• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 49 subscribers
  • Views 864 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Seeking an amplifier suitable for summing multiple inputs in the range of 25KHz to over 1GHz. Gain should not be excessive <5 if possible.

arthur williams73
Intellectual 390 points
Other Parts Discussed in Thread: THS3091, OPA695

Seeking an amplifier suitable for summing multiple inputs in the range of 25KHz to over 1GHz. Gain should not be excessive <5 if possible. The conventional summing amplifier configuration of multiple resistors feeding the inverting input would be desirable. Also as high a supply voltage is desirable to allow large swings due to a crest factor issue.

  • 0
    Guru 140200 points
    Hi Arthur,

    can you show a schematic of your planned circuit?

    Do you need to use 50R technique for the input signal wiring? Or is the circuitry driving the inputs nearby?

    Kai
  • 0 in reply to kai klaas69
    Intellectual 390 points

    The circuit is a classic summing amplifier with three signals from a 50 ohm source going to three resistors connected to the inverting input of an amplifier and a feedback resistor. The issue is finding an amplifier that can be used in the conventional manner at 1GHz. MMIC amplifiers do not have summing inputs.

  • 0 in reply to arthur williams73
    Guru 140200 points
    Hi Arhtur,

    every OPAmp with current feedback technology should work as a summing amplifier. The advantage of using a current feedback OPAmp is explained here:

    www.embedded.com/.../4009914

    Of course, today TI has some much faster OPAmps compared to the discussed THS3091. But nevertheless it will be difficult to handle large signals of more than 1GHz.

    Kai
  • 0 in reply to kai klaas69
    Intellectual 390 points
    Can you recommend a CFB or VFB amplifier suitable for use at 1GHz and above as a summing amplifier?
  • 0 in reply to arthur williams73
    TI__Genius 16735 points
    Hi Arthur,

    Due to the holidays our response are delayed. As Kai mentioned it is going to be difficult to handle large signals more than 1GHz. Someone will get back to you to help you find some solution at the end of the holidays.

    Thanks!
    -Karan
  • 0 in reply to arthur williams73
    Guru 140200 points

    Hi Arthur,

    I would give the OPA695 a try:

    arthur.TSC

    Or you could try one of the ultra fast fully differential OPAmps from TI. They are even faster.

    Kai

  • 0 in reply to kai klaas69
    Intellectual 390 points

    Thank you the OPA695 may be a good possibility. I know the differential amplifiers are faster but I have multiple single-ended inputs to be summed.

    I have another question which falls more in the theoretical. If I have a  white noise-like  random signal having equal intensity at all frequencies up to 1GHz  and a peak-to-peak value of lets say 2V of the envelope, does the amplifier have to pass a 2V sinusoidal signal at 1GHz? The value of the 1GHz  signal buried in the spectrum is much less than 2V but the total signal envelope is 2V. How do you reflect this requirement of a 1GHz signal buried in the spectrum? I hope you understand the issue. Can anyone at TI answer this?  

    Thank you so much.

  • 0 in reply to arthur williams73
    Guru 140200 points

    Hi Arthur,

    the OPAmp must be able to handle the whole signal. But if the noise is broadband noise and the wished signal is 1GHz, you could introduce some high pass filtering. This would decrease the amplitude of the total signal and by this allow the OPAmp to handle the wished 1GHz signal with its small signal capabilities.

    Kai

  • 0 in reply to kai klaas69
    Intellectual 390 points
    I want to amplify the entire signal from 25KHz to 1GHz. What governs the op-amp performance, the low level of the 1GHz tone, or the amplifiers slewing response to the envelope of the composite signal? Lets assume for example the signal consists of 1,000 equal level tones from 25KHz to 1GHz. The 1GHz tone by itself would be a small signal lets say 200mV p-p but the composite could be 2Vp-p. Is the 1GHz tone affected by the slewing caused by the total envelope?
  • 0 in reply to arthur williams73
    Guru 140200 points

    Hi Arthur,

    yes, the 1GHz tone can be effected by the slewing of the total envelope.

    In the first simulation you see a 1Vp square wave of 50kHz (VG2) and a 200mVp sine wave of 100kHz (VG1):

    The output signal is the ideal sum of a square wave and a sine wave.

    In the second simulation you see a 1Vp square wave of 50MHz (VG2) and a 200mVp sine wave of 100MHz (VG1):

    Here the output signal is no longer the ideal sum of a square wave and a sine wave, because the OPA695 cannot follow the edge of square wave due to its limited slew rate. As result the sine wave appears to be distorted.

    Kai