Need help in designing V2I converter

[findsriharsha]

Hi All,

I need to design a V2I converter. Spec is +/-50mV => +/- 1.25uA.

Please suggest a good architecture for this.. I have +5V and -3V supply.

Thanks in advance.

 

[erikl]

What about a 40kΩ resistor? ;-)

Ok, perhaps some more devices:

 

[findsriharsha]

Yes, I initially thought of this circuit. But input range is +/- 50mV with respect to ground!

 

[erikl]

But input range is +/- 50mV with respect to ground!

Sure; that's fine if you connect the resistor to GND.

 

[findsriharsha]

Consider if input is -50mV, then the FB voltage tries to be -50mV which is across the R. BJT will turn off then, isnt it? corret me if I am wrong.. Thanks

 

[erikl]

Yes, you're right, sorry. To get around you could use the following circuit, even saving the transistor:

Its disadvantage, however, is the small complying voltage ("headroom") which is equal to the input voltage. This means the current measurement circuit must have a very low input impedance, i.e. « 40kΩ (< 400Ω for 1% error) -- which also could be achieved by using only the resistor.

This isn't really helpful, I know. We have to look for more complex circuits.

 

[findsriharsha]

Yes, Thanks Erikl!

In the mean time, please post if anyone else has some suggestions...

thanks!

 

[erikl]

We have to look for more complex circuits.

I think the easiest way is to amplify the original input voltage and then generate the required current via an appropriate resistor to GND. With your neg. supply voltage of -3V you should be able to amplify the +/-50mV to +/-2.5V , then create the current via a 2MΩ resistor. I'd suggest to buffer the +/-2.5V output by another opAmp:

 

[findsriharsha]

Final note: this issue has been solved. I implemented a diff pair with source degeneration resistor to achieve the required transconductance.