AD8011 at +-1.5V

Hi, I have an application where I need low power consumption. I have chose the AD8011 and supply it +-1.5V to minimize power consumption.

However I need some bandwith and amplification. I have a signal of 0-0.05V at 20MHz. I amplify it 11 times. And I get about 4MHz of bandwith out.

If I crank up the supply voltage to +-5V, I get the bandwith I need. But too much power loss.

My feedback is 1k:100R. Tried 470R:47R too, and got slightly higher BW.

What do I do to get 20MHz at +-1.5V power supply?

Thanks!

/Anders

  • Does none have experience from this? Or am I missing some vital information? I really need to solve this.

  • 0
    •  Analog Employees 
    on Oct 28, 2020 2:13 AM 1 month ago in reply to AndersII

    Hello,

    Apology if we missed this thread.
    Please let me check onto this.

    Regards.

  • 0
    •  Analog Employees 
    on Oct 29, 2020 5:58 AM 1 month ago in reply to AndersII

    Hi,

    First thing to know here is that the AD8011 is a Current Feedback Topology (CFB).
    Here is a mini-tutorial about CFB amplifiers: https://www.analog.com/media/en/training-seminars/tutorials/MT-034.pdf

    There are several differences between a CFB and Voltage Feedback Amplifier (VFB) and here are two.
    1. One important difference between that two topologies is that the CFB has unbalance input impedance with +IN as High Impedance and -IN as Low Impedance. Where as on the VFB, the input impedance is balance.
    2. The CFB doesn't have constant Gain Bandwidth Product, it has Transimpedance Gain where in most cases, the Feedback Resistance is defined to optimize the bandwidth and need to vary the gain resistor to have different gains.

    Related to this Input Impedance, the CFB should not have a Capacitor feedback as this will cause oscillation on high frequencies. I can see a feedback capacitor in your U1.

    The recommended RF for AD8011 is 1K, and that is not the case on your U2.

    You are aiming for 20Mhz application with +/- 1.5V power.
    We have some VFB that suits this requirement.
    You may check LTC6226/7 or try to check on our selection table for high speed amplifiers and just filter the supply up to 3V range..

    Hope this helps.

    Thanks and regards.

  • Thanks for your reply!

    It is clear that you are correct regarding the second amplifier. The AD8011 was selected by my predecessor for its high BW and still low power consumption. We had issued with the board heating up sensitive components close by.

    The LTC6226 has about 5 times the power consumption. Still <20mW, so it might work. But we need to test in machine. Is there a lower power VFB op amp than your suggestion?

    Regarding the 560 ohm feedback on first amplification stage. This PCB is used in two configurations. One with a lot if light, and one with little. When we have a lot of light, the second amplification stage is bypassed by two optional resistors. At that light intencity, the first amplifier reaches staturation at 1k amplification. That is why 560R is selected instead.

    What would you instead suggest as a solution to half the output amplitude and still keep 1k amplification? Resistive devider on input current?

    Additional question, am I missing balansing on the first amplifier inputs? Should I have 1,2k ohm from positiv input to GND or any such scheme?