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.

Cascading with LMH6881

Other Parts Discussed in Thread: LMH6881, ADS5463

Hi all, 

I'm trying to drive an ADS5463 ADC with a signal that is amplified by the LMH6881 eval board, schematic:  www.ti.com/.../snor004.pdf, providing a single-ended output. However, at the max gain setting on the LMH6881, we still aren't getting as much gain as we'd like. We have tried using a Mini-Circuits RF plug-in amplifier cascaded in series with the output of the LMH, and although amplification is then sufficient, noise is too high (SNR almost 1:1). Our goal is to amplify up to 200MHz. My question: Is there a good way to cascade another high speed amplifier with the output of the LMH without creating too much noise (we need an additional 10-15 dB of gain), or is it best to do all the amplification in one stage? 

Thanks for your thoughts!

  • Hello Jared,

    I would recommend using a 200MHz low-pass filter (maybe a couple in cascade to get higher order) in-between the Mini-circuits RF plug-in amplifier and the ADS5463. The ADC has a 2.3GHz of input bandwidth and without filtering, I would think all the 2.3GHz broadband noise would alias back into the 1st Nyquist band of 200MHz and could be a reason for the SNR to show up as 1:1. Also, do you know the part# for the Mini-circuits RF plug-in amplifier?

    One thing to note is that the LMH6881EVM which has differential to single-ended output balun is intended for ease of interfacing with test equipments only. In actual system, the LMH6881 could interface at the output directly with the ADC (of-course with an anti-alias filter in the middle) eliminating the need of the diff to single output balun and achieve higher gain. I think the same goes at the ADS5463 input EVM that has the two ETC1-1-13t baluns. I believe that could be a reason for seeing lower gain at the LMH6881 output.  

    Best Regards,

    Rohit 

  • Hi Rohit,
    Thanks for your thoughts.
    We are actually measuring just the output of the amplification stage on a 1GHz oscilloscope.
    So the signal chain goes: Voltage signal ---> LMH6881 ---> (Mini-Circuits zx60-3018g) ---> oscilloscope.
    The LMH6881 eval kit gave us a gain of about x9 with the gain setting maximum and when testing with a sine wave input @ 400MHz.
    However, we are now measuring the voltage across an FPGA current sensing resistor, and here are our gain measurements compared to the raw voltage signal across that resistor:

    voltage signal ---> LMH6881 ---> Oscilloscope: x3 gain
    voltage signal ---> Mini-Circuits ---> Oscilloscope: x4 gain
    voltage signal ---> LMH6881 ---> Mini-Circuits ---> Oscilloscope: x12 gain (way too much noise though)

    x9 gain would have been enough but we cannot achieve that with this fpga power signal. Perhaps it's because the power spectrum of the signal contains more power than just a simple sine wave?
  • Hello Jared,

    Could you attach a schematic of exactly how you are driving the LMH6881 output into the FPGA current sensing resistor? Also, it could be helpful to know what is the value of the FPGA current sensing resistor and the maximum current limit which the FPGA input can handle?

    Do you know what is the Scope probe impedance set to? I am wondering that for the voltage signal ---> LMH6881 ---> Mini-Circuits ---> Oscilloscope: x12 gain scenario, the LMH6881 gain is x9 and not x3 which amounts to total gain of x36. I think the too high of a noise you are seeing might be because of that. It would be helpful if you can attach oscilloscope screen-shots of the same.

    One thing I noticed is that the output P1dB compression point of the mini-circuits zx60-3018g is only 13dBm into 50ohm load, which is only 2Vpp(rms). Do you know what amplitude are you measuring at the oscilloscope? It might be possible that you could be running into compression issues at the output of Mini-circuits amplifier and the noise is due to that. Is it possible that you can cascade it in this way: voltage signal ---> Mini-circuits ---> LMH6881 ---> Oscilloscope? You can also check this by programming the LMH6881 to a lower gain setting.

    Best Regards,

    Rohit