You are here

Troubleshooting Noise/Interference

All about the tools and techniques involved in capturing sound, in the studio or on location.

Re: Troubleshooting Noise/Interference

Postby Hugh Robjohns » Tue Dec 10, 2019 7:37 pm

cyrano.mac wrote:I've seen that exact scenario in practice.

With respect, that lethal scenario, and the grounding option I suggested, are very different indeed and completely unrelated.

It's not unusual to have a bad ground.

Obviously, a high ground impedance can occur, but the likelihood depends on where you live (ie. the local electrical safety regulations) and the type of mains supply configuration coming into the building. Its extremely unusual to have a 'bad ground' in the UK, for example. YMMV.

I don't know what's inside these "magical" ground plugs. I hope there's an RC circuit to prevent the worst.

There's nothing magical. It's a piece of wire. An RC circuit would defeat its object. It's not a ground-lift, it's a ground-tie, replicating the hard-ground connection provided in a class-1 device.

Unfortunately, it was the one receptacle that was connected directly to mains, next to the fuse box.

That wouldn't be legal in the UK, and I'm rather surprised it's allowed where you are. :o

It's purpose was to provide light for the electrician when the fuse box shut off.

Have they never heard of torches? :headbang:

Somehow, the thing that was powered by this one connection had bad insulation.

Oh dear... substandard routine maintenance? Any needless loss of life is tragic, and I'm very sorry for your loss. But this story involves multiple catastrophic failures combined with stupidity -- albeit none the fault of your late friend. It's an inherent fact of life that it's always impossible to protect against multiple simultaneous failures of critical parts of the system. That's why we try to engineer systems so that they fail-safe with a single failure.

You should never, ever, connect audio ground to mains ground. Period.

Sorry. That's complete nonsense. There are good scientific reasons why the audio reference ground ~should~ be tied to the mains safety ground in most mains powered systems, and thats been the practical implementation in every professional studio installation I've ever been involved with.

However, multiple connections between the audio reference hround and the mains safety earth can cause problems in some systems, obviously (ground loops), just as an absence of hard grounding can cause problems in others -- especially in systems with unbalanced connections. Having said all that, a connection to mains ground is not always an essential if the system is designed properly -- as in battery powered equipment, for example, or in vehicles, or planes...

People still die every year cause some bozo decides he's smart enough to wire the plug. I see those kinds of errors at least once a week when trouble shooting.

Sad but true... the only cure is education...
User avatar
Hugh Robjohns
Moderator
Posts: 26939
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Troubleshooting Noise/Interference

Postby cyrano.mac » Wed Dec 11, 2019 3:52 am

Sorry, Hugh, it's not extremely unlikely to have a bad ground. Wherever you are. Take old buildings, for instance. I've seen quite a few of those in the UK, with real bad electricity setups. But even if it was rare, it would be very sad to just accept the few casualties.

Electrocution is the fifth leading cause of occupational injury death in the United States:
https://injuryprevention.bmj.com/content/8/4/306

Unfortunately, the UK doesn't even bother to keep stats about electrocution. The numbers in the EU report are extrapolated from other countries' data. Go figure!

The ETCI reported 2,5 million electrocutions per year in the UK of which 200.000 needed to be hospitalised. I never saw any number on deaths, unfortunately. Again, extrapolated data, based on sampling.

Our local regulations are among the most severe, but don't protect from sheer stupidity. About half of a dozen yearly deaths could have been avoided because they were caused by safety equipment not working because of hacks like this.

Failing insulation is very rare, but it is the number one problem proper grounding should protect against.

By bridging audio and mains ground you are creating the environment in which these kind of accidents occur. You are exposing double insulated equipment to grounded equipment through the shield connection, potentially bypassing other safety measures like RC connected ground schemes. You're also potentially making the GFCI useless.
cyrano.mac
Regular
Posts: 102
Joined: Sun Jun 02, 2019 5:00 pm

Re: Troubleshooting Noise/Interference

Postby blinddrew » Wed Dec 11, 2019 9:46 am

The detail is available. Start with hse.gov.uk.
4 industrial deaths in the uk last year related to electricity.
I'm on the train now so can't get the domestic figures as the signal is as unreliable as the train...
But they are tracked and available.
User avatar
blinddrew
Jedi Poster
Posts: 10114
Joined: Sun Jul 05, 2015 12:00 am
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: Troubleshooting Noise/Interference

Postby Hugh Robjohns » Wed Dec 11, 2019 11:27 am

cyrano.mac wrote:Sorry, Hugh, it's not extremely unlikely to have a bad ground.

I can't agree. In well over thirty years of working with mains installations in the UK I've only come across a handful -- certainly less than a dozen -- of occasions where I've discovered problems with mains grounding in building wiring. I'd count that as pretty rare. Obviously YMMV... (Sadly, the same isn't true in some of the far-flung corners of the world where I've worked. Some Malaysian mains wiring still sends shivers down my spine... :shock: )

However, when I'm working out and about I do always carry a tester and I always test sockets (and their ground impedance) before I use them because any safety earth problem is potentially a very serious one.

Unfortunately, the UK doesn't even bother to keep stats about electrocution.

Er... completely untrue. The Health and Safety Executive collect and publish records of all work-related injuries and fatalities, and their causes.

https://www.hse.gov.uk/statistics/causinj/kinds-of-accident.pdf
This 18/19 report states that 4% of 147 fatalities were caused by contact with electricity.

https://www.hse.gov.uk/statistics/pdf/fatalinjuries.pdf
And this one states there were 9 recorded fatalities due to electricity in UK workplaces in 2019.

The UK's Office of National Statistics also publishes data on all manner of things, including recorded deaths related to electricity, with data breakdowns on location (ie. home, residential places, industry, farms and so on). The figures for fatal electrocutions in England from 2001 to 2016 provide a total of 332... in 15 years.

https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/adhocs/007964deathsfromexposuretoelectriccurrentbysexandagegroupengland2001to2016

An older (1998) generic HSE report states: "Each year about 1000 accidents at work involving electric shock or burns are reported to the Health and Safety Executive (HSE). Around 30 of these are fatal. Most of these fatalities arise from contact with overhead or underground power cables...."

http://www.esldatasheets.com/Documents/HSE/ESaY_HSEPub_indg231.pdf

Although these figures are only for England, it's clear that fatal electrocutions is in the low tens, each year, at worst -- and that's tragic, obviously, especially as they are avoidable. But I find it hard to believe there were millions more in Scotland and Wales... so I don't know where your extraordinary figure of 2.8 million came from. It's simply not credible!

Our local regulations are among the most severe, but don't protect from sheer stupidity.

None ever could... as I said, education is the only way forward. Which is why I find it so sad that so many young people consider eschew science and technology these days.

By bridging audio and mains ground you are creating the environment in which these kind of accidents occur.

Er… no. That's utter nonsense... There certainly are situations where ground bonding needs careful consideration, most notably when working with non-standard mains supplies or equipment powered from isolating transformers... but these are not relevant in the hum-seeking context being discussed above.

You are exposing double insulated equipment to grounded equipment through the shield connection...

And what exactly do you think happens when you plug your double-insulated keyboard, or laptop, say, into a Class 1 stage-amp, or a class-1 mixing console, or a class-1 active speaker… ?

Class-1 devices almost always have the audio reference ground tied directly to the star earthing point along with the chassis and mains safety earth. Indeed, this configuration is strongly recommended as the 'best practice' for equipment grounding by the AES, no less.

Check out the diagram on page 5 of the AES48 documention, for example: http://www.aes.org/standards/comments/drafts/aes48-xxxx-190121-cfc.pdf (Sorry this is a draft version... the final document isn't publically accessible on the AES website, but the diagrams and recommendations are the same).

And the only reason some class-1 devices have a 'ground lift' switch with your beloved RC network is to avoid ground-loops when there are multiple interconnected class-1 devices.

All the 'grounding plug' is doing is replacing that tie to a real ground (that would normally be provided by a Class-1 device) in systems which comprise all double-insulated equipment and thus lack a solid ground reference.

I can assure you that I take mains safety extremely seriously -- both personally and in my role as Technical Editor of SOS -- and I certainly wouldn't suggest any practice that represented any potential risk. Bonding the audio reference ground of a double-insulated device to mains earth carries no risk whatsoever with normal working equipment, and nor does increase risks or affect the functionality of any conventional safety devices under fault conditions.
User avatar
Hugh Robjohns
Moderator
Posts: 26939
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Troubleshooting Noise/Interference

Postby ef37a » Wed Dec 11, 2019 12:27 pm

Believe Hugh people!

Ste-c. Now that you know of at least one use for a modest digital test meter (others are...Phantom power check, battery check. You can even signal trace at line level with one) DO get one.

You could also throw a Martindale socket tester into the mix, whole lot should set you back no more than 30quid.

Dave.
ef37a
Jedi Poster
Posts: 11483
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Troubleshooting Noise/Interference

Postby Hugh Robjohns » Wed Dec 11, 2019 1:01 pm

ef37a wrote:You could also throw a Martindale socket tester into the mix, whole lot should set you back no more than 30quid.


The standard tester is a life saver, for sure, but I prefer this slightly more expensive version as it gives a really useful indication of the actual safety ground impedance, rather than just whether or not there's a connection.

https://cpc.farnell.com/martindale-electric/ez150/earth-loop-socket-tester-no-trip/dp/IN04254?st=mains%20socket%20tester

Image
User avatar
Hugh Robjohns
Moderator
Posts: 26939
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Troubleshooting Noise/Interference

Postby cyrano.mac » Wed Dec 11, 2019 10:36 pm

Hugh Robjohns wrote:I can assure you that I take mains safety extremely seriously -- both personally and in my role as Technical Editor of SOS -- and I certainly wouldn't suggest any practice that represented any potential risk. Bonding the audio reference ground of a double-insulated device to mains earth carries no risk whatsoever with normal working equipment, and nor does increase risks or affect the functionality of any conventional safety devices under fault conditions.

Well, then the question is simple, isn't it?

Is there a direct connection in that ground plug device?
cyrano.mac
Regular
Posts: 102
Joined: Sun Jun 02, 2019 5:00 pm

Re: Troubleshooting Noise/Interference

Postby Hugh Robjohns » Wed Dec 11, 2019 10:43 pm

Already answered.

A direct connection to earth in a class II device for functional purposes -- as in the case being discussed earlier -- is permitted in IEC 61140. The relevant paragraph is here:

7.3.2.2 Class II EQUIPMENT may be provided with means for connection to EARTH for functional (as distinct from protective) purposes only where such a need is recognized in the relevant IECstandard. Such means shall be insulated from live parts by DOUBLE or REINFORCED INSULATION.

http://80.229.80.165/tech_info/standards_conformance/standard_pdfs/BSEN61140_protection%20against%20shock.pdf
User avatar
Hugh Robjohns
Moderator
Posts: 26939
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Previous