ProSoundWeb Community

Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => AC Power and Grounding => Topic started by: Jamin Lynch on December 03, 2013, 05:18:12 PM

Title: Power from multiple sub panels
Post by: Jamin Lynch on December 03, 2013, 05:18:12 PM
If 1 sub panel isn't enough, is it generally a bad idea to get power from 2 different sub panels? What problems could there be?
Title: Re: Power from multiple sub panels
Post by: Jeff Bankston on December 03, 2013, 08:54:02 PM
120/240v , 120/208v multiple subpanels are as common as the sun. multiple subpanels get their power from a single service entrance. there will either be what we call switch gear or a distribution panel(s) or a combination of both.

480v , 240v systems can have a single or multiple 120/240 or 120/208 step down transformers. a large single 600kva transformer can supply multiple sub panels from a single ditribution panel. or you might have a single 120/240 or 120/208 transformer supplying single panles. all depends on how the electrical engineer drew it out. i dont see why there would be an issue. a friend works for the studios and runs multiple generators for movie shoots and large music concerts. 
Title: Re: Power from multiple sub panels
Post by: Lyle Williams on December 03, 2013, 10:25:13 PM
Induced noise due to different ground potentials would be my fear.  Fears don't always turn into reality.

I'd manage the problem by keeping clear demarcations on what is powered from what feed.  I definitely wouldn't run a desk and an effects rack from different power! (Too many unbalanced insert connections)

Every signal that crosses from one power system to another should be balanced and may need a xlr ground lift or better still a transformer.
Title: Re: Power from multiple sub panels
Post by: Steve M Smith on December 04, 2013, 02:57:16 AM
The ground connections will all be connected together at some point in the building so I wouldn't worry about that.


Steve.
Title: Re: Power from multiple sub panels
Post by: Jeff Bankston on December 04, 2013, 04:39:28 AM
The ground connections will all be connected together at some point in the building so I wouldn't worry about that.


Steve.
yep, to the cold water pipe within the first 5 feet that its inside the building.
Title: Re: Power from multiple sub panels
Post by: Lyle Williams on December 04, 2013, 05:40:20 AM
The ground connections will all be connected together at some point in the building so I wouldn't worry about that.

Then there is nothing to worry about.  All grounds are perfect and all grounds are equal.
Title: Re: Power from multiple sub panels
Post by: Steve M Smith on December 04, 2013, 07:47:28 AM
Then there is nothing to worry about.  All grounds are perfect and all grounds are equal.

The problems start when you start to add additional grounds such as the shields in link leads causing ground loop problems.


Steve.
Title: Re: Power from multiple sub panels
Post by: TJ (Tom) Cornish on December 04, 2013, 08:31:31 AM
Induced noise due to different ground potentials would be my fear.  Fears don't always turn into reality.

I'd manage the problem by keeping clear demarcations on what is powered from what feed.  I definitely wouldn't run a desk and an effects rack from different power! (Too many unbalanced insert connections)

Every signal that crosses from one power system to another should be balanced and may need a xlr ground lift or better still a transformer.
This is good practice.  Unless there are severe electrical problems, the current-carrying conductors in both the supply side and load side will be separate, so it makes no difference where the power comes from.  Any potential issues would be from currents on ground conductors (which are always there, even if all circuits are sourced from the same panel) affecting poorly-designed equipment with "pin-1" problems.

This can be mitigated to a large degree with the "Poor Man's Distro" concept: bring all of the power together via some means - i.e. multiple rack-mount power strips in the same rack, and then ground them together. 

This can be as simple as scraping the paint off the rack rails and the backs of the rack-mount power strips so the strips are grounded together through the rack rails.  This is perfectly legal, safe, and other than scraping some paint off, doesn't require modifying any equipment. 

Another way to do it is to build a project like this:
http://www.padrick.net/LiveSound/BarDistro/BarDistro.htm

Ground loops are frequently misunderstood.  Here are some things to keep in mind:

- Ground loops are everywhere, and cannot/should not be totally eliminated. 

- Ground loops are generally a harmless result of correct grounding practices - most devices need to be grounded with their electrical supply connections, and many devices that are metallic need to be grounded together to ensure voltage drops don't build up between different pieces of equipment.

- The cause of trouble is current on the ground conductors, caused either by leakage through a device, or more commonly induced from the current-carrying conductors near the ground conductors.

- The magnitude of this ground loop current depends on the length and size of the grounding path, the amount of device leakage, and the amount of current flowing near the current-carrying conductors.  While it may generally be the case that ground loop issues are reduced if circuits come from the same panel, this is not axiomatic.  Say  I have a sub panel at location A with heavy-gauge feeders to the main distribution, and another sub panel at location B near the stage, 200' from the distribution, also with heavy feeders. Most of my audio circuits come from panel A.  If I run a 20A circuit from panel A 200' to the stage through 12-ga wire, it's actually possible that I will have more current on the ground of this long circuit from panel A than if I ran a circuit from panel B which is close by, since the size of the grounding conductors from panel B back to distribution is so much larger than the ground conductor of the long branch circuit from panel A.

- It is never OK to lift the equipment ground - i.e. do not remove the ground pin of any power plug.  This may appear to solve the ground loop problem, but you've now created a huge safety issue, as the primary function of the equipment ground is to conduct fault currents away.  If the equipment ground is missing, this fault current will now go through the signal grounds, and/or the person touching the signal ground - the vocalist, guitar player, etc.


Having a few DIs with ground lift functionality and a few isolation devices like these: http://www.radialeng.com/r2011/twiniso.php should solve most problems.  If you still have trouble, it's time to look at repairing or replacing equipment with pin-1 problems.
Title: Re: Power from multiple sub panels
Post by: Frank DeWitt on December 07, 2013, 11:02:23 AM
This is good practice.
Having a few DIs with ground lift functionality and a few isolation devices like these: http://www.radialeng.com/r2011/twiniso.php should solve most problems.  If you still have trouble, it's time to look at repairing or replacing equipment with pin-1 problems.

+4 on the whole post. 
Title: Re: Power from multiple sub panels
Post by: Lyle Williams on December 07, 2013, 02:57:39 PM
Man those radials are expensive!  I think I'll stick to my cheapie ART boxes.  Not as good but at least I can afford to have enough on hand.
Title: Re: Power from multiple sub panels
Post by: g'bye, Dick Rees on December 07, 2013, 04:16:55 PM

Having a few DIs with ground lift functionality and a few isolation devices like these: http://www.radialeng.com/r2011/twiniso.php should solve most problems.  If you still have trouble, it's time to look at repairing or replacing equipment with pin-1 problems.

I use the EbTech "Hum Eliminator":

http://www.ebtechaudio.com/hedes.html

Model HE2-XLR
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 19, 2013, 09:33:23 AM
My experiments with ground loop induced hum shows that a typical XLR run (20 to 100 ft) will have something on the order of 1 ampere of shield current per volt of ground loop differential voltage. Since I've measured 2 or 3 volts between different grounds in outlets from separate service panels, that implies perhaps 2 or 3 amperes of current flow in your XLR shield. Note that the only limit to this current is the actual resistance of the interconnecting cable's shield resistance plus any connector resistance.

How your sound equipment responds to this shield current assault depends first on the "pin-1 problem" which is the result of manufacturers not running short/direct safety ground paths to the gear chassis, often intermingling their safety ground path with other signals on a circuit board. I've found that some active floor monitors will hum with as little as 100 mA of ground loop current, which can come from even 1/10th of a volt difference between grounds. While other active floor monitors can take 5 or 6 amperes of ground loop current (corresponding to 5 volts ground loop differential) and still be very quiet. However, note that 5 amps of current in the shield makes for a pretty good heater, and I've often wondered why some of my XLR cables would get hot during festivals with strange power setups. Now I know why.

Since these typical ground loop currents are on the order of hundreds of mA, and only occur in the shield with no return through the twisted pair, you can simply clamp a standard clamp-ammeter around the outside of the XLR cable to look for ground loop currents. No need to split out the twisted pair from the shield since the return current path is external to your clamp meter jaws and thus won't cancel out, such as what happens in a power cable. See attached for a graphic.

How much ground loop current to accept is up to you, but know that ground loop differential voltages are not static and will change depending on loading of remote circuits. Thus you can have a perfectly quiet stage during setup and sound check, but when the big coffee makers kick on in the kitchen you can have your floor monitors begin to hum like crazy. As we all know, that makes artists on stage very angry and makes us look like idiots.

That's why I always run audio isolation transformers between output of my drive rack equalizers and the snake feeding my FOH amp rack and active monitor wedges. I have a very nice ISO-8 from Whirlwind in my main rack, as well as keep a few Ebtech Hum Eliminator boxes in my gig box for special situations like needing to run a long XLR cable to a green room for a powered speaker. I now run audio isolation transformers on the outputs of my drive rack as a matter of SOP. I'm a simple man who has to do complex things, so if audio transformers on the outputs means that one less thing to worry about during a gig, them I'm all for it.

This problem does not go away simply because you have a digital board. On several of the X32 and PreSonus installs I've seen lately, there was so much ground loop hum between the analog outputs on the console and the floor wedges and FOH speakers that they installed Whirlwind or Ebtech isolation boxes in the XLR paths. How much you pay for your audio iso boxes is up to you, but realize that the frequency response of your system can be compromised by a cheap transformer. I tend to use the higher priced Whirlwind ISO boxes in my FOH system, and save the less expensive Ebtech Hum Eliminator boxes for floor wedges and green room sends.

One more thing to remember is that XLR connectors with the shield improperly bonded to the shell will short out the ground lift switch in many DI boxes and Isolation boxes that use metal XLR jacks. I've found a bunch of these cables recently in several installation with new imported XLRs. So the first thing I now check when a ground-lift switch doesn't seem to be opening the ground loop is to open up the XLR connector and look for the extra jumper. A little nip-tuck with a pair of wire cutters is the correct fix. I've included a graphic of an improperly shell-bonded shield in an XLR connector.
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 19, 2013, 10:03:31 AM
+4 on the whole post.

Frank, would that be +4 dB or what?  ::)
Title: Re: Power from multiple sub panels
Post by: Jordan Wolf on December 19, 2013, 11:14:55 AM
Mike,

One location I worked at recently just bought a bunch of XLR cables from Monoprice (I think it was their "Premier Series").

I opened one up to check out the solder job (not bad, actually) and found that Pin-1 was bonded to the XLR shell.  Someone's going to have fun snipping all of them...

The cable insulation is also very thick - I'm not a fan.

Just an FYI.

I do check for the XLR Shell bond fairly frequently, and am only occasionally surprised by them.
Title: Re: Power from multiple sub panels
Post by: Kevin Graf on December 19, 2013, 11:37:17 AM
The reason for not connecting portable cable connector shell's to the shield is because with a portable cable the metal shell may come in contact with other metal things that are a  different potential.  Chassis jacks and cable connectors permanently attached to these jacks don't have this problem.  Of-course it's hard to guarantee that the cable will always be used this way.

Tony Waldron has some good papers on big system wiring.

http://www.fragrantsword.com/twaudio/ 
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 19, 2013, 11:54:39 AM
The reason for not connecting portable cable connector shell's to the shield is because with a portable cable the metal shell may come in contact with other metal things that are a  different potential.  Chassis jacks and cable connectors permanently attached to these jacks don't have this problem.  Of-course it's hard to guarantee that the cable will always be used this way.

Tony Waldron has some good papers on big system wiring.

http://www.fragrantsword.com/twaudio/
Yup, I can remember in the early 70's that all my XLR cables were wired with "telescoping shields". That is, one of the XLR connectors (female?) had the shield bonded to the shell, while while on the other end of the cable (male?) the shell was left floating. When you connected two XLR cables together, then the female XLR would "telescope" its shield thru to the male XLR shell.

But I can also remember getting ground loops (and even sparks) when one of these "bonded" XLR shells laid on a metal amp rack. I even saw it spark on a metal covered outdoor stage. I think that the AES standard for XLR wiring changed sometime back then (late 70's?) to eliminate all XLR shell bonding to the shield, in order to eliminate this exact problem. Perhaps the workers soldering these imported cables in sweat shops don't have the latest AES papers to guide them.  ;)   
Title: Re: Power from multiple sub panels
Post by: Steve M Smith on December 19, 2013, 12:57:49 PM
Reading this and other posts here has made me realise how lucky we are with power in the UK compared with what you have in the US.

We usually just turn up and have a choice of three industrial connectors to plug into with no problrms.  The US appears to have an ever increasing* inventory of outlets and wiring arrangements, grounding methods, etc.

Is it due to different states having different rules and/or various legacy systems still in use or being adapted?

(* ever increasing in my knowledge of them rather than in actuality).



Steve.
Title: Re: Power from multiple sub panels
Post by: g'bye, Dick Rees on December 19, 2013, 01:04:30 PM
Reading this and other posts here has made me realise how lucky we are with power in the UK compared with what you have in the US.

We usually just turn up and have a choice of three industrial connectors to plug into with no problrms.  The US appears to have an ever increasing* inventory of outlets and wiring arrangements, grounding methods, etc.

Is it due to different states having different rules and/or various legacy systems still in use or being adapted?

(* ever increasing in my knowledge of them rather than in actuality).



Steve.

 You have shepherds.  We have cowboys.
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 19, 2013, 01:25:02 PM
We usually just turn up and have a choice of three industrial connectors to plug into with no problrms.  The US appears to have an ever increasing* inventory of outlets and wiring arrangements, grounding methods, etc. Is it due to different states having different rules and/or various legacy systems still in use or being adapted?

You have to understand the "states rights" concept to get your head wrapped around the NEC (National Electrical Code) problem. When a new NEC book is released, it's up to each state to determine when and if it will apply to that state. Same goes for each city and county and even each inspector. They will determine what year code book will apply for their district, and if they'll accept all of the code recommendations. Note that I said "recommendations" since the NEC is NOT a legal document, only a suggested set of rules that each district can decide to follow or not.

A good example is the latest requirements for AFCI being installed in bedrooms for new housing (Arc Fault Circuit Interrupters). While many would argue that AFCIs produce nuisance tripping and don't actually save lives from fires, that's a different discussion. However, at least a few of our states (North Carolina and perhaps Indiana, if memory serves) have declared it too expensive for new housing (approx $300 additional on a $150,000 house) and thus would slow the building of new housing. So it's not accepted as part of the local code in some states and is ignored in new builds.

The same thing applies for state speed limits. During the great gasoline shortage and rationing of the 70's the Federal government mandated 55 MPH speed limits on Interstate Highways to save fuel. However, states like Texas (with 80 MPH speed limits) said it was unconstitutional to force them to a 55 MPH speed limit. The only way the federal government could enforce the 55 MPH mandate in those states was to threaten to stop all Federal road grants for highway maintenance and improvements. Of course, money talks so all states adopted 55 MPH for a few years until the gasoline shortage was deemed to be over.   

The NEC has no such financial leverage, so it's a real patchwork of code compliance.
Title: Re: Power from multiple sub panels
Post by: Steve M Smith on December 19, 2013, 01:34:09 PM
I prefer Dick's explanation!

We also had speed limits imposed during the 1970s fuel shortage.  70 MPH on all motorways and dual carriageways.  This was and still is known as the national speed limit.

Before that, these roads had no limit (other than the maximum speed of your car).  It's hard to believe now.


Steve.
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 19, 2013, 01:37:05 PM
I prefer Dick's explanation!
Steve.
I've got to admit, that about sums it up....  ;D
Title: Re: Power from multiple sub panels
Post by: John Sabine on December 20, 2013, 09:12:46 AM
I have a rack with 3 of these that I've carried around for years and almost never used but they save my bacon from time to time.

http://artproaudio.com/art_products/audio_solutions/splitters_patchbays/product/s8-3way/

I originally purchased them to use as a broadcast distribution for a job I did but they have actually proven quite handy. Once I pulled the rack from my trailer because I was tired of carrying it and never needing it. The next job I had was a 1400 person venue with enormous ground hum issues and a staff electrician who was MIA. Back in the trailer it went.
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 20, 2013, 09:39:33 AM
Back in the early days of mixing consoles, there were transformers on the inputs and outputs. This made it pretty easy to simply lift pin-1 on the XLR and create a shield that was terminated on only one end. This resulted in superior CMRR and zero shield currents. However, transformers are expensive, heavy, and prone to magnetic interference. Once op amps became cheap, virtually all consoles went to electronic balanced circuits for both inputs and outputs. And XLR shields are routinely terminated on both cable ends, which introduced ground loop currents into the mix (literally). You can sometimes get away with pin-1 lifts on electronically balanced inputs, but only up to voltage differentials within the common-mode limits of the input electronics. I've not studied this in too much detail, but I think it's a topic worth considering and discussing.

Have any of you experimented with pin-1 lifting of XLR electronically balanced inputs with various ground-loop differential voltages? I have the gear to do some trials using various voltage offsets, but perhaps this has already been done before. If it's unexplored territory I'll start a new thread and gather some data.   
Title: Re: Power from multiple sub panels
Post by: Kevin Graf on December 20, 2013, 10:32:53 AM
But back in the early day's,  the XLR jack pin-1 was incorrectly wired to the audio circuit ground rather than being connected to the chassis/shield creating all kinds of problems.  It has taken 20 years (and counting) to get everyone on-board.
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 20, 2013, 10:53:37 AM
But back in the early day's,  the XLR jack pin-1 was incorrectly wired to the audio circuit ground rather than being connected to the chassis/shield creating all kinds of problems.  It has taken 20 years (and counting) to get everyone on-board.

And still counting. There's plenty of late model active floor wedges with the pin-1 problem. Also a number of studio monitor speakers I've run into that will hum with low ground-loop currents.

But the real question is this... just how much ground-loop current are you willing to accept in your XLR shields? Even if it doesn't create hum in your particular situation, I really don't like the idea of several amps of current flowing through shield paths. While XLR pins are rated for something like 7 amps of current, I've found AV system wiring with 5 amps of ground loop current which seems like a bad idea.
Title: Re: Power from multiple sub panels
Post by: TJ (Tom) Cornish on December 20, 2013, 11:02:28 AM
And still counting. There's plenty of late model active floor wedges with the pin-1 problem. Also a number of studio monitor speakers I've run into that will hum with low ground-loop currents.

But the real question is this... just how much ground-loop current are you willing to accept in your XLR shields? Even if it doesn't create hum in your particular situation, I really don't like the idea of several amps of current flowing through shield paths. While XLR pins are rated for something like 7 amps of current, I've found AV system wiring with 5 amps of ground loop current which seems like a bad idea.
Particularly when that 7 amp XLR pin is connected to a 28-ga snake wire.
Title: Re: Power from multiple sub panels
Post by: Mike Sokol on December 20, 2013, 12:11:17 PM
Last year I did a graduation gig that required I send program audio to a video trailer about 300 ft from the FOH console. And yes, the video trailer pulled its AC power from a separate building on campus with it's own sub panel. There were complaints about all kinds of hum in the video feed the previous years, so I did my simple DI press mult fix. While I have a rather fancy active press mult, it wasn't with me at the time. So I just plugged in a Whirlwind IMP2 from the TRS aux out of the console, did the ground lift, then ran a XLR cable over to the video truck. You can even daisy-chain a number of these passive DIs together to feed several video cameras during a shoot. (See diagram). Since the input impedance of a passive DI is quite high (perhaps 50K or so) you connect a dozen or more of these together without any impedance loading issues.

While a press mult is a nice thing to have, if you're doing small corporate or local political work often you just need to get audio to one or two video cameras. Don't be tempted to just run them a direct XLR feed from the console even for a pro camera since many of them don't play nicely with line-level XLR (even if they claim to). And many pro camera grounds have terrible hash noise from all their on-board switching power supplies. So send them something that looks like a mic-level signal, isolate the shield from your sound system ground via a DI box, and things will be beautiful.
Title: Re: Power from multiple sub panels
Post by: Frank DeWitt on December 20, 2013, 01:05:43 PM
So I just plugged in a Whirlwind IMP2 from the TRS aux out of the console, did the ground lift, then ran a XLR cable over to the video truck. You can even daisy-chain a number of these passive DIs together to feed several video cameras during a shoot. (See diagram). Since the input impedance of a passive DI is quite high (perhaps 50K or so) you cab connect a dozen or more of these together without any impedance loading issues.

This is one of those ideas that is so crazy simple it is wonderful.  I will keep it.

BTW  A Jensen transformer equipped DI box will be about 140 K