[Prune]

I see some transformers use a copper band parallel to the windings but
outside the core, in addition to the usual magnetic shielding around the
core sides or fully enclosing the transformer. My question is if it should
be grounded or not. Here's the information I found on the web:

"There is usually also a circumferential copper shield (or 'flux band')
around the entire transformer. It serves primarily as a radiation shield.
It is often left floating in low cost designs, though it may be connected
to the Secondary ground if desired. If so connected, safety issues may need
to be considered in regards to the requirement of reinforced insulation
between Primary and Secondary, and also the required Primary to Secondary
'creepages' (distance along the insulating surface) and 'clearances'
(shortest distance through air) as applicable. When the transformer uses an
air gap on its outer limbs, the fringing flux emanating from the gap causes
severe eddy current losses in the band. So this band is also usually only
2-4 mil thick. Note that the ends of this band can, and should be, soldered
together, because this is an outer shield, and can never constitute a
shorted turn for the transformer."

It's not clear for me from this. On the one hand, they say it's left
floating in low cost designs, implying it's better to be grounded; on the
other hand, wouldn't it send a bunch of noise into ground? And does
grounding at all affect the shielding ability? Also their mention about
connecting to secondary ground, why put more noise there; shouldn't it go
to chassis ground/earth instead (in case the two grounds are partially
decoupled such as by a ground lifting circuit as used for breaking ground
loops)?

[Eeyore]

Prune wrote:

Note that the ends of this band can, and should be, soldered
together, because this is an outer shield, and can never constitute a
shorted turn for the transformer."

It's a shorted turn for *stray* flux more to the point.


It's not clear for me from this. On the one hand, they say it's left
floating in low cost designs, implying it's better to be grounded; on the
other hand, wouldn't it send a bunch of noise into ground?

Conceivably it might esp with switching supplies.

And does
grounding at all affect the shielding ability?

It would influence any electrostatic shielding effect perhaps. That's best dealt
with by an *internal* band that's not shorted. The band is typically returned to
the primary low side in a switching supply precisely for these noise reasons.


Also their mention about
connecting to secondary ground, why put more noise there; shouldn't it go
to chassis ground/earth instead (in case the two grounds are partially
decoupled such as by a ground lifting circuit as used for breaking ground
loops)?

To be honest. primary and seondary ground *ought* to be the same thing i.e.
ultimately chassis.

Graham

[Prune]

Eeyore wrote in
:

It would influence any electrostatic shielding effect perhaps. That's
best dealt with by an *internal* band that's not shorted. The band is

That's completely different, it's the electrostatic shield between primary
and secondary, which I've seen in non-switching transformers as well; it
prevents electrostatic coupling of mains noise into the secondary. The
outside band is for shielding from stray EMI due to core leakage. I'm not
sure whether that EMI shielding ability could be potentially compromised if
I ground the band.

[Eeyore]

Prune wrote:

Eeyore wrote in
:

It would influence any electrostatic shielding effect perhaps. That's
best dealt with by an *internal* band that's not shorted. The band is

That's completely different, it's the electrostatic shield between primary
and secondary, which I've seen in non-switching transformers as well; it
prevents electrostatic coupling of mains noise into the secondary.

More than that in fact in a switching supply.


The
outside band is for shielding from stray EMI due to core leakage.

It's not primarily for classic *shielding* at all. It's intentionally copper to
keep the resistance low since it's a shunt for stray magnetic flux.


I'm not
sure whether that EMI shielding ability could be potentially compromised if
I ground the band.

Since that's not what it's for - probably not. But you don't really need to
ground it anyway. Why do you think that would help ?

Graham

[Prune]

Eeyore wrote in
:

Since that's not what it's for - probably not. But you don't really
need to ground it anyway. Why do you think that would help ?

Because of the implication of the text I quoted.

[Eeyore]

Prune wrote:

Eeyore wrote in
:

Since that's not what it's for - probably not. But you don't really
need to ground it anyway. Why do you think that would help ?

Because of the implication of the text I quoted.

You mean " It serves primarily as a radiation shield " ?

It actually serves primarily as a magnetic shunt.

Graham

[legg]

On Sun, 30 Jul 2006 01:37:37 +0100, Eeyore
wrote:



Prune wrote:

Eeyore wrote in
:

Since that's not what it's for - probably not. But you don't really
need to ground it anyway. Why do you think that would help ?

Because of the implication of the text I quoted.

You mean " It serves primarily as a radiation shield " ?

It actually serves primarily as a magnetic shunt.

Magnetic stray flux is considered radiation, but the use of the word
shield may seem to be a little misleading. EMI has the two components.

The amount of capacitive current that might be delivered by the flux
band, and it's prefered return path either to primary or secondary,
depend on the transformer winding's layering.

Without a measurable problem to fix, fiddling around with an
undocumented device is likely to be a waste of time, though it's
always a good thing to keep your eyes open.

RL

[Eeyore]

legg wrote:

On Sun, 30 Jul 2006 01:37:37 +0100, Eeyore
wrote:


Prune wrote:

Eeyore wrote in
:

Since that's not what it's for - probably not. But you don't really
need to ground it anyway. Why do you think that would help ?

Because of the implication of the text I quoted.

You mean " It serves primarily as a radiation shield " ?

It actually serves primarily as a magnetic shunt.

Magnetic stray flux is considered radiation,

It's stretching the point a little. A lot even !


but the use of the word
shield may seem to be a little misleading. EMI has the two components.

The amount of capacitive current that might be delivered by the flux
band,

Delivered ? by the flux band ? !

and it's prefered return path either to primary or secondary,
depend on the transformer winding's layering.

How is anything going to *return* there ?


Graham

[Phil Allison]

"Prune" = makes all ordinary ****WITS look like geniuses


I see some transformers use a copper band parallel to the windings but
outside the core, in addition to the usual magnetic shielding around the
core sides or fully enclosing the transformer. My question is if it
should
be grounded or not. Here's the information I found on the web:

"There is usually also a circumferential copper shield (or 'flux band')


** This is the source -

an article on FLYBACK TRANSFORMERS !!!

http://www.national.com/nationaledge/mar04/article.html



It's not clear for me from this.


** Nothing about AC supply, IRON core transformers will be clear from an
article about * SMPS txfmrs* .


On the one hand, they say it's left
floating in low cost designs, implying it's better to be grounded;


** Never seen a copper flux band on an IRON core tranny that was NOT
grounded to the chassis by the simple act of bolting the damn thing down.


FUUUUUKKK OOOFF BLOODY IMBECILE !!!





......... Phil

[Eeyore]

flipper wrote:

On Sun, 30 Jul 2006 05:20:30 +0100, Eeyore
wrote:

legg wrote:

On Sun, 30 Jul 2006 01:37:37 +0100, Eeyore
wrote:


Prune wrote:

Eeyore wrote in
:

Since that's not what it's for - probably not. But you don't really
need to ground it anyway. Why do you think that would help ?

Because of the implication of the text I quoted.

You mean " It serves primarily as a radiation shield " ?

It actually serves primarily as a magnetic shunt.

Magnetic stray flux is considered radiation,

It's stretching the point a little. A lot even !


but the use of the word
shield may seem to be a little misleading. EMI has the two components.

The amount of capacitive current that might be delivered by the flux
band,

Delivered ? by the flux band ? !

and it's prefered return path either to primary or secondary,
depend on the transformer winding's layering.

How is anything going to *return* there ?

By being electrically connected to it (or not), which was the point of
discussion.

*Hardly* !

A flux band by its very nature is loosely connected to the lams !

Graham

[Eeyore]

Phil Allison wrote:

"Prune" = makes all ordinary ****WITS look like geniuses

I see some transformers use a copper band parallel to the windings but
outside the core, in addition to the usual magnetic shielding around the
core sides or fully enclosing the transformer. My question is if it
should
be grounded or not. Here's the information I found on the web:

"There is usually also a circumferential copper shield (or 'flux band')

** This is the source -

an article on FLYBACK TRANSFORMERS !!!

http://www.national.com/nationaledge/mar04/article.html

Did you simply recognise the source ?

Amusing. Actually, the first time I used a flux band was in an smps btw.

Graham

[legg]

On Sun, 30 Jul 2006 05:20:30 +0100, Eeyore
wrote:



legg wrote:

On Sun, 30 Jul 2006 01:37:37 +0100, Eeyore
wrote:


Prune wrote:

Eeyore wrote in
:

Since that's not what it's for - probably not. But you don't really
need to ground it anyway. Why do you think that would help ?

Because of the implication of the text I quoted.

You mean " It serves primarily as a radiation shield " ?

It actually serves primarily as a magnetic shunt.

Magnetic stray flux is considered radiation,

It's stretching the point a little. A lot even !


Rughly speaking, electromagnetic radiation has an E field component
that is 'capacitice' and 'vertically polarized' and an H field which
is 'inductive' and 'horizontally polarized'.


The amount of capacitive current that might be delivered by the flux
band,

Delivered ? by the flux band ? !

and it's prefered return path either to primary or secondary,
depend on the transformer winding's layering.

How is anything going to *return* there ?

Vertically polarized (noise) voltages sources induce capacitive
current flow which will tend to return to the source's reference
terminal. An influenced branch with open circuit terminals will have
reduced current flow, but may exhibit the full impressed voltage.

Horizontally polarized (noise) current sources induce inductive
voltages which will tend to counter the strength of their enclosed
field.

The transformer is in the transmission path of the vertically
polarized component, and the coupled windings are an attempt to
concentrate and optimize the horizontal components of the source.

Oversimplified, analogified and pablified for your personal
consumption, Sir.

RL

[Eeyore]

legg wrote:

Vertically polarized (noise) voltages

WTF do you think a " vertically polarized voltage " is ?

Graham

[Eeyore]

flipper wrote:

The original poster asked "My question is if it should
be grounded or not" and that is the topic of discussion.

It really doesn't matter very much !

Graham

[Eeyore]

flipper wrote:

On Sun, 30 Jul 2006 17:58:36 +0100, Eeyore
wrote:

flipper wrote:

The original poster asked "My question is if it should
be grounded or not" and that is the topic of discussion.

It really doesn't matter very much !

It matters when you go dancing through the posts pretending to be a
know it all without knowing what the topic is.

The OP clearly had some misunderstandings about its function. Better to clarify
what it actually does IMHO.

Graham

[Eeyore]

flipper wrote:

On Sun, 30 Jul 2006 18:31:09 +0100, Eeyore
wrote:


flipper wrote:

On Sun, 30 Jul 2006 17:58:36 +0100, Eeyore
wrote:

flipper wrote:

The original poster asked "My question is if it should
be grounded or not" and that is the topic of discussion.

It really doesn't matter very much !

It matters when you go dancing through the posts pretending to be a
know it all without knowing what the topic is.

The OP clearly had some misunderstandings about its function. Better to clarify
what it actually does IMHO.

Perhaps but you asked a question

Which question did I ask ?


and then pompously declare 'hardly'
when I give you correct facts.

Which facts ?


It's great if you want to help 'clarify' something but I submit it's
easier and a lot more effective if you actually do 'clarify' instead
of playing some variation of the "I'm smart and you're stupid" ego
game.

So, in that light, why don't you go back, read the original post, and
answer the guy's question?

I have done.

Graham

[Eeyore]

flipper wrote:

On Sun, 30 Jul 2006 20:04:48 +0100, Eeyore
wrote:

flipper wrote:

and then pompously declare 'hardly'
when I give you correct facts.

Which facts ?

Oh for Pete's sakes I'm not going through it again.

Reality is, you offered no facts at all.

Graham

[Jerry]

On Sun, 30 Jul 2006 10:57:02 -0400, legg wrote:

On Sun, 30 Jul 2006 05:20:30 +0100, Eeyore
m wrote:



legg wrote:

On Sun, 30 Jul 2006 01:37:37 +0100, Eeyore
wrote:


Prune wrote:

Eeyore wrote in
:

Since that's not what it's for - probably not. But you don't really
need to ground it anyway. Why do you think that would help ?

Because of the implication of the text I quoted.

You mean " It serves primarily as a radiation shield " ?

It actually serves primarily as a magnetic shunt.

Magnetic stray flux is considered radiation,

It's stretching the point a little. A lot even !


Rughly speaking, electromagnetic radiation has an E field component
that is 'capacitice' and 'vertically polarized' and an H field which
is 'inductive' and 'horizontally polarized'.


The amount of capacitive current that might be delivered by the flux
band,

Delivered ? by the flux band ? !

and it's prefered return path either to primary or secondary,
depend on the transformer winding's layering.

How is anything going to *return* there ?

Vertically polarized (noise) voltages sources induce capacitive
current flow which will tend to return to the source's reference
terminal. An influenced branch with open circuit terminals will have
reduced current flow, but may exhibit the full impressed voltage.

Horizontally polarized (noise) current sources induce inductive
voltages which will tend to counter the strength of their enclosed
field.

The transformer is in the transmission path of the vertically
polarized component, and the coupled windings are an attempt to
concentrate and optimize the horizontal components of the source.

Oversimplified, analogified and pablified for your personal
consumption, Sir.

RL

Power transformers made for "hi-fi" amplifiers, scopes
and VTVM's usually had stray flux bands made of .010"
copper which were soldered together at the ends of the
turn. The transformers eminated odd order harmonics
from the gap areas in EI type laminations, especially if
the transformer was butt stacked rather than interleaved
1x1 or 3x3. The amount of emanation depends upon the
flux density of the unit. Some transformers used by Fisher
Radio, Harman Kardon and others had a turn of silicon
steel around the perimeter of the core. That turn would
be 90degrees to the copper turn. The ends of the silicon
steel overlapped and was taped in place.

We measured stray flux by use of a pickup coil connected
to a scope. We measure peak to peak amplitude. The
coil measurements were taken at eight places around the
unit. Excessive stray flux would affect the initial stages
of gain in an amplifier and other devices.

The grounding of the copper band was never an issue. I
have never seen one that has been soldered to a braid then
grounded to the chassis.

Jerry

[robert casey]

The grounding of the copper band was never an issue. I
have never seen one that has been soldered to a braid then
grounded to the chassis.


The copper band likely was grounded, as it touches the chassis in most
power transformer mounting methods. It was incidental, would not make a
difference grounded or not.

[Eeyore]

flipper wrote:

On Mon, 31 Jul 2006 05:07:59 +0100, Eeyore
wrote:


flipper wrote:

On Sun, 30 Jul 2006 20:04:48 +0100, Eeyore
wrote:

flipper wrote:

and then pompously declare 'hardly'
when I give you correct facts.

Which facts ?

Oh for Pete's sakes I'm not going through it again.

Reality is, you offered no facts at all.

Wrong

Perhaps you'd like to remind me of the facts you offered then ?

Graham

[Prune]

Jerry wrote in
:

Radio, Harman Kardon and others had a turn of silicon
steel around the perimeter of the core. That turn would
be 90degrees to the copper turn. The ends of the silicon
steel overlapped and was taped in place.

I have that too, except I used a nickel-based high permeability alloy (but
not mu metal as that saturates too easily). In some transformers I've seen
with both, there was insulation between the copper band and the magnetic
shield, as well as between each of these and the core.

[legg]

On Mon, 31 Jul 2006 06:32:15 -0800, Jerry wrote:

snip
How is anything going to *return* there ?

Vertically polarized (noise) voltages sources induce capacitive
current flow which will tend to return to the source's reference
terminal. An influenced branch with open circuit terminals will have
reduced current flow, but may exhibit the full impressed voltage.

Horizontally polarized (noise) current sources induce inductive
voltages which will tend to counter the strength of their enclosed
field.

The transformer is in the transmission path of the vertically
polarized component, and the coupled windings are an attempt to
concentrate and optimize the horizontal components of the source.

Oversimplified, analogified and pablified for your personal
consumption, Sir.

RL

Power transformers made for "hi-fi" amplifiers, scopes
and VTVM's usually had stray flux bands made of .010"
copper which were soldered together at the ends of the
turn. The transformers eminated odd order harmonics
from the gap areas in EI type laminations, especially if
the transformer was butt stacked rather than interleaved
1x1 or 3x3. The amount of emanation depends upon the
flux density of the unit. Some transformers used by Fisher
Radio, Harman Kardon and others had a turn of silicon
steel around the perimeter of the core. That turn would
be 90degrees to the copper turn. The ends of the silicon
steel overlapped and was taped in place.

We measured stray flux by use of a pickup coil connected
to a scope. We measure peak to peak amplitude. The
coil measurements were taken at eight places around the
unit. Excessive stray flux would affect the initial stages
of gain in an amplifier and other devices.

The grounding of the copper band was never an issue. I
have never seen one that has been soldered to a braid then
grounded to the chassis.

It depends what range of frequencies your observations include.

When dealing with 60Hz magnetics, your normal concern is the
development of outputs that show low noise on their regulated DC
outputs, with signal processing circuitry and hardware in the vicinity
that is similarly not affected by the low frequency magnetic field.
You still see artifacts of rectification at much higher frequencies,
if you look closely enough. The flux band in this case has little work
to do on capacitive feedthrough, unless it's stray RF pickup on bulky
primary connector wiring that passes straight through.

Because the magnetic fields emanate strongly from physical gaps with
well-defined field patterns and decay fairly rapidly with distance,
the orientation of the transformer with respect to victim circuits can
be critical, despite the effects of a flux band.

You will notice that your search coil shows varying sensitivity to the
stray field, dependant on its orientation to these fields. This
sensitivity to orientation becomes more uniform and predictable with
distance - and 'polarization' of the average field becomes apparent.

At higher frequencies, with waveforms carrying large dv/dt, capacitive
feedthrough can be of over-riding importance, and capacitive currents
carried by the flux ban will impress voltages across the elements in
their path of return to the source.

RL.