Return to Answer

There are special analog optocouplers exactly for such purposes you are decribing.

For example see the circuit with an IL300 below.

Such optocouplers have not only one but two photodiodes that are illuminated by the same LED.
The reason for two photodiodes is to use one for a feed-back-loop on the transmitter side that takes care that the photocurrent \$I_{P1}\$ is proportional to the input voltage \$V_{in}\$. The second photodiode on the receiver side will have a photocurrent \$I_{P2}\$ that is proportional to the first one because it is illuminated by the same LED. The photo current on the receiver side is transformed into a voltage signal \$V_{out}\$ by a TIA.
This way an analog voltage signal can be transfered with high linearity across a galvanic isolation.

There are special analog optocouplers exactly for such purposes you are decribing.

For example see the circuit with an IL300 below.

Such optocouplers have not only one but two photodiodes that are illuminated by the same LED.
The reason for two photodiodes is to use one for a feed-back-loop on the transmitter side that takes care that the photocurrent \$I_{P1}\$ is proportional to the input voltage \$V_{in}\$. The second photodiode on the receiver side will have a photocurrent \$I_{P2}\$ that is proportional to the first one because it is illuminated by the same LED. The photo current on the receiver side is transformed into a voltage signal \$V_{out}\$ by a TIA.
This way an analog voltage signal can be transfered with high linearity across a galvanic isolation.

There are also other methods, e.g. using a

• voltage to PWM converter or a
• voltage to frequency converter

and

• a common optocoupler or
• a transformer

to transfer the digital signal to the isolated side and convert it back to a voltage signal.

There are special analog optocouplers exactly for such purposes you are decribing.

For example see the circuit with an IL300 below.

Such optocouplers have not only one but two photodiodes that are illuminated by the same LED.
The reason for two photodiodes is to use one for a feed-back-loop on the transmitter side that takes care that the photocurrent \$I_{P1}\$ is proportional to the input voltage \$V_{in}\$. The second photodiode on the receiver side will have a photocurrent \$I_{P2}\$ that is proportional to the first one because it is illuminated by the same LED. The photo current on the receiver side is transformed into a voltage signal by a TIA.
This way an analog voltage signal can be transfered with high linearity across a galvanic isolation.

There are special analog optocouplers exactly for such purposes you are decribing.

For example see the circuit with an IL300 below.

Such optocouplers have not only one but two photodiodes that are illuminated by the same LED.
The reason for two photodiodes is to use one for a feed-back-loop on the transmitter side that takes care that the photocurrent \$I_{P1}\$ is proportional to the input voltage \$V_{in}\$. The second photodiode on the receiver side will have a photocurrent \$I_{P2}\$ that is proportional to the first one because it is illuminated by the same LED. The photo current on the receiver side is transformed into a voltage signal \$V_{out}\$ by a TIA.
This way an analog voltage signal can be transfered with high linearity across a galvanic isolation.

There are special analog optocouplers exactly for such purposes you are decribing.

They that have not only one but two photodiodes that are illuminated by the same LED.
The reason for two photodiodes is use one for a feed-back-loop on the transmitter side that takes care that the photocurrent is proportional to the input voltage. The second photodiode on the receiver side will have a photocurrent that is proportional to the first one because it is illuminated by the same LED. This way an analog signal can be transfered with high linearity.

An example is the IL300. See e.g. following circuit:

There are special analog optocouplers exactly for such purposes you are decribing.

For example see the circuit with an IL300 below.

Such optocouplers have not only one but two photodiodes that are illuminated by the same LED.
The reason for two photodiodes is to use one for a feed-back-loop on the transmitter side that takes care that the photocurrent \$I_{P1}\$ is proportional to the input voltage \$V_{in}\$. The second photodiode on the receiver side will have a photocurrent \$I_{P2}\$ that is proportional to the first one because it is illuminated by the same LED. The photo current on the receiver side is transformed into a voltage signal by a TIA. 
This way an analog voltage signal can be transfered with high linearity across a galvanic isolation.