Maker Pro
Maker Pro

ground-loop problems

  • Thread starter William Sommerwerck
  • Start date
W

William Sommerwerck

Jan 1, 1970
0
I'd like to thank everyone who took the time to respond to my question. I
was particularly appreciative of those that told me about noise & grounding
problems I've never even heard of.

I'm not quite ready to put everything on one circuit. So I'm going to call
both Parasound and Pioneer to see what insights, if any, they have to offer.

Again, thank you-all for your help.

PS: As for balanced cables being an OTT solution for home installations... I
have five power amps, four of which sit next to the speakers they drive.
Given the distances, and the fact that the amps draw significant current,
and some are on different circuits, it would not be a good idea to use
unbalanced cables.
 
W

William Sommerwerck

Jan 1, 1970
0
I'm not quite ready to put everything on one circuit. So I'm going to
I dunno US electrics, but can't see why that would matter. As far as
audio -- or video -- is concerned there should be only one ground.

Should be -- but isn't. Each device has its own ground. Each ground may have
a slightly different AC potential. This is the cause of "ground loops" when
you connct the devices.

All the equipment in my system is transformer-powered. That, in and of
itself, does not prevent ground loops.

What distances are you talking about? I have unbalanced runs here
of some 80 ft without problems. I do use balanced audio to feed things
round the house - but that is purely to use cheap telephone cable and
avoid crosstalk.

You're fortunate to have no problems with 80' runs. Professionals use
balanced lines for a good -- it avoids problems.
 
W

William Sommerwerck

Jan 1, 1970
0
You're fortunate to have no problems with 80' runs. Professionals use
I'm a sound recordist working in TV so yes I know about balanced
equipment. But its use is more for the common mode rejection of
interference on long and hazardous runs - as well as to provide immunity
from ground loops across a variety of equipment which is likely to be used
in different situations. But shouldn't be needed in a domestic environment
for a permanent installation. And such connections are rare in domestic
equipment. And rightly so, since they add unnecessary electronics at both
input and output -- adding to complexity and cost.

I've never experimented with single-ended and balanced installation, using
the same equipment, so I can't speak from experience. However, balanced
inputs and outputs have become common in "good" equipment, partly because
they add little to the cost, * and partly because -- especially in A/V
systems -- they make things easy for the manufacturer and dealer.

In a system where the amplifiers don't sit right next to control unit (qv,
my system), the designer & installer don't have to worry about hum and
interference. The customer isn't going to be happy if the dealer has to do
extensive troubleshooting to get rid of noise, or (worse) if the house has
to be rewired to reduce the inter-component ground potential.

I remember, ca 1978, installing a relatively simple Crown system in an
equipment rack. We had all kinds of grounding problems. Some of this was
probably due to the Crown equipment itself (surprising, because the same
products were used professional), but we wound up have to completely isolate
the units from the rack. This involved wrapping the screws in vinyl tubes,
and using faucet washers to lift the ears away from the rack. Our first
attempt used black washers, which contain carbon and are conductive. These
were quickly replaced with non-conductive red washers.

* Nor are there any additional electronics.
 
W

William Sommerwerck

Jan 1, 1970
0
Dave Plowman (News) said:
What connectors do they use? There simply wouldn't be room for XLRs
on the back of my AV amp.

Even a small power amplifier has enough room for XLRs. Many control units
(what used to be the "preamplifier") are huge -- mine has ten XLRs on the
back, not to mention more than 50 RCAs, plus S and TOSLINK connectors.

With decent design, you don't ground at each component. You use the
screen of the interconnect to ground everything back to the same point.

I'm afraid to say anything, because I'm not an expert on grounding. But I'm
pretty certain that what you're saying is wrong.

Simply connecting two amplifiers by cable shields does not cause the
amplifiers' grounds to have the same AC potential. This is why components
"far"/"near" each other, not connected/connected to the same outlet,
sometimes/rarely have hum/buzz problems.
 
W

William Sommerwerck

Jan 1, 1970
0
<lots of interesting stuff snipped>

The issue -- in my mind -- is the distinction between "mains" (AC) ground
and audio ground. They're not the same -- at least, not in the U S of A.
This is why it's possible, on devices with three-wire power cables, to
sometimes reduce the hum by lifting the ground on one and "rotating" the
plug. Indeed, the Parasound power amps have a "ground lift" switch on the
back.

I've spoken with Magnolia and Pioneer, and neither had the "Aha!" response I
was hoping for. I'll call Parasound today and ask them.
 
W

William Sommerwerck

Jan 1, 1970
0
The issue -- in my mind -- is the distinction between "mains" (AC) ground
If they're not the same there would be nothing to gain from an earth lift
switch. It's having two non identical paths for the signal ground that
causes an earth loop - you get a differential current flow. But as I
said, it is common for them to be the same. But don't you have 'double
insulated' equipment that doesn't need a safety mains ground? This is
tje norm for Hi-Fi equipment in the UK - even with the higher voltage
here.

Not that I know of. Only power tools (and similar products) are
double-insulated.

And assuming it has the normal transformer supplying the DC - I'm not
sure why reversing the mains lead should have any effect - given an
isolating transformer of this type removes any reference from the supply -
even when one leg of the supply is grounded as in the UK.

This is common experience, not something I've made up. I'm sure other people
in this group can report similar stories.

Could we agree to drop this discussion for the time being, until we can find
a /real/ expert on grounding?
 
W

William Sommerwerck

Jan 1, 1970
0
Could we agree to drop this discussion for the time being,
Fine. But I've certainly got enough 'experience' to know that
balanced connections are total overkill for domestic line level
installations. Different matter with low level signals from
microphones, etc, or for very long runs.
If you've been sold that they're needed, that's your problem.

Well... <ahem> Balanced inputs and outputs have been common on the "better"
audio equipment for around 20 years. They're particularly desirable if
you're a nut about fully balanced circuitry. And you pay for them whether or
not you want them.

Balanced ins and outs became really popular with the introduction of the A/V
system controller, precisely because they eliminate, a priori, grounding and
hum problems, without adding a lot of money to the system's cost.

I previously owned Brand K electronics, which were fully balanced, and I
never had hum problems. I could crank the volume all the way up, well past
normal listening levels, and the system was dead-quiet (other than through
the phono input). Why would I want to use unbalanced lines?

Balanced cables cost only a little more than unbalanced cables. Unlike RCA
plugs (which should have been banned decades ago), they rarely pop loose,
and you can plug and unplug them without getting a "big blast o' hum". *

* The better RCA cables have "long" collars, so the ground is made before
the hot side when plugging in, and vice-versa when pulling the cable.
 
W

William Sommerwerck

Jan 1, 1970
0
This is precisely the kind of discussion I did not want this question to
devolve into.

I don't understand why you think an audio system with components separated
by 20' or more, and connected to different power lines, can be wired with
unbalanced interconnects, on the naive assumption that there will never be
any problems with hum or noise.

For what it's worth, all my prior systems had the power amps sitting next to
the preamps. The components were connected with unbalanced cabling, and I
never had problems with hum, noise, etc. Nor did it ever cross my mind that
any of these systems would be "better" with balanced wiring.

My Parasound components -- and the Apogee electronic crossovers -- have both
balanced and unbalanced inputs and outputs, and can be freely connected as
desired. Other than the fact that balanced cables cost slightly more than
unbalanced -- why would anyone in their right mind use unbalanced cables?
Balanced cabling is in every way "superior".

I'm friends with the manager of the Bellevue Magnolia. I'll ask him what
their installation department's opinion on this matter is, and /why/ they
hold that opinion (whatever it might be).
 
W

William Sommerwerck

Jan 1, 1970
0
For what it's worth, all my prior systems had the power amps sitting
Well they can't be 'better' since you're introducing extra electronics to
balance and unbalance the signal -- not many devices are balanced
internally.

This is not correct. Changing the circuit from an unbalanced input or output
to a balanced input or output does not add a gain stage.

The power amps were designed by John Curl, one of the "inventors" of
full-complementary push-pull amplification. I will ask him what topology the
Parasound A21 uses.

You also have a balanced output radio tuner? Record deck? Etc?

The tuner does, but I use the unbalanced outputs. We were talking about
power amplifier connections, anyway.

If your equipment is designed for balanced operation it's quite likely the
unbalanced connections are an afterthought and the screen connected to
mains ground.

This is common. It is not generally considered poor design.
 
W

William Sommerwerck

Jan 1, 1970
0
If your equipment is designed for balanced operation it's quite
This is common. It is not generally considered poor design.

BIG WHOOPS! I wrote that in a rush.

Balanced inputs are commonly converted to unbalanced by grounding one
side -- to the SIGNAL ground (that is, zero volts in an amplifier with a
balanced power supply), NOT the mains ground.
 
M

Michael Karcher

Jan 1, 1970
0
Dave Plowman (News) said:
And assuming it has the normal transformer supplying the DC - I'm not
sure why reversing the mains lead should have any effect - given an
isolating transformer of this type removes any reference from the supply -
even when one leg of the supply is grounded as in the UK.

Don't forget about capacitive coupling between the windings in the
mains transformer. Depending on transformer geometry, the coupling might
be stronger to one mains lead than to the other. So it is advantageous
to make the stronger coupled one ground.

Regards,
Michael Karcher
 
W

William Sommerwerck

Jan 1, 1970
0
Well they can't be 'better' since you're introducing extra electronics
Who said anything about gain? I said electronics.

bang... bang... bang... bang... [sound of William Sommerwerck banging his
head against a concrete wall] bang... bang... bang... bang...
 
W

William Sommerwerck

Jan 1, 1970
0
Dave Plowman (News) said:
Makes no difference in principle if both the tuner and amps are separate
from the pre-amp. Unless you feel there is some magic length where
balanced suddenly becomes essential?

bang... bang... bang... bang... [repeated sounds of head banging against
concrete wall]

Mr. Plowman, are you aware that you repeatedly change or twist the subject,
without regard to what is actually being discussed? Because you do.

The "magic length" occurs when the power amp is far enough away to be on a
separate circuit. Even if it's on the same circuit, there's no guarantee
that the chassis grounds will be at the same potential. Which is, for me,
The Fundamental Issue (with regard to ground loops.)
 
W

William Sommerwerck

Jan 1, 1970
0
Balanced inputs are commonly converted to unbalanced by grounding
Not much point in using a balanced line if you're going to 'unbalance' it
in this way. Use a transformer to do the job properly.

[More loud and anguished head-banging from yrs. truly.]

I "know people". I'm going to find someone who's a legitimate expert on
audio grounding, and try to get some sort of definitive answer.
 
W

William Sommerwerck

Jan 1, 1970
0
Dave Plowman (News) said:
Balanced inputs are commonly converted to unbalanced by grounding
one side -- to the SIGNAL ground (that is, zero volts in an amplifier
with a balanced power supply), NOT the mains ground.
Not much point in using a balanced line if you're going to 'unbalance'
it in this way. Use a transformer to do the job properly.
[More loud and anguished head-banging from yrs. truly.]
I "know people". I'm going to find someone who's a legitimate expert
on audio grounding, and try to get some sort of definitive answer.
I've already given you chapter and verse which you choose not to
understand. You could try re-reading it more carefully.

Almost everything you've said goes against what I believe to be correct --
and some of it is unquestionably wrong. I have not had any "Aha!" moments of
sudden clarity.

I've already asked two people for advice. I'm going to do some more research
on this, and eventually get back to the group. Case closed, for the time
being.
 
D

David Lesher

Jan 1, 1970
0
I dunno US electrics, but can't see why that would matter. As far as audio
- or video - is concerned there should be only one ground. And that
doesn't really matter what mains circuit things are fed from as there will
be an isolating transformer between it and the DC. Unless it's some crappy
design where it has to be grounded for safety reasons.


The Apollo 11 landing was almost aborted by such an issue.

The LM guidance computer was fielding all kinds of high priority data
from the landing radar and control inputs. It was also, unbeknownst to the
crew, also looking at the rendezvous radar as well. [This so an immediate
Abort to Orbit could be accomplished; if the landing was abandoned.]

If the 2nd radar saw nothing, great, no data. BUT as it happens, it ran
off a different phase of the LM power system, and the little big of noise
between it and the landing radar was enough to be noticed. As a result,
the computer started spitting out 1201 alarms, which was palnned for &
acceptable, but disconcerting to everyone involved.

<http://history.nasa.gov/alsj/a11/a11.1201-pa.html>
 
G

GregS

Jan 1, 1970
0
The Apollo 11 landing was almost aborted by such an issue.
The LM guidance computer was fielding all kinds of high priority data
from the landing radar and control inputs. It was also, unbeknownst to
the crew, also looking at the rendezvous radar as well. [This so an
immediate Abort to Orbit could be accomplished; if the landing was
abandoned.]
If the 2nd radar saw nothing, great, no data. BUT as it happens, it ran
off a different phase of the LM power system, and the little big of
noise between it and the landing radar was enough to be noticed. As a
result, the computer started spitting out 1201 alarms, which was palnned
for & acceptable, but disconcerting to everyone involved.

Do you routinely have different phases running audio equipment in the same
room in a house in the US? This smacks to me of bad practice and isn't
allowed in the UK. Nor would it happen in a pro audio installation in say
a TV etc studio - all the relevant equipment would be on the same phase -
leaving the others for power hungry devices like air con and lighting.

Of course a power hungry large computer system might well be different -
but the pitfalls of this are surely well known?

No, but you can have 180 degree in phase situatuion
between the split 240 transformer, and have 240 volts potential going
to the audio setup. I always like to use only
one leg of the transformer split out on the pole or one breaker. In my opinion
there is a chance of more problems using both sides of the transformer
allthough some audio installations use a split transformer to
reduce posible noise. So its all in how you look at it.

greg
 
G

GregS

Jan 1, 1970
0
"Dave Plowman (News)" said:
I dunno US electrics, but can't see why that would matter. As far as
audio - or video - is concerned there should be only one ground. And
that doesn't really matter what mains circuit things are fed from as
there will be an isolating transformer between it and the DC. Unless
it's some crappy design where it has to be grounded for safety reasons.


The Apollo 11 landing was almost aborted by such an issue.
The LM guidance computer was fielding all kinds of high priority data
from the landing radar and control inputs. It was also, unbeknownst to
the crew, also looking at the rendezvous radar as well. [This so an
immediate Abort to Orbit could be accomplished; if the landing was
abandoned.]
If the 2nd radar saw nothing, great, no data. BUT as it happens, it ran
off a different phase of the LM power system, and the little big of
noise between it and the landing radar was enough to be noticed. As a
result, the computer started spitting out 1201 alarms, which was palnned
for & acceptable, but disconcerting to everyone involved.

Do you routinely have different phases running audio equipment in the same
room in a house in the US? This smacks to me of bad practice and isn't
allowed in the UK. Nor would it happen in a pro audio installation in say
a TV etc studio - all the relevant equipment would be on the same phase -
leaving the others for power hungry devices like air con and lighting.

Of course a power hungry large computer system might well be different -
but the pitfalls of this are surely well known?

No, but you can have 180 degree in phase situatuion
between the split 240 transformer, and have 240 volts potential going
to the audio setup. I always like to use only
one leg of the transformer split out on the pole or one breaker. In my opinion
there is a chance of more problems using both sides of the transformer
allthough some audio installations use a split transformer to
reduce posible noise. So its all in how you look at it.

Really, if you feed two equipment from a separate
breaker, one on one side of the 240 volt transformer
and one on the other, with the transformer center tapped
and becoming neutral, the two sides of the voltage
cancel. Or at least the possible electrostatic
charge is neutralized to zero. But that
would onlt seem to work if both pieces of equipment
had identical power supplies in every respect.

greg
 
D

David Lesher

Jan 1, 1970
0
Do you routinely have different phases running audio equipment in the
same room in a house in the US? This smacks to me of bad practice and
isn't allowed in the UK.

The UK has very different wiring than the US.

Here, a standard circuit is 120v/20A or 2400 watts [at unity PF...]

In the UK; ring wiring supplies 240v/30A, or 7200 watts. Plugs
have internal fuses based on the device.

(The US may well have a 240V/20A outlet for an air conditioner;
where it's 120-N-120.)
 
D

David Nebenzahl

Jan 1, 1970
0
The US philosophy is to only trip breaker in a fault condition and
that means extra capacity.

So what is the UK philosophy in this regard?
 
Top