I have some kind of ground loop in my system as I'm noticing a significant buzzing sound in my recordings. It's not white noise, it's like a cross between a hum and a buzzing sound that is baked into everything I record.

Frontend Chain:
Synths > Speck Electronics X.Sum > API 3122V Preamp > Focusrite Compounder > TL Audio VI1 8 channel Valve Interface > RME Fireface UFX III

Power:

• 1 Torus Power TOT MINI
• 2 Furman P-1800 AR
• 1 Furman PL-8 Series II

The Torus is powering two sets of speakers, my laptop, monitor, and a UAD satellite. The two P-1800's are powering most of the audio gear (the quality pieces), and the PL-8 is powering some of the miscellaneous stuff like the mioXL and a few other sequencers/sound modules.

Each of the four power conditioners are getting plugged into their own outlet in a 4 outlet box.

I don't know what other fixtures in the space are powered off of the line running to this box, but it could be a small exhaust fan, light, fan, etc. (this is a studio shed in my backyard).

I haven't opened the 4 outlet box, but I'm assuming it's grounded properly. I know we should never assume, so I'm going to check this.

USB Cables:

• 1 running from the back of the RME to my laptop (this one is coming out of the back of the rack mounts next to audio and power cables)
• 1 running from my laptop to the front of the mioXL (not interfering with audio cables)
• 2 from the front of the mioXL to a midi keyboard and a sequencer (barely interfering with audio cables)

Cables:

• All synths are connected via brand new Mogami cables
• IEC cables are a blend of new and old
• I'm using AudioQuest NRG cables to power my interface, the X.Sum, API preamp, and the TL Audio VI1 8.

I've done by best to separate audio cables from the IEC cables, but there's definitely some overlap as I have a ton of gear in these racks and the position of the audio outputs and power plugs alternate on different devices so I can't get all audio cables to one side and all power cables to the other. I'm a neat freak, but I'm not sure how I could do this better.

Research:
I did some research here on GS and came across a few tips.

- "Run everything into the same socket" as mentioned here: Yet another question about HUM from a dummy

Does this mean I need to run all of my power conditioners into 1 conditioner and have that single conditioner be the one that's plugged into the single wall outlet?

- "Make sure the chassis is earthed properly" as mentioned here: Yet another question about HUM from a dummy

I'm using those rack mount rolling stands by GRIFFIN and didn't think they needed to be earthed, however the back of the TL Audio VI1 8 says it must be earthed and I do not have it earthed. I feel silly asking this, but how would I earth this unit if I don't have a grounding rod available in my studio?

- "Keep all audio signal cables and signal sources and destinations, away from running parallel with anything which produces the mains cycle" as mentioned here: Noise/hum issues in studio

How are you guys keeping power cables fixed at a right angle running across your audio cables?

- "Everything should have one and only one path to everything else in the system through the ground network" as mentioned here: Noise/hum issues in studio

I think this refers to the first point about running everything into the same (grounded) socket.

Next Steps:
What should be my order of operations with troubleshooting the issue? I'd like to earth the TL Audio VI1 8, but not sure how.

What am I not thinking of?

Thanks for your help!

1. You don't mention what connections are balanced and which are not. Draw out the whole circuit with all of the grounds so that you know where your grounds are going.

2. Start disconnecting. If you pull the synths does the hum go away? If you pull the X.SUM does the hum go away? If you pull the preamp does the hum go away? When you unplug the signal cable, you are unplugging the ground connection too and if you start on one end and move down the chain you will find where the ground loop actually is.

Once you find the loop you know which ground to break.
--scott

Thanks for the response!

1. Ok, I'll make a diagram outlining all of my cables - which are balanced, unbalanced, and when things convert (like running unbalanced into the TL Audio inputs but then coming out as balanced).

I'm not entirely sure what you mean when you say "Draw out the whole circuit with all of the grounds so that you know where your grounds are going." Can you please elaborate on this?

2. My gut told me I had to do this, but I didn't want it to be true. I'm going to start with the TL Audio and then work backwards through the chain because I have a feeling this is the culprit.

Am I overthinking how the unit says "this equipment needs to be earthed"? Isn't it grounded (i.e. earthed) by me powering it via a three pronged power cable? That screw reminds me how a turntable ground gets connected to a mixer, but I don't know where the ground wire from the back of the TL Audio should go in the absence of a copper grounding rod.

Thanks again, I'll report back with my findings.

Everything gets one and only one connection through the ground network to everything else. Never two. Never zero.

The power line ground might be that connection... and if two things are sharing a power line ground, you may have to break the audio ground between them in order to prevent a ground loop.

With modern differential inputs and balanced lines it's easy to break ground loops just by using a ground lift adaptor that carries pins 2 and 3 without carrying pin 1. With unbalanced connections it becomes harder and a transformer might be needed in order to break the ground connection.

But you can only do this in a systematic way when you know where all your ground connections are and where they are going.

Gimmicks like plugging everything into the same outlet don't fix ground loops, they just reduce the physical profile of the loop so that they are less problematic. There is less space between the ground connections so they form less effective antennas to pick up noise. But the real solution is to break the grounds as necessary to prevent forming loops in the first place.
--scott

Thanks for the response, Scott.

I feel like how my pops feels when I explain to him how to copy/paste and then he looks like a deer in headlights

When you say "two things are sharing a power line ground", do you mean two as in 1) power from the box is feeding my power conditioners and 2) power from the same line is feeding something else (like the lights in the room), or do you mean two pieces of gear feeding off of the same power supply?

I still don't know what you mean by "... when you know where all your ground connections are and where they are going."

I don't want to take too much of your time. I think the main takeaway for me right now is do the unplugging and testing like you initially suggested, and to outline all the cables I have utilized in my setup.

This I get, from a conceptual standpoint.

Thank you.

My bet would be on the synths into the line mixer. What happens with only one synth connected to it?

Why would you place your bet on the synths plugged into the line mixer?

From a practical standpoint, I would have to test this use case first and then proceed to testing the units after the X.sum.

I thought it made more sense to work backwards... remove the valve interface first, then the compressor, then the preamp, and then 1 synth at a time from the X.sum.

Do you think I should do the opposite? Remove one synth at a time to see if one of them is causing the issue, and then if not, start backwards as I mentioned above?

The ground of each synth will be connected together at the line mixer and so the synths have multiple ground paths through each other.

Got it. I suppose this is what Scott means when he says The power line ground might be that connection... and if two things are sharing a power line ground, you may have to break the audio ground between them in order to prevent a ground loop.

Is there anything I can do to immediately test this by breaking the audio ground between them without having to remove each synth, one at a time? Like plugging the X.sum into a Morley Hum Exterminator Ground Loop Hum Exterminator or a Morley Humno Noise Eliminator?

Actually, I guess the question is, once I do find the issue (and God willing I will), how do I break the audio ground between the problematic device(s)?

Unplugging all but one synth from the line mixer seems like the simplest test. Is there is any particular reason you can't do so?

I can do that tomorrow morning.

I was just trying to save myself from moving the racks and undoing everything I already worked on, but... I know this is part of the process and if i have to unplug things yet another time, then so be it.

I'll report back with my findings.

TY!

The alternative would be to disconnect the power to all but one synth that has a safety earth connection i.e the ones with three prongs.

Remember rule number 1 ------ life the audio ground ---- not the power ground. Better to live..................

That would definitely be easier. Good idea, thank you!

Yeah, never lift safety earth. Lift audio ground or isolate with transformers.

You got it! Good luck

Which audio ground do you lift when you have a line mixer and 8 pieces of gear plugged into it? The mixer, one of the units, or every unit?

Perhaps leave the mixer as is, and test each unit that is plugged into it and then lift each one that causes the ground loop hum?

I feel like I'm solving a puzzle with you guys haha

TY!

Found one issue with my connections by reading the X.Sum manual.

I currently have the Mix Output jacks of the X.Sum connected to the front-panel 1/4-inch hi-Z/line inputs on the API 3122V however, the manual says all active-balanced output circuits of the X.Sum were designed for balanced lines and should only be connected to balanced inputs.

So, the way I have this currently connected to my preamp is incorrect. I need to use the Mix-A (pre fader) Output Jacks on the Breakout Box to connect to my preamp's 1/4-inch hi-Z/line inputs.

I'll see if this immediately solves the issue although I'm sure I'll have more work to do.

Thanks for all the clues!

a good step

and yes... someone can rightfully say RTFM

Box A is plugged into the power line, so there is a connection from the ground in that box to your panel and then there is a connection from that panel to the ground at Niagara Falls. Everything is groovy.

You plug Box B into Box A, and it has a connection that shares a signal ground but it has no power line ground. So now you have Box A ground connected to Box B ground connected to the panel and everything is groovy.

You plug Box C into Box A, and it is plugged into the power line but the boxes are connected with something that doesn't have any signal ground (like unshielded ethernet). So now Box C is connected to the panel, Box A is connected to the panel, and box B is connected to the panel through box A. Everything is groovy.

Now you plug box C into box B with something that extends the signal ground and all of a sudden you get a hum. Because ground goes from box C over the signal ground to box B, over the signal ground from box B to box A, and then from the power line from box A to the panel and then back from the power line to box C. So now you have a loop and you have formed an antenna that is inductively picking up noise.

The rule: everything has one and only one connection through the ground system to everything else. Ground can be carried on a signal cable, a power cable, or some kind of control cable. Find out where the grounds go and you will find your problem.

(And I do agree that it probably has to do with the synths but do it systematically.)
--scott

Why? Why not take the front panel TRS outputs and plug them into the XLR balanced input of the API? You always want to use a balanced connection if possible, AND a balanced connection gives you an easy option to break a ground loop with a barrel adaptor.
--scott

I read this and felt like my dad finally told me he figured out how to copy/paste! Seriously, I get it now.

Appreciate the humor as well!

TY!

Good news... I took your advice and plugged the balanced master output of the X.Sum into the XLR inputs of the API and BOOM - the ground loop buzz went away!

Here are the test results:

External Instrument at 0 dB Gain_ORIGINAL WIRING - this was how I had things. You can see the buzz creeping in from the low end and the high end.

External Instrument at +24 dB Gain_ORIGINAL WIRING - I boosted the gain of Live's Ext. Instrument to +24dB to emphasize the issue from a visual perspective.

External Instrument at 0 dB Gain_CORRECTED WIRING - here are the results from the balanced cables going into the API. You still see a bit come through on the high end, but now I hear white noise instead of buzz.

External Instrument at +24 dB Gain_CORRECTED WIRING - I boosted the gain of Live's Ext. Instrument to +24dB to emphasize the results from a visual perspective. Again, no buzzing sound - just white noise.

I then proceeded to enable all of the X.Sum's inputs by pressing their corresponding Mix knob on the front panel. No change in either direction (more or less noise/buzz).

I then tried unplugging the power supplies of the synths.

Unplugging Synths at 0dB Gain - looks clean.

Unplugging Synths at +24dB Gain - signal gets dirty on both ends of the spectrum. I can also hear a faint buzz again, but I have the output of my RME volume knob at +6 and the gain of Live's Ext. Instrument set to +24dB. So, this is not really a real life example so I think this residual noise could be normal?!

At this point, I plugged the power cables back into the synths and proceeded to unplug the audio inputs of each one from the X.Sum while everything was cranked (risky, but I wanted to hear if the signal got clearer while I was working). I proceeded to remove the inputs one by one until only one synth was left plugged into the X.Sum. That faint buzz was still there. I removed the final one and the buzz was still there. So, I'm thinking this might be considered as a normal "noise floor"?! Again, I only hear this with the output of my RME volume knob at +6 and the gain of Live's Ext. Instrument set to +24dB - I don't hear the faint buzz under normal gain levels.

Finally, I tested the rest of the gear I have routed in & out of the valve interface. My idea with implementing this interface was to have everything run through tubes. So I capture my recordings through the first 2 inputs, send my Electrix gear through the next 2 so I can have a tube-based filter and delay, and then send my PCM91 through the next 2 so I can have a tube-based reverb.

Here are the test results:

Electrix I/O at 0 dB Gain - you can see this high end noise come through. It's white noise but kind of piercing, and I would like it to go away.

Electrix I/O at +24 dB Gain - a lot of junk comes through when boosting the gain of Live's Ext. Instrument to +24dB. I'm using all balanced cables to run in and out of this gear and to the valve interface. I'm not sure what I can do to improve this signal, but it's not the way I want it. Maybe the Electrix gear is inherently noisy or maybe my units might be a bit faulty?

PCM91 I/O at 0 dB Gain - looks clean (with the input knob on the front of the PCM set to 12 o'clock).

PCM91 I/O at +24 dB Gain - there goes the junk again at both ends of the spectrum (with the input knob on the front of the PCM set to 12 o'clock), but it's mostly noise and I feel like I can easily filter out the extreme lows and highs to get the effect I want.

Miscellaneous Tests:
- Enabling/disabling the Focusrite Compounder had no material effect on the sound
- I have the pad, HP filter, and 3:1 enabled on the preamp

Conclusion:
Not using the correct cables from the X.Sum to the API caused the buzzing sound and changing those to balanced cables solved the buzzing issue.

Under real life conditions (keeping the gain of Live's Ext. Instrument to 0dB) I cannot hear buzz and only some high end noise however, cranking the gain of Live's Ext. Instrument to +24dB brings back a faint level of buzz again. Would you guys consider this normal or is there anything else I can do to improve the signal chain? Would routing all of my synths into DIs before they hit the X.Sum make a material difference?

I consider the PCM91 running through the tubes to have acceptable results, but the Electrix gear is noisy, especially when boosting the gain. Not sure if I can improve this any further and may just have to use a gate.

All pictures are in the order of which I listed them.

Thank you so much for your help!

Back to square one.

I didn't test sound out of the synths, I only listened for the buzz to go away.

Once I set the API front panel buttons to FRONT, the buzz came back. I tried changing the cables from TRS (X.Sum) >XLR (API) to TRS>TRS and no luck.

Then I tried the Mix-A (pre fader) Output Jacks on the Breakout Box to the API. I tried both TS and TRS cables and the buzz is still there.

Then I tried removing devices from the chain:

• X.Sum > Compounder > RME inputs = buzz (with 1 synth plugged into the X.Sum)
• X.Sum > RME inputs = buzz (with 1 synth plugged into the X.Sum)

Removing the synth from the X.Sum inputs and having the X.Sum directly into my RME inputs eliminated the buzz, but obviously this isn't the solution as I can't hear anything from the synths.

I'm also not daisy chaining the power conditioners. Each one is plugged directly into a wall outlet. They are next to an electric baseboard heater panel, but that's not on.

Not sure where to look next. I'm going to unplug the entire rig and rewire to see if I notice anything. I cracked an effen beer too because I thought I could work on music, but no... more cable work!!!

Here is the circuit.

Synths:
All synths are connected to the X.sum via Mogami TS to TS cables.

Frontend Recording Chain:
I'm using the X.sum Mix Output jacks.
- Out of the X.sum (TRS) into the front-panel 1/4-inch hi-Z/line inputs on the API 3122V (TRS)
- I have the following functions enables on the API: FRONT, PAD, HP, and 3:1 (screenshot of manual is attached)

The API feeds the Focusrite Compounder.
- Out of the API (XLR) into the line input on the Compounder (TRS)
- The input level on the Compounder is set to +4

The Compounder feeds to the TL Audio Valve Interface.
- Out of the Compounder (XLR) into the TL Audio (TRS)

I capture through my RME A/D.
- Out of the TL Audio (TRS) into the RME (TRS)

Power Distribution:
First Furman P-1800 AR
- The 3 analog outlets power my RME, API, and the TL Audio valve interface via Audioquest power cables
- The other 5 outlets power the Compounder, Electrix FilterFactory, Electrix MoFX, Lexicon PCM91, and a Waldorf Q rack
- The Waldorf is plugged into a Lencent European adapter before going into the Furman
- The main power line goes directly into outlet 1/4

Second Furman P-1800 AR
- Powers almost everything on that smaller rack and the X.sum
- The main power line goes directly into outlet 2/4

Furman PL-8 Series II
- Powers the mioXL, DarkTime sequencer and Novation Drumstation via their normal adapters
- The main power line goes directly into outlet 3/4

Torus TOT (not pictured)
- Powers two sets of speakers, my laptop, and a UAD satellite
- The main power line goes directly into outlet 4/4

My computer monitor is plugged directly into another outlet on the other side of the room.

Other I/O:
- RME outputs 5/6 (TRS) go into FilterFactory inputs (TRS)
- FilterFactory outputs (TRS) feeds MoFX inputs (TRS)
- MoFX outputs (TRS) feeds the TL Audio inputs (TRS)
- The TL Audio outputs (TRS) go into RME inputs 1/2 (TRS)

- RME outputs 6/8 (TRS) go into PCM91 inputs (XLR)
- The input level on the PCM is set to the balanced position
- PCM91 outputs (XLR) feeds the TL Audio inputs (TRS)
- The TL Audio outputs (TRS) go into RME inputs 3/4 (TRS)
________________________________________________________________
I don't know where my grounds are getting loused up here.

At this point, the only things I can think of is:

- I need a pair of TRS to TS cables to connect the X.sum (TRS) to the API (TS)
- I need to insert a barrel adaptor in between the connections of the X.sum and API
- I need to plug the synths into DI boxes before they hit the X.sum, and use the ground lift feature wherever appropriate. Alternatively, I can try injecting a Morley Humno Noise Eliminator to the synth power cables, and/or the X.sum, API, etc.

Any additional leads to help me troubleshoot is very much appreciated. Since we're on GS, picture attached.

My DJ rig is on the other side of the studio. Another four outlets are taken up.

I'm actually using outlets in all four corners of the room. Maybe the gear I have plugged into the other outlets is causing issues with the outlet of my production gear, even though those outlets are not being used at the same time?

Could the rolling racks touching each other cause an issue? Maybe one of the rack mounted units is touching the rack and causing an issue?

The 2 Furman conditioners in the big rack are butted up next to each other. Maybe that's causing an issue?

I can also try switching the power cables into different conditioners to see if that makes a difference. Also, I have to double check, but I believe four adapters are two-pronged (mioXL, DrumStation, and both Doepfer units). Perhaps those are the culprits so I will try removing/moving them around.

Maybe the USB cables I have connected to the computer from the RME, and from the computer to the mioXL is causing issues.

Apparently I'm not connected to Niagara Falls, so everything is not groovy!

I also ordered a Morley Humno Noise Eliminator.

Okay. Removing the synth from the X.Sum input makes the noise go away. So now you are assured that the synth ground is part of the loop. (Which you pretty much knew already but now you really know because you measured it).

Next step is to go up the chain:

Is the X.Sum itself grounded in any way other than through the signal cable? Is it grounded through the rack to the rest of the equipment? Does it have a three-prong power plug?

It really helps to just draw yourself a diagram showing where all the grounds go.
--scott