[LAU] ground loop hell

Roger gurusonic at gmail.com
Mon Feb 16 00:50:11 UTC 2015


On 11/02/15 13:58, David Christensen wrote:
> On 02/10/2015 05:43 PM, Len Ovens wrote:
>> As an aside... I don't think I will need it at this point, The isolation
>> boxes that are easy to get are "DI" boxes tha go from high impedance
>> unball to low impedance bal. Would these still work to go from a
>> ballanced output to an unballanced input?
>
> DI boxes are designed to connect an unbalanced output to a balanced 
> input.  I've also pondered running DI boxes "backwards". From a 
> technical view, I'd need to study the schematics of the driving 
> equipment, the DI box, and the driven equipment to convince myself 
> that it could work and not damage anything.  I might try calling 
> technical support.  From a execution view, I'd need the right adapter 
> cables (such as XLR-F to XLR-F).  But from a conservative/ safety 
> view, I'd rather hook everything up the way the equipment 
> manufacturers tell me to and not have to worry about it.
>
>
>> Or do I need a one to one
>> isolator? Or does it depend on the powered monitor?
>
> I'd look in the monitor manual and follow the manufacturer's 
> instructions.  If you don't have a manual and can't get direction from 
> the manufacturer, I'd use an isolation transformer with suitable 
> connectors and cables.
>
>
>> How does this one look: http://www.rolls.com/pdf/M_HE18.pdf
>
> That looks useful.
>
>
>> Having "solved" my problem with a single power source, would it still be
>> worth while isolating all stage amps from the mixer?
>
> If the mixer has balanced outputs, the stage amplifiers have balanced 
> inputs, and both are powered from the same supply, isolation 
> transformers should not be necessary.  I'd run balanced cables from 
> the mixer to the amps and call it good.  That said, good isolation 
> transformers will cause a small lose of signal and fidelity, but they 
> might provide protection for certain equipment failures or operator 
> errors.
>
>
>> I think I will run
>> ballanced audio through the snake and do the cable adapting at the stage
>> end anyway.
>
> It is best to keep balanced signals in balanced cables for as far as 
> practical.
>
>
>> Right now there is a TS plugged directly into a TRS mixer
>> output which means the signal minus is grounded.
>
> I wouldn't plug a TS cable into an balanced TRS output.  First, you 
> are shorting one half of the balanced driver output circuit. Second, 
> you are injecting chassis ground noise into the audio circuits.
Agreed. If you run balanced -> unbalanced, it's probably best to run 
shielded twisted core from the output and just leave the shield 
disconnected at the input. It's discussed in those Rane docs I posted.


More information about the Linux-audio-user mailing list