How to read isolated 4-20 mA

Thread Starter

avijit36

Joined Jul 22, 2011
4
I have to read 4-20 mA output from six water pressure transmitters and six water flow sensors. As, in case of flow sensors (sink type) this mA output is not flowing towards ground, I cannot just drop the current in a 250 ohm resister and measure the voltage (1-5 v) across it in respect of ground.

I made the attached circuit and when tested individually (either only PT or only FS) it gives linear and correct output voltage at Op-AMP (OP 07) o/p (with 24v GND and 9v GND isolated). But whenever I tried to use the both (PT and FS), the digital displays showed erroneous results.

What may be the reason and how to recover it?
 

Attachments

ErnieM

Joined Apr 24, 2011
8,377
You did not connect the 24V and 9V return but use them both in the same circuit, so they will find a common connection; this is going to be thru the 4 resistors on the + amp inputs: with the FS shunt sitting up around 24V and the PT shunt around 5V (full scale), then the difference of these voltages is where the 9V return sits.

The current going into these resistors is adding to both FS and PT sensor current.

The common mode voltage at the FS amp is 12V and thus outside the amp's operational common mode voltage range.

You can check this by running separate 9V supplies for each amp.

A solution to all this is to use an instrumentation amplifier which has a much greater common mode voltage range.

I'd also look into connecting these returns somewhere unless you absolutely need to isolate them: they are not as isolated as you may think.

 

John P

Joined Oct 14, 2008
2,025
If you've got a power supply on the instrument side, you can build a circuit with voltage-to-frequency converters and send the data via optocouplers. Or you could put in a microntroller and send the data over a single channel as packets. I don't think you'll ever get anywhere with entirely analog circuits if the grounds aren't connected.
 
Top