Current measurement - op-amp as buffer, input to ADC of microcontroller

Thread Starter

Akshatha Venkatesh

Joined Mar 28, 2017
7
We want to measure current in the range 1 to 40 uA , and a 100kohm shunt resistor is being used to convert the current into voltage to be fed into the microcontroller. I'm trying to understand why they have selected R1 to be 100kohm , since the lowest voltage reading will be 0.1 and highest 4V. If a larger value for R1 could have been selected , the full range of the microcontroller could have been utilised, but it is not being done in the above circuit , is there any specific reason for this ? I think some parameter of the op-amp is in play here. The op-amp is being used as a buffer in the above circuit , what parameter of the op-amp as buffer is not allowing us to use the whole 5V range of the mi

MOD: Rotated and Cleaned your image.
IMG_20180415_230736.jpg
 

Hymie

Joined Mar 30, 2018
1,277
We want to measure current in the range 1 to 40 uA , and a 100kohm shunt resistor is being used to convert the current into voltage to be fed into the microcontroller. I'm trying to understand why they have selected R1 to be 100kohm , since the lowest voltage reading will be 0.1 and highest 4V. If a larger value for R1 could have been selected , the full range of the microcontroller could have been utilised, but it is not being done in the above circuit , is there any specific reason for this ? I think some parameter of the op-amp is in play here. The op-amp is being used as a buffer in the above circuit , what parameter of the op-amp as buffer is not allowing us to use the whole 5V range of the mi

MOD: Rotated and Cleaned your image.
View attachment 150568
It appears to me that R0 and R1 are configured as a dividing resistor network in this circuit – as simple as that.
 

cmartinez

Joined Jan 17, 2007
8,218
It appears to me that R0 and R1 are configured as a dividing resistor network in this circuit – as simple as that.
Yes ... conventions dictate that R0 should be oriented vertically, with the supply tagged at the top. That would make the circuit a lot easier to read. And btw, R2 is there to limit current into the device, which I find a little odd, since the voltage divider already has very high values and would make the use of R2 moot. And Eric has a point, a conventional OPA won't give you the full range you want.
 

Thread Starter

Akshatha Venkatesh

Joined Mar 28, 2017
7
Yes ... conventions dictate that R0 should be oriented vertically, with the supply tagged at the top. That would make the circuit a lot easier to read. And btw, R2 is there to limit current into the device, which I find a little odd, since the voltage divider already has very high values and would make the use of R2 moot. And Eric has a point, a conventional OPA won't give you the full range you want.
Why won't the conventional OPA give the full range , what parameter of the OPA is preventing it from giving the full range ?
 
Try to use a transimpeadance amplifier if you can. An OP-amp in a current to voltage converter configuration. You;ll get much better results. The TIA would need to be grounded.
 

cmartinez

Joined Jan 17, 2007
8,218
Why won't the conventional OPA give the full range , what parameter of the OPA is preventing it from giving the full range ?
Well, ordinary OPAs have that limitation, that is, they have losses in them that won't allow for a full range output. But your question is now irrelevant, since you've stated that you're using an LMV772QMMX, which according to its datasheet is of the rail-to-rail type, so it should output a full range like you want.
 

danadak

Joined Mar 10, 2018
4,057
Why won't the conventional OPA give the full range , what parameter of the OPA is preventing it from giving the full range ?


If the negative rail is ground, you can see if the input is brought below Vbe_T1+Visource then input shuts off.
If in turn thats the feedback input, the inverting input, then the feedback loop loses control. Note NPN inputs do
common mode to their positive rail, conversely PNPs to their negative rail.

Output side



Here we can see that if negative rail is ground, the output NPN cannot pull any closer
to ground than Vcesat_npn. Schemes such as place a load R in parallel with output
to ground, the R could pull the V lower but at cost of increased Pdiss in the circuit.
Note mosfet outputs can pull to the rail, but with no current flow thru them, otherwise
their Rdson x Iload limits output swing.

Regards, Dana.
 
Top