[grc] A bit of trouble...

[al davis]

On Sun, 30 Oct 2016 19:33:12 -0700
Lanny Cotler via grc <grc at maillist.peak.org> wrote:

> 1. I heard it.  soft voices and music (I clearly understood the
> voices, recognized them.) And a loud (louder than the voices) buzz.
> (it did not sound like just AC hum.)

> 2. when I placed my hand near (a couple of inches) the Amp power
> cord; the buzz changed.

> 3. The only remedy I had was to turn off our transmitter.
>       a. that reduced the buzz to almost not noticeable, and
>      b. no KLLG signal was heard at all.

> 4. Later, after the transmitter was off, their sound guy was cleaning
> up the stage (before the musicians appeared.)  When he bundled all
> the power cords together, the buzz disappeared.

> This sounds (to me & the sound guy) like a bad ground.


Wiring, grounding, bad equipment design ..  could be any or all of
that.  It could be any of the equipment in the place, including
something that doesn't seem to be bothered.

The common wisdom of grounding one end of a shield to avoid ground
loops isn't always so wise.  At low frequencies (60 Hz) the wavelength
is long, a ground loop can cause hum because of induced current, so
break the loop to fix it.  At 100 MHz, the wavelength is 3 meters, so
shields grounded at one end act as antennas, so ground it everywhere
and make as many loops as you can.

It could be the transmitter's grounding.  Poor transmitter ground could
lead to RF all over, and also it would be prone to lightning damage.

Some equipment does not have proper RF bypassing on the inputs.  This
is common on consumer stuff, even more common on niche market gadgets.
Even some equipment marketed for broadcast does not have RF bypassing
on the inputs.  Look at the schematic.

The very common one-op-amp "differential" input stage, with no
bypassing, is especially prone to this problem.  The variant with a
gain trim that shunts the input (with no bypass) is really asking for
trouble.

5532 op-amps are more prone to RF problems than the FET opamps like
LF353 or TL072.  If the input stage is a 5532, you may be able to fix
the RF problem by changing the opamp to a LF353.