Ground loops

[[email protected]]

Hello,

I am observing this affect on a few components whereby their frequency
response shows the 60Hz power line frequency. This seems to be
isolated to just components from a few date codes. However, this
does indicate that there is interference with the 60Hz frequency with
some parts. I have tried to tie the chassis of the unit under test,
the measurement unit and the unit under test to ensure that they are
at the same ground potential, but the problem still persists. While
the circuit is powered down, the chassis of the unit under test, the
unit under test and the measurement unit have no resistance between
them. But when I power up the circuit, there exists a negligible
resistance of upto 20 ohm between the chassis of the unit under test
and the uunit under test and the measurement unit and the unit under
test. Does this create the possibility of ground loops? Could the
60Hz power line frequency that I am seeing be related to ground loops?

I have also tried to run the experiment with/without an isolation
transformer, but the problem still persists. Any ideas on where the
60Hz power line frequency could be creeping in? If the failing date
code is run at a different location, it does not exhibit the 60Hz
power line frequency. Therefore, it is something in setup that is
causing parts with some date codes to capture the 60Hz power line
frequency.

Please help.

 

[Jamie]

Hello,

I am observing this affect on a few components whereby their frequency
response shows the 60Hz power line frequency. This seems to be
isolated to just components from a few date codes. However, this
does indicate that there is interference with the 60Hz frequency with
some parts. I have tried to tie the chassis of the unit under test,
the measurement unit and the unit under test to ensure that they are
at the same ground potential, but the problem still persists. While
the circuit is powered down, the chassis of the unit under test, the
unit under test and the measurement unit have no resistance between
them. But when I power up the circuit, there exists a negligible
resistance of upto 20 ohm between the chassis of the unit under test
and the uunit under test and the measurement unit and the unit under
test. Does this create the possibility of ground loops? Could the
60Hz power line frequency that I am seeing be related to ground loops?

I have also tried to run the experiment with/without an isolation
transformer, but the problem still persists. Any ideas on where the
60Hz power line frequency could be creeping in? If the failing date
code is run at a different location, it does not exhibit the 60Hz
power line frequency. Therefore, it is something in setup that is
causing parts with some date codes to capture the 60Hz power line
frequency.

Please help.

i would first check to make sure the bad locations have a real earth
ground in the receptacle.
You didn't say exactly what this component is ? it would shed a lot
more light on it.
It would appear to me that maybe you have some inductive problems
generating EMF, but that is just a guess.
I doubt that your meter is actually reading 20 ohms, it most likely
is getting energy from it's leads near by.

 

[[email protected]]

i would first check to make sure the bad locations have a real earth
ground in the receptacle.
You didn't say exactly what this component is ? it would shed a lot
more light on it.
It would appear to me that maybe you have some inductive problems
generating EMF, but that is just a guess.
I doubt that your meter is actually reading 20 ohms, it most likely
is getting energy from it's leads near by.

--
"I'm never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.http://webpages.charter.net/jamie_5- Hide quoted text -

- Show quoted text -

The device is an audio sample rate converter

 

[martin griffith]

On 11 Feb 2007 08:16:16 -0800, in sci.electronics.design

Hello,

I am observing this affect on a few components whereby their frequency
response shows the 60Hz power line frequency. This seems to be
isolated to just components from a few date codes.

Please help.

what sort of components, transformers, inductors, flux capacitors?

the 60hz, how many dB's worse is it, and what are the specs for 60Hz
rejection?


martin

 

[john jardine]

Hello,

I am observing this affect on a few components whereby their frequency
response shows the 60Hz power line frequency. This seems to be
isolated to just components from a few date codes. However, this
does indicate that there is interference with the 60Hz frequency with
some parts. I have tried to tie the chassis of the unit under test,
the measurement unit and the unit under test to ensure that they are
at the same ground potential, but the problem still persists. While
the circuit is powered down, the chassis of the unit under test, the
unit under test and the measurement unit have no resistance between
them. But when I power up the circuit, there exists a negligible
resistance of upto 20 ohm between the chassis of the unit under test
and the uunit under test and the measurement unit and the unit under
test. Does this create the possibility of ground loops? Could the
60Hz power line frequency that I am seeing be related to ground loops?

I have also tried to run the experiment with/without an isolation
transformer, but the problem still persists. Any ideas on where the
60Hz power line frequency could be creeping in? If the failing date
code is run at a different location, it does not exhibit the 60Hz
power line frequency. Therefore, it is something in setup that is
causing parts with some date codes to capture the 60Hz power line
frequency.

Please help.

Mains frequency injection (and harmonics through MHz) is a real problem
down at the mV, uV level. It creeps in everywhere, via all paths and all
mechanisms. Textbook 'solutions' such as isolation transformers, filters,
star points, baluns, line breakers, shielding etc, provide little help and
more usually waste time and effort.

Effective cure is use of battery supplied or isolated power supplies with
only the signal commons galvanically connected.
Maybe worth adding that signal sources themselves are prone to injecting
mains noise along with the wanted reference signals. That 50ohm BNC output
socket may be delivering say 100uV of signal on it's centre pin and 10uV
mains noise through it's body.
(Cleanest signal I've ever seen came from an el cheapo 'Levell TG200' with
it's internal switched mode battery adapter . Worst, came from a Wavetek
function gen').
john