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: Stephen Swaffer on February 15, 2014, 11:09:41 PM

Title: Grounding and RFI
Post by: Stephen Swaffer on February 15, 2014, 11:09:41 PM
It has been discussed that an "earth ground" is primarily for lightning protection.  The Egc (equipment grounding conductor) bonded to the neutral at one place exists to prevent potential differential on metal equipment and to trip overcurrent devices in case of  fault.

Since radio transmission depends on the earth for transmission, how important is a good earth ground to preventing RFI in audio systems?
Title: Re: Grounding and RFI
Post by: Mike Sokol on February 15, 2014, 11:29:11 PM
It has been discussed that an "earth ground" is primarily for lightning protection.  The Egc (equipment grounding conductor) bonded to the neutral at one place exists to prevent potential differential on metal equipment and to trip overcurrent devices in case of  fault.

Since radio transmission depends on the earth for transmission, how important is a good earth ground to preventing RFI in audio systems?

I do know that any type of corrosion in the earth ground or shield of a sound system will act like a detector and can demodulate AM radio stations into program audio. Also note that AM radio station quarter-wavelengths can be on the order of 125 to 175 ft, which works out to a 100-ft snake plus a 50-ft subsnake plus a 25-ft mic cable. But while both these scenarios will cause RFI, I don't think that either really answer your RFI grounding question.

I'm guessing that RFI mitigation is more about adequate shielding than earth grounding, but I could be full of it. Time to put on the thinking cap. What do the rest of you guys know about this?
Title: Re: Grounding and RFI
Post by: Frank Koenig on February 16, 2014, 01:47:17 AM
Since radio transmission depends on the earth for transmission

Radio transmission does not, in general, depend on the earth. Some antennas, however, use the earth as a counterpoise, such as traditional AM broadcast antennas operating in the .55 to 1.6 MHz band, where the intended propagation is by ground wave. They are often located in areas with high ground conductivity, such as salt marshes. Radio transmission to and from spacecraft, aircraft, and wireless mics for that matter, works just fine, no earth required.

Protection from radio frequency interference (RFI) relies mostly on shielding and filters on connections that penetrate the shield.

--Frank
Title: Re: Grounding and RFI
Post by: Kevin Graf on February 16, 2014, 10:33:27 AM
a] Frank's explanation is most excellent.

b] Jim Brown, being both AES past committee chair on EMI/RFI and a Ham radio operator, often covers connections to dirt in his many papers.

http://www.audiosystemsgroup.com/publish.htm

c] This paper has a section on connections to dirt:
"Testing for Radio-Frequency Common Impedance Coupling (the "Pin 1 Problem") in Microphones and Other Audio Equipment"
The author has shown that a primary cause of VHF and UHF interference to professional condenser microphones is inadequate termination within the microphone of the shield of the microphone's output wiring, a fault commonly known as the pin 1 problem. Tests using only audio frequency test signals generally fail to expose susceptibility to radio frequency (RF) interference. Simple RF tests for pin 1 problems in microphones and other audio equipment are described that correlate well with EMI observed in the field. Preprint Number: 5897 Convention: 115 (September 2003) Author: Jim Brown 
http://www.audiosystemsgroup.com/AESPaperNYPin1-ASGWeb.pdf
Title: Re: Grounding and RFI
Post by: Kevin Graf on February 16, 2014, 10:47:30 AM
It has been discussed that an "earth ground" is primarily for lightning protection.
Yes the GEC (Grounding Electrode Conductor) system is for lightning protection and power company (POCO) high-voltage failures.

Quote
The EGC (Equipment Grounding Conductor) bonded to the neutral at one place exists to prevent potential differential on metal equipment and to trip overcurrent devices in case of  fault.
[...................]
Most definitely!
However in cases like Festivals and swimming pools the connection to dirt is also part of the  EGC (Equipment Grounding Conductor) system.
 
Title: Re: Grounding and RFI
Post by: Mike Sokol on February 16, 2014, 12:25:51 PM
Protection from radio frequency interference (RFI) relies mostly on shielding and filters on connections that penetrate the shield.

--Frank

I've used the Neutrik EMC XLR connector which includes an internal filter to keep RFI out of your mixing console. These are great for keeping AM radio out of old Mackie mixing consoles. BTW: With all due respect to our southern neighbors, this far-off sounding radio noise in a mix is often called "Mexican Radio" by my students. That nickname probably comes from some of the huge radio stations just south of the border that leak into the US.

See http://www.neutrik.com/en/xlr/xlr-cable-connectors/emc-series/ (http://www.neutrik.com/en/xlr/xlr-cable-connectors/emc-series/) for the link.

Title: Re: Grounding and RFI
Post by: John Roberts {JR} on February 16, 2014, 12:45:26 PM
I've used the Neutrik EMC XLR connector which includes an internal filter to keep RFI out of your mixing console. These are great for keeping AM radio out of old Mackie mixing consoles. BTW: With all due respect to our southern neighbors, this far-off sounding radio noise in a mix is often called "Mexican Radio" by my students. That nickname probably comes from some of the huge radio stations just south of the border that leak into the US.

See http://www.neutrik.com/en/xlr/xlr-cable-connectors/emc-series/ (http://www.neutrik.com/en/xlr/xlr-cable-connectors/emc-series/) for the link.

Yes it is interesting when a product with an arguable design flaw (input stage rectification) becomes so popular that a connector company designs and sells a band-aid for it.

Appreciate all the good answers already posted.

In the context of RF is may be worth mentioning that a low impedance ground path for DC or LF, may not be a low impedance at RF frequency. RF detection inside products is mostly a relative thing, the RF on the hot vs the RF on the ground. Truly common mode RF should be ignored.

JR
Title: Re: Grounding and RFI
Post by: Stephen Swaffer on February 16, 2014, 03:42:22 PM
Part of the reason I asked this is that I have fought RFI on our church choir mics for years. 

For a while it was quite noticeable-we had a Mackie 32-8 mixer.  Was going to build low pass filters as recomended by Mackie (not sure why they just didn't incorporate them themselves?) but then the Mackie became to unreliable so we replaced it with an A & H GL3200.  I expected RFI to go away, and it did improve a LOT, but still have trouble on one choir mic-but maybe Mike answered that.  If I recall correctly, the choir mic snake is 100 feet long, plus the mic in question has a 25 mic cord and the Shure MX212s come with a 30 foot cord-or 155 feet.

JR mentions a DC/LF ground may not necessarily be a good RF ground.  An RF ground needs to be as short as possible, but running a direct, shorter ground to the input box of the snake would create a ground loop-and violate the principle of a "star" topology for grounding?
Title: Re: Grounding and RFI
Post by: John Roberts {JR} on February 16, 2014, 04:52:24 PM
Part of the reason I asked this is that I have fought RFI on our church choir mics for years. 

For a while it was quite noticeable-we had a Mackie 32-8 mixer.  Was going to build low pass filters as recomended by Mackie (not sure why they just didn't incorporate them themselves?)
Mainly because they did not realize that the design was marginal until they got a bunch of them out into the field.
Quote
but then the Mackie became to unreliable so we replaced it with an A & H GL3200.  I expected RFI to go away, and it did improve a LOT, but still have trouble on one choir mic-but maybe Mike answered that.  If I recall correctly, the choir mic snake is 100 feet long, plus the mic in question has a 25 mic cord and the Shure MX212s come with a 30 foot cord-or 155 feet.

JR mentions a DC/LF ground may not necessarily be a good RF ground.  An RF ground needs to be as short as possible, but running a direct, shorter ground to the input box of the snake would create a ground loop-and violate the principle of a "star" topology for grounding?

 ;D ;D  I love it when I hear design theory regurgitated like that.

It's a little more complicated than that and not sure we need to get into circuit design.

My point is that for RF frequencies an apparent ground may not be very low Z. They make antennas out of wire so RF voltage varies over the length of the antenna wire.

Keeping RF out of audio is mostly about design inside the product input.

Grounds are about human safety. For properly designed professional gear interfaces not a signal quality issue. 

JR
Title: Re: Grounding and RFI
Post by: Stephen Swaffer on February 16, 2014, 11:52:14 PM
So where would you look to deal with the problem?  Admittedly the RF is weak, (but if I want to listen to the local news I don't need a radio)- grounds/shielding all appear to be fine.  Known issues with the A & H design?  The Shure MX212?  Maybe just going with the aforementioned connector would be a good starting point?

Not trying to get into circuit design, just trying to understand what I can do to deal with the problem without redesigning the input circuit on the mixer-and still follow good install/field practices.
Title: Re: Grounding and RFI
Post by: Tim Padrick on February 17, 2014, 12:53:30 AM
Make sure that Pin 1 is connected to the XLR case ONLY where two cables connect together (e.g. mic cable to snake), never where a cable connects to a piece of gear.  Check all cables and snakes on both ends.  And make sure that none of the cases that connect to Pin1 touch anything metal (e.g. truss).
Title: Re: Grounding and RFI
Post by: Jordan Wolf on February 17, 2014, 09:00:12 AM
Make sure that Pin 1 is connected to the XLR case ONLY where two cables connect together (e.g. mic cable to snake), never where a cable connects to a piece of gear.  Check all cables and snakes on both ends.  And make sure that none of the cases that connect to Pin1 touch anything metal (e.g. truss).
+1

If Pin 1 is connected to the shell, it's usually only on the male end.  However, if your cables haven't been checked out in a while, you might as well look at both ends to ensure everything is proper.
Title: Re: Grounding and RFI
Post by: Kevin Graf on February 17, 2014, 11:31:20 AM
As I wrote in reply #3: (and in many other threads)
Jim Brown, being both AES past committee chair on EMI/RFI and a Ham radio operator, often covers interference in his many papers.
http://www.audiosystemsgroup.com/publish.htm

*********************************************

This paper may be the best place to start, although there are several others on interference.
Don't let the title fool you, this started as an audio system paper, the a lot of Ham information was add:

"A Ham's Guide to RFI, Ferrites, Baluns, and Audio Interfacing"

by Jim Brown K9YC
Audio Systems Group, Inc.
http://audiosystemsgroup.com

The basis of this tutorial is a combination of my engineering education, 55 years in ham radio, my
work as vice-chair of the AES Standards Committee working group on EMC, and extensive research
on RFI in the pro audio world where I’ve made my living.

http://www.audiosystemsgroup.com/RFI-Ham.pdf
Title: Re: Grounding and RFI
Post by: Frank DeWitt on February 17, 2014, 08:11:54 PM
Part of the reason I asked this is that I have fought RFI on our church choir mics for years. 


Here is a link to a thread on the topic.  That link has links to other threads.  Bottom line, ferrite beads.
http://forums.prosoundweb.com/index.php/topic,142374.msg1319364.html#msg1319364
Title: Re: Grounding and RFI
Post by: Kevin Graf on February 17, 2014, 09:20:52 PM
There is a lot more to ferrites than just popping one on the cable. See the links in my reply #12.
Jim Brown's Ham group sometimes makes bulk purchases of up to 1000 chokes.
Title: Re: Grounding and RFI
Post by: Lyle Williams on February 18, 2014, 08:42:04 AM
Series inductance (eg small beads on each lead, not the clamp over the cable jacket style) to keep the RF out and shunt capacitance to give it somewhere else to go...
Title: Re: Grounding and RFI
Post by: Frank Koenig on February 18, 2014, 02:31:47 PM
Series inductance (eg small beads on each lead, not the clamp over the cable jacket style) to keep the RF out and shunt capacitance to give it somewhere else to go...

The beauty of ferrite beads is that they're so lossy that, as circuit elements, they act more like frequency-dependent resistors than inductors, so that the Q (ratio of total energy stored to energy dissipated per cycle) of the lowpass stays low, which is generally good.

Without getting deep into the art of RFI filter design, some considerations include selecting a ferrite material appropriate for the frequency range, and selecting a shunt capacitor big enough to attenuate low RF without dragging down the signal while having low enough series inductance to remain effective for high RF. The physical layout with respect to the shield must be such that the RF doesn't leak around the whole shebang, and it has to fit.

--Frank
Title: Re: Grounding and RFI
Post by: Mike Sokol on February 18, 2014, 07:05:39 PM
Just to throw another bone in the soup, it's also possible for RFI to sneak into speaker wires. You wouldn't think there would be enough RF energy to hear on a speaker, and that's true. However, if you have an amplifier which suffers from TIM (Transient Inter-Modulation Distortion) then any RF that creeps in the backdoor through the speaker wires can get into the negative feedback loop of the amplifier output stage. TIM was caused mainly by poor amplifier designs with too low of a slew rate in the negative feedback loop. Audio transients would overload the negative feedback loop's slew rate, which caused a splattering of distortion only during these transients, but which could not be detected by steady state tone testing. I figured this out about 30-years ago when my Dynaco Power amps of the day would pick-up CB radios in the street and "broadcast" them into my speakers during those tender love songs. The fix was adding small disc capacitors on the speaker wires where they connected to the amp outputs, effectively shorting out the RF before it got into the amp where it overloaded the negative feedback circuits. 

RFI will make you crazy since it can sneak into so many places and cause a lot of trouble.
Title: Re: Grounding and RFI
Post by: John Roberts {JR} on February 18, 2014, 08:49:56 PM
Just to throw another bone in the soup, it's also possible for RFI to sneak into speaker wires. You wouldn't think there would be enough RF energy to hear on a speaker, and that's true. However, if you have an amplifier which suffers from TIM (Transient Inter-Modulation Distortion) then any RF that creeps in the backdoor through the speaker wires can get into the negative feedback loop of the amplifier output stage. TIM was caused mainly by poor amplifier designs with too low of a slew rate in the negative feedback loop. Audio transients would overload the negative feedback loop's slew rate, which caused a splattering of distortion only during these transients, but which could not be detected by steady state tone testing. I figured this out about 30-years ago when my Dynaco Power amps of the day would pick-up CB radios in the street and "broadcast" them into my speakers during those tender love songs. The fix was adding small disc capacitors on the speaker wires where they connected to the amp outputs, effectively shorting out the RF before it got into the amp where it overloaded the negative feedback circuits. 

RFI will make you crazy since it can sneak into so many places and cause a lot of trouble.

Not to quibble with teacher but TIM is an overly esoteric explanation for a simple and fairly well understood mechanism by many (perhaps not the young David Hafler of Dynaco) audio engineers. It's about the slew rate of the active amplifier (not the negative feedback loop which is passive and doesn't have a slew rate.) When the active amplifier is not fast enough to keep up with the too-fast input signal the output slews as fast as it can but still distorts the output. When that too-fast input is audio modulated RF, that distortion decodes (aka rectifies ) the RF into radio station audio.

In the case of power amplifiers that often have small inductors in series with their final output stage to isolate capacitive loads (for stability), RF can snake backwards into the should-be low Z output that is now high Z at radio frequency and back down the (passive) feedback network to the amplifier input. The amplifier does not know or care how the too fast signal got into it's input, but tries to keep up and can't.

This is not very common, but does happen. Sometimes the speaker run is just the magic length for some strong local radio station. Sometimes the amp is not very fast. 

Note: It is possible to design an amp that harmlessly rolls off too-fast input signals, but not every amp designer got that memo, either.

JR
Title: Re: Grounding and RFI
Post by: Mike Sokol on February 18, 2014, 09:01:31 PM
Not to quibble with teacher...

JR, I'm not really a teacher... just a student trying to figure it all out. Lot's of fun, isn't it?  8)
Title: Re: Grounding and RFI
Post by: Frank Koenig on February 19, 2014, 11:58:27 AM
TIM is an overly esoteric explanation for a simple and fairly well understood mechanism by many

This is exactly right (of course!). The whole TIM brouhaha was discovered (invented?) in the late '70s when folks noticed that some amplifiers -- not just power amps -- had significant IM when at least one of the input signals was near the upper end of the pass-band.

AES even came up with a standard test involving the sum of a sine wave and a square wave and a certain accounting of the resulting intermodulation products. I set this up in a lab at Stanford where one of the profs took pity on me and let me use a very nice (analog) HP spectrum analyzer to test a bunch of amps, both large and small. I recall that the 5534 op amps, which were the hot new thing, did very well.

The Crown DC300A was considered by many to be the villain at the time (I did not get to test one) and I believe Yamaha came up with their "Natural Sound" line to differentiate themselves in the marketplace.

--Frank
Title: Re: Grounding and RFI
Post by: John Roberts {JR} on February 19, 2014, 12:26:11 PM
This is exactly right (of course!). The whole TIM brouhaha was discovered (invented?) in the late '70s when folks noticed that some amplifiers -- not just power amps -- had significant IM when at least one of the input signals was near the upper end of the pass-band.

AES even came up with a standard test involving the sum of a sine wave and a square wave and a certain accounting of the resulting intermodulation products. I set this up in a lab at Stanford where one of the profs took pity on me and let me use a very nice (analog) HP spectrum analyzer to test a bunch of amps, both large and small. I recall that the 5534 op amps, which were the hot new thing, did very well.

The Crown DC300A was considered by many to be the villain at the time (I did not get to test one) and I believe Yamaha came up with their "Natural Sound" line to differentiate themselves in the marketplace.

--Frank
[esoteric veer]
For some reason audio has always attracted a lot of quasi science..... IIRC Matti Otolla was the primary TIM pimp (there were a couple other superfluous new distortions floated around that time). There was a funny comment in an interview with an old graybeard tube designer (Baxandall?) who was involved with the HF radar signal processing design during WWII remarking how this (speed related distortion) was well known to those skilled in the art back in the '40s.

The bottom line is that early solid state power transistors were as slow as dirt (2n3055 anyone?), so the slewing speed of early SS audio amplifiers were marginal. The more power the more voltage swing and more apparent the speed issue.  Marshal Leach an EE prof at Ga Tech published a short but seminal AES paper about how to bandwidth or rise-rime limit an active NF amplifier such that it could not be slew limited for any valid (unclipped) input voltage. Too fast signals are just harmlessly rolled off.  As the semiconductor technology evolved the devices became faster, and that created another group of snake oil inclined marketing types to over sell slew rate (not unlike over selling damping factor and other poorly understood specs). Perhaps ironic, a well designed amplifier can not be slew rate limited, and some amplifier makers would have to defeat that part of the circuitry to measure a big slew rate number for the sales manager to publish on the data sheet. The customer is always right and they wanted faster slew rates even though they did not have a clue about how much they actually needed.   :o

Enough is good, more than enough, is just more, not necessarily better.

There was a sub genre of self-appointed audiophile experts ginning up these faux design problems. like their livelihood depended on it, because it did.  8)

JR   [/esoteric veer]
Title: Re: Grounding and RFI
Post by: Stephen Swaffer on February 19, 2014, 12:36:56 PM
JR, I'm not really a teacher... just a student trying to figure it all out. Lot's of fun, isn't it?  8)

Can't speak for Mike-but I know I was a lot smarter 25 years ago....seems the more  learn the less I know.
Title: Re: Grounding and RFI
Post by: Mike Sokol on February 19, 2014, 12:55:27 PM
The bottom line is that early solid state power transistors were as slow as dirt (2n3055 anyone?), so the slewing speed of early SS audio amplifiers were marginal.

But back to my original point. Under the right set of circumstances (poorly designed amps, cables just the right length, etc...) it's possible to have the speaker cables act like an antenna feeding RFI into the amplifier's circuitry with is then detected (rectified) and output as noise, or even DJ's talking. Of course, this type of interference also happens on the front end with poorly shielded/bonded snakes and XLR cables with mic preamps that aren't playing nicely.

RFI is a cruel master....
Title: Re: Grounding and RFI
Post by: Stephen Swaffer on February 25, 2014, 10:08:53 PM
I have experienced the most difficulty with RFI on phantom powered condenser mics. Is this typical?  What causes this?
Title: Re: Grounding and RFI
Post by: John Roberts {JR} on February 25, 2014, 10:23:25 PM
I have experienced the most difficulty with RFI on phantom powered condenser mics. Is this typical?  What causes this?

I do not like to make sweeping generalities, but there are many different phantom powered mics, with sundry different circuitry inside the mic. It's not hard to imagine something as simple as an emitter follower rectifying RF if it is large enough in the cable (>500mV). 

Passive microphones do not have any active semiconductor junctions so will pretty much ignore RF. So will well designed mics.  8)

JR
Title: Re: Grounding and RFI
Post by: Steve M Smith on February 26, 2014, 02:48:23 AM
and output as noise, or even DJ's talking

Is there a difference?!!!


Steve.