DIY power cable design considerations

When two three prong devices are plugged into a duplex outlet, there is a ground loop developed, and both pieces of equipment are now relying on a common bonding point.
If one of those devices is radiating as a result of either internal generated EMI, or asymmetrically filtering line noise to the grounding safety conductor, shielding a line cord could change the coupling to the victim unit.

This is simple EMI/EMC material.

The details of which end a shield should be connected to, 1audio and others can detail that ..

Me, I like to make things up...where's the fun in bog standard understandings??😉

John
ps. My work enviro requires all this and more. The motion control stuff by itself is crazy, 25Khz PWM into stepper motors, where coil to case capacitance forces the hf hash to get back to the drivers either by grounds or cable trays. Figuring out the path of the hash is extremely important, as are mitigation techniques. Line cord shields, while seemingly ridiculous for a long transmission line, are sometimes needed to fix local ground loop coupling.
 
As a test engineer of 40+ years' experience, to me it makes no sense to shield just the last few feet of a long transmission line. If noise is an issue on a power line input to a device, I'd specify a power line filter, rather than depending on a cord to do that. The shielded cord certainly won't hurt anything, though. Another problem with the cord is that if it's a detachable one, then it's possible that some other unshielded cord could be accidentally used instead, and there went your 'filter'.
Another thing; when I ordered a fancy piece of test equipment such as the now old HP4195 500Mhz spectrum analyzer for the test lab at work, the analyzer came with a bog-standard detachable power cord--nothing fancy about it at all. Other expensive equipment such as that by Keithley and Tektronix, which I also ordered, also came with similarly straightforward power cords. So, if that kind of equipment can meet its very demanding specifications with power supplied by such cords, then it should be a piece of cake for an audio component, with its comparatively minimal frequency range of 20Hz to 20kHz to meet its spec. An ultra fancy cord would certainly (hopefully) not hurt the equipment's performance, but it most assuredly wouldn't help it either. What a fancy cord will do, however, is to assuage your beliefs about it, enabling you to then enjoy the music on the equipment it powers. I'm done with this.
 
Last edited:
Sorry, I don't see the relationship between device output frequency and EMI/RFI noise sources and or EMI/RFI noise affected devices.

Please see the attached to help better understand some of the real issues related to what people are trying to filter out with power cords.
 

Attachments

Maybe this will get you started. Let me know if you need more.

It is well known that a forward biased semiconductor junction can act as a small signal modulator/demodulator. EMI/RFI ingress into semiconductor devices in audio amplfiers can be demodulated/remodulated with the audio signal being processed. This is well known by opamp manufacturers, who take steps in more recent opamps to minimize the problem.

Some measurements by PMA on the subject:
https://www.audiosciencereview.com/...ed-signal-with-1mhz-carrier.20828/post-689704

https://www.audiosciencereview.com/...-the-lm4562-lme497x0-family.10687/post-296533


Some EDN articles:
https://www.edn.com/review-tool-measures-power-line-emi/
https://www.electronicdesign.com/te...l-in-electronics-electromagnetic-interference


My explanation of Bybees:
https://www.diyaudio.com/community/threads/funniest-snake-oil-theories.234829/post-7154011
https://www.diyaudio.com/community/threads/funniest-snake-oil-theories.234829/post-7153981
https://www.diyaudio.com/community/threads/funniest-snake-oil-theories.234829/post-7154881


Quote from "Grounds for Grounding, 2nd Ed." (IEEE publication):
The AC-powered computer and peripherals are interconnected through their respective electrical safety ground wires within the building’s power distribution network. A voltage drop across any impedance of the grounding system within the facility may result in a possible ground potential difference between equipment. Computers may also be connected to peripherals via unbalanced RS-232 data communications cables. Multiple ground paths frequently exist. Ground loops caused by RS-232 links between devices may cause computer lockups due to power line transients and electrical noise coupling onto the interface through the ground reference line. Similar results commonly occur within multimedia systems (audio and video) with TV screens and a DVD player interconnected via an HDMI cable, for instance.


Quote from the first attached paper:
Electronic equipments that we use in our day to day life are constantly being exposed to Electromagnetic Interference present in the environment. These interferences can be in the form of lightning, electromagnetic pulses, RF noise, etc which gets coupled to a product directly through AC / DC power lines and input / output lines leading to the equipment malfunctioning.

EDIT: Second Article attached

EDIT 2: I would also refer you to the post by @1audio https://www.diyaudio.com/community/threads/diy-power-cable-design-considerations.405286/post-7506640
 

Attachments

Last edited:
The experts generally don't have a problem knowing what's happening and how to fix it.

The OP doesn't exactly need a BSEE degree to understand all these things, but definitely needs much of the knowledge learned in getting a BSEE.
Every conductor has resistance and inductance, and every two conductors have capacitance between them. We (or you) don't need to always take into account all these things (it's complicated enough when we do), but it takes experience to have a good idea of when to ignore one or more of these. And sometimes you're wrong, and have to go back and see what you need to take into account.

Yes, high-end test equipment comes with "ordinary" power cords (three-conductor, two mains current and a safety ground) with the socked thing on the equipment end, initially called "HP cords" and then they were used for microcomputers and everything that needs a ground. It's the engineers who use these that know when they can all be plugged in to the wall, and when one or more needs to be run off an isolation transformer(s).
 
It is already uncontroversial that RFI incursion can have audible effects. Please search for the words "veiled" or "grainy" in the attached document. Then search for "Blame" and read the next few slides that follow.
 

Attachments

Last edited:
That presentation (I'm not even halfway through, though I've seen a similar, shorter one from Mr. Whitlock) reminds me that "we" have discussed using balanced interconnects DECADES ago, which would solve a lot of (but not all) problems regarding hum and ground loops. I recall an article discussing balanced connections in consumer stereo equipment in the 1990s in what was it, "Audio Electronics" or "AudioXpress," the magazine title that was previously "Audio Amateur," but nothing changed among manufacturers. Mini XLR connectors would be ideal for this, but audiophiles still believe gold-plated RCA connectors to be the "gold standard."

Bill Whitlock certainly knows "everything" in this area. I recall in one of his articles (either on the Jensen site or a recent edition of Handbook for Sound Engineers) he writes of putting 60Hz current between the ground connection of a balanced input of equipment and its ground connection of the power cable and measuring it on the audio output to test its immunity from ground loop currents, a possible/likely problem with not-well-designed equipment using balanced connections.

ETA: I see he discusses the Pin 1 Problem and the "Hummer" device to detect it.
 
Last edited:
I posted that Whitlock presentation (which might be thought of as a white paper on the great benefits of audio coupling transformers) for a particular reason. Just as a reminder, Mr. Whitlock, while very well versed on the subject of grounding and shielding, does come from a certain business perspective. He shows transformers can be an excellent solution, but he says nothing about problems with transformers and what adverse effects they may have on sound. Just sayin' 🙂
 
Threads are by their nature a conversation, brief digressions usually tend to be deemed okay so long as they don't go on too long.

The subject was about power cords. That fact that I have to educate people about grounding and shielding for them to start to understanding about power cords is unfortunate. Also unfortunate many still don't seem to get it.

To cut to the chase, power cords as filters were at one time a precursor to filtered power inlet modules. Now the situation is partially fixed, and at the same time become more complicated due to things like class D amplifiers and SMPS, among other technologies.
 
Last edited:
Wait! isn't this thread about power cables?
How did it jump to sensitive input stages of semiconductors?
he writes of putting 60Hz current between the ground connection of a balanced input of equipment and its ground connection of the power cable and measuring it on the audio output to test its immunity from ground loop currents, a possible/likely problem with not-well-designed equipment using balanced connections.
ETA: I see he discusses the Pin 1 Problem and the "Hummer" device to detect it.
Yep, that is the John Windt "Hummer Tester"
He wrote a 1994/94 Audio Engineering Society paper.
"An Easily Implemented Procedure for Identifying Potential Electromagnetic Compatibility Problems in New Equipment and Existing Systems: The Hummer Test"
Still available as a small part of a great 85 page AES Journal for $15.