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.

About OPA552 linearity (Howland current pump)

Other Parts Discussed in Thread: OPA552, OPA551, OPA445

Dear Sirs

My customer use OPA552 as a modified Howland current pump. (like the following two figure)

They give the input voltage (+Vin) and get Iout.

But the linearity is bad near 0uA. (+Vin is near 0~0.1V)

input voltage 0.3V=>Iout 28uA,

input voltage 0.2V=>Iout 18uA,

input voltage 0.1V=>Iout  6uA,no good,

How could them improve near input voltage 0.1V? Thanks

BRS

Nat

 

  • Hello Nat,

    The nonlinearity near zero volts is due to the fact that the OPA552 is being powered by a single +60 V supply. It is most often powered with dual supplies. When you attempt to drive the OPA552 input voltage to 0 V, its lower common-mode voltage limit (V-) + 2.5 V is violated, and the low-level output voltage has bottomed out as well. The amplifier no longer operates in linear mode and that is why the linearity suffers near zero volts.

    A TINA simulation shows the problem with the common-mode violation and how the output voltage and load current become non-linear as the common-mode voltage drops below the lower limit of (V-) + 2.5 V, or +2.5 V in this case when V- is 0 V. By comparison, when the OPA552 is powered with +/-30 V, dual supplies and the output voltage and load current show good linearity through the 0 V crossing.

    The OPA552 cannot be expected to provide good linearity when it is operated in non-linear input and output regions. There isn't any way to correct this with the single supply arrangement. I would suggest using a dual supply arrangement. That could consist of traditional dual supplies such as +/-30 V, or by substituting the negative supply with an charge pump inverter or boost/buck regulator powered from the positive supply. The Power Management Products group could advise the best supply solution.

    Regards, Thomas

    PA - Linear Applications Engineering

  • Dear Sirs

    They has test Vs=+-30V, But it si still no good.

    Another issue, in the Howland current pump.

    Could the current accuracy achieve 10uA?

    Thanks

    BRS

    Nat 

  • Hi Nat,

    I am quite surprised that the OPA552 linearity performance was not improved when dual supplies were applied. Clearly, there are the issues I pointed out when the Improved Howland Current Pump was used with a single supply configuration, but that should have been corrected with the use of dual supplies. It would be interesting to see the measured results they obtained with the dual supplies.

    Although the Howland current pump is most often used at current levels much higher than 6 uA, it should be possible to apply it at low current levels too. But I am not sure if there is something occurring that we don't understand when one tries to apply a power operationl amplifer such as the OPA552 at a very low output current level. The amplifier is normally used for higher current loads.

    Has the OPA551 been tied in the dual supply circuit? The OPA552 is designed to be used for voltage gains of 5 V/V, and greater. In the Improved Howland Current Pump the primary function is a current source, but there is voltage gain that is a function of the input voltage and the load resistance. If you look back at the TINA Vout vs. input voltage plot that I provided for the dual supply configuration it does appear that the voltage gain does drop down below 5 V/V with the 10 k load resistor. It could be even lower with a lower load resistance. It might be worthwhile to test the circuit with the OPA551 just in case there is some stability issue that is being overlooked.

    There is another operational amplifer that I would suggest evaluating, the OPA445. It is a high voltage opeational amplifier rated for supplies up to +/-45 V, but its output current is only rated at +/-15 mA (min) which is still way more than the application requires. Although the OPA551/OPA552 do boast very low THD+N, the OPA445 is even lower. Do note that the THD+N conditions are not the same for the two devices. It might prove worthwhile to evaluate the OPA445 in the circuit and see if there is improvement in the linearity.

    http://www.ti.com/lit/ds/symlink/opa445.pdf

    Regards, Thomas

    PA - Linear Applications Engineering

  • Hi Nat,

    if you have not seen it,  read "AN-1515 A Comprehensive Study of the Howland Current Pump", written by Bob Pease.

     http://www.ti.com/lit/an/snoa474a/snoa474a.pdf

    Regards,

     

  • Dear Sirs
    Thank you for your reply~

    BRS
    Nat