Reading micro-amp current output of a sensor

[justind000]

Hello,

I have a handmade dissolved oxygen sensor, as described here https://wiki.metropolia.fi/display/sensor/Dissolved+Oxygen+Sensors#DissolvedOxygenSensors-GalvanicElectrode.

After looking at the values it outputs with a DMM at 0% and 100% oxygen saturation, my particular sensor is outputting current in the range of 0 to 25 microamps, and the voltage eventually stabilizes around 500+ mV.

I have looked at a couple different ways to measure this small current. Most circuits seem to place importance on interfering as little as possible with the current and voltage, however that isn't a requirement for reading this sensor. I am considering using an INA219 http://www.ti.com/lit/ds/symlink/ina219.pdf with a 100 or 1000 ohm shunt resistor since voltage drop doesn't matter. The INA219 has a built in op amp and high resolution ADC, so this is the simplest way that I have found, but I am here to ask if there are any other approaches to measuring such a small current when interference isn't an issue and also if the high shunt resistor approach is a reasonable solution for what I'm trying to do.

Thanks for your help.

[tszaboo]

The INA219 is not suitable for this. It has 20uA input bias current.
You want an instrumentation amplifier on that shunt. Something like a INA333, which only has 0.2 nA input bias.
Or check out the uCurrent, made by Dave.

[jaromir]

While INA333 will work, I think it's a bit of luxury here. Since sense resistor is grounded by one leg, so you can amplify it by any CMOS/JFET R-R opamp in non-inverting configuration(*), like NCS325SN(**). Amplify the signal to level required by your ADC. If you use 1k resistor, you have 0-25mV signal for 0-25uA sensor response. Set gain to 100, you get 0-2,5V. Maximal offset voltage of NCS325SN is 50uV, after amplification 5mV (maximal offset error 0,2%). If you are OK with that, you are golden, if not, you have to search for better opamp or use larger resistor value - if that can be tolerated. For 5k, you need 5x less amplification and get 5x less error (max 0,04%).
You can then process the voltage with suitable ADC. If you are OK with 0,2%-ish error mentioned before, you may want to use 10-12b ADC, common in almost any MCU designed in last 10 years; alternatively use dedicated ADC, like MCP3201.

* http://www.radio-electronics.com/info/circuits/opamp_non_inverting/op_amp_non-inverting.php
** https://www.mouser.sk/ProductDetail/ON-Semiconductor/NCS325SN2T1G/?qs=sGAEpiMZZMtCHixnSjNA6NXlPpRsFEjtLsZx4V9NZw8%3d

[hirada]

* http://www.radio-electronics.com/info/circuits/opamp_non_inverting/op_amp_non-inverting.php

Slightly off topic, but in that link under the "Non-inverting amplifier using single supply" headline rather at the end, would that also be considered as biasing the opamp into Class A operation? I do not want to hijack the thread, so if the answer is "no", that'll be a different topic one day.

[jaromir]

I do not want to hijack the thread

You already did.

if the answer is "no", that'll be a different topic one day.

No.

[tszaboo]

While INA333 will work, I think it's a bit of luxury here.

It is a one-off project.

[justind000]

Thanks for your recommendations. Given the cost of the INA333, ~4 and the relatively simple amplifier circuit, it seems like the op amp is the simpler approach.

[danadak]

TIA -

https://www.digikey.com/en/articles/techzone/2017/jun/how-to-design-stable-transimpedance-amplifiers-automotive-medical-systems

http://www.linear.com/solutions/Transimpedance_Amplifiers

https://www.planetanalog.com/document.asp?doc_id=527534


Regards, Dana.

[Zero999]

Hello,

I have a handmade dissolved oxygen sensor, as described here https://wiki.metropolia.fi/display/sensor/Dissolved+Oxygen+Sensors#DissolvedOxygenSensors-GalvanicElectrode.

After looking at the values it outputs with a DMM at 0% and 100% oxygen saturation, my particular sensor is outputting current in the range of 0 to 25 microamps, and the voltage eventually stabilizes around 500+ mV.

I have looked at a couple different ways to measure this small current. Most circuits seem to place importance on interfering as little as possible with the current and voltage, however that isn't a requirement for reading this sensor. I am considering using an INA219 http://www.ti.com/lit/ds/symlink/ina219.pdf with a 100 or 1000 ohm shunt resistor since voltage drop doesn't matter. The INA219 has a built in op amp and high resolution ADC, so this is the simplest way that I have found, but I am here to ask if there are any other approaches to measuring such a small current when interference isn't an issue and also if the high shunt resistor approach is a reasonable solution for what I'm trying to do.

Thanks for your help.

Presumably you're measuring the current, not the voltage, so you want a current to voltage converter, with as lower input impedance as possible.

Perhaps a MOSFET input op-amp such as the MCP602 or would the offset voltage be a problem?