Crazy Power & Noise Saga
Dec 29, 2019 at 2:48 PM Thread Starter Post #1 of 16

chrisdrop

1000+ Head-Fier
Joined
Jun 22, 2019
Posts
1,110
Likes
2,709
Location
London, United Kingdom
For ~4 months, I have been trying to get to the bottom of a painful noise problem in my setup(s). It has been a very persistent and difficult to nail down issue. (Or I am a few sandwiches short of a picnic and quite slow). With hindsight, I do feel like I should have found it easier/ sooner. Live and learn! It is good to stay humble :disappointed_relieved:

As someone on another thread said; "That's the insidious thing about noise: You don't know it's there, until you get rid of it!". So true it is!

I wanted to share the saga here, on the chance that it might benefit someone else. I'd also like to consolidate my update to the many people who helped me. By the way; it sucked :head_bandage:. I am sure it (and I) sucked even more for every person who listened to me whinge about it for months!

Of note, I am in London in the UK. Power issues seem worse in Europe. Advice and products are also less plentiful than in the US. Much of the commentary is specific to my environment here.

Big thanks to: @ZRW0, @2359glenn, @JazzVinyl, @GDuss, @mordy, @teknorob23, @hypnos1, @OctavianH. Others have made helpful comments and if I left your name out - my apologies. It has been a long ride!

The problem:

In short; Two independent, highest quality, valve-based headphone amplifiers had unexpected noise. In some cases they were “bearable noisy” in other cases they had “unbearable buzzing”. Different valves had different noise levels. The issue was present with nothing connected to the amp (except headphones) - not even a source. The issue was invariant to volume.

To be 100% clear, both of the amps are 100% excellent and issue free. There is/ was a clear problem in my home power environment. One amp is a Feliks Euforia and the other is a Glenn OTL, both with stellar makers and reputations - deservedly so.

The punch line; the (a?) solution was an automatic voltage regulator. Perhaps there will be “better" ones of those in my future and/ or more "power products". I am now obsessed with lowering the noise floor limbo-style .. "how low can you go?”. I am very keen to have as silent an environment as possible.

It took a lot of experimentation (fumbling) to get to the end. It also took quite a lot of conversation.

What hypotheses were there throughout the journey:
  • A grounding issue. A 100Hz buzz signature pointed to that.
  • Ground loop (factored out early)
  • EMI/ RFI in the environment. I live in London, maybe it is crazy with the stuff!
  • Very distorted mains power
  • DC on the AC line
What was the journey:
  • I purchased a perfectly kept and beautiful Feliks Eufora from another great head-fier. Exciting! It eventually came to me and I got to listening ...
  • At some point early on I noticed that the Euforia did not have a “black” background (when using the stock valves). I was puzzled by this. It was usable, but certainly, there was an audible hum when the music was off or quiet.
  • Using very highly regarded valves popular on the Euforia thread, which the seller had been regularly using, the buzz was not usable. This really puzzled me. The seller and he was absolutely awesome at helping to investigate the issue. You have my apologies for being sucked into my issue!
  • I brought the Euforia to another location and plugged it in. I got the same issue. I assumed something was wrong with the amp and decided to send it back to Feliks Audio to debug. Off the amp went to Feliks Audio.
  • As fate would have it, I had been trying to get on the Glenn amp waiting list. Fortune favoured me and for reasons I will spare you, an amp to my specs became available from Glenn! I had to say yes to getting a GOTL. The Euforia - perhaps it would go to my office, 2nd listening location. Perhaps (crazy?) I’d keep both at home? Who knows. Glenn sent me the amp and after much ado about customs, importing and waiting, the amp came home!
  • Guess what. I think you can see where this is going … same story! The amp also sounded amazing, but it did have some noise. Higher gain valves produced a lot of noise. The amp was 100% new and worked perfectly through testing and use. I was now 100% sure it was an issue at my place.
  • I did a lot of valve experimentation. One puzzling session was with @GDuss and @JazzVinyl. We thought there were some noisy valves. We moved them around and the noise stayed on one side? Crazy. Eventually, this was explained; the vibration on the transformer side of the amp vibrated the tubes on that side of the amp, generating noise. Higher gain output tubes manifest the noise the most.
  • Experiments/ changes:
    • Upgrade all power cables (I assembled top-quality UP-OCC cables with top quality parts.). Most parts from Hificollective.
    • Power off everything in the house, leaving just the socket for the amp. Same.
    • (Temporarily) Ground-lifted power to the amp (from power strip to device) [DO NOT DO THIS]
    • Sent the Euforia amp to the manufacturer - absolutely no issues to be found!
    • With an electrician; installed a dedicated power circuit with an additional dedicated ground line, using high-grade low gauge (fat) power cabling and top quality sockets, etc. Parts from MCRU
    • .. and many products
  • What products did I buy (and almost exclusively return - sorry)?
    • AirlinkTransformers conditioning balanced power supply
    • Isotek Evo3 Syncro Uni
    • Furman Ac 210 A E HIFI Power Filter Mesh Filter Power Distribution AC Power Conditioner
    • Isol-8 Powerline Axis
    • iFi AC iPurifier
    • Tacima Mains Conditioner
    • ... probably more that I am forgetting ...
  • Other oddball items:
    • EMI Meter to measure the EMI/ RFI environment. There is no hard-core EMI happening FYI.
    • Put a copper spike in-ground outside and ran a copper cable to amp casing. This helped someone on the Elise thread at some point.
    • Went nuts with ferrite chokes. Nada.
    • Put isolation feet on the amp (really 1st 4x 1/2 tennis balls as feet, then cheap springy ones off of Amazon!). This actually helped a lot! The noise dampened down quite a bit (but it was still there).
    • Try tube dampers (Herbies and Golden Dragon tube rings). These helped a bit. Recommended.
    • The Spectroid audio spectrum analyser app was really handy to see the 100Hz cycles kicking off the amps. I could point it at the amp, or shove my phone into the headphones and see the noise-signature.
    • Power Line EMI Dirty Electricity Meter to measure the mains noise. Only detects higher frequency "dirty mains", not distortion lower in the audible region.
  • What vendors did I talk to for guidance/ product advice? Many. I think this is a comprehensive list of exceptionally generous and helpful people.
    • Mike at Puritan Audio was excellent and very willing to try to help. He offered to come to my house from his place and measure my mains distortion. He does not live close to me!
    • Peter at Igloo Audio was extremely helpful. He let me come to his place with my kit to see how it was there https://www.iglooaudio.co.uk/
    • Mike at AirlinkTransformers https://airlinktransformers.com/. Although I returned their unit, they were immensely helpful in trying to diagnose the issue.
    • PS Audio in the US and their UK distributor (most people suggested not using a regenerator, although this may have worked (expensive too))
    • Nic at Isol-8 about their product line.
    • Adam at Futureshop was helpful when I returned an Isol-8 unit.
    • David at MCRU was helpful with my dedicated circuit cabling and parts.
    • Helpful support people at IsoTek. FYI they advised their regenerators were not for any amplifier products, only sources.
    • Enrico at Sonic Syndicate here in London. He came by with an engineer. They had been planning on getting hold of some loaner units of very high end (expensive) power conditioning units.
    • My local power provider was largely baffled and unable to help
    • I was planning an engagement with Ben Duncan who was very pleasant and well informed.
    • Peter at Russ Andrews
All these vendors and professionals were exceptionally generous with their time and very helpful. They were also uniformly flummoxed (along with me) with respect to my issue. All of them were also pretty clear about what their products would not do, and what they are intended to do. I loved that. So many vendors basically said; "this won’t help" or “this is not for that”. The overwhelming collective believe was that it was a ground issue of some sort FYI.

How did I get to the end?

While on my 2-week holiday break (presently), I spent a good 2 days reading on this thread and a related thread on another forum. While a Topaz isolation transformer is "unobtanium" (especially in Europe), I decided to try a similar setup with lesser isolation transformers. I ordered yet-another-isolation transformer (although I had tried one, with balanced power, I thought I'd try a different kind) and this; APC LE1200I Line-R 1200VA Automatic Voltage Regulator, Input 230V. Now, at work, in my professional life - I have a research team. We run a lot of experiments. For this exercise I have been pretty good about 1) documenting experiments/ results and 2) trying to isolate experiments/ effects (change 1 thing at a time). Before putting the chain of things together, I just put the 1 unit in between my socket and the amp. Then I set to listening. Quiet. Wow. In shock and disbelief. After a while, I put some power valves in that had been unusably too noisy. They were quiet. Wow! I did that a few more times. I has now been the better part of a day - quiet. Finally.

Much of this post has been cut-n-pasted from my notes on this topic. Given that I had communication with various helpful parties (as you see), I needed to keep clear information with and for them. So - this post was an easy job.

What next?

Really? Enjoy listening even more! I may still try a Torus Power TOT MAX AVR UK. This is a well-made isolation transformer with automatic voltage regulation. Why? I believe the higher quality the parts, the bigger the impact. I don’t care what anyone else says, power is insanely important and a key determinant of the quality of what comes out of your setup.

Power cables made a direct and audible difference in my setup through all of this. Right now this voltage regulator is using a crap IEC cable, so I need to re-terminate a good cable with a male IEC end to plug into the thing probably. I wish there was nothing in between the dedicated mains power line and the amp, but - beggars can't be choosers!

Final Thoughts

Through all the experiments several things made positive audible differences, even though they didn’t fix my issue. The dedicated mains power circuit made a clear and positive difference. Isolating your system away from noisy things in your house is undeniably helpful. As a recent video from @Monsterzero agrees - high-quality power cables absolutely made a positive difference.

The captured collective knowledge here on head-fi was immensely helpful. Years of people working through issues and trying to enhance their sound-systems was truly useful. Finally, the community has been excellent. So many people have been generous with their time and tried to help me get to a happy place. Thanks. Well, that was a long post. Sorry about that.

Best,
Chris

N.B. - Even if I am the only one who ever reads this, the catharsis has been worth it!

... and since this is all about listening, this post brought to you by James Blake's - Assume Form album. Here is the one I am on now...
 
Last edited:
Dec 29, 2019 at 3:14 PM Post #2 of 16
Hi chrisdrop,
Glad that you finally found a solution to your noise problem!
If I understand correctly, the problem was a fluctuation in the mains voltage? Did you measure the potential voltage swings?
 
Dec 29, 2019 at 4:10 PM Post #3 of 16
Hi chrisdrop,
Glad that you finally found a solution to your noise problem!
If I understand correctly, the problem was a fluctuation in the mains voltage? Did you measure the potential voltage swings?

Either 1) persistent high voltage or 2) big voltage swings as you say.

I didn't measure the swings but the voltage was on average around 245v+ as measured by a probably imprecise socket tester. The voltage regulator is always bringing the voltage down to 230v fwiw.
 
Dec 29, 2019 at 5:30 PM Post #4 of 16
Did a little reading - in the past the UK used 240V and other European countries 230V.
The more recent standard changed it to 230V for everybody. However, a +/- swing of 10% is acceptable.
"All modern equipment will therefore be able to accept 230V +/-10% i.e. 207-253V."

Based on the above I am wondering if your problem is something else than the actual voltage but the main thing is of course that your equipment is humphrey lol!
 
Dec 29, 2019 at 7:53 PM Post #5 of 16
Your dedication, persistence, and experimental skills demonstrated by this process are truly amazing. Given that you also wrote a dissertation to start this thread (whether for catharsis or not), you should be awarded a doctoral degree in noise isolation for all this work :). Congrats.
 
Dec 30, 2019 at 2:21 AM Post #6 of 16
Did a little reading - in the past the UK used 240V and other European countries 230V.
The more recent standard changed it to 230V for everybody. However, a +/- swing of 10% is acceptable.
"All modern equipment will therefore be able to accept 230V +/-10% i.e. 207-253V."

Based on the above I am wondering if your problem is something else than the actual voltage but the main thing is of course that your equipment is humphrey lol!

Precisely because of the UK power "acceptable swing range" is why I had a low prior that a voltage regulator would have a positive effect. It was a "systematic flail" on the long list of things to try. (Believe it or not, there were still more things on that list!). I wish I understood deeply why this voltage regulator solved my noise. Empirically - it sorts the hum. Theoretically - it is less clear why. Effect-wise; it made the transformers on both amps vibrate, (and the power supply chokes on at least one). That made the valves nearby vibrate - generating audible hum. Nearby output valves with higher gain were especially affected. That is the effect of non-voltage-regulated power in my environment. At some point, I will see if I can better measure the raw mains power. I would have liked to get an oscilloscope on the mains power directly.

On another note @mordy perhaps you have given me a new name for my Glenn amp?
 
Dec 30, 2019 at 1:16 PM Post #9 of 16
Fascinating and VERY thought provoking - THANKS!!


  • Helpful support people at IsoTek. FYI they advised their regenerators were not for any amplifier products, only sources.

Any idea WHY not use regenerators for amps?!
PS Audio seems to recommend plugging EVERYTHING to their power stations (which I agree are SUPER expensive...).
 
Dec 30, 2019 at 1:19 PM Post #10 of 16
FYI - after hearing this, one of the vendors that I had been speaking to also said that a while ago he had a customer experience this same issue in the UK with Croft amplifiers. Apparently not unique to me :)

Seems like this would be a more common issue than is reported, so maybe it's common and just not reported? Or maybe others have a higher tolerance for tube-related noise than you do? Or maybe they just can't hear it?

In any case, I remember one of the first questions you asked me after you got the GOTL was whether my amp made an audible noise without even having headphones on. Mine does not (at least not that I can hear) but yours did. From your discussion, this audible noise is now gone? I found your discussion of vibration-induced tube noise interesting (in relation to its ability to generate the audible hum), and it may address why the springs you put under the amp helped so much (which you already covered).
 
Dec 30, 2019 at 1:57 PM Post #12 of 16
Fascinating and VERY thought provoking - THANKS!!
Any idea WHY not use regenerators for amps?!
PS Audio seems to recommend plugging EVERYTHING to their power stations (which I agree are SUPER expensive...).

Some people say regenerators are great. Some say they "bleach the sound". I am not really sure as I have not experienced the + or the -. One of the local hifi-installers who was kind enough to come by with an engineer did say he thought we should try a high-powered regenerator. PS Audio certainly suggested it would be an improvement to my environment. Conceptually - I think the regenerators are very appealing, at a sane price! I may try one of these: https://www.powerinspired.com/new-ac-power-regenerator-ag1500s/ @hypnos1 had an earlier version of this unit in his power chain (at least not long ago, he still probably does!).

Here is a direct quote from Isotek support: "Our regenerators (Genesis One, Mosaic Genesis, Genesis) are ONLY for source components and NOT for amplified products like amplifiers, active loudspeakers, active subwoofers. "

I suspect the power demands of headphone amps is significantly less than some of those.

I am amazed by the amount of change/ improvement the power chain can account for.
 
Last edited:
Dec 30, 2019 at 2:06 PM Post #13 of 16
Seems like this would be a more common issue than is reported, so maybe it's common and just not reported? Or maybe others have a higher tolerance for tube-related noise than you do? Or maybe they just can't hear it?

In any case, I remember one of the first questions you asked me after you got the GOTL was whether my amp made an audible noise without even having headphones on. Mine does not (at least not that I can hear) but yours did. From your discussion, this audible noise is now gone? I found your discussion of vibration-induced tube noise interesting (in relation to its ability to generate the audible hum), and it may address why the springs you put under the amp helped so much (which you already covered).

It is possible that with closed headphones the hum is more audible. That said, there are some valves that anyone would be able to hear, to the level of rendering the valves unusable.

I have put the headphones on my wife's head and my 17 y/o's head and they could certainly hear the hum. There were quieter times for sure, but never for long (sadly!). Low gain outputs helped manage it.

I don't understand why this issue is not more common. I don't even really understand what my problem was exactly! All I really know is that plugging into the AVR unit sorted he hum (and that the list of things before it didn't sort it).

A free method to see the 0 volume noise floor in your headphones is to put the Spectroid audio spectrum analyser app into your headphones and see what it looks like (with the amp on but nothing playing). It was a graphical version of my noise-floor that guided me. Seeing a persistent 100Hz hum (1st harmonic of 50Hz power cycle) is what kept the focus on power issues (ground issues in particular).
 
Dec 31, 2019 at 7:30 AM Post #14 of 16
While I am happy, I continue!

I thought I'd show some pictures of line readings using one of these devices.

Items used:
All are connected to a dedicated home mains circuit using top quality cable & sockets.

Caveats:
  • I am not sure how much these things actually matter, but the fact that they are visible/ measurable is interesting. It is generally measuring high-frequency noise in the frequency range 10 KHz - 10 MHz
  • I am not sure how accurate these things are. Specifically, the voltage doesn't really agree with the AVR unit's intended output voltage.
  • Readings of < 50 are considered no issue, > 50 are considered to be "a bit noisy and may impact sensitive equipment".
Reading 1. Plug into the plain power strip, connected to the wall socket. It shows the voltage at around 243v. I had seen it as high as 250v, but not more than that. It shows 68mV of "line-noise". The value jumps around +/- 10, but this is around the normal value (notwithstanding a pics later).

20191231_112803.jpg

Reading 2: Wall -> APC AVR unit -> Isol-8 PowerLine Axis (power strip). Shows about the same to a bit more general line-noise (~70mV) as the other path, but the voltage is closer to 220v. It is pretty stable, but I do question the accuracy due to the fact that the AVR is trying to keep the voltage at 230v. The line noise did spike a reasonable amount, however. More later...

20191231_112357.jpg

Reading 3: When I plugged my DAC (RME ADI II) into the powerline strip, there was a lot more noise - always > 100mV (I assume coming from the DC switching power supply).

20191231_112118.jpg

Reading 4: Here is the noise when adding the DE2 Dirty Electricity EMF Filter. It brought the noise right down to around 15mV (as intended). Again - not sure if this matters, but it does what it says on the tin.

20191231_112133.jpg

Reading 5: Unfortunately, I can't upload my short video of "spiking" behaviour but there was some significant spiking up to between 150-200 mV. Wall -> AVR -> Powerline strip w/ DAC plugged in. I couldn't get a good picture because it was spiking and returning the level. The plug-in filter did eliminate the spiking.

Again, I don't know if this has any real impact, but I thought - why not add it to this ridiculous story !?!?

A final comment; several people and articles mentioned "dimmer switches" as being noise adders to hi-fi systems. I took some readings on the normal mains system (not the dedicated circuit) and turning my kitchen dimmers on accounted for + 10-15mV of added powerline noise.
 

Attachments

  • 20191231_112219.jpg
    20191231_112219.jpg
    445.3 KB · Views: 0
  • 20191231_112048.jpg
    20191231_112048.jpg
    341.9 KB · Views: 0
Last edited:

Users who are viewing this thread

Back
Top