Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Ryan C. Davis]

Just a quick question, after reading about separate FOH and Monitor mixes in this thread: http://forums.prosoundweb.com/index.php/topic,138124.10/topicseen.html

When you guys talk about having a separate Engineer mixing FOH and monitors I've gathered that that happens on two different boards in two different parts in the venue. I guess what I'm trying understand is how the signals from the on-stage sources make it to two mixers. Do they go through a distribution amp, or is it just as simple as splitting the XLR Feeds? It seems like splitting the two would noticeably reduce the level if done passively (wouldn't that be a current divider?). Just not sure about the most common and best engineering practices (I guess those two could be totally different ) on systems like this since I have no experience with a setup that elaborate.

Educate me please & TIA!

[Patrick Tracy]

A simple XLR splitter is often used. I have a 16-channel splitter with an input box and two tails. There are ground lift switches for each channel affecting the "split" tail while the main tail is always grounded. Sometimes the split is done with transformers, such as when it goes to a remote recording truck that may be on different power, but for most of us it's just a passive split.

Modern interconnection works by voltage detection rather than power transfer. There is very little signal loss.

[Jordan Wolf]

Ryan,

Most systems utilize passive splitters that have one set of parallel outputs and one or more sets of transformer-isolated outputs.  The convention I normally see is: MON gets the parallel split, to provide phantom power (it doesn't pass through transformers), and FOH & Broadcast/Recording/Whatever Else gets the iso feeds.

Unless you're splitting a mic-level signal more than 3x, I don't think there are too many issues.  Line level feeds, I'd feel comfortable splitting up to, say…6x or so unbuffered (no D.A.).  If you use an active splitter, you can split the signals almost as many times as your budget allows, er…you want to.

As long as everything is balanced and fed off of the same power source (or properly isolated from each other), everything should play nice and hum/buzz should be a nonissue.

[Mike Sokol]

Just a quick question, after reading about separate FOH and Monitor mixes in this thread: http://forums.prosoundweb.com/index.php/topic,138124.10/topicseen.html

When you guys talk about having a separate Engineer mixing FOH and monitors I've gathered that that happens on two different boards in two different parts in the venue. I guess what I'm trying understand is how the signals from the on-stage sources make it to two mixers. Do they go through a distribution amp, or is it just as simple as splitting the XLR Feeds? It seems like splitting the two would noticeably reduce the level if done passively (wouldn't that be a current divider?). Just not sure about the most common and best engineering practices (I guess those two could be totally different ) on systems like this since I have no experience with a setup that elaborate.

Educate me please & TIA!

Here's the basics. When you have a low-impedance output such as a microphone or DI box, it's typically capable of driving a 600 ohm load. That's from the good old days of Ma Bell, when telephone systems were all 600 ohms. But modern mixers don't have anywhere near that low of input impedance. Most are in the 10K Ohm range. That implies that a single mic would be able to drive something like 16 mixer inputs (10,000 divided by 600 equals 16.67), but of course we don't need to drive 16 different mixers, we only need to drive two mixers in your case, FOH and Monitor.

But there's more than just letting the signal go down two paths... you also need to isolate phantom power and lift ground loops. Of course, phantom power is the 48 volt DC bias that's applied on pins 2 & 3 of the mixer inputs, which then powers condenser mics and active DI boxes. If you use a simple Y-splitter cable from a mic to feed two mixers, any channel with phantom engaged will back-feed to the other mixer, reducing the phantom voltage and shutting down the mic. You "could" turn on the phantom power on the same channel from each mixer, but that's not generally done. There's a simple way to block DC voltage while allowing AC signal to flow, which is a capacitor. So in a simple splitter snake you will have one of the set of mixer inputs be direct wired to the mic output, and that mixer output is the one that supplies phantom power, while the other output is capacitor isolated from the phantom power. Typically the hard wired output with a phantom path is reserved for the monitor console since the operator is much closer to the stage and knows which mics/DI'ls need phantom or not.   

But we've still not addressed the issue of ground loops, which can induce all sorts of hum into a PA system. Let's not get into the details of how and why ground loops occur (I could write a book on that) but focus on how to get rid of them. While there are various power grounding techniques and XLR pin-1 lifts available, the gold standard is to use an audio transformer split to feed one of the mixing consoles (usually FOH) while the other output is the direct wired one feeding the monitor mixer. This allows the introduction of a true ground lift on the center-tap of the audio transformer, which eliminates the hard-copper bond in the shields between the mixing consoles. The result is that ground loop currents are eliminated, along with that pesky hum.

These splitters are also available in active DA (distribution amps) that have powered outputs, most useful when you need even more splits for the broadcast board and whatever else you like. Plus an active split is better at feeding very long cable runs (over 250 ft total) that will start to sound dull using passive splits simply due to the capacitance of the twisted pair cables in the snake(s) shorting out the high frequencies of the audio signal.

So you can get a snake with a built in hard-wire or transformer isolated split. Or you can get a splitter box with 16 to 32 inputs and 32 to 64 outputs. You can even get active splitters with 4 or more mic level outputs per each mic input. But, of course, the price goes up with each level of technology.   

Here's a really good tutorial from Whirlwind about how splitter boxes and snakes work: http://whirlwindusa.com/support/tech-articles/microphone-splitters/

[Charlie Zureki]

But there's more than just letting the signal go down two paths... you also need to isolate phantom power and lift ground loops.

  Not necessarily.  But, it is easy to "isolate" phantom power by using a transformer.  Sometimes we encounter ground loops, but, it is not a given.  Lifting pin #1 is easy with a lift adapter, or jumper cable with pin #1 not connected,... that is... if the snake does not a have a pin lift per channel. 


 Of course, phantom power is the 48 volt DC bias that's applied on pins 2 & 3 of the mixer inputs, which then powers condenser mics and active DI boxes. If you use a simple Y-splitter cable from a mic to feed two mixers, any channel with phantom engaged will back-feed to the other mixer, reducing the phantom voltage and shutting down the mic.


Huh?  feedback.... the voltages do not add or necessarily subtract as the phantom supplies would be in parallel.  (providing the correct polarity was observed)

 You "could" turn on the phantom power on the same channel from each mixer, but that's not generally done.

  Having Phantom engaged on both consoles  rarily causes any problems

[Mike Sokol]

  Not necessarily.  But, it is easy to "isolate" phantom power by using a transformer.  Sometimes we encounter ground loops, but, it is not a given.  Lifting pin #1 is easy with a lift adapter, or jumper cable with pin #1 not connected,... that is... if the snake does not a have a pin lift per channel. 


 Of course, phantom power is the 48 volt DC bias that's applied on pins 2 & 3 of the mixer inputs, which then powers condenser mics and active DI boxes. If you use a simple Y-splitter cable from a mic to feed two mixers, any channel with phantom engaged will back-feed to the other mixer, reducing the phantom voltage and shutting down the mic.


Huh?  feedback.... the voltages do not add or necessarily subtract as the phantom supplies would be in parallel.  (providing the correct polarity was observed)

 You "could" turn on the phantom power on the same channel from each mixer, but that's not generally done.

  Having Phantom engaged on both consoles  rarily causes any problems

Last things first: It's not that phantom voltage adds or subtracts, it's that there are 10K build-out resistors which feed the mic. If you using a simple Y-cable split, then the 10K resistor in the board with phantom on is bonded to the input pin(s) of the second board. If that board's phantom is off, then your mics are being powered from the center-tap of two 10K resistors in series, and likely only receiving 24 volts. Some mikes don't care and will run off 24 volts, but old-school mics need the full 48 volts to operate. Yes, you can turn on the phantom in both consoles, and it will work, but we don't like that much for concert work since that's one more thing to remember during setup. It won't hurt anything... we just don't like to do it due to more error possibilities.

As far as ground loops, I've been doing ground loop experiments in my studio lab for the last 2 years, and dealing with them on stage for the last 40 years. A properly designed snake with transformer isolated audio splits will never pull any ground loop current, and thus can't hum. Pin-1 XLR lifts will work in many circumstances, but their success is predicated on the CMRR (Common Mode Rejection Ratio) of the balanced input stage of your console(s). If your console manufacturer used first grade laser trimmed resistors for the input stage, then it will work most of the time. If your console manufacturer used throw-away parts from Radio Shack, then Pin-1 lifts won't work at all. Ground loops are really make-up currents that the shield in your XLR cables are forced to endure because the power outlet wiring was incorrect to begin with. 

FYI: Here's a link to an earlier thread where I discuss grounding issues in building power outlets. http://forums.prosoundweb.com/index.php/topic,137760.0.html   

And below is a graphic from my No~Shock~Zone seminar where I demonstrate how to test for ground loops with a standard clamp meter. Note that since the actual currents are only flowing in the shield of the XLR cable(s) or snake channels, not the twisted pair(s), there's no need to split out the twisted pair from the shield to test with a clamp meter. You just place a clamp meter around the outside of the cable. Many ground loop voltage gradients will be up to a volt or two, but even 1/10 of a volt will make some powered monitors hum. This is largely due to a design issue known for the last 20+ years called the pin-1 problem where the XLR shield connection is routed through the circuit board first on its way to the chassis ground. Too much ground loop current can even cause these circuit board traces to overheat and melt.

Note that you'll likely see about 1 amp of ground loop current per volt of ground differential, so any clamp meter that will resolve down to 0.1 amps will work.  You can even clamp around a group of cables to eliminate all of them as ground loop hum candidates.

[Charlie Zureki]

Last things first: It's not that phantom voltage adds or subtracts, it's that there are 10K build-out resistors which feed the mic. If you using a simple Y-cable split, then the 10K resistor in the board with phantom on is bonded to the input pin(s) of the second board. If that board's phantom is off, then your mics are being powered from the center-tap of two 10K resistors in series, and likely only receiving 24 volts. Some mikes don't care and will run off 24 volts, but old-school mics need the full 48 volts to operate. Yes, you can turn on the phantom in both consoles, and it will work, but we don't like that much for concert work since that's one more thing to remember during setup. It won't hurt anything... we just don't like to do it due to more error possibilities.

   Yes, I agree with you in the above scenario...if we're talking about cheap consoles, but, this problem has been overcome in more modern console designs, and... the people using cheaper consoles are less likely to be using the older mics that you're referring to.

  FWIW... I have never heard of any Sound System Contractor being stumped by the low voltage divider obtained by rigging two consoles together, and in conjunction with older Mics in your example.  The people that generally use these older (expensive) mics, have higher quality consoles in their Live system.  If so... the first thing they'd do is toggle a phantom switch...

As far as ground loops, I've been doing ground loop experiments in my studio lab for the last 2 years, and dealing with them on stage for the last 40 years. A properly designed snake with transformer isolated audio splits will never pull any ground loop current, and thus can't hum. Pin-1 XLR lifts will work in many circumstances, but their success is predicated on the CMRR (Common Mode Rejection Ratio) of the balanced input stage of your console(s). If your console manufacturer used first grade laser trimmed resistors for the input stage, then it will work most of the time. If your console manufacturer used throw-away parts from Radio Shack, then Pin-1 lifts won't work at all. Ground loops are really make-up currents that the shield in your XLR cables are forced to endure because the power outlet wiring was incorrect to begin with. 

  Yes And No...  While I agree with the idea of using transformer isolation to rid ground loop hums, it is not always necessary.  The only time I've observed problems with a lack of common mode rejection was when the inputs were not balanced and the hum was caused by another high voltage inductive force. The ground loop currents that you speak of are hardly near 1 amp at 1 volt, it happens, but, I'd think its generally half or less.  Ground loops can happen even though the electrician has performed with diligence.  A ground loop is caused a difference in potential.... that is all.   This is the reason most foh power is provided from the same electrical feed as the monitor and system gear.... to try and keep this potential at a minimum.  Some of these ground loop problems can be solved with inspecting the electrical connections to ensure a tight connection.



FYI: Here's a link to an earlier thread where I discuss grounding issues in building power outlets. http://forums.prosoundweb.com/index.php/topic,137760.0.html   

And below is a graphic from my No~Shock~Zone seminar where I demonstrate how to test for ground loops with a standard clamp meter. Note that since the actual currents are only flowing in the shield of the XLR cable(s) or snake channels, not the twisted pair(s), there's no need to split out the twisted pair from the shield to test with a clamp meter. You just place a clamp meter around the outside of the cable. Many ground loop voltage gradients will be up to a volt or two, but even 1/10 of a volt will make some powered monitors hum. This is largely due to a design issue known for the last 20+ years called the pin-1 problem where the XLR shield connection is routed through the circuit board first on its way to the chassis ground. Too much ground loop current can even cause these circuit board traces to overheat and melt.

Note that you'll likely see about 1 amp of ground loop current per volt of ground differential, so any clamp meter that will resolve down to 0.1 amps will work.  You can even clamp around a group of cables to eliminate all of them as ground loop hum candidates.

   Yes, this is a good method of troubleshooting, but, the operators must understand that most of these problems are easily resolved by inspecting cables, connections and striving for a common power source and ground.

  Hammer

[Mike Sokol]

The ground loop currents that you speak of are hardly near 1 amp at 1 volt, it happens, but, I'd think its generally half or less. 

I think that many audible ground loop hums are cause by current on the order of 100 mA or so. My lab experiments using a Glo-Melt transformer allows me to create ground loops up to 3 volts at will between any two pieces of audio gear. My cable resistance measurements and empirical data gathered on the bench and in the field both concur with my initial model that ground loop currents will be approximately 1 amp per volt of ground differential between the two piece of gear.
     

Ground loops can happen even though the electrician has performed with diligence.  A ground loop is caused a difference in potential.... that is all.   This is the reason most foh power is provided from the same electrical feed as the monitor and system gear.... to try and keep this potential at a minimum.  Some of these ground loop problems can be solved with inspecting the electrical connections to ensure a tight connection.

Well, sorta kinda. I believe that most ground loop differential voltages and the resultant currents which cause hum are the result of mis-informed electricians and inspectors Double Bonding the Ground and Neutral busses in sub panels throughout the building. Any voltage drop in a hot wire will be mirrored in the neutral, which then is reflected in the safety ground.  This has been against code for at least 20 years, but electricians still do it out either of ignorance or trying to make a "better ground". I've seen this occur in brand new building installations within the last year, and many inspectors I've talked to say their local ordinances let them ignore that part of the code. Seems crazy, but that's what they tell me.

   Yes, this is a good method of troubleshooting, but, the operators must understand that most of these problems are easily resolved by inspecting cables, connections and striving for a common power source and ground.

Back in the day when I was playing in a different club each week, I would run a long extension cord from the power amp rack back to the mixing console. Of course, it didn't hum because my ground voltages were exactly the same between the power amps and the console. But as I began doing larger shows, and especially when I'm sending a mix to a broadcast truck, I can't take the chance that somebody downstream is going to contaminate my grounds and make everything hum. So I always send transformer isolated audio to remote trucks. There's just no way to keep all the grounds at the same level. In a perfect world, perhaps... but not at the gigs I do.

So if it doesn't hum, you don't need audio isolation transformers in your split. But if you can't make the hum stop, a good audio isolation transformer is your best friend.

[Charlie Zureki]

Back in the day when I was playing in a different club each week, I would run a long extension cord from the power amp rack back to the mixing console. Of course, it didn't hum because my ground voltages were exactly the same between the power amps and the console. But as I began doing larger shows, and especially when I'm sending a mix to a broadcast truck, I can't take the chance that somebody downstream is going to contaminate my grounds and make everything hum. So I always send transformer isolated audio to remote trucks. There's just no way to keep all the grounds at the same level. In a perfect world, perhaps... but not at the gigs I do.

   

   Yes,  I always use transformer isolation between  Corpy's Video World, and Video Broadcast Trucks. It's just easier.. and, I rather be doing something else than fixing a problem that could have been prevented..... (

So if it doesn't hum, you don't need audio isolation transformers in your split. But if you can't make the hum stop, a good audio isolation transformer is your best friend.

   Agreed...   

   My only reluctance to insisting on mandatory Transformer Isolation devices was that this being the Lab Lounge,  many may find the cost for purchasing a snake with Iso could find it prohibitive.

    Cheers,

    Hammer

[Greg_Cameron]

Here's the basics. When you have a low-impedance output such as a microphone or DI box, it's typically capable of driving a 600 ohm load. That's from the good old days of Ma Bell, when telephone systems were all 600 ohms. But modern mixers don't have anywhere near that low of input impedance. Most are in the 10K Ohm range. That implies that a single mic would be able to drive something like 16 mixer inputs (10,000 divided by 600 equals 16.67), but of course we don't need to drive 16 different mixers, we only need to drive two mixers in your case, FOH and Monitor.

You might want to double check that. Most mic preamp inputs on mixers are somewhere between 1.2k ohms to 2.5k ohms as opposed to the line inputs where are usually somewhere around 10k ohms to 20k ohms. So loading can be more of a factor with mics. So far I haven't personally noted an audible difference doing a 3 or even 4 way passive split, though it's a violation of the 10:1 loading guideline. But it is worth noting that mic pres have substantially less input impedance compared to line inputs.

[ProSoundWeb Community]