Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

High Speed Buffers

  1. Oct 8, 2014 #1
    Hey guys,

    So I have been running some simulations where I am creating a BPF around the FM band and when I measure the signal on the output, it's much lower than my original signal. So I thought if I put a buffer between the LPF and HPF to electrically isolate the circuits this would help, but it didn't appear to be any improvement. In fact it just seemed to make the output look more "messed up". My only guess is that the buffer I am using (LT1395) is not fast enough? Any ideas?

    Does anyone have any favorite high speed op amps (buffers) they use for the FM band (~100 MHz) or above?

    Thanks
     
  2. jcsd
  3. Oct 9, 2014 #2

    berkeman

    User Avatar

    Staff: Mentor

    This is all in simulation, or have you actually built this on a PCB or breadboard?

    That is a pretty fast opamp for small signals. What size signal are you running through your filter? Can you post your whole circuit diagram as a PDF or screenshot?
     
  4. Oct 10, 2014 #3
    Right now it is all simulation but I would like to build it but wanted to make sure the simulation worked first.
    The simulation is only a 1V signal but real application would be a bit larger (5V). I did 1V in the simulation because I wanted to verify the -3dB point was were I wanted it to be, but it never actually reaches -3dB, the highest I see is -9 dB. I have attached the circuit.
     

    Attached Files:

    • BPF.PNG
      BPF.PNG
      File size:
      2.9 KB
      Views:
      76
  5. Oct 10, 2014 #4

    berkeman

    User Avatar

    Staff: Mentor

    I'd suggest just testing the buffer circuit by itself first, so you can characterize its performance versus frequency and versus input signal amplitude.

    If you just drive it with the AC source and monitor its output voltage, what do you get for a small 10mV input signal? What about 100mV and 1V?
     
  6. Oct 12, 2014 #5

    NascentOxygen

    User Avatar

    Staff: Mentor

    At first glance I'm wondering about the practicality of some component choices. For example, you show a 1.47pF in your LP stage, but was this determined having in mind the 2pF (typ) input capacitance of the 1395? Similar considerations pertain to HP component choice, so that you'll probably need another buffer at the output. I'm not familiar with current feedback amplifiers, but I'd be surprised if these factors can be dismissed.

    Good luck with your experimenting.

    For the information of others, a LT1395 datasheet not requiring a pdf viewer is available here: http://www.datasheetlib.com/datasheet/225628/lt1395_linear-technology.html
     
  7. Oct 15, 2014 #6
    I would recommend sticking to a 50 ohm environment: Design your LPF and HPF with 50 ohm matched inputs/outputs, and for buffering use an RF gain block such as the Minicircuits ERA-3 (very nice flat response from DC to 3GHz, 2.7dB noise figure, for $1.50).
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook