Simple Earth Loop problem - can someone help ?

[[email protected]]

Hi,

I have an earth loop problem (or at least a hum problem and I'm
assuing it's an earth loop issue). I have a PC that I use at home and
want to play live with (using midi keyboards to play virtual
instruments). I have the PC connected to my old Sony amp at home with
a 3.5mm jack to 5 pin DIN cable and experience no hum at all.

When I tried connecting the PC to the amp using a 3.5mm jack to phono
(cinch) lead, I had terrible mains hum. No big deal, I just carry on
using the DIN connection instead. Why do I get hum using phono
connectors and none using the DIN connector ? (all wiring is sound,
all connections good). Do DIN input circuits have resistors or
something that kills the earth loop ?

However, now I'm testing the live setup and try connecting the PC to
our PA system. In short... ARRGHH !! Mains hum in abundance !!
I've tried connecting it to the Tape-In (phono) inputs on the PA and
into a Hi-Z 1/4" jack on one of the channels - same thing - lots of
mains hum. Incidentally, the hum volume is unaffected by the gain
pots on the tape-in / channel - only the master volume makes a
difference.

I tried connecting the PC and the PA to the same electrical outlet, no
change. I've obviously got an earth loop through the shielding of the
connection between the PC and the PA.

What can I do about this (that doesn't involve buying anything !). Is
it possible to put a resistor in series with the shield ? ...or will
this degrade the sound quality / level ?

Any (cheap) ideas would be very welcome ! :-)

Many thanks,

Kev.

[DJ TecThreat]

> wrote in message
...
> Hi,
>
> I have an earth loop problem (or at least a hum problem and I'm
> assuing it's an earth loop issue). I have a PC that I use at home and
> want to play live with (using midi keyboards to play virtual
> instruments). I have the PC connected to my old Sony amp at home with
> a 3.5mm jack to 5 pin DIN cable and experience no hum at all.
>
> When I tried connecting the PC to the amp using a 3.5mm jack to phono
> (cinch) lead, I had terrible mains hum. No big deal, I just carry on
> using the DIN connection instead. Why do I get hum using phono
> connectors and none using the DIN connector ? (all wiring is sound,
> all connections good). Do DIN input circuits have resistors or
> something that kills the earth loop ?
>
> However, now I'm testing the live setup and try connecting the PC to
> our PA system. In short... ARRGHH !! Mains hum in abundance !!
> I've tried connecting it to the Tape-In (phono) inputs on the PA and
> into a Hi-Z 1/4" jack on one of the channels - same thing - lots of
> mains hum. Incidentally, the hum volume is unaffected by the gain
> pots on the tape-in / channel - only the master volume makes a
> difference.
>
> I tried connecting the PC and the PA to the same electrical outlet, no
> change. I've obviously got an earth loop through the shielding of the
> connection between the PC and the PA.
>
> What can I do about this (that doesn't involve buying anything !). Is
> it possible to put a resistor in series with the shield ? ...or will
> this degrade the sound quality / level ?
>
> Any (cheap) ideas would be very welcome ! :-)
>
> Many thanks,
>
> Kev.

Is the case of the mixer metal? If it is, take a length of decent guage
wire (16-14 guage) single conductor and loosen a screw on the mixer's frame,
wrap the wire on the screen and tighten it up then do the same on the other
end of the wire to a case screw on the computer. This _may_ relieve the
hum. I had the same problem with a rack mount server i use for lighting
control. I had to use it to do video playback, once and when i jacked it
into my dj board it hummed real bad. Grounding the machine to the mixer
fixed it. (my mixer has thumb screws on the back for grounding turntables).

DJ TecThreat
The Real Threat

[Scott Dorsey]

> wrote:
>
>When I tried connecting the PC to the amp using a 3.5mm jack to phono
>(cinch) lead, I had terrible mains hum. No big deal, I just carry on
>using the DIN connection instead. Why do I get hum using phono
>connectors and none using the DIN connector ? (all wiring is sound,
>all connections good). Do DIN input circuits have resistors or
>something that kills the earth loop ?

The DIN input circuit might be differential or have transformer
isolation.

>However, now I'm testing the live setup and try connecting the PC to
>our PA system. In short... ARRGHH !! Mains hum in abundance !!
>I've tried connecting it to the Tape-In (phono) inputs on the PA and
>into a Hi-Z 1/4" jack on one of the channels - same thing - lots of
>mains hum. Incidentally, the hum volume is unaffected by the gain
>pots on the tape-in / channel - only the master volume makes a
>difference.

So, get a DI box. That's what they are for. You can even buy some
DI boxes with built-in 1/8" plugs for computer soundcards if you're
willing to spend more money than getting a pair of IMP-2s and cables.

>I tried connecting the PC and the PA to the same electrical outlet, no
>change. I've obviously got an earth loop through the shielding of the
>connection between the PC and the PA.
>
>What can I do about this (that doesn't involve buying anything !). Is
>it possible to put a resistor in series with the shield ? ...or will
>this degrade the sound quality / level ?

You could try going into an XLR on the PA system and lifting the shield
completely. But you really, really want to balance the line if you are
going any length. The DI box is your friend.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Tim Padrick]

You don't want a DI box unless you want to run it into a mic input. If you
want to run it into a line input you want a 1:1 isolation transformer.
Scroll to the bottom of http://www.padrick.net/LiveSound/Interfacing.htm for
links to two such products (one inexpensive, one not inexpensive).

"Scott Dorsey" > wrote in message
...
> > wrote:
> >
> >When I tried connecting the PC to the amp using a 3.5mm jack to phono
> >(cinch) lead, I had terrible mains hum. No big deal, I just carry on
> >using the DIN connection instead. Why do I get hum using phono
> >connectors and none using the DIN connector ? (all wiring is sound,
> >all connections good). Do DIN input circuits have resistors or
> >something that kills the earth loop ?
>
> The DIN input circuit might be differential or have transformer
> isolation.
>
> >However, now I'm testing the live setup and try connecting the PC to
> >our PA system. In short... ARRGHH !! Mains hum in abundance !!
> >I've tried connecting it to the Tape-In (phono) inputs on the PA and
> >into a Hi-Z 1/4" jack on one of the channels - same thing - lots of
> >mains hum. Incidentally, the hum volume is unaffected by the gain
> >pots on the tape-in / channel - only the master volume makes a
> >difference.
>
> So, get a DI box. That's what they are for. You can even buy some
> DI boxes with built-in 1/8" plugs for computer soundcards if you're
> willing to spend more money than getting a pair of IMP-2s and cables.
>
> >I tried connecting the PC and the PA to the same electrical outlet, no
> >change. I've obviously got an earth loop through the shielding of the
> >connection between the PC and the PA.
> >
> >What can I do about this (that doesn't involve buying anything !). Is
> >it possible to put a resistor in series with the shield ? ...or will
> >this degrade the sound quality / level ?
>
> You could try going into an XLR on the PA system and lifting the shield
> completely. But you really, really want to balance the line if you are
> going any length. The DI box is your friend.
> --scott
>
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."

[Pooh Bear]

Tim Padrick wrote:

> You don't want a DI box unless you want to run it into a mic input. If you
> want to run it into a line input you want a 1:1 isolation transformer.
> Scroll to the bottom of http://www.padrick.net/LiveSound/Interfacing.htm for
> links to two such products (one inexpensive, one not inexpensive).

Uh ?

Since when were 'DI boxes' exclusively mic level ?

Please ignore the above advice.


Graham

[Pooh Bear]

wrote:

> Hi,
>
> I have an earth loop problem (or at least a hum problem and I'm
> assuing it's an earth loop issue). I have a PC that I use at home and
> want to play live with (using midi keyboards to play virtual
> instruments). I have the PC connected to my old Sony amp at home with
> a 3.5mm jack to 5 pin DIN cable and experience no hum at all.
>
> When I tried connecting the PC to the amp using a 3.5mm jack to phono
> (cinch) lead, I had terrible mains hum. No big deal, I just carry on
> using the DIN connection instead. Why do I get hum using phono
> connectors and none using the DIN connector ? (all wiring is sound,
> all connections good). Do DIN input circuits have resistors or
> something that kills the earth loop ?
>
> However, now I'm testing the live setup and try connecting the PC to
> our PA system. In short... ARRGHH !! Mains hum in abundance !!
> I've tried connecting it to the Tape-In (phono) inputs on the PA and
> into a Hi-Z 1/4" jack on one of the channels - same thing - lots of
> mains hum. Incidentally, the hum volume is unaffected by the gain
> pots on the tape-in / channel - only the master volume makes a
> difference.
>
> I tried connecting the PC and the PA to the same electrical outlet, no
> change. I've obviously got an earth loop through the shielding of the
> connection between the PC and the PA.
>
> What can I do about this (that doesn't involve buying anything !). Is
> it possible to put a resistor in series with the shield ? ...or will
> this degrade the sound quality / level ?
>
> Any (cheap) ideas would be very welcome ! :-)

I suspect that you're missing out some critical info.

Is the 'PC' a laptop btw ?


Graham

[[email protected]]

On Sun, 18 Jul 2004 10:38:02 +0100, Pooh Bear
> wrote:

>I suspect that you're missing out some critical info.
>
>Is the 'PC' a laptop btw ?
>
>
>Graham

Hi Graham,

No, the PC is a PC :-)

I've been experimenting a bit to see what reduces the hum etc., and I
tried disconnecting the shield on the cable from the PC to the PA (at
the PC end) - that helped a bit, but not hugely. I also tried
powering the PC with a mains cable with the earth disconnected, with
and without another heavy gauge wire connecting the chassis of the PC
to the chassis of the PA. (Before I get warned about powering gear
with no earth, I was just testing - I would never, ever use any gear
that wasn't fully earthed AND had an RCD breaker on ! I'm not a fan
of 240v up the arm !). Once again, only minor improvements - the
majority of the mains hum is still there. If it helps, I have
established that I only have to touch the outside (shield) of the
phono jack to the outside of the phono socket on the PA to get the
hum. (i.e. not actually plugging it in, just connecting the shield
only, with the other end connected to the PC).

Nothing I've tried so far has made a significant difference. Yet,
what is confusing me is how I get no hum at all when connecting the PC
to my home amplifier using the old DIN socket on the amp, yet get lots
of hum if I connect it via the phono 'tape in' sockets on the amp. I
may take a peek at the wiring in my home amp to see the route the DIN
signal takes - I think there are a couple of resistors in there the
last time I had the lid off.

I am tempted to try a DI box, but I'm not sure this is what I need...
I don't know the impedance of the line-out on the soundcard (Audigy),
but the PA manual says the Tape-in connections are "high impedance,
200mV sensitivity". I need to connect the PC to the Tape-in as it's
the only free input once the rest of the band are plugged in !

Would a DI box help here ? After all, I would be taking an unbalanced
3.5mm line-out from the soundcard to the DI box (1/4" jack ?) and then
a balanced XLR output from the DI to an unbalanced phono tape-in input
on the PA. I thought DI boxes were only for taking unbalanced Hi-Z
inputs and turning them into balanced Low-Z outputs ?
If I use a DI box to connect a Hi-Z output to a Hi-Z input, won't I
get strange results ?

BTW Graham, I'm still trying to source a tranny for that Studiomaster
Powerpack - I've had one quote to make one from the spec sheet - £100
- £120 !!! I can see it ending up in the skip sometime soon !

Cheers,

Kev.

[Michael R. Kesti]

Pooh Bear wrote:

>Tim Padrick wrote:
>
>> You don't want a DI box unless you want to run it into a mic input. If you
>> want to run it into a line input you want a 1:1 isolation transformer.
>> Scroll to the bottom of http://www.padrick.net/LiveSound/Interfacing.htm for
>> links to two such products (one inexpensive, one not inexpensive).
>
>Uh ?
>
>Since when were 'DI boxes' exclusively mic level ?

They're not.

>Please ignore the above advice.

Except that Tim's advice is valid.

A typical transformer DI box uses a transformer with a 12:1 turns ratio.
This allows it to do the impedance transformation that allows one to run
a hi-Z source down a long line, but not without the price of a signal level
drop of about 22 dB. This level drop is typically made up for by applying
the output of the DI to a mic input so that the level may be restored,
hopefully without adding too much noise.

The OP asked for a way to connect the line output of his computer's
sound card to a line input of his PA system. A 1:1 transformer is, as
Tim said, more appropriate than a DI unless the DI's output is to be
connected to a mic input.

Of course, the OP also specified that he requires a solution "that doesn't
involve buying anything" so there is, in this case, no correct answer!

--
================================================== ======================
Michael Kesti | "And like, one and one don't make
| two, one and one make one."
| - The Who, Bargain

[Chris Hornbeck]

On Sun, 18 Jul 2004 18:56:55 GMT, wrote:

> once the rest of the band are plugged in !

Chris Hornbeck

[[email protected]]

It's OK - I've found a (temporary) solution. I've taken the digital
optical-out on the soundcard to the optical-in my minidisc recorder,
connected this to the PA with a 3.5mm jack to phono lead, press record
on the minidisc (without subsequently pressing play and actually
recording), and the MD does the optical - line conversion for me, and
giving me complete isolation !! What a difference in sound quality !!
:-) :-)

It would be nice to find a proper solution to this for the long term,
but for my immediate requirements, this setup will do.

Cheers,

Kev.

[Pooh Bear]

wrote:

> It's OK - I've found a (temporary) solution. I've taken the digital
> optical-out on the soundcard to the optical-in my minidisc recorder,
> connected this to the PA with a 3.5mm jack to phono lead, press record
> on the minidisc (without subsequently pressing play and actually
> recording), and the MD does the optical - line conversion for me, and
> giving me complete isolation !! What a difference in sound quality !!
> :-) :-)
>
> It would be nice to find a proper solution to this for the long term,
> but for my immediate requirements, this setup will do.

Ah - OK.

The optical out is providing the same ( galvanic ) isolation that a DI box
would give. I.e. the ground connection is not common to the sending and
receiving equipment.

Don't fret too much over impedances and levels. Almost any audio gear has
low(ish) output impedance and medium to high input impedance. Operating
levels that aren't mic level ( so-called line ) typically may vary from
around 100mV to about a Volt. Any respectable DI box designer knows this
and all should work fine in your application.


Graham

[Tim Padrick]

Were are back to my original post in which (via a link) I suggested that you
need http://www.jensen-transformers.com/ci2rr.html or
http://www.ebtechaudio.com/he-2des.html

"Pooh Bear" > wrote in message
...
> wrote:
>
> > It's OK - I've found a (temporary) solution. I've taken the digital
> > optical-out on the soundcard to the optical-in my minidisc recorder,
> > connected this to the PA with a 3.5mm jack to phono lead, press record
> > on the minidisc (without subsequently pressing play and actually
> > recording), and the MD does the optical - line conversion for me, and
> > giving me complete isolation !! What a difference in sound quality !!
> > :-) :-)
> >
> > It would be nice to find a proper solution to this for the long term,
> > but for my immediate requirements, this setup will do.
>
> Ah - OK.
>
> The optical out is providing the same ( galvanic ) isolation that a DI box
> would give. I.e. the ground connection is not common to the sending and
> receiving equipment.
>
> Don't fret too much over impedances and levels. Almost any audio gear has
> low(ish) output impedance and medium to high input impedance. Operating
> levels that aren't mic level ( so-called line ) typically may vary from
> around 100mV to about a Volt. Any respectable DI box designer knows this
> and all should work fine in your application.
>
>
> Graham
>

[Ralph Staub]

Pooh Bear wrote:

> Ah - OK.
>
> The optical out is providing the same ( galvanic ) isolation that a DI box
> would give. I.e. the ground connection is not common to the sending and
> receiving equipment.

Correct.
>
> Don't fret too much over impedances and levels. Almost any audio gear has
> low(ish) output impedance and medium to high input impedance.

Also Correct.

> Operating
> levels that aren't mic level ( so-called line ) typically may vary from
> around 100mV to about a Volt. Any respectable DI box designer knows this
> and all should work fine in your application.

Painfully incorrect. Respectable DI box designers know that there's
always someone who will buy their product and hook it up incorrectly.
All Passive (the kind without an amplifier in them) DI's will reduce
gain by about 20dB or so. A DI receives (up to) line level input and
produces an output much lower than line level. When plugged in to a line
input, the signal will be quiet and distant (much like you should be on
this topic, Graham). A 1:1 transformer is much like DI, but is designed
to connect line level outputs to line level inputs as the OP described.
It's no more expensive than a DI, no harder to find than a DI, and is
clearly the correct choice for the OP's connection, assuming he needs a
transformer at all (see http://www.rane.com/note110.html ).

Graham, I wish you'd quit telling people to do things incorrectly, then
telling them to ignore proper advice, like Tim Padrick gave. What's yer'
friggin' problem, anyway?

Ralph
Ralph

[Pooh Bear]

Ralph Staub wrote:

> Pooh Bear wrote:
>
> > Ah - OK.
> >
> > The optical out is providing the same ( galvanic ) isolation that a DI box
> > would give. I.e. the ground connection is not common to the sending and
> > receiving equipment.
>
> Correct.
> >
> > Don't fret too much over impedances and levels. Almost any audio gear has
> > low(ish) output impedance and medium to high input impedance.
>
> Also Correct.
>
> > Operating
> > levels that aren't mic level ( so-called line ) typically may vary from
> > around 100mV to about a Volt. Any respectable DI box designer knows this
> > and all should work fine in your application.
>
> Painfully incorrect. Respectable DI box designers know that there's
> always someone who will buy their product and hook it up incorrectly.
> All Passive (the kind without an amplifier in them) DI's will reduce
> gain by about 20dB or so. A DI receives (up to) line level input and
> produces an output much lower than line level. When plugged in to a line
> input, the signal will be quiet and distant (much like you should be on
> this topic, Graham). A 1:1 transformer is much like DI, but is designed
> to connect line level outputs to line level inputs as the OP described.
> It's no more expensive than a DI, no harder to find than a DI, and is
> clearly the correct choice for the OP's connection, assuming he needs a
> transformer at all (see http://www.rane.com/note110.html ).

A. He would appear to need a transformer since it was galvanic isolation that
cured the problem ( as you just agreed ! ).

B. Referring ppl to tech pages like Ranes may well be technically accurate
but the OP didn't indicate any great desire to start getting involved in
complicated solutions. It's all very well for an audio tech to say " get a 1:1
transfomer " when you know where to find these things. I disagree with your " no
harder to find " assertion and you won't to my knowledge easily find a ready
boxed and ready to go 1:1 transformer.

C. My bet is that a simple DI box will provide an adequate solution in this
application. The OP hasn't specced levels and gain adjust but IME a DI is likely
to work fine. It's certainly worth trying before searching for a more time
consuming solution.


> Graham, I wish you'd quit telling people to do things incorrectly, then
> telling them to ignore proper advice, like Tim Padrick gave.

There is nothing 'improper' about checking out a simple DI box to see if it
fixes the problem. Don't overcomplicate the issue.

> What's yer'
> friggin' problem, anyway?

I don't have one. What's yours ?


Graham

[Ralph Staub]

Pooh Bear wrote:
> A. He would appear to need a transformer since it was galvanic isolation that
> cured the problem ( as you just agreed ! ).

Yes, perhaps, but not a 12:1 transformer, a 1:1 transformer.

> B. Referring ppl to tech pages like Ranes may well be technically accurate
> but the OP didn't indicate any great desire to start getting involved in
> complicated solutions.

Had you read the links, you'd know they offered a variety of
transformerless solutions that cost nothing but a little time and solder
because the OP didn't want to buy anything.

> It's all very well for an audio tech to say " get a 1:1
> transfomer " when you know where to find these things. I disagree with your " no
> harder to find " assertion and you won't to my knowledge easily find a ready
> boxed and ready to go 1:1 transformer.

Same place you'd find a DI; your local music store or sound shop. Mine
are a whirlwind product, but many others provide the same thing. You
obviously don't have any in your stock since you don't know where to
find them or what their proper usage is. Albeit a weak solution, here's
a link that's just right for you...
http://www.radioshack.com/product.asp?catalog%5Fname=CTLG&category%5Fname=CTLG%5F011%5F002%5F016%5F000&product%5Fid=273%2D1374
That was so tough to find...

> C. My bet is that a simple DI box will provide an adequate solution in this
> application. The OP hasn't specced levels and gain adjust but IME a DI is likely
> to work fine. It's certainly worth trying before searching for a more time
> consuming solution.

"If" he's got gain to make up for what he looses, then yes. The best
your solution offers is a maybe. Why would you dis Tim's more
appropriate solution when you don't know anything about it?

> There is nothing 'improper' about checking out a simple DI box to see if it
> fixes the problem. Don't overcomplicate the issue.

You don't use a DI to drive a Line level input. That's improper. I'm
sorry that's too complex for you.

>>What's yer'
>>friggin' problem, anyway?
>
>
> I don't have one. What's yours ?

People that give poor advice and dis proper advice.

Ralph

[sycochkn]

I think the solution he found is the best one.

Bob

"Ralph Staub" > wrote in message
...
> Pooh Bear wrote:
> > A. He would appear to need a transformer since it was galvanic
isolation that
> > cured the problem ( as you just agreed ! ).
>
> Yes, perhaps, but not a 12:1 transformer, a 1:1 transformer.
>
> > B. Referring ppl to tech pages like Ranes may well be technically
accurate
> > but the OP didn't indicate any great desire to start getting involved in
> > complicated solutions.
>
> Had you read the links, you'd know they offered a variety of
> transformerless solutions that cost nothing but a little time and solder
> because the OP didn't want to buy anything.
>
> > It's all very well for an audio tech to say " get a 1:1
> > transfomer " when you know where to find these things. I disagree with
your " no
> > harder to find " assertion and you won't to my knowledge easily find a
ready
> > boxed and ready to go 1:1 transformer.
>
> Same place you'd find a DI; your local music store or sound shop. Mine
> are a whirlwind product, but many others provide the same thing. You
> obviously don't have any in your stock since you don't know where to
> find them or what their proper usage is. Albeit a weak solution, here's
> a link that's just right for you...
>
http://www.radioshack.com/product.asp?catalog%5Fname=CTLG&category%5Fname=CTLG%5F011%5F002%5F016%5F000&product%5Fid=273%2D1374
> That was so tough to find...
>
> > C. My bet is that a simple DI box will provide an adequate solution
in this
> > application. The OP hasn't specced levels and gain adjust but IME a DI
is likely
> > to work fine. It's certainly worth trying before searching for a more
time
> > consuming solution.
>
> "If" he's got gain to make up for what he looses, then yes. The best
> your solution offers is a maybe. Why would you dis Tim's more
> appropriate solution when you don't know anything about it?
>
> > There is nothing 'improper' about checking out a simple DI box to see if
it
> > fixes the problem. Don't overcomplicate the issue.
>
> You don't use a DI to drive a Line level input. That's improper. I'm
> sorry that's too complex for you.
>
> >>What's yer'
> >>friggin' problem, anyway?
> >
> >
> > I don't have one. What's yours ?
>
> People that give poor advice and dis proper advice.
>
> Ralph
>

[Pooh Bear]

Ralph Staub wrote:

> You don't use a DI to drive a Line level input. That's improper.

That's ********.

> I'm
> sorry that's too complex for you.

Depends on your personal definition of a 'DI'.

Suggesting that 'DIs' are mic level only is a false generalisation and a misuse of the term.



Graham

[Ralph Staub]

Pooh Bear wrote:
>
> That's ********.
>
>

> Depends on your personal definition of a 'DI'.
>
> Suggesting that 'DIs' are mic level only is a false generalisation and a misuse of the term.

I'm not suggesting anything, only stating fact. A passive DI does not
generate a line level output without a significantly greater than line
level (usually more than +20dBu) input. If you've got +20dBu going in,
you MIGHT get 0dBu put if your lucky. If you have an input that handles
a -20 to -30dBu, (that's instrument level, not line level, BTW) then and
only then it has a chance.