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.

OPA454 power consumption in normal operation

Other Parts Discussed in Thread: OPA454, OPA141

Hi Everyone,

we need some help concerning the setup attached as file:

1665.current_meas.pdf

OPA454 is used in a constant current measurement set up for high resistor values > 1MOhm. Without attaching a resistor (Rx) everything is fine. A quiescent current of round about 4mA @100V can be measured  for each opamp as stated in the datasheet.

But after inserting a resistor power consumption increases more than I would expect. Normally I would expect a current of more ore less quiescent current plus current flowing through load which is defined by Uri/Ri in our case.

That would mean a max. current of additional 10uA@100V when selecting lowest measuremt range (160k).

But when I take a look at my current meter I can see a current consumption of round about 7mA for each OPA454. That seems to me much too high for that case. With respect to the information in the datasheet I cannot  find a good explanaition for this issue.The only one I could think of is that internal power consumption of OPA454 increases by this even when driving very high impedance loads but I cannot belive that this is true.

I also removed ectrenal protection circuits and the additional OPA141 and the analog switch to be sure that this has no influece but nothing changed. So it must have to do with the Opamp and external circuit itself.

Any idea what goes wrong ??

Any help is very appreciated

Thanks, Christian

 

  • Hi!

    Interesting problem in an interesting application. Even if I cannot read the schematic (too bad resolution), I think that the opamp does not have any real output current because it just drives a MOST. Just two thoughts: Does the output voltage have influence on the quiscent current? This can only be answered by TI engineers. And, is your current souce stable (you directly drive a MOST gate)?

    Best regards,
    Edwin Krasser

  • Hi

    here are some additional information concerning my problem:

    There was a mistake in my circuit, a power supply was missing at the OP for Vref so there wer some strange effects. After correcting situation looks like this:

    For 10uA and 1uA my circuit works as expected when regarding power consumption. My current meter shows nearly 4mA plus something when operating. But when I chose the lowest possible current of round aboout 95nA power consumption rises as described before.

  • Hi again,

    situation is different as decribed before. Right there was a mistake that also lead to wrong results but difference in power consumption has nothing to do with the selected measurement range but it has to do with the attached load.

    When attaching a load power consumption is a function af attached resistor. The smaller the resistor the smaller opamp'soutput voltage and the higher power consumption.

    At the first view this seema to be logic but it is still to much rise of current with respect to the load current that is flowing.

  • Hi Nikoaus,

    There are two things I would check regarding your OPA454 circuit:

    1. Place a low capacitance O-scope probe at the OPA454 output pin and run the circuit at the different current levels. See if an oscillation is observed at the lowest current source setting where the operating current is increasing. It could be the MOSFET input capacitance is loading the output and reducing the phase margin. When an operational amplifier oscillates the operating current often increases.
    2. It is a little difficult to see some details in your schematic. Be sure the OPA454 input drive level is within the specified common-mode voltage range at each current level. I think I see a label that shows 1.5 V being applied at the non-inverting input. The data sheet lists a typical common-mode voltage range of 1.5 V from the supply rails, but also states it can be as high as 2.5 V from the supply rails. I am not certain how the OPA454 quiescent current is affected by CMV level, however, it is conceivable the one is affecting the other if it outside the specified range.

    Regards, Thomas

    PA - Linear Applications Engineering

  • Thanks for your reply.

    I cannot see any significant oscillation at the output of OPA454. But I also thought of that so I worked with lead compensation but without succes. Current consumption is untouched.

    I changed reference voltage from 1.5V to 2.5V to check influence but also without success. Beahivour stay the same. But I could see, that range with lowes current is only working with in input voltgae of 2.5V with lower voltages this range is not working properly.

    Any other idea or sugeestion of what i could do or try ...?

    Attached you will find the schematic in a higher resolution.

    7271.current_meas.pdf

  • Hi!

    First of all, I have to say that I cannot find an obvious reason for this strange behaviour. The only thing that I can suggest is to remove the BAV99W clamping between GND and +100 V (D700 at ISO_MEAS_A). The BAV99s have just a maximum reverse voltage of 80 V.

    Best regards and good luck
    Edwin Krasser

  • Hi Nikolaus,

    Thank you for the higher resolution schematic. I can now see your OPA454 circuit details more clearly. I am studying it for clues to the increased supply current.

    I am surprised that you didn't find any oscillation present. Did you use a wide bandwidth O-scope, or DSO? Someimes an oscillation can be near the unity-gain cross frequency, or in the case of a parasitic oscillation at very high frequencies.

    Regards, Thomas

    PA - Linear Applications Engineering

     

  • Hi again,

    sorry for my late reply but I had some days off.

    Concerning the protection diode I can say that we use an BAV99W from NXP- type which is rated for 100V reverse voltage. But anyhow thanks for the advice which I whave to note in schematic more clearly to avoid problems.

    I cross checked the behaviour of the output pin of OPA454 and I could not see a ringing that is responsible for the raise of power consumption. Of course I could see some damped HF ringing but that has something to do with GND loops of the probe and changes when probe is grounded at another point of the pcb. I used a 1GHz O. scope to measure.

    I also tried to put a capacitor between output and negative input but bevaviour did not change at all.

    Up to now I cannot find any good reason for that.

    Anyhow It seems as if we hace to change Vref to 2.5V to guarantee proper operation in highest measurement range because some of the PCB with the opamp do not work with VCM= 1.5V at lowest current.

    Regards, Christian