mr_monster
Member level 4
I'm starting a curve tracer design and the 1st fundemental circuit I need to test is a bi-polar current source I can control digitally as the project is micro-controller based. After looking at all kind of different solutions I've found the good old Howland current pump in an application sheet. Using the DAC (by ADI) or some other high resolution DAC I can get a high resolutions (300nA using 16bit). This design is off course bi-polar so it can drive both NPN and PNP transistors.
Off course there are a few problems with the design, like having to match the resistors (probably 0.1%) and that the opamp can't push more then about 40mA which will translate to 5mA max. step value in my design but perhaps I can live with that as I am not indenting the test devices which are not small signal.
Can anyone think of a reason why this would not be a good circuit if at the output I'll have a current sensing resistor to have closed loop for real-time control over the output? It seems like this circuit is favored in text books but not really used that much.
Off course there are a few problems with the design, like having to match the resistors (probably 0.1%) and that the opamp can't push more then about 40mA which will translate to 5mA max. step value in my design but perhaps I can live with that as I am not indenting the test devices which are not small signal.
Can anyone think of a reason why this would not be a good circuit if at the output I'll have a current sensing resistor to have closed loop for real-time control over the output? It seems like this circuit is favored in text books but not really used that much.