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: Ivan Beaver on July 30, 2016, 07:11:35 pm

Title: 8KHz or higher ground loop-true story
Post by: Ivan Beaver on July 30, 2016, 07:11:35 pm
I just saw the locked thread on 8K ground loop.  So started a new one.

Here is a TRUE story.

It started about 18yrs ago.

There were some Community 920s (not important-but just for reference) installed in a Church.

About once a year the highs would go out in them.  They never pushed the system hard.

Each time the HF drivers were fine, but the protection bubs were blown. 

We replaced them about once a year for several years.  I got tired of that, and during one replacement I decided to look a bit deeper.

The cabinets were biamped on Crown amps (again not important, but for reference).

There were 3 cabinets covering 3 different areas of seating (the Church was shaped like a T with very distinct different seating areas), but all driven from the same Community processor (analog with amplifier feedback sensors).

There was a very strong around 60KHz sine wave present on the outputs of the HF channels of the amps (each speaker had its own stereo amp).

I started "playing around" and found that when I flipped the GND switch on the 50 watt 70V amp for the lobby (TOA I think but not sure), the 60Khz went away completely.  I went back and forth several times to be sure, and it followed every time.

The speakers are still there (to be replaced this year), and they have not had a problem for the last 18 or so years, yet once a year before then they would go out.

The impedance of the drivers was so high (due to the self inductance) that they were not seeing much power so they were fine. I am not sure what part of the circuit the bulbs were in (before or after the crossover), but all they saw was the voltage and over time they blew.

Sometimes you have to look a bit deeper.
Title: Re: 8KHz or higher ground loop-true story
Post by: Stephen Swaffer on July 30, 2016, 11:13:42 pm
So you got my curiousity up so I looked at RF bands. Having lived in Michigan's UP  years ago I am aware of the low frequency sub communications antenna systems built there.  60 khz is too high for sub communications- but 8 khz is listed as a frequency that might be used by navigation.

Did you ever find the root cause of the 60 kHz or just flip the ground switch and call it good?
Title: Re: 8KHz or higher ground loop-true story
Post by: Mac Kerr on July 31, 2016, 12:58:30 am
So you got my curiousity up so I looked at RF bands. Having lived in Michigan's UP  years ago I am aware of the low frequency sub communications antenna systems built there.  60 khz is too high for sub communications- but 8 khz is listed as a frequency that might be used by navigation.

Are you sure you meant 8kHz? One of my professors in college was the director of the Navy Underwater Sound Lab, who developed that transmitter to talk to submarines, and IIRC it was an 8Hz carrier. The antenna was 75m miles by 75 miles. Long wavelengths have good penetration.

I expect the ground somehow set up an oscillation in the overly wide band amp. I once blew up half a rack full of Spectrosonics amps because of improper grounding that caused them to go into oscillation up around that frequency. No sound came out, only smoke.

Mac
Title: Re: 8KHz or higher ground loop-true story
Post by: TJ (Tom) Cornish on July 31, 2016, 07:45:01 am
So you got my curiousity up so I looked at RF bands. Having lived in Michigan's UP  years ago I am aware of the low frequency sub communications antenna systems built there.  60 khz is too high for sub communications- but 8 khz is listed as a frequency that might be used by navigation.

Did you ever find the root cause of the 60 kHz or just flip the ground switch and call it good?
60Khz is the NIST WWVB radio time station in Boulder, CO.
Title: Re: 8KHz or higher ground loop-true story
Post by: TJ (Tom) Cornish on July 31, 2016, 07:56:40 am
Are you sure you meant 8kHz? One of my professors in college was the director of the Navy Underwater Sound Lab, who developed that transmitter to talk to submarines, and IIRC it was an 8Hz carrier. The antenna was 75m miles by 75 miles. Long wavelengths have good penetration.

I expect the ground somehow set up an oscillation in the overly wide band amp. I once blew up half a rack full of Spectrosonics amps because of improper grounding that caused them to go into oscillation up around that frequency. No sound came out, only smoke.

Mac
The Navy has apparently played with a lot of different frequencies. This site lists a bunch of LF and VLF transmitters, many around 17KHz, but a couple that operate down to single digit Hz.

http://navy-radio.com/xmtr-vlf.htm

I'm not sure where this guy falls on the line of enthusiast -> obsessed, but his site is very fun, and judging by the gear in his basement that he has collected and restored, he knows his subject material.
Title: Re: 8KHz or higher ground loop-true story
Post by: Ivan Beaver on July 31, 2016, 04:00:30 pm


Did you ever find the root cause of the 60 kHz or just flip the ground switch and call it good?
When I found it, we were already late getting home-it was the last day of the year, my daughter was singing in a Christmas production that night etc etc.

I was happy it went away and we left.

It worked fine for about 2 decades after that.  The whole system (speakers and electronics) is being replaced soon. 

I do still wonder how a ground switch could cause that high of an oscillation though.  Without the scope, there is no way I could have found the solution. 

I can't say why I flipped that ground switch.  Maybe it was a quick thing to try.  I did not suspect grounding to be an issue.  All of the other wiring was done properly by my company.

I have a "knack" for targeting problems.  Something just draws me to it.
Title: Re: 8KHz or higher ground loop-true story
Post by: Mike Sokol on August 01, 2016, 06:53:02 am
I do still wonder how a ground switch could cause that high of an oscillation though.  Without the scope, there is no way I could have found the solution.

I've seen a similar high-freq oscillation caused by a DVD player with a Y cable summing the the stereo outputs to mono. I don't remember the brand, but it created an ultrasonic "whine" which took out the horn drivers at a church I was working on.

Also, many pieces of A-V gear have internal ground planes which are connected to the chassis ground with some sort of capacitor-filter circuit. For instance, old-school Sony video camera XLR inputs will behave badly if you accidentally plug in an XLR cable with a "grounded shell". That is, pin-1 to shell jumper. I've heard this create a nasty high-freq "hash" from them, but I could imagine there being ultrasonic content as well. The problem is that circuit ground is sometimes not chassis ground. I believe this is what also caused early MacBook laptop computers to inject all kinds of power supply hash into a sound system if you just connected the 1/8" headphone output into 1/4" inputs on the mixing console. New Mac's have corrected that after much complaining by pro-sound guys (I think I was the first to bring this problem formally to Apple).
Title: Re: 8KHz or higher ground loop-true story
Post by: Stephen Kirby on August 01, 2016, 07:05:55 pm
I believe this is what also caused early MacBook laptop computers to inject all kinds of power supply hash into a sound system if you just connected the 1/8" headphone output into 1/4" inputs on the mixing console. New Mac's have corrected that after much complaining by pro-sound guys (I think I was the first to bring this problem formally to Apple).
That explains a dance studio I did recitals for.  The Mac laptop had this odd swirling static in the PA.  Everything on the same ground point.  I ended up isolating it with a Radial passive stereo DI which I now use regardless.  I thought it might be related to the Mac not having a real safety ground.  I can still feel leakage from my Air on business trips to places with 220VAC.  Didn't get a chance to glom onto a meter on my last trip to China to measure it against their 3rd pin.  But I will on a future trip.  It does seem to come and go randomly.
Title: Re: 8KHz or higher ground loop-true story
Post by: Mike Sokol on August 01, 2016, 08:00:36 pm
The Mac laptop had this odd swirling static in the PA.  Everything on the same ground point.  I ended up isolating it with a Radial passive stereo DI which I now use regardless.
Back when Mac Titanium G4's and OSX came out, I had Apple as a sponsor of my surround-sound demonstrations. In fact, I taught classes to the Apple Core Audio and Final Cut Pro groups about surround-sound editing and processing. I discovered at that point that all Titanium G4 computers would create this horrible "hash" when their 1/8" headphone jack was directly connected to TS unbalanced inputs on a mixing console. This noise was not there when listening to the jack with headphones. The solution was some sort of stereo DI which provided a balanced connection to the mixing console, and isolated the laptop ground from the console ground. In fact, if you flipped the ground lift switch on the DI to "ground" the noise would come back. I discussed this with Apple engineers a few times, and even wrote an article about it which Apple posted on their tech support site as a fix. Eventually, later model Mac laptops seemed to have been corrected. But I always use a Whirlwind pcDI or podDI for my laptop connections.   
Title: Re: 8KHz or higher ground loop-true story
Post by: Mac Kerr on August 01, 2016, 08:16:49 pm
But I always use a Whirlwind pcDI or podDI for my laptop connections.

For less money, and at least as good performance you can use a Peavey USB-P interface (https://www.amazon.com/Peavey-3001370-USB-Audio-Interface/dp/B004A4PSEU). Balanced transformer isolated line level out, no ground noise problems.

Mac
Title: Re: 8KHz or higher ground loop-true story
Post by: Mike Sokol on August 01, 2016, 08:52:12 pm
For less money, and at least as good performance you can use a Peavey USB-P interface (https://www.amazon.com/Peavey-3001370-USB-Audio-Interface/dp/B004A4PSEU). Balanced transformer isolated line level out, no ground noise problems.

Mac

Yup, that should work as well. The trick is to get the chassis ground of the laptop completely isolated from the chassis ground of the mixer. And this isn't just an Apple problem. I had a number of my church sound seminar attendees tell me they had sent their Dell laptops back multiple times when they first discovered the noise in their sound systems.

BTW: I saw the same problem going the other way. That is, when they took an aux send from the console and fed it to the soundcard on a computer to record it, there was a lot of hum and hash. I added an Ebtech Hum Eliminator on that audio path, and the hum/hash in the recording went away. https://www.amazon.com/EBTECH-HE-2-Ebtech-Hum-Eliminator/dp/B0002E4YH4/ref=sr_1_sc_1?s=musical-instruments&ie=UTF8&qid=1470098884&sr=1-1-spell&keywords=Ebtech+Hum+Eominator

When dealing with computer audio it's best to use transformer isolation if possible.