Guitar amp buzz even with balanced power supply. Help!

[RogerAF]

I installed a recording studio in a new house for a client. He bought a balancing transformer and mounted it on the wall of the garage which was below the control room. It was installed by a "licensed electrician" (insurance reasons) and the center tap went to the ground at the meter. That thing buzzed so loud you heard it throughout the house. Not electrical buzz, but actual physical vibration transferred into the structure. I told my client that it needed to be mounted outside on a separate concrete base, in a weatherproof housing. But I don't think that's your problem.
When I guilt my studio I took a 240V feed from the meter and ran it to a a pair of 20A Power Line Noise Filters, added MOVs for surge protection. I used 3-wire cable for each circuit and added 6' copper earth ground rods 10' apart--one for the lighting circuit and the other for the studio outlets. All outlets were isolated ground types. All the wires were in rigid conduit that was grounded at the breaker box. Only the conduit was grounded at the breaker box. All outlets and lighting grounds went to the earth-rods.
That worked beautifully! My dad's shop was next to the studio and he could run any power tool or table saw and there was no noise in my monitors or in the guitar amps.
IMHO the whole idea of "Balanced AC Power" is a kind of audiophool-ishness. Especially when it comes to guitar amps. Our amps in the USA are designed to have a hot leg, a neutral leg, and a ground leg. The ground leg connects to chassis ground and insures protection from electrical malfunctions (including electrocution). The guitar cable's shield is connected to chassis ground and thus to mains and meter (earth) ground. If it's a tube amp, chances are that the -V side (after the rectifier) is also connected to chassis ground. The neutral wire is only connected ground at the meter. It then "floats" throughout the building. The little Power Line Noise Filters I use are rated at 120/240VAC @ 20A. They have a neutral leg, a hot leg, and a ground/case leg. They used to be $5 each, but now I have to go to ebay and they run about $15.
Your balanced power transformer should have a center tapped secondary that IS connected to earth ground at least at the meter. Or you can install another 6' copper rod into the earth in your studio and hook all your studio outlets to it. That goes then connects to the center tap of the transformer secondary and NOT to the meter's earth ground.
To me, it sounds like a grounding issue. And radio pickup also indicates that. The whole idea of a balanced supply of 120VAC means that each leg is at 60VAC when measured to ground. And ground voltages can change from one end of a building to another end. The military specs grounding rods at 10' centers throughout all warehouses. All outlets connect to them. They all connect to each other, and to the earth at the meter. So make sure your transformer has a center-tap on the secondary, and that the ct is grounded to the earth. (Tech Tip 1: Check that any screws or nuts that make connections to your transformer and to your outlets are all tight. And that you use crimp connectors on the wire ends where they connect to the outlet. Do not just wrap the wire around a screw.) (Tech tip 2: Get an outlet strip that has Power Line Noise Filtering--not just Surge Protection--try that on an outlet. If you have to, run an extension cord to an outlet NOT connected to your balancing transformer. What happens then?)
On a side note: I once did a session at a studio with my Les Paul and small Marshall. Things were going along fine (I thought) I had my headphones on. But when I took them off, I could hear the playback coming from my amp. I called one of the engineers in and we eventually discovered that my headphone mix was somehow getting into my guitar and thus to the amp. His solution? He soldered a wire to a copper penny which I put against my skin under my belt. The other end was soldered to the cover of my guitar cord plug. That fixed things, but no one ever figured out how that problem even existed in the first place. It only happened when the headphones were on my ears or down around my neck. Go figure!

 

[JohnRoberts]

On a side note: I once did a session at a studio with my Les Paul and small Marshall. Things were going along fine (I thought) I had my headphones on. But when I took them off, I could hear the playback coming from my amp. I called one of the engineers in and we eventually discovered that my headphone mix was somehow getting into my guitar and thus to the amp. His solution? He soldered a wire to a copper penny which I put against my skin under my belt. The other end was soldered to the cover of my guitar cord plug. That fixed things, but no one ever figured out how that problem even existed in the first place. It only happened when the headphones were on my ears or down around my neck. Go figure!

Do not do stuff like that... You can buy commercial grounding straps that will effectively ground meat puppets, but through enough series resistance that you do not expose yourself (or friends) to personal harm.

JR

 

[Brian Roth]

FWIW:

"I used 3-wire cable for each circuit and added 6' copper earth ground rods 10' apart--one for the lighting circuit and the other for the studio outlets. All outlets were isolated ground types. All the wires were in rigid conduit that was grounded at the breaker box. Only the conduit was grounded at the breaker box. All outlets and lighting grounds went to the earth-rods."

That is a clear violation of USA building/electrical codes. ALL "grounding" wires have to be bonded to a single point in a structure and NOT to random "earth-rods". In addition, it all has to be bonded to any/all metallic water and gas pipes in the structure.

Bri

 

[abbey road d enfer]

no one ever figured out how that problem even existed in the first place. It only happened when the headphones were on my ears or down around my neck. Go figure!

It makes me strongly suspect the strings were not properly grounded to the guitar.

 

[abbey road d enfer]

That is a clear violation of USA building/electrical codes. ALL "grounding" wires have to be bonded to a single point in a structure and NOT to random "earth-rods".

Is that an obligation when there is no direct connection to the grid? In Europe you can create a secondary earth when using an iso xfmr.

 

[JohnRoberts]

It makes me strongly suspect the strings were not properly grounded to the guitar.

not to veer too far into human safety concerns, but you can ground guitar strings et all through a capacitor to likewise reduce shock hazards. Fairly common for guitars to provide a ground path for energized microphones.

JR

 

[RogerAF]

Do not do stuff like that... You can buy commercial grounding straps that will effectively ground meat puppets, but through enough series resistance that you do not expose yourself (or friends) to personal harm.

JR

At the time I was an ignorant musician. I just wanted to get my guitar part recorded without my headphone mix added to the track. I was just happy that it worked. BTW the amp had recently been overhauled and the electrical ground was connected. Later, when I knew enough to be dangerous, I took the amp apart and started modding it until it was horrible. I eventually traded it to a friend who knew all about Marshalls and he restored it.

That is a clear violation of USA building/electrical codes. ALL "grounding" wires have to be bonded to a single point in a structure and NOT to random "earth-rods". In addition, it all has to be bonded to any/all metallic water and gas pipes in the structure.

Bri

I beg to differ, although you may be right--now. Honolulu had a plethora of military warehouses. The conduits ran overhead and there is a quad box at every post, with a ground rod embedded in the slab. The only code I am aware of is that the neutral wire connects to ground only at the meter. Connecting piping also to ground would tend to render your point moot, as any pipe that went in the earth would serve as a "random earth rod". My studio was in a basement with a floor at least 12' below the level of the meter ground. I felt there was a good chance of a difference in voltage potential in the ground. Also, I wanted to separate lighting ground from audio ground. That way my dimmers did not buzz the monitors. What can I say? It worked.

It makes me strongly suspect the strings were not properly grounded to the guitar.

That's the mystery. I had replaced the pickups and found a broken ground wire. I made sure to replace it. Letting go of the strings in some places I played, would cause the amp to buzz more (it was a Marshall after all). So the strings had a good connection to the cable shield, chassis ground, and power ground, but I don't know about the studio ground. The studio was built in a house and belonged to an award winning producer and was pretty impressive, but who wired the house in the first place is unknown. The other part of the mystery is that when the tech (who wired up the penny) was trying to figure out what was going on, he moved the headphones away from me and the sound of the mix in the amp, went away. So it was some kind of body capacitance that was causing the bleed. At least that's the theory he and I came up with later. (Later after I graduated from tech school and worked in the tech department of a stage, sound, and lighting company, he was another tech in the same department. We became partners doing recording studio installations and repairs.)

 

[ccaudle]

In Europe you can create a secondary earth when using an iso xfmr.

I believe that is also the case in the US if the iso transformer is permanently mounted as part of the building wiring. Not allowed if the iso transformer is part of a movable power filter or distribution box of some kind.

 

[Brian Roth]

I now hope that a licensed master electrician (or a properly licensed Electrical Engineer) will chime in here.

I see two different system design conditions.

1. A sub-panel installed for the studio. I am almost certain that sub-panel will require its own individual grounding conductor that goes back to the service entrance ground. There, and only there, the grounding and neutral conductors are bonded together. IOW, the sub-panel will require individual conductors for neutral and grounding. The neutral and grounding conductors cannot be tied together at the sub-panel. But, I'm pretty certain the NEC doesn't "care" if you add an additional "ground rod" at the the isolated grounding bus in the sub-panel, as long as it's all tied back to the service entrance.

2. What the NEC describes as a "derived system" which would include an iso transformer (but would include a generator, wind or solar supply). It's a bit hazy for me here, but I recall a project from decades ago. It was a studio in (effectively) the basement of a high rise. The building had 480/277 VAC bus bars through out. In multiple places, transformers were installed to derive 208/120 VAC (Wye connected secondaries) to mainly supply "ordinary" branch circuits via their own breaker boxes. In those cases, the grounding and neutral conductors HAD to be connected in those derived power breaker boxes to ensure safety. I am almost certain that EVERY grounding point in the building were all connected to a individual grounding bus that ran throughout.

Bri

 

[RogerAF]

I'll mention a further note here: My dad's house in Honolulu, was a large wooden structure with 3 stories. Each floor had its own breaker box and the basement had 2, plus 1 when I installed my studio. The meter and earthing rod were installed by the power company. There was a breaker panel for each interior breaker box. The wire to the meter came off a transformer on the street above. It was 2 large conductor wires twisted around a steel cable. The cable was the ground wire and derived from the CT of the street transformer. The 2 large conductors were 240VAC between them. 120VAC was between either conductor and ground/neutral. Only at the meter earth rod was neutral connected to ground. So really this is a balanced 240VAC system with a center tapped ground. And neutral is derived from ground, but only at the meter power earth point.

Last story (I promise): My first recording studio job was down a few blocks from my house. The equipment was all MCI. A JH636 console and an older 2" 24 Track. When I took over the job, the wires had all been pulled through recently installed conduit, but not yet connected to the Tuschel connectors for the console. The tape machine was connected, as were the monitors. There was the faintest of buzzing from the tape heads. Raising or lowering the head shields didn't seem to make any difference. The noise was so faint that any audio drowned it out, still... not something you want getting on your master tape. The noise was loudest near a quad outlet box that the previous tech had disconnected (in an effort to get rid of the noise). We found that spinning the machine and moving it away from the old outlet could make the sound less, so we found the quietest position and the tape machine lived there.
A few years went by and I had to go back and deal with some issue that required me to crawl under the house below the control room. While under there, I noticed a piece of abandoned conduit that went at a 45 degree angle to the area near the console power supply rack. This piece of conduit was not connected to anything except the old quad box in the control room. No wires were in it, but it was in my way, so I removed it, did my thing and went back in the control room. The buzz was gone. Just like that.
My conclusion is that because the console power supplies and the 2" tape power supplies were always on, there were large fields (electric and magnetic) surrounding the conduits. All the newer conduits were running at 90 degrees to each other. This piece of conduit at 45 degrees to the east/west and north/south newer conduits, was in just the right position to couple to the radiated fields and re-radiate into the control room from the abandoned quad box. Let me try to give a clear picture; there was a steel quad box coming out of the floor and below the floor its conduit went for 8' or 10' to dangle below the console power supply rack. Just a long floating antenna coming out of the floor into the control room! It should be noted that this house had only the earth rod at the meter box. So effectively both ground and neutral were floating throughout the structure. The meter was at the opposite corner of the building. About as far from the studio as it could get. Perhaps another earth rod to ground below the console power rack would have helped, but I wasn't quite as hip to such things back then. I kinda understood capacitance, but inductance was very mysterious (which it still is). Well, life is for learning and I hope to always be doing some of that.

 

[Dan Elleson]

Hi. I don't have time to read the entire conversation so forgive me if I'm doubling up on this.

I'm not an electrician, but an electronic engineer and studio tech, and as I understand it, connecting multiple earth /ground steaks is incredibly dangerous because the earth has high electrical resistance and when you get lightning strikes near the stakes, you get a huge voltage gradient across the surrounding area, and the voltage between any of those stakes and therefore the ground circuits, will be huge, which is potentially deadly and destructive, rather than safe.

Aside from stray electromagnetic fields from nearby lighting dimmers and power transformers, or things like that affecting guitar pickups, every hum or buzz I've solved has been an interconnection problem between instruments and equipment, or poor wiring and layout in the instruments and devices themselves, like electric pianos and the like, which typically have poor wiring layouts and mechanical construction.

There's no panacea, it requires testing, consideration and experimentation. If you haven't read these 2 papers, they're good all round references for tackling these issues. Just in case they help...

https://www.jensen-transformers.com/wp-content/uploads/2014/08/generic-seminar.pdf
https://www.ranecommercial.com/legacy/pdf/ranenotes/Sound_System_Interconnection.pdf

 

[Matt Syson]

Using a dedicated transformer to provide 'balanced power' is largely unnecessary as ,almost by definition, the power supplied to your gear is at least current balanced. It is CURRENT flow that creates the magnetic 'hum' field that CAN be induced into other conductors. In France especially (I have not looked at other regions recently) ALL circuits are fed from 'current balance' breakers (ELCB or other terminology) so for every circuit the LIVE current through the breaker MUST have the corresponding 'Neutral' (return) current, thus by implication all circuits must be balanced to LESS than 30milliamps otherwise the breakers would keep tripping out.. Mains cords to gear tend to be twisted pairs (or triples when they have an earth) so the magnetically radiated 'hum' field will be minimal. Some old stuff might have 2 (or 3) core 'oval' profile cord but as the conductors are in close proximity their radiated field will be small. An isolating (balancing) transformer will also raise the impedance of the mains supply thus 'distortion' of the mains caused by rectifiers will be more pronounced. Poor wiring practice in a building, where the live and it's neutral return do not stay physically close (Ideally twisted as a pair but this would be unusual) is likely a 'source' of unwanted hum/buzz issues.
After all, most people use balanced mics simply to gain some freedom from induced interference so the same 'rules' of twisting pairs should be used with mains wiring although electrical 'code' is more concerned with safety rather than interference.
Matt S

 

[RogerAF]

There's no panacea, it requires testing, consideration and experimentation. If you haven't read these 2 papers, they're good all round references for tackling these issues. Just case they help...

Jensen Transformers Rule! I was able to access the first paper, but got an error message on the second. Still, the first is a long read, but the information is well worth it. I may have made some errors when I wired up my studio with separate earth rods. But I will stand by a couple of things: 1) It was extremely quiet and the light dimmer did not get into any audio or video. I had little halogen track lights and a solid state dimmer. Also there was no interference from power tool motors running in the shop next door. 2) Before installing this distribution system, I would get a small shock from touching a mic or my guitar strings while standing barefoot on the slab. I was using a single power line noise filter plugged into an outlet in the shop, with a 50' extension cord. (Large--12 gauge). This also blocked motor noise, but if I stepped out of my flip-flops I got surprised! After installing separate earth rods at the level of the slab, this shocking behaviour went away. So it may have been the wrong approach, but it was successful none the less.
P.S. Thanks for that paper. I got the Jensen catalog back in the late 90s and it is full of circuits and app notes. This new (ish) paper makes a nice little brother (or sister, or ??? insert PC ID label here).
Cheers,
Roger

 

[Matt Syson]

Equalising the potential between all and various parts is a good practice for safety at least. the issue for 'noise' (hum/buzz) starts when you consider the physical routing of all the grounding paths (be it building frame, water/gas pipes etc and THEN the physical routing of the grounds and 'reference' connections of all instruments, gear and mics etc. Combined this is a heck of a complicated 'web' for which spiders would be jealous. Also consider that as switchmode supplies and mobile phones etc are increasingly present each section of 'conductor' (a few inches when considering microwaves from mobile phones) represents an aerial either transmitting or receiving. ALL of this simply adds to the confusion so while there are good practices to 'minimise' interference guaranteeing it is a LOT more complicated. transformers CAN be a part of a solution but not necessarily because they can be 'balanced but because they also represent a low pass filter which is passive so less likely to present a 'semiconductor' connection which can demodulate high frequency interference that has 'modulation' superimposed. You can't hear microwave frequencies but you can hear microwave frequency signal that is modulated at an audio frequency (Basic AM radio style).
Matt S

 

[ccaudle]

Using a dedicated transformer to provide 'balanced power' is largely unnecessary as ,almost by definition, the power supplied to your gear is at least current balanced.

The one situation that may make sense is if the leakage current through the filter capacitors in line entry EMI filters is causing problems for high impedance unbalanced inputs (e.g. guitar amps), balanced power feed can reduce that leakage current through reduced voltage across each cap, and partial cancellation at the earth connection point of those caps.

connecting multiple earth /ground steaks [sic] is incredibly dangerous

Multiple ground rods are allowed if all rods are connected together with proper gauge wired connections. You are correct that multiple unconnected ground rods are a safety hazard.
And the neutral to ground connection must always be at the main power entrance, if there are multiple ground rods the neutral to ground connection cannot be made at any closest ground rod, it must run back to the power entrance.