I built this system 32 years ago, and have regularly maintained and upgraded
it over that span of time. The System consists of multiple amplifiers in a
rack fed with 240vac, split 120vac for the low power equipment, which
included preamps and electronic crossover networks.

I recently started using a spare Hafler 500 amp (one of two spares) for
Surround Sound amplification, driven directly off the outputs of my DVD
player's discreet 5.1 channel outputs. Speakers are floating and the amp
outputs are not connected to any AC ground other than the speakers
themselves. It has worked fine since last summer.

System was working normally the night of 12/31. Problem occurred 20 minutes
into a movie, shortly before midnight.

Problem manifested as a loud popping noise, dominant in right surround
channel, but heard in both surround channels, at about 6 cps-it sounded like
a square wave. This blew the B+ fuse and the N-channel MOSFETs in the left
channel of the surround sound amp, a Hafler 500, even though the popping
issued from the right channel with greater amplitude.

After repair, amp works perfectly on bench. No oscillation, perfectly
stable.

Reinstalled in rack, oscillation starts when left input is connected-even
with connected equipment turned off. If the DVD player surround output is
connected to the left input on the amp, oscillation occurs. However, if the
power to the entire video equipment is disconnected, the DVD player can be
connected to the amp without adverse effects. This suggests some sort of
loop between AC power line and amp audio in/out.

Amplifier inputs are at same ground potential and both connected to amp
chassis.

When left channel is connected and popping occurs, a large voltage
difference between chassis and rack appears. This was discovered because
arcing was observed between chassis and rack when amp was loosely sitting in
rack. This was repeated several times to verify that it was not imagination.

When amp switched on and off (before speaker relay has a chance to close),
4-5 seconds later, a loud popping sound is heard in the right and possibly
left speakers, and chassis voltage difference appears. (Note, speakers are
theoretically NOT connected to amp as the relays are supposed to be open.)

Problem persists no matter which branch circuit amp is connected to.

Speakers are not connected to building ground; they are floating, as their
only connection is the amp's output terminals. The only persistent
connection to the speakers is the chassis ground (the relay only disconnects
the 'hot' side). If one side is disconnected and the ground side is moved to
a different potential, how can substantial current flow in the voice coil?

Even stranger is that when a battery-powered WAV player is connected to the
amp inputs, the amp plays fine. If the portable player is connected to the
left input and the DVD or the mixing board is connected to the right input,
the portable player's signal can be heard in BOTH outputs.

The suspect amplifier was replaced with an identical Hafler 500, known to be
working, and it too exhibits the same motorboating problem when some
AC-powered signal source is connected to its LEFT channel input. Connecting
to the RIGHT channel input, it behaves normally.

The offending motorboating noise appears at the INPUT jack of the amplifier
WHEN THE SPEAKER RELAY CONNECTS THE SPEAKERS. The output of the amplifier is
a replica of the input signal, only larger.

If a tone signal larger than 100mV is applied to the amp input from the
mixing board, the motoboating stops-however the sine signal becomes an
overshooting (rising edge) square wave with a rounded trailing edge.

Thought: it seems as if somehow the AC power circuits are acting as a
positive feedback loop for the audio circuit. There seems to be common mode
conduction going on, but cannot figure out why the common mode path appeared
without any user intervention, during a movie. No one touched the racks when
the oscillation started for the first time.

Neither can I explain how two independent amplifier channels behave like
they are connected together as mono when one input is connected to the
mixing board or the DVD player, while the other is connected to the portable
player. The Hafler 500 employs no means of switching to mono with one or two
inputs connected. The spare Hafler 500 behaves the same way, so I'm inclined
to believe the problem lies elsewhere.

When I run the amp in the shop, it works normally. When I feed either of the
suspect audio sources to the amp WHILE IN THE SHOP it works normally. The
common denominator seems to be that the amp misbehaves only when in the
rack. Even when electrically isolated from the rack rails with rubber and
cardboard strips.



I have been working on this problem all night and all day today, but am no
closer to a solution. I even checked the main breaker panel and tightened
all the ground screws. No other electrical problems in the house. The rest
of the sound system works normally. It is just the one amplifier used for
the surround that suddenly doesn't like external sources connected to its
LEFT channel input.



None of this makes sense, but I tell you that I could not make this up if I
thought about it all night! It's just too incredible. All the standard
troubleshooting methods were used, but the common aspect is anything
connected to AC power, WHEN THE AMP IS IN THE RACK. It's happy on the shop
bench with any source. But not when sitting in the rack, even isolated from
the rack, or even with extention cords bringing AC power from another branch
circuit. The proximity to the rack seems to be a factor.



In fifty years of working in electronics, I've not run into anything like
this in the field of broadcast engineering. But here it is in my own home
studio and all the classic remedies are ineffective.



Anyone got any ideas on this mystery?


--


Take care,



Mark & Mary Ann Weiss



VIDEO PRODUCTION . FILM SCANNING . DVD MASTERING . AUDIO RESTORATION
Hear my Kurzweil Creations at: www.dv-clips.com/theater.htm
www.basspig.com The Bass Pig's Lair - 15,000 Watts of Driving Stereo!
Business sites at:
www.mwcomms.com
www.adventuresinanimemusic.com
-

Almost 24 hours TO THE MINUTE since this problem mysteriously appeared, it
vanished!

I had pulled the amp from the rack 20 minutes ago, leaving slack on the
wiring enough to set it on a crate in front of the rack. I powered it up and
it worked normally.

So the next thing I do is take an alligator lead and cautiously touch the
amp chassis with one end and the rack with the other, fully expecting
fireworks. Not a thing happened.

So the next thing I do is carefully lift the entire amp and sssllooooowwlly
slide it into the rack space where it normally sits. Again, NOTHING unusual
happens. So now I'm really wondering... The next thing I do is put one screw
in. Still no fireworks. I put the rest of the screws in. It's still working!

Next test: I power off the amp and power it back on. Still works.

Now let's try the other surround source (at this point I have it working
with the front end mixer output, something it would not do for the last 20
hours.) I swap the input connections with the output of the DVD player's
surround outputs. I power the amp up and... no popping. Just surround
program coming out of both channels as it should.


The bottom line is, this goes down in my "X-Files" folder. It's something I
cannot explain. For exactly 24 hours, no matter what I did, I could not
solve the "feedback loop" or whatever phenomenon was occuring. Now it's
working and all I did was install the amp while it was powered on. The other
7 or 8 times I installed it in the last 24 hours it was powered off. Go
figure. I'm not a mystical man, but this occurance is deeply disturbing
because I have not found the cause of the problem, and it cleared up just as
quickly and mysteriously as it appeared last night.