optical isolation transformer for audio

[kambule]

I just read about the effect of bad input grounding on hum.

And I read that an isolation transformer (copper + iron core) helps to reduce to the hum caused by bad connecting arrangement between two equipment.

Would anyone tell if it is possible to use optical isolator (LED + light sensitive diode) to achieve a similar result of an ordinary isolation transformer ?? I think linearlity is a major issue here.



thanks


kb

 

[Pano]

I think linearlity is a major issue here.

I believe you are right. And noise, and maybe dynamic range.

It seems like such a good idea that if it were practicle, it would be done more often.

FWIW, every movie theater has optical pickups in the projectors. But they use a constant light source. That source has always been a normal filiment lamp (bulb). But now changing over to red LED. So I guess that the optocoupler may not work well because of the light source, not the pickup.

 

[Nordic]

The exception would be transfer of music in the digital domain...

 

[Geoff H]

Another way of breaking the earth loop is to place 10 ohm resisters on the signal earth leads at the input to the power amp and preamp.

Has no effect on audio but the heavy earth current flows back through the mains wiring, not the signal wiring.

Linearity is a big issue with opto coupling. Tried it a few years ago. Telephone quality.

 

[dangus]

There are "isolation amplifers" that use capacitive coupling between inputs and outputs. I don't know how suitable they are for audio... I've used them in equipment but never thought to send audio through and listen to them. Check out the Burr-Brown ISO120 family.

But for the original purpose, better grounding and maybe balanced inputs would be a better solution.

 

[Bgt]

`just use optical Spdiff, than a DAC and voila...isolated audio

 

[Geoff H]

Well that would help. Some of us don't want to convert analogue to digital. Each conversion process adds distortion to the signal. (do manufactures quoted distortion levels at 10Khz?)

The problem starts when several devices are interconnected, resulting in several earth loops, rather than earthing to a single point. So we need to fix the problem, not find a work arround.

IF you were certain the power transformers were double insulated, it would be OK to remove the earth from the devices. BUT do not do this unless the power transformers (or supplies) ARE double insulated, tested, and you are qualified to do the modification. An error could be lethal.

 

[chrispenycate]

Simply balancing inputs (active balancing, and it doesn't have to be too precise) can greatly improve the situation; balance the drive amplifiers and use decent balanced cable, and you can send the stuff kilometers. Even dynamic microphones (microvolt levels) into a balanced input can be used with tens of meters of cable in an environment which includes thyristor dimmers, and practically total rejection. Of course, a differential input will never be quite as quiet as a single ended, but except for turntable inputs, good, low noise op-amps generally give adequate performance without needing esoteric, boost before comparing circuits; indeed, in many commercially available preamps they are essentially using a differential input, comparing wit local ground.

 

[Bgt]

Originally posted by Geoff H Well that would help. Some of us don't want to convert analogue to digital. Each conversion process adds distortion to the signal. (do manufactures quoted distortion levels at 10Khz?)

If you have all digital Spdiff optical outputs f.i. PC/CDP/Digital recorder, than it is very easy to do. Converting analoque to digital and than back again is no option.