Help with mystery wire from AC shielded tranformer

[Don Pearce]

On Sat, 07 Apr 2007 13:22:53 -0400, wrote:

>
>Hi all,
>
>I found a few lower power shielded transformers at a flea market, and
>they do 120 to 40 CT VAC, so perfect for small bipolar power
>supplies.. They're steel-shelled with 4 mounting screws on the bottom
>and the transformer is potted into the shell, so I can't get a look at
>it without destroying it.
>
>There's 6 wires coming out of it, the 5 identified and a mysterious
>grey one which is insulated from all the others ( inf. ohms) but reads
>5 different voltages ranging from 2.8 - 65 VAC when measured against
>each of the other ones. This one and all the others are insulatd from
>the case, so its not a case ground wire.
>
>So:
>
>1. Could this wire connect to an internal shield to minimize
>capacitive coupling or is it something else?
>
>2. If the shield, are these voltage readings normal for such a
>connection?
>
>3. If the shield, should I connect this wire directly to AC earth
>ground, or to chassis ground?
>
>I'll just ignore it if I can't figure it out, but if it is some kind
>of shielding it would be nice to take advantage of it properly.
>
>Thanks for any help.
>
>u

That is certainly a connection to a foil shield between primary and
secondary. It needs to connect to a decent signal ground. If your
layout follows normal practice, the speaker ground will probably be
the most convenient point.

d

--
Pearce Consulting
http://www.pearce.uk.com

[[email protected]]

Hi all,

I found a few lower power shielded transformers at a flea market, and
they do 120 to 40 CT VAC, so perfect for small bipolar power
supplies.. They're steel-shelled with 4 mounting screws on the bottom
and the transformer is potted into the shell, so I can't get a look at
it without destroying it.

There's 6 wires coming out of it, the 5 identified and a mysterious
grey one which is insulated from all the others ( inf. ohms) but reads
5 different voltages ranging from 2.8 - 65 VAC when measured against
each of the other ones. This one and all the others are insulatd from
the case, so its not a case ground wire.

So:

1. Could this wire connect to an internal shield to minimize
capacitive coupling or is it something else?

2. If the shield, are these voltage readings normal for such a
connection?

3. If the shield, should I connect this wire directly to AC earth
ground, or to chassis ground?

I'll just ignore it if I can't figure it out, but if it is some kind
of shielding it would be nice to take advantage of it properly.

Thanks for any help.

u

[Richard Crowley]

> wrote ...
> I found a few lower power shielded transformers at a flea market, and
> they do 120 to 40 CT VAC, so perfect for small bipolar power
> supplies.. They're steel-shelled with 4 mounting screws on the bottom
> and the transformer is potted into the shell, so I can't get a look at
> it without destroying it.
>
> There's 6 wires coming out of it, the 5 identified and a mysterious
> grey one which is insulated from all the others ( inf. ohms) but reads
> 5 different voltages ranging from 2.8 - 65 VAC when measured against
> each of the other ones. This one and all the others are insulatd from
> the case, so its not a case ground wire.
>
> So:
>
> 1. Could this wire connect to an internal shield to minimize
> capacitive coupling or is it something else?

Seems likely, unless it is a (broken) tap on one of the
other windings?

> 2. If the shield, are these voltage readings normal for such a
> connection?

How are you measuring it? If it is with a high-impedance
meter (which is likely) then no telling what kind of capacitive
coupling you may be getting. Measure it with a 1K ohm
"load" and see what kind of voltages you are getting.

> 3. If the shield, should I connect this wire directly to AC earth
> ground, or to chassis ground?

I would try connecting to ground if it actually makes
an improvement, otherwise, just let it float.

[[email protected]]

>
>That is certainly a connection to a foil shield between primary and
>secondary. It needs to connect to a decent signal ground. If your
>layout follows normal practice, the speaker ground will probably be
>the most convenient point.
>
>d

Thanks Don. Pardon my ignorance, but would connecting it to the DC
ground of the system it's powering not "contaminate" that ground?
That's why I was considering connecting it to AC earth.

One of the projects I'm considering using it for is a stand-alone
self-powered subwoofer preamp/LPfilter. In that context, would it
still be best to connect to the DC ground?

TIA

u

[Don Pearce]

On Sat, 07 Apr 2007 13:54:09 -0400, wrote:

>
>>
>>That is certainly a connection to a foil shield between primary and
>>secondary. It needs to connect to a decent signal ground. If your
>>layout follows normal practice, the speaker ground will probably be
>>the most convenient point.
>>
>>d
>
>Thanks Don. Pardon my ignorance, but would connecting it to the DC
>ground of the system it's powering not "contaminate" that ground?
>That's why I was considering connecting it to AC earth.
>

No. What you are doing is forcing the foil to be electrically quiet
with respect to the signal ground, which in the end is all that
matters.

>One of the projects I'm considering using it for is a stand-alone
>self-powered subwoofer preamp/LPfilter. In that context, would it
>still be best to connect to the DC ground?
>
>TIA
>
>u
Still the same, but I am surprised you are considering this for such a
project. Subwoofers are ideal territory for switching power supplies.
You won't be able to make a very beefy one with just 45 watts
available.

d

--
Pearce Consulting
http://www.pearce.uk.com

[Scott Dorsey]

> wrote:
>
>There's 6 wires coming out of it, the 5 identified and a mysterious
>grey one which is insulated from all the others ( inf. ohms) but reads
>5 different voltages ranging from 2.8 - 65 VAC when measured against
>each of the other ones. This one and all the others are insulatd from
>the case, so its not a case ground wire.

Measured with WHAT? A high-impedance VTVM or a 1Kohm/volt multimeter?

>So:
>
>1. Could this wire connect to an internal shield to minimize
>capacitive coupling or is it something else?

Yes, it could be, especially if these came out of medical equipment.
Are the two windings split on either side of a divided bobbin, or are
they wrapped one on top of the other like most filament transformers?

>2. If the shield, are these voltage readings normal for such a
>connection?

Depends on what you measured it with. If you measured it with a mechanical
meter, it's not normal at all. If you measured it with a DVM that can see
tiny leakage voltages, it is.

>3. If the shield, should I connect this wire directly to AC earth
>ground, or to chassis ground?

THIS is a hard question to answer, because both of those grounds should be
the same. There is a discussion in Henry Ott's book on interference.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

>>One of the projects I'm considering using it for is a stand-alone
>>self-powered subwoofer preamp/LPfilter. In that context, would it
>>still be best to connect to the DC ground?
>>
>>TIA
>>
>>u
>Still the same, but I am surprised you are considering this for such a
>project. Subwoofers are ideal territory for switching power supplies.
>You won't be able to make a very beefy one with just 45 watts
>available.
>
>d

Perhaps I haven't explained myself properly. I'm planning on using
this transformer to power a PRE-amplifier and low-pass filter (a
rather small circuit consuming less than 200ma). I take the output
from my sound card to this circuit, and then to stand-alone amp and
speaker. So I'm not powering the amp from this transformer.

Meaning that the xformer's shield would connect to signal ground at
the bipolar power supply's output's ground, which is the defacto
signal ground. Correct?

Thanks for you patience, knowledge and time.

u

[[email protected]]

>> 1. Could this wire connect to an internal shield to minimize
>> capacitive coupling or is it something else?
>
>Seems likely, unless it is a (broken) tap on one of the
>other windings?

I've tested 3 of transformers and they're all the same.

>> 2. If the shield, are these voltage readings normal for such a
>> connection?
>
>How are you measuring it? If it is with a high-impedance
>meter (which is likely) then no telling what kind of capacitive
>coupling you may be getting. Measure it with a 1K ohm
>"load" and see what kind of voltages you are getting.

Yep you were right, thanks. I used a Hi-Z VOM and I suspected that
these currents were in fact tiny. Measuring across a load I got 0 VAC
accross all the wires and the grey one..

>
>> 3. If the shield, should I connect this wire directly to AC earth
>> ground, or to chassis ground?
>
>I would try connecting to ground if it actually makes
>an improvement, otherwise, just let it float.

Thought there might be a standard way to connect this shield. I'll
experiment and see.

Thanks for the tips!

u

[Don Pearce]

On Sat, 07 Apr 2007 16:03:29 -0400, wrote:

>
>>>One of the projects I'm considering using it for is a stand-alone
>>>self-powered subwoofer preamp/LPfilter. In that context, would it
>>>still be best to connect to the DC ground?
>>>
>>>TIA
>>>
>>>u
>>Still the same, but I am surprised you are considering this for such a
>>project. Subwoofers are ideal territory for switching power supplies.
>>You won't be able to make a very beefy one with just 45 watts
>>available.
>>
>>d
>
>Perhaps I haven't explained myself properly. I'm planning on using
>this transformer to power a PRE-amplifier and low-pass filter (a
>rather small circuit consuming less than 200ma). I take the output
>from my sound card to this circuit, and then to stand-alone amp and
>speaker. So I'm not powering the amp from this transformer.
>
>Meaning that the xformer's shield would connect to signal ground at
>the bipolar power supply's output's ground, which is the defacto
>signal ground. Correct?
>
>Thanks for you patience, knowledge and time.
>
>u
>

Sorry - got my wires crossed. That sounds fine. Go ahead and use the
signal ground as I suggested (but not the speaker, clearly!). If you
are using star point grounding for minimum hum and noise, ground it
there.

d

--
Pearce Consulting
http://www.pearce.uk.com

[Mark]

On Apr 8, 1:21 am, (Don Pearce) wrote:
> On Sat, 07 Apr 2007 16:03:29 -0400, wrote:
>
> >>>One of the projects I'm considering using it for is a stand-alone
> >>>self-powered subwoofer preamp/LPfilter. In that context, would it
> >>>still be best to connect to the DC ground?
>
> >>>TIA
>
> >>>u
> >>Still the same, but I am surprised you are considering this for such a
> >>project. Subwoofers are ideal territory for switching power supplies.
> >>You won't be able to make a very beefy one with just 45 watts
> >>available.
>
> >>d
>
> >Perhaps I haven't explained myself properly. I'm planning on using
> >this transformer to power a PRE-amplifier and low-pass filter (a
> >rather small circuit consuming less than 200ma). I take the output
> >from my sound card to this circuit, and then to stand-alone amp and
> >speaker. So I'm not powering the amp from this transformer.
>
> >Meaning that the xformer's shield would connect to signal ground at
> >the bipolar power supply's output's ground, which is the defacto
> >signal ground. Correct?
>
> >Thanks for you patience, knowledge and time.
>
> >u
>
> Sorry - got my wires crossed. That sounds fine. Go ahead and use the
> signal ground as I suggested (but not the speaker, clearly!). If you
> are using star point grounding for minimum hum and noise, ground it
> there.
>
> d
>
> --
> Pearce Consultinghttp://www.pearce.uk.com- Hide quoted text -
>
> - Show quoted text -

as far as the audio goes, it won't make a big difference which ground
you connect it to or even leave it floating.

But for saftey reasons, I'd connect it to the GREEN wire saftey ground
i.e. the AC line cord ground. If the transformer should fail, it may
short the AC winding to the internal shield and you are best off then
if that is connected to the saftey ground.

Mark