OPA596: Improved Howland Current Pump output current with soft edges

Hi Binit,

Binit Panda said:

I am having an issue with avoiding rounding off of my output current waveform.

Do you mean that you would like to have sharper step transition where the current follows the voltage in waveforms? What is the current transition and current amplitude in timing requirements?

What is your max. input differential voltage and common mode voltage? I'd like to understand the constant current requirements before I am making certain suggestions. 

Best,

Raymond

Hi Raymond, 

Yes, I am looking for sharper step transitions. I am trying to deliver 20mA and the timing requirement is under 80 uS. Currently, the rise time is around 350uS.

In terms of input differential voltage is around 450uV and common mode voltage is around 30V for max load of 1.5Kohm. My Vin is 5V to -5V for 5ms each phase. 

Hope this helps 

Hi Binit,

Binit Panda said:

I am trying to deliver 20mA and the timing requirement is under 80 uS

This is going to tough for our HVA product line. The differential input is +/-5V and output is going to be +/-0.1V. Assume the output current follows the output voltage in step transition, the 0.2V/100V/usec = 0.002usec (the slew rate is defined between 10% to 90% of rising or falling edge). In theory, 100V/usec is good enough, if it is the op amp's natural slew rate. The stated 100V/usec is boosted slew rate. Although the rising/falling edges are sharp (10% to 90% by definition), the op amp is returned to its natural slew rate after it is reached the output setpoint --> this may take extra usecs to reach the final setpoint. 

You would need to op amp with high natural slew rate, say OPA828 or similar or op amps from HSAMPS (High Speed Op Amp Product lines). 

Binit Panda said:

In terms of input differential voltage is around 450uV and common mode voltage is around 30V for max load of 1.5Kohm. My Vin is 5V to -5V for 5ms each phase. 

0.2V per 80usec is your definition. What is your current source application? You mentioned the supply rail needs to be >30V, because of 20mA*1.5kΩ. For Howland current pump, the differential input is +/-5Vdc and the current source is bidirectional at the load (common mode voltage at input can be GND). Please confirm the design requirements, and I will try to find you a fast response current source and circuit. 

Best,

Raymond