• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

RF Interference in Speaker Cables??? (video)

JimWeir

Member
Joined
Jun 10, 2020
Messages
48
Likes
75
Very Interesting stuff! Your argument reinforces the findings described in tomchr's excellent post about conformity bias tests by Asch and later Greg Burns. In 2005 Burn's measured brainwave activity to prove that conformity bias alters the perception of test subjects at the neurological level, that is, "social conformity literally causes their brains to rewrite reality."

I previously thought of bias as nearly imperceptible prejudice that nudges us towards conclusions that are not objective. Your references and Asch/Burns suggest it can be far far stronger than that, totally upending our ability to perceive reality.

Closer to audio, Toole at Harmon compared blind testing uncertainty with mono vs stereo. Stereo listening greatly expands uncertainty in blind testing relative to mono. In mono, almost the same signal is a both ears, the brain does not have to contrive a 3-D space.
I do all my DYI speaker development listening in mono on one speaker until I’m comparing to versions. Then the signal is still in mono and the speakers are side by side.
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,874
Location
Santa Fe, NM
I'm not curious enough about this since I never use that dimmer when I'm listening; I have relay controlled lights for that which are either on or off and there is no buzz then. The buzz is just something I notice when I enter the room to fetch something and turn that dimmer controlled light on. Those horns are insanely sensitive, especially since they are run off an active crossover before the power amp, so the horns are directly connected to the amp. I've run them for fun off of a small transistor radio (with a capacitor high pass) and they went extremely loud.
Interestingly, in the latest online review of the Wilson speaker in Stereophile, JA notes a buzz problem similar to the one I experience (see Footnote #1 at the end of the measurements).

Wilson Review Measurements
 

Cbdb2

Major Contributor
Joined
Sep 8, 2019
Messages
1,553
Likes
1,534
Location
Vancouver
Great video. One error I noticed, maybe someone already pointed it out, but any voltage applied to the output of an amp is reduced by the feedback divider the same amount as the gain of the amp,before it reaches the negative input so you dont need to add that 27db? of gain ( you cant mix dbv and dbm directly also) So the inpact is 27db less than Amir said. Not to mention the output zobel and stray capacitance. And impedance matching, which matters at RF, I doubt the amps output impedance at RF is the same as the cable. My one question, that -55dbm, what load, 50ohms? The feedback is voltage so just wondering what that was?
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
before it reaches the negative input so you dont need to add that 27db?
Actually if you put RF in the output of an Amp the resulting demodulated noise amplitude has not mush todo with its LF gain.
At high frequency there can be enough phase delay that negative feedback becomes positive feedback?

My one question, that -55dbm, what load, 50ohms?
Thats also waht i thought.
 

PeteL

Major Contributor
Joined
Jun 1, 2020
Messages
3,303
Likes
3,846
Great video. One error I noticed, maybe someone already pointed it out, but any voltage applied to the output of an amp is reduced by the feedback divider the same amount as the gain of the amp,before it reaches the negative input so you dont need to add that 27db? of gain ( you cant mix dbv and dbm directly also) So the inpact is 27db less than Amir said. Not to mention the output zobel and stray capacitance. And impedance matching, which matters at RF, I doubt the amps output impedance at RF is the same as the cable. My one question, that -55dbm, what load, 50ohms? The feedback is voltage so just wondering what that was?
I don't know if we may say "an error" In the context, that was hypothetical, like even if for some reason the gain was to be applied, it would still be a very small value, altough I understand what you mean it was not meant to show what happens in real life.
 

Cbdb2

Major Contributor
Joined
Sep 8, 2019
Messages
1,553
Likes
1,534
Location
Vancouver
Actually if you put RF in the output of an Amp the resulting demodulated noise amplitude has not mush todo with its LF gain.
At high frequency there can be enough phase delay that negative feedback becomes positive feedback?
.

Sure, depends on the amp. I was just thinking about the resistive divider between output and neg input, most amps don't have much else between, but there are always parasitics. It would be interesting to measure. And audio amps don't amplify RF so as mentioned its the modulation into the audio band that matters.
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,891
Likes
16,696
Location
Monument, CO
There is also bandlimiting (filtering) of the feedback signal itself to help reject RF... Sometimes more effective than others.
 

Francis Vaughan

Addicted to Fun and Learning
Forum Donor
Joined
Dec 6, 2018
Messages
933
Likes
4,697
Location
Adelaide Australia
One error I noticed, maybe someone already pointed it out, but any voltage applied to the output of an amp is reduced by the feedback divider the same amount as the gain of the amp,before it reaches the negative input so you dont need to add that 27db? of gain
The basics were ignored! The speaker cable is connected to the output of a power Amp. Those have a DF figure that define their output resistance. A mediocre figure would be 150. That equals to
an output resistance of 8/150=0.053 ohms. the higher the DF the lower the output resistance.

No.

Damping factor is a near meaningless term and doesn't define the output impedance, it is defined using the output impedance and calculated relative to the speaker's impedance. It is a marketing, not an engineering metric. The output impedance of a feedback design is the intrinsic impedance of the output stage divided by the feedback factor.

But to feedback. This is critical. Feedback does not work at every frequency. As frequency rises the effect of feedback drops. This occurs for a range of reasons, but is required to be designed in to avoid instability. (This is control theory 101, and is true for any negative feedback system, whether electronic, mechanical or whatever.) At radio frequencies the feedback network has no effect. Indeed at frequencies much higher than audio you will see every feedback amplifier's output impedance rise until it reaches the non-feedback value. At RF energies, the impedance seen at the speaker terminals can be almost anything you care and will vary wildly with frequency.

So the cables are picking the signals, they're going against the 'flow' to the amplifier, they're getting transformed into the audible range and amplified and back into the speakers?

Yes. There is no such thing as "against the flow". As described in a number of posts earlier, amplifiers can and do receive perfectly audible AM transmissions. All you need to do is to couple the RF energy into part of the circuit where there is some form of diode effect (basically any transistor will do) and the audio component will be demodulated and appear as an audio frequency signal. The amplifier will treat it just the same as the desired audio signals, and send it, amplified, to the output.

RF comes in a huge range of wavelengths. Not unakin to the range of wavelengths we deal with in audio (just travelling at c, not the speed of sound in air). AM radio stations have wavelength of the order of a mile, down to 5GHz WiFi which are 6cm. The manner in which these travel in and around equipment is very different. As the wavelength drops they behave more and more weirdly. As the old saw goes, to a proper RF engineer, anything that will go down a wire is DC. RF will travel down things you don't expect. Making equipment tolerant to RF energy isn't trivial, and is an ongoing battle. Those FCC certifications you see on gear are a mix of requirements. Manufactures of RF emitting gear are required to ensure that they don't emit energy anywhere except in the designed for range of power and frequency. Avoiding emitting harmonics of your carrier frequency is the sort of thing they are worried about. But your consumer audio gear is simply required to cope with whatever it sees assuming all the emitting devices are operating according to their specification and rating.

Calculating things like impedances for RF against audio components is not trivial. You can't use the audio frequency values. At any given radio frequency something like a voice coil could have nearly any impedance. Crossover components similarly so. Self inductance in capacitors and inter-winding capacitance in inductors will typically result in total inversion of component characteristics. And as the frequency rises, audio value components just look like a dead short, and then swap to looking like an open circuit, and so on. As the wavelengths get short relative the lengths of wire the wires look like transmission lines and energy reflects.

Bottom line. RF is really difficult. Properly testing gear takes million dollar test facilities and lot of time. Some idiot on the innertubes claiming that braiding wires is a cure all is a perfect example of the Dunning Kruger effect. So stupid that what he babbles is not even wrong.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,648
Likes
240,779
Location
Seattle Area
Great video. One error I noticed, maybe someone already pointed it out, but any voltage applied to the output of an amp is reduced by the feedback divider the same amount as the gain of the amp,before it reaches the negative input so you dont need to add that 27db?
Yeh, I thought about that after I finished the video. So reality is way worse than I showed.
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
Francis Vaughan Exelent post,thanks!

In addition to filtering the input and power pins, amplifier outputs also need to be protected from EMI/RFI, especially if they must drive long lengths of cable, which act as antennas. RF signals received on an output line can couple back into the amplifier input where it is rectified, and appears again on the output as an offset shift. A resistor and/or ferrite bead, or both, in series with the output is the simplest and least expensive output filter, as shown in Figure 7 (upper circuit).
its abut opamps but its also true for power amps
https://www.analog.com/media/en/training-seminars/tutorials/MT-096.pdf
The most common op amp response to EMI is a shift inthe DC offset voltage that appears at the op ampoutput. Conversion of a high-frequency EMI signal toDC is the result of the nonlinear behavior of the internaldiodes, formed by silicon p-n junctions inside thedevice, especially the ESD diode. This behavior isreferred to as rectificationbecause an AC signal isconverted to DC. The RF signal rectification generatesa small DC voltage in the op amp circuitry. When thisrectification occurs in the op amp signal path, the effectis amplified and appears as a DC offset at the op ampoutput. This effect is undesirable because it adds to theoffset error. EMIRR is a useful metric to describe how effectively anop amp rejects rectifying EMI.
http://ww1.microchip.com/downloads/en/appnotes/00001767a.pdf
 

Cbdb2

Major Contributor
Joined
Sep 8, 2019
Messages
1,553
Likes
1,534
Location
Vancouver
Yes the voltage is about -78db below 1 watt into 8ohms. But this means very little as explained, RF is tricky. And a measurement at the amps output with a speaker attached will probably change the numbers. RF is impedance matching sensitive, so putting the 50ohms in the circuit might also effect the numbers. Etc. RF, where an open wire can look like a short or vice versa. Gotta love it.
 
Last edited:

respice finem

Major Contributor
Joined
Feb 1, 2021
Messages
1,867
Likes
3,777
Speaking of Floyd Toole, probably everybody here knows his term "circle of confusion" https://seanolive.blogspot.com/2009/10/audios-circle-of-confusion.html
When I first read it, my spontaneous thought was "OK, but then, does real Hi-Fi even exist, with so many sources of error along the way"? I'm afraid, if we would want really high fidelity, all parts of the circle would need to be addressed. That would include building unified studios, recording rooms, and listening rooms, which is obviously impossible. The "room factor" alone is that big a cause of error, followed closely by speakers, that the whole electronics, cables etc. problems and "problems" are dwarfed by it. Just a thought...
 

ezra_s

Active Member
Forum Donor
Joined
Mar 23, 2020
Messages
293
Likes
327
Location
Spain
Not only I liked the video a lot as instructive, I also appreciate that it goes straight to the point putting aside the politically correct bs.
 
Last edited:

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,703
Location
Hampshire

jbrown

Member
Forum Donor
Joined
Oct 21, 2019
Messages
20
Likes
25
Location
Florida
Cross-sectional area is the main driver, but you only have to get above AWG 16 for it to no longer matter.

Oldie but goodie:
https://www.soundandvision.com/content/speaker-cables-can-you-hear-difference

Tom
I agree with you that it is old, but not that it is good. I am asking about empirical data, not opinions and guesses. Surely the spinarama can detect whether there is a significant difference between 24 gauge speaker cable and 12 gauge speaker cable, And possibly if there is a difference between fancy multi conductor braided cable or just parallel lamp cord.
 
Top Bottom