How to read low currents

[Yaroooo]

How I can read low currents like between 0.1mA to 3mA?

Classic shunt plus amplifier involve use of high resistors. In example with 200gain amp and 3.3mA of current I should use a 5R resistor to get 3.3V suitable output for used MCU. But 5R is a very high resistor.

Since I'd like to make a fast test (it's not a project), I really like a solution like integrated amp+shunt like INA series, but can find one suitable for this purpose.

Thanks in advance for tips!

[TimFox]

Why is 5R a high value?  The important factor is the "burden voltage", which is the voltage across the resistor with the maximum current flowing through it.  For 3.3 mA and 5  , that is 17 mV, which gives you your calculated 3.3 V after the x200 gain amplifier.

[Howardlong]

Measuring 100uA should be easy, when I read the title, I though you might be thinking in terms of single digit nA or pA!

You don't mention what kind of dynamic range this is for, what accuracy, what resolution, or what sampling rate.

Is it to measure a low voltage power supply? If so, one solution is to place the shunt before a low Iq regulator: there are a number of LDOs that offer Iq in the order of tens of nA, so you can measure the voltage drop across a shunt in front of the regulator for the current range you mention. If you want to measure higher dynamic range, you can use two shunts in series, the sensitive one with a low leakage diode (or transistor as a diode with base & collector connected) across it, and take two measurements.

There are plenty of INA current sense devices, you need to figure out things like what offset is acceptable, what gain, accuracy, drift, directionality, high- or low-side, and how it's presented, eg as a voltage or as a digital interface like SPI or I2C.

I'm curious, you say it's a "fast test" and "not a project", what is wrong with a multimeter, or is this for your own academic purposes?

[Yaroooo]

Actually I've tested an INA240 which I've already used for higher current measurement with a R002 shunt. It correctly worked with this configuration but when I've placed a 5R as shunt I had a really strange output (like a 100khz PWM with 500mV of amplitude) that didn't changed with current changes on load. I just wanted to check an MCU consumption to be sure that I'm in datasheet range.

Maybe there's something that I didn't connect correctly. I should re-check.

[Howardlong]

Actually I've tested an INA240 which I've already used for higher current measurement with a R002 shunt. It correctly worked with this configuration but when I've placed a 5R as shunt I had a really strange output (like a 100khz PWM with 500mV of amplitude) that didn't changed with current changes on load. I just wanted to check an MCU consumption to be sure that I'm in datasheet range.

Maybe there's something that I didn't connect correctly. I should re-check.

Well, 500mV is 0.5uA on a 200V/V gain across a 5 ohm resistor.

However the bandwidth of the device is of that order too: are you capacitively loading the output?

Furthermore, are you using filtering resistors and a filter cap between the shunt & the amplifier inputs? Usually these are of the order of 10 ohms each and about 0.1 to 1uF depending on your filtering requirements.

[harerod]

Yaroooo, we don't know much about your actual requirements and circuit topology.
Your measurement range is low enough to be compensated by an OPAMP or INA output, though.
If your shunt was connected to ground, a single OPAMP configured as a "Feedback Ammeter" might be helpful. For more complex requirements (high-side measurement) you might want to add more OPAMPs or use an INA.

https://www.tek.com/document/handbook/low-level-measurements-handbook - FIGURE 1-10: Feedback Ammeter. Your original "very high" 5R shunt would have been between the V1-terminals and has been removed.

[TimFox]

The inverting current-to-voltage amplifier above is the classic way to measure current with zero burden voltage.
Note that it is typical to use a split power supply, depending on the polarities of the current to be measured and the common-mode voltage range of the amplifier.
Also, avoid having large capacitance across the input to the amplifier, unless you deal with it carefully.
For 3.3 mA in, 3.3 V out, the feedback resistor would be 1 k, and the amplifier needs to be able to source 3.3 mA.

[David Hess]

Since I'd like to make a fast test (it's not a project), I really like a solution like integrated amp+shunt like INA series, but can find one suitable for this purpose.

Integrated current monitors do not really solve that problem if you want lower shunt resistance.  They are just more integrated.

How I can read low currents like between 0.1mA to 3mA?

Classic shunt plus amplifier involve use of high resistors. In example with 200gain amp and 3.3mA of current I should use a 5R resistor to get 3.3V suitable output for used MCU. But 5R is a very high resistor.

Either use a low noise precision operational amplifier, some kind of chopper stabilized one or precision bipolar, so that the gain can be as high as possible and the shunt resistance can be as low as possible, or use the transimpedance (current to voltage) configuration for effectively zero shunt resistance.