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.

High Speed OPamp for DDS application

Other Parts Discussed in Thread: OPA684, OPA2684, OPA695, VCA824

Hi,

for my project thesis I'm using the AD9952 which creates a sine wave output. After  rf-transformer and filter the signal amplitude should be 1V.

Now I want to go down to 0.1V with a voltage divider and then on demand go up to 4,5V with switchable resistors.

The problem is the high frequency of max 100MHz.

So the amplification of the OPamp should reach from 1 to nearly 100 in the range from 0 to 100MHz.

The output should deliver enough current for 50 Ohm impedance matching.

I have a dual power supply of +5V and would be great if it  the amp was offset compensated.

 

Does anybody know a suitable device for this application?

Thanks.

Peter

  • Hi Peter,

    The OPA684 should fit your need quite nicely.  In order to achieve 100V/V with 100MHz bandwidth, you might want to consider using 2 amplifiers in series each with a maximum gain of 10V/V.

    Note that the gain resistor will need to be change as the OPA684, OPA2684 for the dual, is a current feedback amplifier in which the feedback element is also the compensation element.  So for the OPA684 it should stay at 1kohm.

    For the 50ohm load, 4V into 50ohm is ±80mA.  The OPAx684 will be sufficient here.

  • Hi Xavier,

    thank you very much for your suggestion. As an alternative I found the OPA695 with a greater bandwidth. I have no experience with current feedback amplifiers, so I dont know

    why Rf is 1kohm with the OPA684 and any other value with the OPA695.

    Is it possible to get a gain of 1/2 if using the inverting amplifier or would it run unstable below a gain of 1?

    What limits the allowed value of Rf and Rg if not the gain bandwidth product?

    Where can I see if the device is offset compensated at the output or how can I reach offset compensation?

     

     

    Kind regards

    Peter

  • Peter,

    Here are a couple of TI app notes that discuss current feedback op amps:

    Voltage Feedback vs. Current Feedback Op Amps - http://focus.ti.com/general/docs/litabsmultiplefilelist.tsp?literatureNumber=slva051

    Current Feedback Amplifier Analysis And Compensation - http://focus.ti.com/general/docs/litabsmultiplefilelist.tsp?literatureNumber=sloa021a

    In a nutshell, while the gain and bandwidth cannot be set independently in voltage feedback (VFB) op amps, the gain and bandwidth CAN be set independently in current feedback (CFB) op amps. In a CFB amp, the bandwidth is set by the feedback resistor Rf, while the gain is set by the ratio of the gain and feedback resistors (Rg and Rf). The app notes above explain why the feedback resistor is part of the compensation and sets the stability of a CFB amp.

    The OPA695 datasheet does have recommended values for Rf to ensure stability. Other than providing the Rf values for the typical gains in the electrical characteristics (p.3), the datasheet shows a graph of recommended Rf vs. gain on Figure 16 on p.24.

    Yes, it is possible to get a gain of less than 1/2 in the inverting configuration, but I would recommend starting with at least an Rf of 500Ω, increasing the value (and Rg to maintain the gain) until you get a satisfactorily flat frequency response.

    From reading the previous posts though, it sounds like you are planning to adjust the Rg or Rf on the fly to adjust the gain. Do you plan on using an analog switch of some kind to switch between different feedback resistor paths? We do not recommend doing this as analog switches will add capacitance into the feedback network which can cause stability issues and oscillations.

    You may want to look into the VCA82x parts, for example, the VCA824, which is a variable gain amplifier with >40dB gain range. The gain is controlled with an analog voltage input (Vg).