Question for Galen about power cables

Be nice.

I think he means well, and just draws a line saying what he feels is “audible” or measurable in a system. My data is correct, but the argument is how can it be heard or measured to show it might be audible. The approcah is awkward, I agree. The data he seems to feel is out of thin air is plenty well documanted by IEEE amd MIL-C-17 requirements for RF testing.

OK, should RF be filtered at all, seems to be his approach. For that we need proof it does NOT need to be attenuated. Are analog circuits tossed to the wind with too much RF. This assumes the stages aren’t RF filtered at the signal input, and the DC supply doesn’t have by pass caps for RF. Still, to lower the RCA loop DCR we take care of RF, too, and to 100 dB attenuation. That is 100% correct.

Power cords DO filter RF, passively, but is this required if the power supply does it too? AGAIN, we take care of this with the proper selection of dielectrics such that it isn’t a problem…or in his view a solution to a problem he feels doesn’t need to be addressed. No big deal either way for DISCUSSION but to say the other approach isn’t “working” is 100% WRONG. BAV power cables suppress RF based on the swept impedance values. That attenuates anything at those frequencies.

A nicer, “does it matter” approach and with power supply RF rejection values would have been informative and a benefit to the group. If an analog power supply is 110 dB RF attenuation, is anything needed in front of that with the power supply cord? That assumes we are talking INGRESS verses EGRESS out of the supply, a common issue with switching supplies. Here the power cord can benefit FCC and EC emissions tests for digital devices. Most thing are fine with INGRESS but can have EGRESS issues where a shielded power cord is a benefit.

Best,
Galen

2 Likes

Some use TRANSFER IMPEDANCE versus SEED. SEED is in dB, where transfer impedance is the opposite way to look at shielding. It is in milli-ohm/mtr usually. Notice Transfer Impedance it is opposite SEED. Transfer Impedance describes the way a shields surface IMPEDANCE looks with frequency, the lower the better. SEED looks at the ATTENUATION through the shield in a coaxial test chamber, the higher the dB the better.

SEED looks at the top of the shields reference value relative to the inside signal value and is a RATIO in dB or attenuation.

Transfer Impedance is a shield’s surface Impedance to ground at frequency and is an ohm value per unit length. Lower is better to remove RF to ground.

They both test the same basic function, though, how well and where does a shield really work.

Both are very real and well documented and used. Both methods show as you go up in frequency the shield effectiveness gets worse. Those shorter wavelengths capacitively couple easier as frequency rises. That’s why you want 100% optical coverage to block the electrical paths to the inner shield surface where it now sees the signal wire. It is primarily inductive coupled at lower freq. and capacitively coupled at high frequencies.

Once you get low enough, it is a MAGNETIC field and not an ELECTRIC field. Now we need materials that shield field lines, or have a low permeability to magnetic flux lines. If a magnet sticks to it, it has low permeability to flux lines and will shield magnetic fields as is shown in the illustration below.

image

image

Best,
Galen

2 Likes

Apparently your mommy and daddy didn’t teach you manners. So let me try:
When one starts a conversation with sophomoric arrogance, it’s most likely that the other party will just shut down and go off to do something more worthwhile than give you a second more of their time.
Does that make sense to you?

5 Likes

SMH.

You are obviously very knowledgeable.

However, your demeanor (hostility) and inability (refusal?) to advance and focus the discussion toward a common understanding of relevant points or foster a constructive dialogue is frustratingly unfortunate.

You could be an asset to this community - really (instead of behaving like an a**).

A shame…

6 Likes

audio2analysis

Stop being so silly. AUDIO is a “signal” same as any other. Ingress and egress attenuation ratio on a shields are perfectly symmetrical either way. Convention is ingress. It doesn’t matter what the “spec” area is military or otherwise, it is a FREQUENCY and as such MIL-C-17 counts same as any other. A good shield for RF is a double braid. And this experience counts as evidence as to what should be reliably used.

100 dB is the ideal SN for excellent noise floor in your equipment. To get that, and retain it, you’d need good noise rejection on down the line. If you want to have a worse noise floor, so be it. Like you said, a good shield isn’t expensive. A double braid gets you 20 dB better attenuation for not a lot more. But to get that level a 100 dB shield is a good place to start to make sure you keep what you paid for. You can’t get a better in use SN out better than what you put in. A single braid isn’t enough, see the charts.

If a digital device can pass the FCC tests (home spec is tougher, as we know) with the worst case cord, it can have an “open” socket like an IEC. If not, like you said the cord needs to be captive to the device. Not sure what your beef is on that and most devices fail EGRESS, not ingress as you and I both point out. Egress signal are far more aggressive.

Sure, RF will impact high impedance more than low impedance inputs. There isn’t much energy there to work at low impedance’s. Not sure your angst about that. Again, we both stated RF is better addressed in the power supply and DC rails RF shunts to ground. We have many a power cord that FAILS egress and needs shielding like voltage controlled motors for one and digital power supplies for two, that are used for audio. Look, for the FCC compliance label.

You say “p.s. if you are worried about Egress, again, that low impedance AC cable is not your friend.” All good shields have a low Transfer Impedance and are grounded at both ends, SPG are a disaster. AC cables AWG benefits are more related to the ground plane pumping.

High Impedance, will have less ripple. I stated that, yes? Ripple is directly related to the current draw from the supply.

Again you say , “IF you are already rejecting feedback, which is poor engineering, then you are already toast”. That is YOUR opinion. I never said to not use feedback any more than to never use it. Lab amps can use it well but it is a time based function. It improves some things and hurts others. It can be a real lazy way to make an amp. I can get you in contact with Barry Thornton the chief audio engineer for MANY audio companies such as SAE and he can read the ropes on feedback. There are many a zero feedback amplifiers on the market that work very well for audio with excellent specs. Don’t put biases into design anchoring. Both designs can be good or bad.

All supplies are “regulated” to some degree loose or tight or you won’t get DC. You can use zero feedback design with regulated supplies. Regulation isn’t the domain of required feedback. Not sure what you are suggesting here. Ripple is controlled, not eliminated and the line to define how good is good enough isn’t yours to make. The test needs to be accurate, the spec is up to the user to adopt or not and based on downstream effects to support or reject a single point of performance variation.

People have bigger issues with poor ground noise than peak current demands from the power supply. Higher current aggravates that ground noise. Sure, spread that demand out and it turn ripple into a voltage sag. You can adjust the cords AWG based on current draw ratio and control ground pumping attributes. That doesn’t mean a huge cord all the time.

Ingress and egress test the exact same attenuation, but the egress and ingress signals are vastly different so saying one is OK does NOT say the other one is. How can you suggest that? A shield is only a RATIO of attenuation, not the end amount, that depends on the external or internal noise source amplitude. How can you state, “since one can’t have one without the other”? No, it doesn’t.

I gave you the data, you still have nothing to show that it is “wrong” except you feel the line in the sand is where you put it and it isn’t. The customer gets to decide WHICH variables are effecting them. Ground noise where lower AWG cords can help but better to see where the ground is lifted or high impedance cords for RF issues and better to use quality supplies so RF isn’t the main issue to address. I think we agree on that.

You have really strong anchoring bias to what you feel is “adequate” verses better than that. The cords data is given as are all ICONOCLAST cords and you still complain because the line misses your definition of good enough. Sorry, but the world can make better cable if they can and in time, the standard will get better and the price drops. Would you STILL buy lower tested standards at the same price? Of course not. THAT is the only fact here.

I gave you the charts and data, but in return you just state an opinion it is “fact” that a variable moves this way or that when you damn well know the argument is HOW FAR it has to move to make a difference, not what makes it move. Bring that data, please. Your entire rebuttal is engineering generalities. Even a spec makes concessions to where the limit needs to be to get a product out the door. It isn’t ever “perfect” by any means.

Engineers made the o-rings for the shuttle, too, and many other disasters. Enough said there.

“Oh, it sounds very impressive, but all very meaningless.” Except that the exact same data I supplied you is used to apply meaning to ALL cable products, you just don’t like the spec lines other than your own. It’s right, but you don’t like it. Fine. We get it.

Best,
Galen Gareis

12 Likes

Way to go Galen…

This one oughta be fun…
Would be entertaining to see him expound to a bumble
bee why it can’t fly with the same attitude that he delivers
here…

The result would be irritated bumble bees and these don’t
sting just once…but deliver multiple stings…oouchy…

It’s just the numbers and science…

Oh well …
Thanks Galen,

Best wishes

2 Likes

Way to go Galen,

You have single handedly shown what a bunch of sycophants this website attracts.

Yes you showed graphs. You also referenced a shielding specification THAT IS DESIGNED FOR TRANSMISSION OF RF AT POWER … which is not what an AC cable is.

Your whole posts are just sad obfuscation to take advantage of the ignorant. If you were a licensed engineer, I would be reporting you to the licensing body.

All fluff, no substance. Not one iota of proof anything you claim makes any difference, heck, not one iota of actual analysis that it would make an iota of difference.

Truly pathetic Galen. A real engineer would be embarrassed by what you have become.

@elk - how much more of this do we have to tolerate?

:poop: :poop: :poop:

I have suspended @ audio2analysis for six months after previously privately mailing him as to his actions.

It is unfortunate. He is bright, capable, and could have been a great asset to the forum. Instead he chose a different path.

12 Likes

@Elk
:smiling_face_with_three_hearts: :smiling_face_with_three_hearts: :smiling_face_with_three_hearts:

Thank you elk…you have done us a great service…!!

Congratulation elk for doing us this much needed service…!!!

Best wishes

1 Like

@rower30 - It’s coincidental that you mentioned braid for power cables. I recently purchased a roll (below) that’s sitting on my bench waiting to become a power cable. The cable is going to consist of 9 AWG for neutral and positive (two 12 AWG wires for each in a star quad), a 10 AWG ground wrapped opposite the Star winding, and then covered in the braid below followed by Techflex (it’s gotta be presentable). It’s should be a rather stout build.

1 Like

We’ll have to be on the lookout for his nom de plume.

3 Likes

I’m sure his language will give him a away. But we have our attack Elk at work . . .

1 Like

“Yes you showed graphs. You also referenced a shielding specification THAT IS DESIGNED FOR TRANSMISSION OF RF AT POWER … which is not what an AC cable is.”

Say what? A shield is a shield. It attenuates RF around a cable (ingress or egress) same on an AC cable, digital Ethernet cable or coaxial CATV cable. It is the same test. There is no such thing as a test for just power cable. FCC test limits yes. The shield characteristics are universal to the design. SEED or transfer impedance are about the shield’s attenuation ratio chacteristics, not the signal it is protecting or the environment. A shielded power cord with RF on it has to be looked at as the ratio of the signal to the RF noise. Do we need more RF attenuation? Funny that the advent of shielding tests was for CATV coaxial cable, and not power AC. Yes, the limits will vary to the application not how a shield reaches these limits or the design necesary to get there.

My take is you are talking about VFD cable and yes, those have digital drive data on them with limits. That doesn’t change the performance of how a shield works, though. Shields aren’t only about RF at power and cord EGRESS as you try to imply.

Notice we DO NOT sell shielded power cables! I do not see it as a benefit to users and it is costly. Passive RF attenuation is plenty fine for AC systems with quality power supplies. I’ve stated OVER and OVER until we can detect signal changes based on the amps power delivery, it is hard to say a power cord has an influence to a repeatable standard, and not just variations in perceptions by end users. Better can be better, and we push the boundaries to see what happens. Some want to try those boundaries so we provide the proper tests we can do and properly let the user know what the tests are and how they work.

The data is 100% correct and you have done NOTHING to show otherwise. You want to cement the specs where you feel they are “adequate” but no better. Really? So that’s where you hide? As soon as a spec line is legitamatley moved you cry foul.

PE engineers are boxed into fixed specs. They aren’t allowed to change the “reference” on their own. This makes sense so you aren’t sued. As long as you do it per spec, a disaster isn’t your fault when it fails, and fails they do. The specs are changed all the time to further protect liability. You know it and I know it. Engineering isn’t ever completely right. Get the “wrongs” lined up and BOOM. We have major spec revisions.

Entire SM fiber optic installations across parts of the country that FAILED BW testing? Why? All the PE driven specs were followed but the true nature of SM BW wasn’t understood by the likes of people who think the BOOK is the final answer and it IS NOT. The guidelines are nothing more than limiting liability to make sure we have no fault insurance. They aren’t really about what is right so much as it protect engineering companies from what we don’t know. That is needed, or we’d live in the dark ages, I agree. But never for a second think the book is “right” all the time.

There are places where we can move the specs above or below the FROZEN requirements. I’ve stated over and over to this crowd that it can’t sound better if it isn’t truly better measuring but…we are getting to be VERY good and the costs to try the design is high. I ONLY design to calculated and measured improvements. Who else does that? Who else offers TPC as a reference to keep cost down, and offers the exact same design with alternatives to users who can now honestly try to hear what the beef is about? Who published ALL the design data on their cable to prove HOW and WHY a design WORKS? Who tests every cable sold for R, L and C to show they do get the paper specs in real practice and as a finished assembly? To you this is “meaningless” fine. To many it is a truly better cable option to try. Something they don’t have except with ICONOCLAST.

ICONOCLAST is specifically designed to allow user’s to try TRULY better design fair and square with industry accepted MEASUREMENTS but with better than industry accepted spec limits that are not cost driven. We have no liability the bridge will fall, the shuttle will explode or the fiber link will fail to deliver. We can push for better designs that can eventually improve the PE accepted “limits” for a product once it is proven reliable and repeatable to a set standard of measurement.

“Not one iota of proof anything you claim makes any difference, heck, not one iota of actual analysis that it would make an iota of difference.”

Really? Have you even read the papers and looked at the TESTED results? Your beef is you feel it is past “good enough”? Your PE mentality limits are not the final answer are they? The physics used to make ICONOCLAST are ages old, but we push it past cost convention designs so people can try a “better” cable. If it doesn’t work for you send it back…we sell all the cost driven alternatives that are tested the exact same way. The better cable specs stay in place wheather you like them or not.

So far, you have brought nothing to show why even the “PE accepted” limits apply and why those limits should go no farther. Many PE specs are MINIMUMS. But does using better than that say it is “meaningless” and you should be sued?

If you want to try to prove better R, L and C are against the law go for it. If you want try to prove ground plane differential isn’t proportional to the wire length between grounds, go for it. If you want to try to prove Vp linearity isn’t changed by cable designs and fixed by the physics at the frequency extremes, go for it. If you want to try to prove dielectrics don’t passively attenuate RF go for it it. If you want to try to prove smaller wires aren’t more current coherent across frequency go for it. I could go on.

You and I know you can’t, as these attributes are all accepted properties of the physical world that we have defined today. Your beef is just how good does an attrubute need to be to be a benefit. That’s fine, but you also need to show, as I explained earlier, that if say a power supply has 400 dB of RF from A to B frequency attenuation, the power cord needs none…so don’t worry about it. That’s a help. That the power cord DOES have passive RF attenuation isn’t a reportable offense is it? I just report what that attribute is and how it is measured. I also put specs into perspective, so no, I’m not deceiving anyone this way am I?

My concern with power cords is NOISE, either ground loop or pumping with some systems, nor RF. This is why we sell, unshielded cords. And no, this does not always mean a giant 10 AWG cord. The current in the cord can be proportionally ratioed to lower wattage devices. But, if we have ground loop noise, we can TRY a larger AWG cord and see if it is mitigated. Ground pumping isn’t very commonly heard. Best solution, fix the ground plane. I hope we agree on that.

It seems mighty odd that all the data and explainations given, you feel it is deceptive? But we’re supposed to accept your, “that’s a fact” answers with no fact shown? This is a data driven discussion, at least my posts are so you’re welcome to bring the data to put the proper spec into a PE reference mentality. No one will admonish you for placing the “minimum” standards and how they were agreed upon.

Help define what your accepted limits are and why. No, I have no reason to ban you. You kind of ban yourself with pure arguments that if it means nothing to you, it must mean nothing to the rest of us. OK, explain and show it.

Many interchange cables and hear changes (not power or digital cables in my case). The speaker / amp/ cable system does get changed as a system. The static (we use a system in dynamic models that aren’t available) spice modeling number do show that changing the R, L and C changes the performance. Who gets to say when it is good enough, though?

I am agnostic to feelings, and use just the data in my designs. Some chide me for not using carbon nano tube jackets or what have you. I only use proven calaculation and measurements. Same as a PE would. I just push the “maximums” as far as I can and to where even I say it is getting too expensive to go farther. Would farther but measurable have meaning? We won’t know until we get there.

Best,
Galen

6 Likes

mic-drop-drop-the-mic

2 Likes

Don’t they all seem to be one person? One personality?

2 Likes

I like to think upping the “background quiet” in cables, signals and components is the point of all of audiophile tweaking.

Yeah. I think it’s Sybil.

4 Likes