Noise pickup in audio transformer

[cdh7]

Hi,

I'm currently working on an audio circuit that seems to be picking up noise in an isolation transformer and hope to get some good advice from you all. It's not my design and the circuit board is complete, so I need a good "band-aid" fix.

An on-board op-amp amplifies a 1Vpp audio signal a bit, then sends it through a DC-blocking capacitor to a 600 ohm isolation transformer. The other end of the transformer goes through about 30 cm of cable to a commercial audio amplifier. I don't have any information about the amplifier, but I guess it has a relatively high input impedance. As the amplifier is intended for use in an automotive environment, I suspect it also has balanced inputs for noise suppression.

The audio circuit is on the same board as a single board computer, so there are noise problems. The audio circuitry itself is all okay and sounds nice, but the isolation transformer is located above a ground pin on the connector that powers the computer. I believe the digital ground currents from the PC are magnetically coupling into the transformer and making that typical PC squealing noise on the audio output.

I built a small test amplifier with a pair of headphones to listen to different points in the circuit. On the PCB before the transformer, I verified the sound is good. After the transformer though, I hear noise.

Are my suspicions correct that the squealing sound comes from inductive coupling of these ground currents, and is there any easy way to improve the situation? Could this be some kind of impedance matching issue between the transformer and the commercial amplifier?

I'm attaching a schematic showing the setup. Thanks for any tips!

Regards,
Chris

 

[FvM]

To achieve high dynamic, an audio transformer needs a magnetic shield, typically a box made of a high permeability
material, e.g. "µ-metal". If your device hasn't an effective shield, it can't be used in an enviroment of interfering AC magnetic
fields. The most severe problem are mains transformers, but I guess pulsating ground currents generated by a car alternator
or ignition system can be bad as well.

If the amplifier already has a differential input, you most likely get better performance without a transformer.

A box of ordinary sheet steel (e.g. a Teko RF enclosure) has a certain magnetic shielding effect, but it's poor compared to true
magnetic shielding material.