[LAU] Behringer and Linux

[Patrick Shirkey]

On Tue, April 7, 2015 12:54 am, Chris Caudle wrote:
>> "ground loop" is almost certainly the problem.
>
> A problem with "ground loop" is symptomatic of a fundamental
> misunderstanding of Ohm's law and is a problem of poor layout in the
> mixer.
>

To clarify, I didn't say switching power supplies cause ground loops, I
DID say they will run at different switching frequencies.  Those
frequencies have fair odds of being close and the (quite common) combined
leakage will beat at audible frequencies.

The second hypothesis talks about a "ground loop", but not through the
wall outlet power grounds, but analog and digital ground references in TWO
devices being different.  A USB metal connection will make their digital
"ground" reference the same, but the two analog ones now probably will not
be.  Disconnect the USB connection and analog "ground" goes back to where
it is designed to be.


> Fons has explained reasonably well for a short email where the previous
> email goes wrong, but there really isn't time to go into full detail of
> why the problem should not exist at all.
>
> The shortest version I can come up with is that any two pieces of
> equipment will have some voltage difference between their cases.
> Connecting the equipment together allows current to flow between the
> equipment along the shields of connecting cables, and across whatever
> conductors form a path for that current to get back to the source (form a
> complete circuit).  That path could include chassis, cables, printed
> circuit board traces, anything conductive.  Any current flowing in a
> conductor will have an associated voltage drop described by Ohm's law
> (voltage is the product of current and impedance), and the job of the
> design engineer is to understand and control where that current will be
> flowing, and make sure that no audio circuitry uses one of those
> conductive paths as a reference conductor for the audio.
>
>> Both of these, are not really Behringer problems.
>
> They most certainly are.
>
> The design engineers at Behringer failed that layout task for at least the
> headphone amplifier in the mixer Fons described.

You haven't proven that is the case so unless you have absolute proof you
shouldn't be claiming that they have absolutely failed.

Fons hasn't definitively proven his rebuttal either.

Either way I'm sure the issue could be resolved by donating the device for
analysis at LAC.


> It is two decades too late to make excuses for incompetence in dealing
> with that particular problem.
>

It does seem that some people just hate Behringer, for what reason I don't
know.  Yes, the early Behringer Eurorack and Eurodesk mixers had power
supply problems but that was pre-Xenyx and at least ten years ago. This
was also before Behringer controlled all its own manufacturing at
"Behringer City".

Failures of Behringer gear are exceedingly rare, and then only under
severe abuse.  The abuse it does take and keeps operating is noteworthy.

Behringer deserves some great credits.  They are good designs (designed
for clever automated manufacturing and intrinsic reliability) at amazingly
low prices.

The people on this mailing list suggesting that Behringer's engineers
don't know how to lay out a proper circuit board are seriously confused.
One has to question what on earth is their motivation to smear good
companies and people?


> This is what most consider the turning point in highlighting the problem,
> a full issue in 1995 (twenty years ago this June!) of the Journal of the
> Audio Engineer Society dealing specifically with grounding and shielding
> concerns, still available as a special publication from the AES today:
> http://www.aes.org/publications/specialpubs/journal_issues.cfm
>
> SHIELDS AND GROUNDS: SAFETY, POWER MAINS, STUDIO, CABLE AND EQUIPMENT,
> (special excerpt) The June 1995 issue of the Journal was a definitive and
> comprehensive collection of information on this important topic. The seven
> papers by Neil Muncy and other experts in the field have been reprinted
> into a convenient guide for designers and practitioners.
>
> And of course following the recommendations in AES standard 48 go a long
> way toward alleviating the problem.  Only available for one decade, so
> perhaps we could excuse the designers for not being familiar with it yet:
> http://www.aes.org/publications/standards/search.cfm?docID=44
>
> AES48-2005 (r2010): AES standard on interconnections - Grounding and EMC
> practices - Shields of connectors in audio equipment containing active
> circuitry
> Printing Date: 2010-07-08
> Publication History: Pub. 2005; Reaffirmed 2010
> Abstract: This standard specifies requirements for the termination, within
> audio equipment, of the shields of cables supporting interconnections with
> other equipment, taking into account measures commonly necessary for the
> preservation of EMC (electromagnetic compatibility) at both audio and
> radio frequencies. The shielding (or screening) of audio equipment,
> cables, and microphones can be critical for EMC. The improper connection
> of these shields can cause common-impedance coupling in equipment. From XL
> connector usage, where Pin 1 is standardised as the designated shield
> contact, this has
> been identified as the Pin 1 problem.
>
> Bill Whitlock of Jensen Transformers has written a lot of good material (a
> couple of the articles in the June 1995 JAES), but a recent web page
> redesign at Jensen seems to make his online material difficult to find by
> requiring a login to download the app notes.
> The app notes at Rane are nearly as good, and are a good place to start
> while waiting for your copy of the June '95 JAES to arrive in the post:
>
> http://www.rane.com/note151.html
> http://www.rane.com/note110.html
> http://www.rane.com/pdf/whitlock.pdf
>
> This stuff isn't brain surgery, and it's well past time to stop making
> excuses for manufacturers who won't get it right.
>

So we have at least two competing theories based on Fon's very basic and
off hand description of the issue. It will be very interesting to find out
which one is correct or if it is something else completely different.

BTW, I note that Fons mentioned he found the problem while working at a
research lab. I sure hope there are no conflicts of interest there as that
would be extremely embarrassing.



--
Patrick Shirkey
Boost Hardware Ltd