This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Minimizing power consumption while attenuating signal and driving an ADC

Other Parts Discussed in Thread: THS4131, ADS1271, THS4521

I am currently using the THS4131 as a driver for the ADS1271 much like figure 70 in the datasheet.  However I now want to reduce power consumption and was looking at the THS4521, but the THS4131operates from a ±15V supply, the input signal is ±10V while the THS4521 operates from a single +5V supply.  Essentially I need to attenuate my signal and drive the ADC while reducing power.   Is it feasible to use the THS4521 in place of the THS4131 with the ±10V signal?  If so, are there other component changes I need to make? 

  • Karen,

    Yes, this is a feasible design.  The THS421 operated from a single 5V power supply and with < 1/10th the quiescent current can be used to convert the ±10V input to the appropriate voltage levels for the ADS1271.  The circuit shown in  figure 70 of the datasheet ADS1271 will work, however I would suggest increasing the resistor values so the feedback resistors (RF) are 1k ohm and the input resistors (RG) are 4.02k ohm. For best phase margin, place a 2.67k ohm resistor (RT) directly accross the input pins of the THS4521 (pins 1 & 8).  For further information on this circuit see TI Analog Applications Journal article Using fully differential op amps as attenuators, Part 1: Differential bipolar input signals .

    Another design consideration for this circuit is the physical size of the input resistor can impact performance.  I worked on a  circuit with similar attenuator configuration to take a 32Vpp differential input and reduce the signal level to 5Vpp differential for input to an ADC. Originally we used 0402 size resistors and saw bad THD and noise performance. I found that increasing the physical size of the input resistor to 1206 size fixed the problem (I used resistors made by Rohm , but I think other brands will also work). My theory is there is a non-linearity due to the voltage coefficient of the input resistor or there is a non-linearity due to heating of the input resistor. 

     

    Respectfully,

    Jim Karki