Proper wiring to avoid ground loops

[mrtnasty]

Hi, have 60hz hum in my recording setup, and I am suspecting it is due to ground loops. I have a preamp and compressor that have their chassis grounded to earth, and also both have both input and output xlr pin 1 grounded to the chassis. If I remove the grounds on xlr pin 1 from both outputs, is that the proper way to break the ground loops in my setup? Also somewhat related, my microphone does not have xlr pin 1 connected to the mic body, should I make that connection? Thanks.

 

[Khron]

my microphone does not have xlr pin 1 connected to the mic body, should I make that connection?

Yes, definitely.

 

[Ike Zimbel]

+1. I would add that once you ground the microphone case, your other "ground loops" might just disappear.

 

[fazer]

Very common to have loops especially with sound cards and pro gear. I usually lift outputs and keep inputs grounded except mic lines need ground all the way through for case ground on microphones and phantom power.

You can connect a .01 mfd cal to shield and pin 1 for an RF shield at higher frequency.

 

[mrtnasty]

Very common to have loops especially with sound cards and pro gear. I usually lift outputs and keep inputs grounded except mic lines need ground all the way through for case ground on microphones and phantom power.

You can connect a .01 mfd cal to shield and pin 1 for an RF shield at higher frequency.

Yea I noticed that my computer is a huge source of EMI, if I put a mic cable and put it close to the computer, there is an incredible amount of noise. How do I make sure the usb cable is grounded? Im using a Focusrite Clarett.

 

[Khron]

How do I make sure the usb cable is grounded?

I should expect the USB connectors to have their shields and ground pins directly and firmly connected to ground at each respective end (computer and interface), from the factory.

if I put a mic cable and put it close to the computer, there is an incredible amount of noise

Is that with a microphone connected, or not?

 

[ccaudle]

If I remove the grounds on xlr pin 1 from both outputs, is that the proper way to break the ground loops in my setup?

No, pin 1 tied to chassis is a proper shield connection.
What most people call "ground loops" is a mistake in routing the shield connections (XLR pin1 ) across the audio circuitry so that shield currents have a path through the audio reference node (aka "circuit ground").
If that is the cause of your problem you may be able to correct it, but you need to be confident of your analysis and not afraid to cut into connector pins and PCBs.

I usually lift outputs and keep inputs grounded

If the shield is connected at only one end it should be connected at the output end and lifted at the input end. Bill Whitlock showed why this is the case in a very clear diagram in his 1995 paper "Balanced Lines in Audio: Fact, Fiction, and Transformers."
You can find the same diagram in several of his presentations online. Short version is if connected at the output end the output driver circuit and the shield are both driven by the same common mode voltage, so the cable capacitance does not have any effect. If the receiver end is connected, then the common mode noise voltage driving the shield is different than that driving the output circuitry, so any imbalance in cable capacitance causes common mode to differential mode conversion.

 

[ccaudle]

Who cares about bill wilcock and his rigged cable to try to prove you had to put a ground on both ends for the shield to work. It has been done ever since shielded wire came into existence. It only needs one side grounded.

You obviously did not read or at least did not understand the material. The point was that if you only connect one end (no disagreement that connection at one end only is fine for low frequency shielding, the only issue is RF performance), the proper connection point for the shield is at the output end, not the input end.

 

[CJ]

Eggnog?

 

[ccaudle]

Eggnog?

Sounds good, I think I might. Good suggestion CJ.

 

[mrtnasty]

It makes no difference on the microphone. your ground loop is more likely from the output of your outboard to the interface, which you would lift the gnd at the computer plug so you don't radiate the computer ground.

Whats the gear you are connecting together?
I'll tell you the best way to connect them.

I’ve got a ribbon mic (currently pin 1 lifted) to a tube preamp to a tube limiter to a passive eq, all transformer coupled, I wired up all power cables and Xlr connectors and grounded all grounds to the chassis. Thanks!

 

[mrtnasty]

Eggnog merry Christmas everyone and thanks for your help!

 

[JohnRoberts]

Who cares about bill wilcock and his rigged cable to try to prove you had to put a ground on both ends for the shield to work. It has been done ever since shielded wire came into existence. It only needs one side grounded.

The text books I have sitting around on topic were published back in the 70s (like Ott), so maybe you have some newer information?

The general consensus AFAIK is that HF shielding effectiveness requires ground bonding both ends of signal cable shields. There are hybrid RF shield approaches that ground the second shield end through a small cap to provide a low impedance bond at RF but high impedance at hum frequencies to avoid corrupting ground current flows between chassis.

JR

PS: Bill Whitlock has probably forgotten more about grounding than I ever knew, but even I know to ground both shield ends for RF rejection.

 

[Newmarket]

Who cares about bill wilcock and his rigged cable to try to prove you had to put a ground on both ends for the shield to work. It has been done ever since shielded wire came into existence. It only needs one side grounded.

Someone has a grudge

 

[Newmarket]

I should expect the USB connectors to have their shields and ground pins directly and firmly connected to ground at each respective end (computer and interface), from the factory...

You'd reasonably think so. BUT IIRC the USB standards themselves require that the cable screen is hard connected at one end only. For "ground loop" avoidance. I have discussed this several times in EMC labs etc. And on occasion I've specified making a hard connection to chassis. Generally to give better immunity to rfi of some sort.
Caveat: experience here is with USB 2. Not sure if standard differs on this with USB 3.x, USB C connector, Thunderbolt.
Also filtering on USB interfaces is often minimal compared to what is optimum (see Wurth literature on this).

Merry Christmas

 

[Toobme]

That interface - in our edit suites at school, we use those, and I’ve had the best (least hum) performance with a ferrite bead on the usb cable.
The older gen 2 Avid mbox had this issue - the bead didn’t work in that case, but it was the same hum on both units. USB and ground maybe?

 

[JohnRoberts]

No, Its been done for several decades inside equipment all the time. Electronic theory doesn't change when its passing out the XLR jack.

Of course it is a hugely different circumstance.

Bonding different equipment chassis grounds together can allow current flows between those two chassis that could corrupt signal ground integrity in poorly designed audio gear. The well inspected "pin 1 problem" shows how to protect signal ground integrity when dealing with shield ground connections.

JR

 

[MidnightArrakis]

Hi, have 60hz hum in my recording setup, and I am suspecting it is due to ground loops.

[have 60hz hum in my recording setup and I am suspecting it is due to ground loops] -- Years and years ago when I built and wired my music mixing production studio, I was generally worried about inadvertently creating some ground-loops with so many balanced and unbalanced cables going between everything everywhere. I knew that there had to be some kind of solution to this problem without having to resort into inspecting how every cable and piece of equipment was grounded and wired and after researching this issue a bit.....I found the answer.....well, at least for me.

>> BALANCED-POWER << -- While I am more than certain that there will be those on this forum who will probably completely debunk and/or decry the use or validity of using "Balanced-Power", but at least within my mixing studio, my audio is -- DEAD SILENT -- even at high gain levels. I can barely even hear any hiss coming out of my monitors with the faders up on my mixing console and the console's output control set at the 3 o'clock position. DEAD QUIET!!!

I have three 8-channel snakes running from my portable digital recorder rack to the inputs of my mixing console. Then, there is a TON of cabling running between the various console AUX and CHANNEL sends/returns and my signal-processing rack (shown as the bottom image below). And, guess what!!! >> NO HUM!!! <<

And.....I seriously doubt it was just "dumb luck" that I happened to interconnect and wire-up everything correctly!!! (Although that certainly is a real possibility)!!!

> TOP -- Mic-Preamp Rack // BOTTOM -- Digital Recorder Rack


>> IMAGE -- View of Earth photographed from the Moon photo-mural (Moon horizon is seen at the mid-woofer level)
>> TOP -- DynAudio BM15A Studio Monitors // BOTTOM -- MACKIE 32*8 Mixing Console


>> SONY 1U CD-PLAYER // TC ELECTRONIC "FINALIZER" // SONY CD MASTERING RECORDER
// PRESONUS 8-CHANNEL COMPRESSOR/LIMITER "X3" // TC ELECTRONIC "TRIPLE-C" MULTI-BAND COMPRESSOR


>> LOTS OF TC ELECTRONIC SIGNAL-PROCESSING EQUIPMENT (INCLUDING 7 MORE "TRIPLE-C" COMPRESSOR/LIMITERS)
>> The "Balanced-Power" chassis is mounted at the bottom portion of this signal-processing rack and is not shown here:


/

 

[JohnRoberts]

We were having a discussion about single point shielding John.
why would it be different in a rack patch compared to shielding a run inside?

BTW, Merry Christmas.

The OP was asking about hum in his recording setup....

JR

 

[RCAguy]

Even “professional” gear - recording or replay - with balanced i\o can have a “pin 1 problem.” Caused by the internal signal common circuit not properly connected to the enclosure & cable shield ground once, at the PS reference point. (I return an astonishing amount of new gear that cannot pass the test of injecting 1v of 60Hz AC to any pin 1.) It can also help to ground a balanced shield at the source (disconnect at the destination except in a high RF environment, then through 0.02uF).