Raised earths and resulting potential differences

S

Stoatgobbler

Guest
Hi there,

I'm building a JL Hood hifi amplifier and to construct the power
supplies I've mounted two toroidal transformers in a case, one on top of
the other. Each toroid has two secondary windings which I'm going to use
to create two separate, independent, regulated +22v - 0 - -22v power
supplies. To make these power supplies truly independent (and to reduce
the chances of earth loop hum) I've "lifted" each power supply 0v common
from the chassis earth via 60ohm resistors.

On testing the power supplies I've noticed that the greater the current
I pull out of each power supply the greater the potential difference
that appears across the 60ohm resistor (chassis earth to power supply 0v
line). when I pull 1.1A from the power supply then I have nearly 17v
across the 60ohm resistor. This voltage does not seem to vary with
raising the upper toroidal away from the lower so I don't think the size
of this voltage is to do with interactions between the two (but would
stand to be corrected).

I've never "played" with this, I did expect a voltage across it (due to
eddy currents within the transformer normally "lost" in heating the iron
core in a conventional E I type transformer?) but I thought toroidals
minimised this and the voltages resulting from this would be
nominal....17v across 60ohm is not nominal.

Is this normal or is something going on?

Any advice or comments would be appreciated.

Stoat
 
On 2009-02-21, Stoatgobbler <stoatgobbler@minasmorgul.demon.co.uk> wrote:
Hi there,

I'm building a JL Hood hifi amplifier and to construct the power
supplies I've mounted two toroidal transformers in a case, one on top of
the other. Each toroid has two secondary windings which I'm going to use
to create two separate, independent, regulated +22v - 0 - -22v power
supplies. To make these power supplies truly independent (and to reduce
the chances of earth loop hum) I've "lifted" each power supply 0v common
from the chassis earth via 60ohm resistors.

On testing the power supplies I've noticed that the greater the current
I pull out of each power supply the greater the potential difference
that appears across the 60ohm resistor (chassis earth to power supply 0v
line). when I pull 1.1A from the power supply then I have nearly 17v
across the 60ohm resistor. This voltage does not seem to vary with
raising the upper toroidal away from the lower so I don't think the size
of this voltage is to do with interactions between the two (but would
stand to be corrected).

I've never "played" with this, I did expect a voltage across it (due to
eddy currents within the transformer normally "lost" in heating the iron
core in a conventional E I type transformer?) but I thought toroidals
minimised this and the voltages resulting from this would be
nominal....17v across 60ohm is not nominal.

Is this normal or is something going on?
Click to expand...
somethings going on.

AC or DC ?
frequency?

Any advice or comments would be appreciated.
Click to expand...
can't make any usefull guesses without knowing what it was and what
you've done to it.
 
Jasen Betts wrote:
On 2009-02-21, Stoatgobbler <stoatgobbler@minasmorgul.demon.co.uk> wrote:
Hi there,

I'm building a JL Hood hifi amplifier and to construct the power
supplies I've mounted two toroidal transformers in a case, one on top of
the other. Each toroid has two secondary windings which I'm going to use
to create two separate, independent, regulated +22v - 0 - -22v power
supplies. To make these power supplies truly independent (and to reduce
the chances of earth loop hum) I've "lifted" each power supply 0v common
from the chassis earth via 60ohm resistors.

On testing the power supplies I've noticed that the greater the current
I pull out of each power supply the greater the potential difference
that appears across the 60ohm resistor (chassis earth to power supply 0v
line). when I pull 1.1A from the power supply then I have nearly 17v
across the 60ohm resistor. This voltage does not seem to vary with
raising the upper toroidal away from the lower so I don't think the size
of this voltage is to do with interactions between the two (but would
stand to be corrected).

I've never "played" with this, I did expect a voltage across it (due to
eddy currents within the transformer normally "lost" in heating the iron
core in a conventional E I type transformer?) but I thought toroidals
minimised this and the voltages resulting from this would be
nominal....17v across 60ohm is not nominal.

Is this normal or is something going on?

somethings going on.

AC or DC ?
frequency?

Any advice or comments would be appreciated.

can't make any usefull guesses without knowing what it was and what
you've done to it.


Two toroidal transformers, both inputs 250vac 50hz, both outputs +/-
Click to expand...
37vdc (no load) regulated to +/-22vdc with a 0v centre. Simple power
supply, split rail regulated, circuit is here >
http://www.tcaas.btinternet.co.uk/jlhlm3x7cb.htm
 
On Sun, 22 Feb 2009 16:43:45 +0000, Stoatgobbler
<stoatgobbler@minasmorgul.demon.co.uk> wrote:

Jasen Betts wrote:
On 2009-02-21, Stoatgobbler <stoatgobbler@minasmorgul.demon.co.uk> wrote:
Hi there,

I'm building a JL Hood hifi amplifier and to construct the power
supplies I've mounted two toroidal transformers in a case, one on top of
the other. Each toroid has two secondary windings which I'm going to use
to create two separate, independent, regulated +22v - 0 - -22v power
supplies. To make these power supplies truly independent (and to reduce
the chances of earth loop hum) I've "lifted" each power supply 0v common
from the chassis earth via 60ohm resistors.

On testing the power supplies I've noticed that the greater the current
I pull out of each power supply the greater the potential difference
that appears across the 60ohm resistor (chassis earth to power supply 0v
line). when I pull 1.1A from the power supply then I have nearly 17v
across the 60ohm resistor. This voltage does not seem to vary with
raising the upper toroidal away from the lower so I don't think the size
of this voltage is to do with interactions between the two (but would
stand to be corrected).

I've never "played" with this, I did expect a voltage across it (due to
eddy currents within the transformer normally "lost" in heating the iron
core in a conventional E I type transformer?) but I thought toroidals
minimised this and the voltages resulting from this would be
nominal....17v across 60ohm is not nominal.

Is this normal or is something going on?

somethings going on.

AC or DC ?
frequency?

Any advice or comments would be appreciated.

can't make any usefull guesses without knowing what it was and what
you've done to it.


Two toroidal transformers, both inputs 250vac 50hz, both outputs +/-
37vdc (no load) regulated to +/-22vdc with a 0v centre. Simple power
supply, split rail regulated, circuit is here
http://www.tcaas.btinternet.co.uk/jlhlm3x7cb.htm
Click to expand...

A full schematic including transformers and circuits would be more
useful, but the two grounds (common 0V) are isolated from each other
with a total of 120 ohms? Why islolate the grounds from each other?
I'd tie the two common 0V points together, then run them together to
teh chassies as needed.
 
On Feb 21, 7:12 am, Stoatgobbler
<stoatgobb...@minasmorgul.demon.co.uk> wrote:
Hi there,

I'm building a JL Hood hifi amplifier and to construct the power
supplies I've mounted two toroidal transformers in a case, one on top of
the other. Each toroid has two secondary windings which I'm going to use
to create two separate, independent, regulated +22v - 0 - -22v power
supplies. To make these power supplies truly independent (and to reduce
the chances of earth loop hum) I've "lifted" each power supply 0v common
from the chassis earth via 60ohm resistors.

On testing the power supplies I've noticed that the greater the current
I pull out of each power supply the greater the potential difference
that appears across the 60ohm resistor (chassis earth to power supply 0v
line). when I pull 1.1A from the power supply then I have nearly 17v
across the 60ohm resistor. This voltage does not seem to vary with
raising the upper toroidal away from the lower so I don't think the size
of this voltage is to do with interactions between the two (but would
stand to be corrected).

I've never "played" with this, I did expect a voltage across it (due to
eddy currents within the transformer normally "lost" in heating the iron
core in a conventional E I type transformer?) but I thought toroidals
minimised this and the voltages resulting from this would be
nominal....17v across 60ohm is not nominal.

Is this normal or is something going on?

Any advice or comments would be appreciated.

Stoat
Click to expand...
The part you're overlooking is these transformers have primary
connected to neutral, but secondaries are floating (until you connect
them to ground with the resistor).

That means it is merely coincidence if you measure 0V difference
relative to ground when the circuit isn't powered, as there's simply
nothing generating a charge to make it different than ground. Once
you put current through the circuit, the return is through the ground,
or center of the secondaries. It's normal, as it's carrying current.

Now what you might want to do is remove that 60 ohm resistor and use a
more conventional grounding arrangement with a smaller value resistor
and a capacitor. Ground the chassis directly to the ground wire from
your outlet through the power cord, assuming grounded outlets and 3
wire cord which it should be as a metal chassis is being used. Use a
ground loop-breaker config between amp ground and chassis ground, that
being a roughly 5-10 Ohm 5W resistor in parallel with a 0.1uF XY
rated, generally about 250V /110VAC line or ~630V/220VAC line
capacitor between amp ground and chassis. The point at which this
resistor and capacitor are taken from amp ground to the chassis should
be a connection made in the middle of the star grounding system we'll
assume you are using.
 
emailaddress@insightbb.com wrote:
On Feb 21, 7:12 am, Stoatgobbler
stoatgobb...@minasmorgul.demon.co.uk> wrote:
Hi there,

I'm building a JL Hood hifi amplifier and to construct the power
supplies I've mounted two toroidal transformers in a case, one on top of
the other. Each toroid has two secondary windings which I'm going to use
to create two separate, independent, regulated +22v - 0 - -22v power
supplies. To make these power supplies truly independent (and to reduce
the chances of earth loop hum) I've "lifted" each power supply 0v common
from the chassis earth via 60ohm resistors.

On testing the power supplies I've noticed that the greater the current
I pull out of each power supply the greater the potential difference
that appears across the 60ohm resistor (chassis earth to power supply 0v
line). when I pull 1.1A from the power supply then I have nearly 17v
across the 60ohm resistor. This voltage does not seem to vary with
raising the upper toroidal away from the lower so I don't think the size
of this voltage is to do with interactions between the two (but would
stand to be corrected).

I've never "played" with this, I did expect a voltage across it (due to
eddy currents within the transformer normally "lost" in heating the iron
core in a conventional E I type transformer?) but I thought toroidals
minimised this and the voltages resulting from this would be
nominal....17v across 60ohm is not nominal.

Is this normal or is something going on?

Any advice or comments would be appreciated.

Stoat

The part you're overlooking is these transformers have primary
connected to neutral, but secondaries are floating (until you connect
them to ground with the resistor).

That means it is merely coincidence if you measure 0V difference
relative to ground when the circuit isn't powered, as there's simply
nothing generating a charge to make it different than ground. Once
you put current through the circuit, the return is through the ground,
or center of the secondaries. It's normal, as it's carrying current.

Now what you might want to do is remove that 60 ohm resistor and use a
more conventional grounding arrangement with a smaller value resistor
and a capacitor. Ground the chassis directly to the ground wire from
your outlet through the power cord, assuming grounded outlets and 3
wire cord which it should be as a metal chassis is being used. Use a
ground loop-breaker config between amp ground and chassis ground, that
being a roughly 5-10 Ohm 5W resistor in parallel with a 0.1uF XY
rated, generally about 250V /110VAC line or ~630V/220VAC line
capacitor between amp ground and chassis. The point at which this
resistor and capacitor are taken from amp ground to the chassis should
be a connection made in the middle of the star grounding system we'll
assume you are using.
Click to expand...

Many thanks for this, you refer to this as a more conventional
methodology though this is a new idea to me (putting a capacitor into
the ground lift). Would you mind pointing me in the direction of any
literature on this please, I'd like to read up on it.

Many thanks.

Stoat
 
On Mar 9, 1:49 pm, Stoatgobbler <stoatgobb...@minasmorgul.demon.co.uk>
wrote:
emailaddr...@insightbb.com wrote:
On Feb 21, 7:12 am, Stoatgobbler
stoatgobb...@minasmorgul.demon.co.uk> wrote:
Hi there,

I'm building a JL Hood hifi amplifier and to construct the power
supplies I've mounted two toroidal transformers in a case, one on top of
the other. Each toroid has two secondary windings which I'm going to use
to create two separate, independent, regulated +22v - 0 - -22v power
supplies. To make these power supplies truly independent (and to reduce
the chances of earth loop hum) I've "lifted" each power supply 0v common
from the chassis earth via 60ohm resistors.

On testing the power supplies I've noticed that the greater the current
I pull out of each power supply the greater the potential difference
that appears across the 60ohm resistor (chassis earth to power supply 0v
line). when I pull 1.1A from the power supply then I have nearly 17v
across the 60ohm resistor. This voltage does not seem to vary with
raising the upper toroidal away from the lower so I don't think the size
of this voltage is to do with interactions between the two (but would
stand to be corrected).

I've never "played" with this, I did expect a voltage across it (due to
eddy currents within the transformer normally "lost" in heating the iron
core in a conventional E I type transformer?) but I thought toroidals
minimised this and the voltages resulting from this would be
nominal....17v across 60ohm is not nominal.

Is this normal or is something going on?

Any advice or comments would be appreciated.

Stoat

The part you're overlooking is these transformers have primary
connected to neutral, but secondaries are floating (until you connect
them to ground with the resistor).

That means it is merely coincidence if you measure 0V difference
relative to ground when the circuit isn't powered, as there's simply
nothing generating a charge to make it different than ground.  Once
you put current through the circuit, the return is through the ground,
or center of the secondaries.  It's normal, as it's carrying current.

Now what you might want to do is remove that 60 ohm resistor and use a
more conventional grounding arrangement with a smaller value resistor
and a capacitor.  Ground the chassis directly to the ground wire from
your outlet through the power cord, assuming grounded outlets and 3
wire cord which it should be as a metal chassis is being used.  Use a
ground loop-breaker config between amp ground and chassis ground, that
being a roughly 5-10 Ohm 5W resistor in parallel with a 0.1uF XY
rated, generally about 250V /110VAC line or ~630V/220VAC line
capacitor between amp ground and chassis.  The point at which this
resistor and capacitor are taken from amp ground to the chassis should
be a connection made in the middle of the star grounding system we'll
assume you are using.

Many thanks for this, you refer to this as a more conventional
methodology though this is a new idea to  me (putting a capacitor into
the ground lift). Would you mind pointing me in the direction of any
literature on this please, I'd like to read up on it.

Many thanks.

Stoat
Click to expand...
Not to play favorites on articles (many are written by audiophiles who
have strange ideas), you will find several with this search:

http://www.google.com/search?q=ground+loop+breaker+amp
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top