I am wrong about this, please show me.
I can't believe RLC diehards sometimes. They test with a sine or square and think transmission line theory covers all the bases even if they know that there are and will be frequency deviations caused by deviations with any one of the three. Then call on creds to give an air of infallibility.
This test could settle this debate once and for all.
@Myles
I dunno about you Myles but changes in FR emphasis is what I hear that differs between my two long time PCs. For the record these are the only PCs with which I heard actual performance gains. Many of my former PCs just looked cool but I'd be damned if I could tell any difference whatsoever. Again only two actually made things worse. Yes they are minor. Minor enough that the differences only became readily apparent with the addition of 10 wall panels. In service of the music however, maybe not so minor.
Again, such a test would be put up or shut up time. Question is....who'll be brave enough?
But is everything we perceive as FR anomolies actually FR deviations? John Curl pointed out that some types of distortion manifest themselves as apparent "FR" issues when actually they're not!
Jack
So do I Jack.. Same with several folks under blind conditions.. My view and let's be clear on that is that there are differences between components many of these audible... I have no problem with that notion.. Not at all. What I have a problem with are the differences that cannot be sustained under blind testing or the refusal to accept the fact that one's ears can be fooled ...or that most "tweaks" simply don't work.
Now I am utterly confused here .. Isn't it a measuring tool that give us the waterfall plots? aka Cumulative Spectral Decays. They are obtained/derived from the impulse response of a system, one of the most used algorithm for the purpose is STFT (Short Time Fourier transform) . Impulse response measurments are used for amps as well, thus measurements such as slew rate and rise time ...
So do I Jack.. Same with several folks under blind conditions.. My view and let's be clear on that is that there are differences between components many of these audible... I have no problem with that notion.. Not at all. What I have a problem with are the differences that cannot be sustained under blind testing or the refusal to accept the fact that one's ears can be fooled ...or that most "tweaks" simply don't work.
Now I am utterly confused here .. Isn't it a measuring tool that give us the waterfall plots? aka Cumulative Spectral Decays. They are obtained/derived from the impulse response of a system, one of the most used algorithm for the purpose is STFT (Short Time Fourier transform) . Impulse response measurments are used for amps as well, thus measurements such as slew rate and rise time ...
Myles, I'm not talking about distortion but true output of pulsed pink and white noise with one cord (manufacturer supplied) versus another.
Frantz, No debate on your first paragraph. I accept that my ears can be fooled. That's why stereo is my passion after all. Blind testing in my experience can also fool the ears so that's where we can happily part ways as buddies
On the second part. YES!!!! That is the testing I'm talking about except FFT and STFT are done after output from the speaker not the amplifier under test. This brings in way too many variables. Mainly that we are no longer in the electrical domain, it has been "transduced".
Impulse response measurements are (again correct me if I'm wrong) done with a specified frequency into a dummy load. This can not compare to the data available via FFT and STFT which shows all frequencies simultaneously. Being a telecom engineer I think you see where I'm heading with this. No dummy load since load affects frequency. Questions are which, when, where and by how much. The challenge is that I really don't think you can get an accurate measure of pink or white noise out because you would I think need a dynamic load to provide the back EMF expected for each frequency driven at the same time as it would in the real world. That's why I'm thinking in pulses too. I'd like to see slew and rise time across the board not just 1kHz. Now we hear circuit changes or parameter changes like say changing the bias or replacing certain parts. But really, if you scan all the literature no manufacturer or DIYer has ever shown graphically how even rough 1/3 octave bands are affected.
Should it be possible, one would categorically prove or disprove that the power cord is not just part of the circuit but actually one that affects more than just sufficient power to the audio circuit. Such a test would also be useful for many other things like modeling and simulation in various stages of design and implementation.
Frantz, No debate on your first paragraph. I accept that my ears can be fooled. That's why stereo is my passion after all. Blind testing in my experience can also fool the ears so that's where we can happily part ways as buddies
On the second part. YES!!!! That is the testing I'm talking about except FFT and STFT are done after output from the speaker not the amplifier under test. This brings in way too many variables. Mainly that we are no longer in the electrical domain, it has been "transduced".
Impulse response measurements are (again correct me if I'm wrong) done with a specified frequency into a dummy load. This can not compare to the data available via FFT and STFT which shows all frequencies simultaneously. Being a telecom engineer I think you see where I'm heading with this. No dummy load since load affects frequency. Questions are which, when, where and by how much. The challenge is that I really don't think you can get an accurate measure of pink or white noise out because you would I think need a dynamic load to provide the back EMF expected for each frequency driven at the same time as it would in the real world. That's why I'm thinking in pulses too. I'd like to see slew and rise time across the board not just 1kHz. Now we hear circuit changes or parameter changes like say changing the bias or replacing certain parts. But really, if you scan all the literature no manufacturer or DIYer has ever shown graphically how even rough 1/3 octave bands are affected.
Should it be possible, one would categorically prove or disprove that the power cord is not just part of the circuit but actually one that affects more than just sufficient power to the audio circuit. Such a test would also be useful for many other things like modeling and simulation in various stages of design and implementation.
I admit that as best I can recall I have never performed TDR or s-parameter testing using a VNA (vector network analyzer) on a power cord. Actually, I did, many, many years ago, but don't recall finding out anything earthshaking in the process. As for various time- and frequency-domain tests over a wide frequency range, I assume it is routinely done for interconnects, but perhaps not power cables. Other than coupled noise (RFI/EMI/whatever) there should be nothing other than 60 Hz (50 Hz, whatever) on the power cable, assuming an adequate power supply. The isolation from the line in most power supplies is awfully high. I wonder what would happen if we took a "normal" cable and shielded it well? I suppose we'd have to shield it back to the generator in the power plant...
I have to think that if changing a power cable makes a noticeable difference in the sound, then either (a) the stock cord is well undersized, (b) there's a lot of "noise" in your room and using a better-shielded cord from some sort of conditioner/noise filter to the component helps, and/or (c) the power supply is poorly designed. I suspect the latter if differences are actually found in listening tests and measurements.
Whatever! - Don
I have to think that if changing a power cable makes a noticeable difference in the sound, then either (a) the stock cord is well undersized, (b) there's a lot of "noise" in your room and using a better-shielded cord from some sort of conditioner/noise filter to the component helps, and/or (c) the power supply is poorly designed. I suspect the latter if differences are actually found in listening tests and measurements.
Whatever! - Don
Jack
Impulse response suppose ALL frequencies possible from DC to infinity .. The device under test will naturally act as filter ...
It remains possible however to use an STFT or any FFT algorithm to analyze the impulse response and thus deliver a CSD plot from it .. In electronics that may not be too useful for transducers it indeed is..
Impulse response suppose ALL frequencies possible from DC to infinity .. The device under test will naturally act as filter ...
It remains possible however to use an STFT or any FFT algorithm to analyze the impulse response and thus deliver a CSD plot from it .. In electronics that may not be too useful for transducers it indeed is..
i've read all the way thru this thread. i've been thinking that with my Equi-tech isolation transformer my agenda for power cords likely needs to be re-thought. in any case; i need to approach power cords with an open mind.
starting about 10 years ago i was a bit of a power cord junkie. i started with the Shunyata King Cobra, the the Elrods, and then the Jena Labs Fundamental One's with the in-line conditioners i have now. i tried lots of different power cords and spent a good amount of time doing listening comparisons. right now i have mostly the Jena Labs PC's and also a few Telwire PC's in a few places.
i'm thinking that with the Equi=tech it might be best to choose one type cable and do everything that way. since reducing noise is dramatically less an issue than it might have been earlier, any input on my direction would be appreciated. my guess is that my previous power cord comparison experience has little value now.
i'll point out that in addition to the Equi=tech i also have 10 gauge Romex home runs to each outlet, and the outlets are all the Oyaide R-1's with the WPC-Z carbon fibre covers. so this is just a PC issue.
but with 3 tt's, 2 phono stages, 2 RTR decks, a tape repro unit, stereo amp, and 2 active subwoofers in my speakers i do have a number of power cords to add.
starting about 10 years ago i was a bit of a power cord junkie. i started with the Shunyata King Cobra, the the Elrods, and then the Jena Labs Fundamental One's with the in-line conditioners i have now. i tried lots of different power cords and spent a good amount of time doing listening comparisons. right now i have mostly the Jena Labs PC's and also a few Telwire PC's in a few places.
i'm thinking that with the Equi=tech it might be best to choose one type cable and do everything that way. since reducing noise is dramatically less an issue than it might have been earlier, any input on my direction would be appreciated. my guess is that my previous power cord comparison experience has little value now.
i'll point out that in addition to the Equi=tech i also have 10 gauge Romex home runs to each outlet, and the outlets are all the Oyaide R-1's with the WPC-Z carbon fibre covers. so this is just a PC issue.
but with 3 tt's, 2 phono stages, 2 RTR decks, a tape repro unit, stereo amp, and 2 active subwoofers in my speakers i do have a number of power cords to add.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |