Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

How to optically couple Audio signal to an isolated circuit.

Status
Not open for further replies.

AndreyG

Full Member level 4
Joined
May 10, 2010
Messages
196
Helped
27
Reputation
54
Reaction score
27
Trophy points
1,308
Location
Vancouver, Canada
Activity points
2,946
Here is challenging little problem:

I have a circuit (current source) which I want to drive with combination of four independent analog signals. Sources of signals will be sound cards. The frequencies will be very low - fraction of Hz.

Because they are four independent sources I want to isolate the signals. I want to use optocouplers.
I want to make audio signal to drive optocoupler LED. And nothing else on the primary side of coupling circuit.
So far I cannot figure out how to do it. There are several complications: coupling should be reasonably linear, incoming signal is bipolar and I do not want to have to add power sources to power up primary side of coupling circuit (I have four such circuits which would mean 4 isolated power supplies – too much!) And the circuit should be inexpensive.

I discovered isolating amplifiers by Analog Devices that do all that I need. They are linear, work with bipolar signal, generate power on primary side from secondary side. But they cost upwards of $80 a piece (I need 4).

There are isolating amplifiers at $6 a piece but all require isolated power supply on primary side.

I discovered highly linear arrangements with opto couplers made with two integrated photo diodes providing feedback on primary side. Very neat but requires power supply on primary side.

I can connect LED’s of two optocouplers oriented oppositely and combine outputs from corresponding photo transistors – this will give me bipolar response to bipolar input. This does not require power supply on primary side. But I have a gap (dead zone) in the middle from ~-1 to ~+1Volts when diodes do not conduct.

The latest approach comes the closest to what I want. I can leave with some non-linearity no problem. The question is: how to eliminate the dead zone around zero when diodes do not conduct?
 

Can the atypical sound card deliver sub-sonic sound?

Use a battery power supply.

For the LEDs, forward bias them..................
 

>>Can the atypical sound card deliver sub-sonic sound?
PC sound card does. iPhone sound card - no.

>>Use a battery power supply.
This will require 4 batteries. Should be a way to do without
 

You need to get some power to the primary side and there aren't many options for that without using some form of power supply from the secondary to the primary. You could make a simple DC-DC converter with four separate output windings on the transformer.

Edit: Perhaps something like this using a transformer with four identical isolated outputs.
 
Last edited:

I have done this before using linear opto-couplers from HP and a small DC-DC converter (NME0505 I think) for each channel. You don't need a bipolar input, just bias the LED side to half conduction and let the signal add or subtract from that condition.

Brian.
 
If you need decent performance then you'll need active circuitry for both sides, which requires a power supply. I would recommend sigma delta modulation, or using linearized photodiode array.
 

If they're PC soundcards connected to PCs, then you can use power from the PCs. There's almost always at least one spare MOLEX connector inside a PC, that has connections for +5V, +12V and ground. (Ground relative to that PC and it's soundcard)
 

Thank you everybody for answering my questions. This gave me breadth of ideas.

Andrey
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top