MEASUREMENTS: Do power cables make a difference with audio amplifiers?

Alright guys, as you probably know, here at the Musings, I'm not in general a "cable believer". This doesn't mean I'm a "cable denier", after all, I need them in order to hear something from the system :-). Over the years I have written about silly cable claims, and here's the summary post looking at all the cable varieties I've measured (I will of course include this article in there as well).

With measurements dating back to 2013, I have neither heard nor seen any evidence to compel me to recant my opinions on this matter. Furthermore, I have not seen any new articles exploring the topic with any kind of depth in the audiophile press. No audio cable company has produced material demonstrating a believable, sonic benefits for their $$$$ products. Despite countless ads in magazines, I see a lack of accountability from the mainstream press to investigate; it's really just about maintaining the status quo and promoting sales of high-margin luxury products as far as I can tell. No magazine wants to purposely publish stuff that results in loss of advertising revenue from a whole segment of products, right?

Up to now, the power cable measurements I've published have been with low-power devices (ie. measuring DAC output quality with different power cables), but what happens with higher current demands, when the device needs >100W from the wall socket? In the last few months, since I've been working on amplifier measurements, I have had the opportunity to try measuring power amplifier output using different power cables.

As you can see in the image above, I have what can be constituted as an "audiophile power cable" on the right - a nice looking, thick, "HiFi Audiophile Power Cord" with advertised 4N oxygen-free copper (99.997%, OFC), 10AWG conductors, of typical 6.5'/2.0m length. The connectors are rather robust Hubbell HBL8215C, "hospital grade" plugs.

Here's a better look at the cable - rather pretty, I think, in a manly kind of way that might impress friends and family:

In contrast, behold on the left "The Nastiest Audio Power Cable In The World". I found this yellow 50' extension cable in the garage as I was getting ready to put up some Christmas lights in December. The 3-prong male end is tarnished after being outside over the years. On the other end, I used the thinnest IEC cable I have - a standard, generic 18AWG, 2m wire I got from an old AV receiver.

Yuck... Tarnished, dirty 50' garden extension cable in the garage.

I ran this test as I was wrapping up my Emotiva XPA-1L measurements because this Class AB monoblock is the most power-hungry of my amps here at home. When idle, this amplifier sucks up 100W of power based on my Kill-A-Watt meter and when asked to play a 100W sine wave into 4Ω, this amplifier will easily approach 250W power consumption! Surely, if power cables make a clearly audible difference, these two wires should be literally worlds apart in terms of quality and should also easily show differences.

Here's what the "test bench" looked like with the "Nasty Cable" plugged into the Emotiva amplifier:

Notice that this testing took place in my basement right beside that fridge. I temporarily unplugged the fridge during this testing to remove a potential variable.

As you can see to the right, my Autoranger and RME ADI-2 Pro FS are being powered off lithium batteries to keep noise low on the measurement side.

Before posting test results, consider these questions. What do you think I'll find in terms of differences when I ask the amplifier to reproduce test signals at various power levels? Do you think there will be extra noise detected from the amplifier output between the "Nasty" 50+' and "Audiophile" 6' power cables? Should there be distortions like more harmonics or intermodulation tones? Would using the "Nasty" cable reduced power output? All quite reasonable possibilities to wonder and ask about.

Let's answer some questions one by one...

1. I have heard some people claim that cables will change frequency response... For example, an "audiophile" cable can make bass notes even better, typically "more powerful". The high end might sound "sweeter" and "more extended".

Well, as you can see, at a typical 1W output which is not an unusual, normal playback level for home systems, there's no difference between the two cables. Notice how precise the overlay is out to 40kHz! at 1.002kHz where I put the cursor, it looks like the Audiophile Cable must be below 72.75dB and got rounded down to 72.7dB whereas the Nasty cable rounded up to 72.8dB.

2. The Nasty Cable, being >50-feet must result in higher resistance and impede the amount of current to the amplifier. This would reduce the amplifier output when all else is the same.

Well this one seems to have some objective support. Remember that I'm measuring output voltage across the 4Ω load through the Autoranger so the dB scale on the left is relative to how much attenuation the device applied; what's important is to look at the difference between the two cables with the Autoranger applying exactly the same attenuation.

Indeed, we see that the amp output with the Audiophile Cable is visibly higher. That's an interesting finding, but remember, we're still looking at just 0.1dB-level difference at 1kHz! So, despite the amplifier drawing >200W from the wall socket, I'm simply not impressed that this small change could result in the type of audible differences some people claim can be heard with power cables.

3. I believe that I hear a difference in distortion - an Audiophile Cable "of course" results in lower noise and distortion, right?

Click on the image above to examine closer. See any difference at a typical 1W output level with amount of harmonic distortion across the spectrum? I don't.

Fine, how about looking at the FFT at 1W and also pushing this to 100W:

The graphs above are the FFTs at 1kHz with the amplifier playing at 1W and 100W into 4Ω using either the Audiophile or Nasty cable. First, notice that there was no difference in the amount of 60Hz hum. Even with a 50+' cable, it's not picking up anything unusual (I have some compact florescent pot lights in the room turned on). In fact, even if I plugged in the fridge, nothing changed (not shown). This amplifier appears quite resilient to electrical hum.

If anything, with the Audiophile Cable and the amp playing the 100W signal, there's some very low level "needles" around the 2nd harmonic not seen with the Nasty Cable!

If you look at the dBFS levels on the top right of each FFT, we see that the Nasty Cable has a slightly lower output level - this is consistent with the 100W frequency response graph in section (2). Now if we drill down to the highlighted THD+N results on the top left, we see that indeed, the Nasty Cable achieves slightly worse distortion numbers. Notice though that the THD+N amount is only something like 0.2dB difference between the two cables! While there is some inter-test variability, when I captured a few of these results, there is a general tendency towards mildly superior THD+N with the Audiophile Cable.

0.2dB difference in THD+N? That's all, folks...

4. But Archimago... It's all in the transients!

Okay. Let's see then if out TIM test tone that especially looks for distortion at the edges of square waves with >90kHz bandwidth (signal samplerate at 192kHz) shows a difference:

Nope... Nothing all that exciting. Remember what this test represents. It is asking the amplifier to continuously reproduce a 1kHz square wave where the edges are transitioning suddenly from about +6V to -6V (and vice versa) - >12V transients, 2000 time/s - into my 4Ω resistor while having a 12kHz sine wave overlapping. If this is reproduced poorly, we'll see noise and "sidebands" develop. While there are some sidebands with this amplifier on either side of 12kHz, notice that they are low (below -90dB from the 1kHz peak), and it made no difference whether I'm using the Nasty or Audiophile cable. 

5. Fine. I bet you'll see an abnormality with the more complex multitone and intermodulation tests!

Let's go straight into the triple-tone signal and look at total distortion & noise:

With 3 tones, there are ample opportunities for us to look for intermodulation sidebands as well as other distortions (including Class AB crossover distortion components) and of course worsening of the noise floor. Again, I'm measuring this at 6.32V into 4Ω, approximately 10W continuously which is louder than one typically would be asking the amplifier to output when listening at normal music playback levels at home.

Notice that I am able to show slightly worse TD+N with the Nasty Cable. What is key is the magnitude difference - again, only a 0.2dB difference. This is consistent with the worsening in THD+N above. I did this test switching back and forth three times using the two cables and can say that I found this difference to be real and replicable.

Conclusions...

It's good to see that my measurement system was sensitive enough to show a very slight (~0.1dB) amplitude reduction when the amplifier was pushed to 100W playback with the Nasty Cable. Likewise, there is a 0.2dB worsening of THD+N and TD+N when I used the poor cable. This gives us a good idea of the magnitude of differences one might find between power cables even though this test used an extremely poor outdoor garden cable for comparison :-).

As you can see, I didn't spend much money on that "HiFi Audiophile Power Cord" - ~US$20 ;-). Whether I used this 6' cable with claimed 10AWG 4N OFC conductors or an atrocious 50' utility garden extension cable with tarnished connectors plus 6' 18AWG IEC cable to the amplifier, it simply does not make a difference to any significant degree.

Some audiophile friends I'm sure will be impressed by the girth of the Audiophile Cable and how good it looks connected to my Hypex NC252MP amplifier these days. Nothing wrong with having nice looking window dressing (one of the "non-utilitarian benefits" we can spend money on)! However, there are probably a few thousand other things I'd put higher up on the list of purchases before I'd spend >$100 (never mind $1000+) on a single power cable in the belief that it affects "sound quality".

Over the years, audiophiles have been exposed to many articles and discussion threads about this topic. Just recently, I saw this thread on the Steve Hoffman Forum, and in there, we have a link to this Q&A "Why Power Cables Make a Difference" by Caelin Gabriel of Shunyata Research. Notice how we've always been treated to words. Words that allege significant differences or tries to make the topic complicated by dragging up variables like inductance, capacitance, impedance, EMI, etc. without useful context to the point where the reader is either numbed to death or falls prey to fear, uncertainty, and doubt thinking that this guy is right and there's some magical, remarkable quality to these cables that will enhance sound magnificently!

If a cable company (come on Shunyata, AudioQuest, Synergistic, Kimber, Furutech, Crystal Cable, Audience, and others) seriously wants to promote honest discussion, just show us some results from DACs or amplifiers that demonstrate the differences they speak of using their cables. Show us the magnitude and under which conditions they find a difference whether due to special cable geometries, OCC copper, silver conductors, monocrystals/filaments, specialty connectors, tuning "bullets", fancy dielectrics, super shielding, etc. Likewise, companies that sell expensive power conditioners claiming audible improvements - consider this breathless sales pitch video with drama and back story, check out ~10:00-14:00 for manufacturer claims - need to consider the same.

Remember what Carl Sagan said about "extraordinary claims..." The claims being made about power cables have poor face validity even without objective data. Unless there is evidence beyond testimony, we're truly just wasting our breaths and keystrokes incessantly discussing or disputing claims. Keystrokes are cheap and anyone can claim that they heard anything online. Most of the time, at best, I believe folks are just expressing a subjective opinion in good faith, at worst I wouldn't be surprised if there are occasions where testimonies have been written falsely to advertise products.

I think most reasonable audiophiles at this time in history recognize that the "high end audio cable" industry is really nothing more than about selling "impressive" looking luxury products (plenty of inexpensive good looking ones out there these days). On the odd occasion, perhaps a more expensive cable might be less susceptible to noise and if one is using extremely thin cables, a thicker one might reduce a bottleneck with power-hungry devices. I believe audiophiles and music lovers should just be reasonable and follow straight forward electrical principles (here's a 10' 14AWG Monoprice for example, or a 6' 14AWG Pangea Audio for a few bucks more), turn up the BS detector when you see the ads of 4-figure power cables claiming apparent sonic improvements.

As I've also said before, remember that there's nothing wrong with buying expensive products if it looks good to you; just be clear about what you're buying it for and be able to show humor regarding one's neurotic beliefs if this be the case.

------------------------

Okay guys and gals, remember, get me your impressions for the "Can you hear high harmonic distortion?" blind test started last week!

I had a quick peek at a handful of submissions and can say that at least 1 person got the order from least to most amount of distortion correct :-). It's clearly doable and like I said before, should be much easier to hear a difference than a "hi-res" vs. "standard res" listening test!

Despite all that's going on in the world, I wish you all a good February, and still enjoying the music.