MEASUREMENTS: Experiments in audio component grounding - using a bus bar & HumX. And on the last The Cranberries album, audio quality legacy...

You may remember a number of years ago, I talked about reducing interference I was experiencing with my Emotiva XSP-1 pre-amp. The issue had to do with the audio system picking up interference and noise through the pre-amp's "Home Theater Bypass" unbalanced RCA input when connected to my AV receiver. This "HT Bypass" mode is used as a conduit for the front channels and subwoofer out when the AV receiver is in use, typically when I watch movies with multichannel sound.

While this was a bit of a hassle for me, sometimes out of these hurdles and limitations, one is provided with opportunities to explore things like the 8kHz USB PHY packet noise which seeped into my system from the TEAC UD-501 DAC (not an issues these days with my Oppo UDP-205). Playing with things like the Corning Optical USB 3 extender allowed me to lower the noise level. Furthermore, I was also able to show that different USB hubs affected the severity of that 8kHz noise. Remember that much of this investigation was prompted in those days when devices like the AudioQuest USB Jitterbug and the silly single-port-hub known as the UpTone Audio USB Regen were being hyped up by certain websites and forums. To this day, I have not seen any evidence that this stuff improved things like noise level and jitter with reasonable asynchronous USB DACs.

Since those days, my system has changed a bit. Instead of my Onkyo TX-NR1009 receiver, I actually "downgraded" to an overall less expensive but still excellent sounding Yamaha RX-V781 receiver which gained me object-oriented Atmos and dts:X processing. Unfortunately, I could still hear hum with the Yamaha receiver connected to the Emotiva preamp and with a powerful gaming HTPC nearby, I could hear interference coming through when that's turned on. Remember, as demonstrated previously, a good DAC should not show excess noise even when hooked up to a noisy streaming device like this computer. Most of the time, this is not an issue since the hum isn't terribly loud and the only time I boot up the HTPC and use the AV receiver's multichannel output is for gaming rather than listening to hi-fi audio so a little bit of noise from the computer's electrical activity isn't really a problem.

Nonetheless, even if it's not generally an issue, ideally it would be nice to deal with these ground noises and hum as best we can. Today, let's demonstrate something one could try without much expense at all to address the noise... Using a grounding bus bar.

Remember that ground loops are caused when there is a difference between the ground reference potential of two or more devices in the system. In audio circuits this typically shows up as stray 50/60Hz hum transmitted through the unbalanced interconnect shield (eg. ground line of RCA cable) between devices. A "bus bar" is basically a convenient metal "central" connection point to which you can attach the ground lines. Usually we find these bus bars installed in one's electrical panel / distribution board in our homes.

Before getting into all this, don't forget to practice good cable "hygiene" to reduce interference and hum. Use shielded cables. Connect all devices to a common ground such as the same power strip or power conditioner. Try not to run power and interconnect cables in parallel. Use balanced cables when possible.

But sometimes even if one practices good audio hygiene, as I tried to do in my situation, in complex component systems with multiple pieces and where there are devices that have floating ground, you might still run into issues. So, here on the left below, is a bus bar I got for less than US$20 and on the right an even less expensive <US$10 one that I found locally made of a chunk of copper:

Copper as a material is a significantly better conductor than aluminum or brass (we're talking about 5x lower resistivity), but empirically both of these worked fine in my testing. In fact, the brass one on the left empirically performed marginally better so the measurements that follow will be with that one. I stuck on a couple of soft footers to prevent any scratching of my hardwood flooring. The bar on the left also included a number of yellow heat shrink ring terminals which I didn't bother using.

Connection was easy, run the ground nut from each of the devices I wanted to connect to the bus bar using thick wiring... In this case, I used 3 to 4 feet of 12G speaker cable! What we want to do is to create a very low resistance path from chassis ground into the bus bar which ultimately leads back to my power conditioner (the Belkin PureAV PF60 I wrote about a few years back).

Once it's all connected, the bus bar looks like this sitting on the floor behind my equipment rack:

As you can see, I connected the ground lug from the Belkin PureAV PF60, the Yamaha RX-V781 receiver (right image), Emotiva XSP-1 preamp, and my Technics SL-1200 M3D turntable to this bus bar grounding point.

And what did this do to the sound? Why, subjectively "blacker blacks", even "more silent silences" of course! :-) Easy to say... But let's take a few steps back and use measurements to document and verify the process and of course the effect of what was done.

As a start, let's look at what the "chain" of devices looks like in my home theater room (at least the most relevant pieces) before adding the bus bar and extra grounding:

As you can see, what a maze of cabling of various types! Home audio and theater systems can get complicated rather quickly once we start adding all kinds of cables like ethernet, HDMI, alongside the balanced (XLR) and unbalanced (RCA) analog connections.

My main concern is the yellow unbalanced audio connection between the Yamaha and my Emotiva preamp which represents the "HT Bypass" input to the Emotiva. With the Yamaha receiver turned on but not sending any audio to the preamp, if I take my RME ADI-2 Pro FS ADC to measure the balanced output from the Emotiva preamp (blue) which normally goes to my monoblock amps, with the receiver volume pushed way up to 0dB (which will accentuate the noise level), we can see that there is indeed a 60Hz hum intruding on what should be silence coming out from the Emotiva XSP-1 pre-amp to my amplifiers:

As you can see, other than some ultrasonic noise especially a -99dBFS peak up at 48kHz, the next strongest tone is the 60Hz hum at -101dBFS. Then there are the harmonics including 120/180/240/300Hz, etc...

Since this is likely ground loop hum, if we use the ground bus bar and tie together the PureAV power conditioner, Yamaha receiver, Emotiva preamp, and Technics turntable as shown in the photograph above, this is now what it all looks like:

And here's the noise floor with the extra grounding in place:

As you can see, the 60Hz hum has now dropped to almost -116dB. This is more than 14dB of attenuation which is very significant. The audible 60Hz hum is now rendered inaudible even with volume pushed up to maximum when connected to my monoblock amps. To better see the change, here's an overlay of the noise floor before and after:

The level of many of the harmonics has dropped in concert with the primary at 60Hz. Note that not all the noise spikes have been reduced which suggests that some noise is simply inherent in the Yamaha unbalanced output and likely not ground related.

Okay, so not exactly "blacker blacks" noise floor across the spectrum, but the most distracting hum has been reduced significantly.

Notice in my title that I also mentioned the Ebtech HumX. At around US$75, it's a bit expensive for basically a couple of rectifier diodes back-to-back as you can see of the innards here. They do work for ground hum but there is some controversy as to safety; better than a cheater plug with safety ground totally disconnected in any event.

Here's mine:

There are many places one could use this and I'll show you one strategic location that has worked for me. Probably the noisiest electrical component in my sound room is the Home Theater PC which when turned on can cause audible distortions seeping into that sensitive "HT Bypass" RCA input. To show you the effect of the computer noise, when we monitor the noise floor from the pre-amp output while the computer is booting, for a few seconds, we see some bursts of noise coming through looking like this even with the grounding bus bar in place:

Low level "brush" of noise evident as a nearby computer boots up. Noise appears to be correlated with CPU activity, fan speed and hard drive spin-up during the boot process.

With this test, I've also turned on the Vizio 75" TV in my room. What if we now insert the HumX between the wall socket and the computer in hopes of reducing hum further? Remember that the HumX is rated for up to 720W (6A here for N. America's 120V outlet) so unless your computer/device is quite power-hungry, this should be fine. Here's what the connections look like:

Here's the noise floor plot while booting pre- and post- HumX:

As you can see, the HumX was able to prevent a good amount of the noise from the HTPC seeping into the audio input through the ground. Of course, there are many potential sources of noise from a computer depending on the components active inside! As I have done in the past, what if we use FurMark to stress out the GPU?

Here's what happens when I run FurMark for 5 minutes, stressing the nVidia GTX 1080 GPU in my computer and again examine the noise level from the Emotiva preamp with and without the HumX connected to the computer:

Again, as we can see, the HumX is able to reduce the amount of noise picked up by the AV receiver to Emotiva preamp through the unbalanced "HT Bypass" connection. Notice that the HumX in this case attenuated but could not remove most of the noise. Perhaps some of this noise is being conducted through the hot and neutral power lines, or perhaps the HDMI connection to the AV receiver rather than through ground.

Summary:

1. REMEMBER: What I have shown here applies only to the very sensitive analogue unbalanced RCA input from an AV receiver to the "Home Theater Bypass" input of the Emotiva XSP-1 preamp, typically when listening to multichannel material decoded in the AV receiver. Do not take this as a general endorsement of the technique being useful for all systems. I don't want to hear about people going out and doing stuff like this unless you know you have a hum/noise problem and grounding the devices together might improve the noise level.

How do you know if there's a problem? Well, if you hear noise of course! Especially that characteristic 50/60Hz hum. Also, another way to know if there's likely a ground noise issue is to grab a cheap "cheater" plug and can convince yourself that noise is lowered by temporarily disconnecting the ground line to various devices.

Just to prove that doing this made no difference to the usual Emotiva XSP-1 pre-amp XLR and RCA inputs, here are some graphs of the noise floor of the RCA input from my Sony CD/SACD player, and XLR input from the Oppo UDP-205 with and without running extra ground wires from the metal chassis to the bus bar. The Emotiva preamp volume was set to -20dB which when I listen to music would usually be quite loud, achieving 85+dB SPL peaks (I typically do not listen to music this loud). So as not to remove potential sources of noise, the TV, HTPC and Yamaha AV receiver were all turned on while recording these measurements.

As you can see, the Emotiva XSP-1 does a great job with maintaining low noise level. The extra grounding with the bus bar made no difference with the RCA and XLR audio inputs from the CD player and bluray/DAC.

2. For the noisy "HT Bypass" input, despite my preamp and AV receiver being plugged into the Belkin PureAV power console, I was able to reduce the amount of hum further by using this grounding bus bar connection to the ground lug of each of these devices using low resistance, larger gauge wiring (12 gauge speaker wire).

This is a measurable effect which was quite substantial. Close to 15dB attenuation of the 60Hz hum is an audible effect in my room (when ambient noise low). Previously, I could detect that hum from the listening position at normal listening volume, now no more.

3. The Ebtech HumX is able to reduce noise further, in my situation by strategically targeting the HTPC. Furthermore, it should be a safer option than using a "cheater plug".

Note that the effect of the HumX, while present, was not a massive difference. As perfectionistic audiophiles, I could still say that for <$100, the difference is real and the expenditure has value.

While the noise level in my system from the HTPC can be heard, compared to the magnitude of the 60Hz hum, it was not terrible even with FurMark running for awhile hammering the GPU. Furthermore, remember that this is an Intel i7-3770K CPU + nVidia GTX 1080 combo system which is rather power hungry. These days with more efficient CPUs and likely a less powerful GPU in most HTPC set-ups, the interference should be much less.

4. If you're wondering whether RCA interconnect cables could make a difference, yes, they do have an effect though modest. Remember that in the past, I had measured the effect of cables and had shown that, as expected, noise levels through analogue RCA cables could vary. For a noisy connection like this "HT Bypass" prone to interference, you can try "dual shielded" cables often used for long lengths (such as connecting a subwoofer to the receiver). I picked up some inexpensive 6-feet "Mediabridge ULTRA" cables (<US$10 each) for connecting the pre-outs of the receiver to the Emotiva preamp:

Shown are two 6' Mediabridge ULTRA cables. For my system, I bought 3 lengths - R + L fronts and subwoofer passthru from receiver to preamp's "HT bypass".

The cables are a bit thicker than the standard shielded Radio Shack RCA I was using before. I can show you the difference in noise level before I added the ground bus bar between the 2 cables:

Notice that there's a lot of overlap but the dual shielded cable did show slight attenuation of the higher noise peaks in my system (I took a few measurements and was able to reproduce the subtle differences). I would not say it's significant but if one wanted to squeeze a few dB's here and there, it's certainly inexpensive to do so. Of course, if you are currently using unshielded cables to start with, the difference could be more substantial.

Over the years I have seen audiophile companies selling grounding devices of all sorts. Last week for example, I showed a picture of the Telos Grounding Noise Reducer while going through The Adelphi in Singapore. Notice how on their web page, they claim that this grounding device resulted in "300%-500% improvement" without qualifying the claim whatsoever! Here's a subjective review of the Telos if you're curious about testimonials out there.

Elsewhere, we see devices like these rather bizarre, earthy-looking wood Entreq "Groundboxes". I don't think there's anything wrong with calling the things bizarre since according to this review (which itself is an example of how pure subjectivism easily goes awry into fictional audiophile voodoo), the devices somehow utilize proprietary "natural earth minerals" in the way it functions. It would be fascinating to see what's inside those boxes! For even more fascinating review/discussion of this product, I refer you to Rafe Arnott's article when he was still writing for Part-Time Audiophile last year. Am I reading this right? The designer of these devices (Per-Olof) was a Swedish farmer who serendipitously learned about "real Earth" through geologists who happened to be building a bridge nearby and this was how the special combination of "minerals" were discovered?!

Similarly, a few years back, Synergistic Research was at it again with their "active" Grounding Block and "high definition ground cables". Hmmm, what makes ground cables "high definition" since it's not like anyone is listening to the current passing to ground to judge the definition!? As usual, there was also the audiophile review posted online by an apparently "golden eared" reviewer plus numerous links to sites promoting the device listed on the Synergistic web page. I think it's fair to be cautious about whether one should trust these websites.

While these devices might have an ability to reduce ground noise in some situations, how much more are they doing compared with a $20 bus bar and large gauge speaker cables as ground wires I'm showing here? The price figures I saw was that the larger Telos Active Grounding Box is asking for US$5000, plus US$200 for each ground cable! Synergistic is asking US$600 for the basic Grounding Block, and US$3000 for the "active" larger version. Entreq's "mineral"-utilizing devices range from US$399 to US$2700, plus US$99 for their "Eartha" ground cables. So, can any of these reviewers actually back up their claims of significant audible effect with any measurements? We are only looking at noise differences - Arnott's review even specifically mentioned "slow bleeding-out of the noise floor" - aren't we? Shouldn't that be measurable?

[For the record, the measured improvement in noise floor I'm showing above from connecting the grounds is instantaneous. There is no "slow bleeding-out"! Electricity does not work like that. No "break-in" needed.]

Such is the neurosis of audiophilia. Better cables. Better AC conditioning. Better ground devices. Even better sound when you "tithe" to the company of choice with more $$$$! I hope that the majority of rational audiophiles recognize the importance of evidence rather than just words... And appreciate that without such evidence, reviews such as those above are really nothing more than glorified vacuous endorsements.

Here's hoping you guys aren't experiencing any issue with ground noise. And if you are, perhaps something like what I'm doing with the bus bar could be helpful.

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

For those who are fans of The Cranberries, check out their last album In The End featuring vocals from the late Dolores O'Riordan released this past week. Certainly an enjoyable album in many ways.

I really hate harping on this, but if nobody says anything, nothing changes. Tell me, why did this album have to be DR5? Like many bands, notice that some of their biggest hits and most loved songs were released back in the days before extreme dynamic range compression. Songs like "Linger" and "Dreams" were from their first album Everybody Else Is Doing It, So Why Can't We? from 1993 - with a DR13 average value. And "Zombie" came from the 1994 album No Need To Argue averaging DR11. Those songs sound great and will be forever remembered as essential tracks for this band.

I contend that the Loudness War had a part in killing off the music industry revenues in the late 90's and early 2000's by destroying sound quality just as much as changes in technology played a role. Highly compressed albums consciously or unconsciously are heard as unnatural, and uncomfortable to listen to long term. Listeners lose interest quickly due to fatigue and move on or just play the music in the background since they don't capture our attention due to the lack of dynamics. This is not good especially as the average attention span seems to be diminishing over the years.

If what I'm suggesting is true, isn't it self defeating for bands and record labels to purposely damage their product like this? Isn't it contradictory for musicians who take pride in their craft to release products that inherently could be making many listeners potentially dislike their music, perhaps even dislike them as musicians for sounding like "that"? How can any band aim to promote a loyal fan base if perhaps unwittingly yet inherently the products the consumer buys with his/her hard-earned dollars were created to only "hook" them in for a few listens until they're fatigued?

IMO a very sad juxtaposition of what has happened to much of the popular music in the last 26 years. And no... "In The End" is far from what one might expect might be a loud, distorted hard rock track!

Sadly, I see this as part of the culture of our times. Who cares about quality in music anymore? Who cares about longevity? Or integrity? Just make a buck... As fast and as much as you can! Push the next album. The audio equivalent of "fast food", or candies with double portions of artificial sweeteners, please.

This goes way beyond trivial arguments like whether 320kbps MP3 compression sounds good, whether PCM or DSD is better, or whether hi-res audio has any value. These are but minor distractions considering that the corruption of sound quality was purposely "baked in" when the music itself was made in the studio prior to release to the public.

What surprises me is actually how little the mainstream audiophile press highlights the issue. Sometimes they even accept this as somehow okay. They would rather that we be distracted with the next "great" audio format, impress upon us just how "terrible" MP3 is, discuss the miraculous sound quality of multi-thousand dollar DACs, and of course why one needs hundreds if not thousands of dollars worth of cabling. As an Industry, these myths are there to keep the dollars flowing. Of course, sadly, the truth is that none of this ultimately will remedy what is essentially "garbage in, garbage out" from the recording industry itself.

I guess a part of me had wished that with this album as their lead singer's posthumous release, The Cranberries could have tried making a recording that sounded more natural as a final tribute to Ms. O'Riordan's beautiful vocals to accentuate the emotional content and for the instruments to "breathe". To me, as an audiophile and music lover, that would have been a better "end". Perhaps poetically brought the sound quality full circle. As is, the album sounds at best just fair and another entry among the countless missed opportunities for something of greater sonic quality and legacy.

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

Remember everyone, only a few days left to get me your blind test results! Keep track of your entry as next week I'll reveal what devices I used for the samples. Also, I'll be closing the streaming subscription poll around the same time; stand up and be counted.

Hope you're enjoying the music!

BTW: Just watched Avengers: Endgame today. Good movie and fitting character arcs. Personally a good sign when 3 hours passed by without feeling dragged out! Check it out if superheroes and special effects blockbusters are your thing. :-)