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.

LMH6522: Impedance Matching

Part Number: LMH6522
Other Parts Discussed in Thread: ADC34J45, , LMH6521, TIDA-00073, ADS41B49, TINA-TI

Hello,

My design will include the ADC34J45 that will have JESD204B interface to ALTERA ARRIA 10 FPGA.

The four analog inputs will be amplified by the LMH6522 amplifier.

There are 2 frequency possibilities of analog signal inputs 189MHz or 236MHz.

The analog inputs parameters are :

  • single ended
  • Frequency: 189MHz or 236MHz.
  • BW: 30MHz
  • 50 Ohm.

The sample rate is 108MHz or 135MHz (Nyquist4) respectively to the input frequency.

Can you please advise what is the recommended impedance matching between the LMH6522 and the ADC34J45 and the recommended impedance matching between the

single ended input signal (50 Ohm) and the LMH6522?

Thank you,

David Siksik

  • Hello David,

    The single ended to differential conversion is most easily accomplished by a transformer.  On the LMH6522 EVM we use the JTX-2-10T transformer.  This provides the input impedance matching as well as the single ended to differential conversion. 

    As for the filter, I recommend using a 200 Ohm filter.  I believe this provides the best balance between the needs of the amplifier and the ADC.  Filter design and construction is a bit of an art.  I usually design a filter between 5% to 10% higher in response than desired and then it needs to be tuned on the PCB once it's built.  The reason to design the filter with a higher response is that the PCB solder pads will tend to detune the filter to a lower frequency.  This happens even if the ground and power planes are removed from beneath the filter.  

    I have linked two potentially useful reference designs below.  They use the LMH6521 which is a dual DVGA with very similar functionality to the LMH6522.

    Regards,

    Loren

     

  • Hello Loren,
    If I take the TIDA-00073 as reference design and with the help of the application note SBAA195 and TINA I will match the filter to a centre frequency of 235MHz, BW40MHz , can I say that the ADS41B49 equivalent input circuit is the same as in the ADC34J45?
    Thank you,
    David
  • Hi David,

    I don't think the ADS41B49 and ADC34J45 have the same equivalent input circuit. It looks like the ADS41B49 has differential input impedance of 10k||3.5pF at dc, whereas the ADC34J45 has a differential input impedance of 6.5k||5.2pF at dc. I would recommend you to post further questions on the data converters in the High Speed Data Converter Forum:

    Best Regards,

    Rohit

  • Hello David,

    There is a 2pF capacitance difference between the two ADCs. You should be able to adjust your filter to compensate.

    Otherwise the two ADCs do have very similar inputs.  

    Regards,

    Loren

  • Hello Loren

    According to application not SBAA195 I understood that it will be easy to take the filter in the example,change it first to the impedance of my ADC (ADC34J45) and then change the centre frequency to 235M.

    So do I have first to change the shunt capacitor from 13.5p to 10.5p (2pF difference between the 2 ADCs)?

    How do I then move the centre frequency from 184M to 235M? Do I have to redesingn all the filter?

    Thank you,

    David

  • Hi David,

    It is possible to change the shunt capacitor from 13.5p to 10.5p and use it at the ADC34J45 front-end instead of the ADS41B49. However, the center frequency will still remain at 184MHz. You would need to change the filter components in-order for 235MHz center frequency.

    I have attached a Chebyshev band-pass filter below with 0.01 ripple that you could use at 235MHz center frequency with TINA-TI file:234MHz_BandpassFilter.TSC . The last shunt-arm capacitor needs to be adjusted to the ADC34J45 differential input capacitance of 5pF as mentioned in the App-Note SBAA195.

    You could also reduce the filter component count by making the shunt-arm inductor and capacitors differential. As a result, the two shunt arm 9nH inductors could be replaced by single 18nH inductor; whereas the two shunt 50pF capacitor could be replaced by single 25pF cap. The 18nH differential inductor would stay the same on the last shunt-arm; whereas you could use 20pF in the last shunt arm (adjusted for the differential input cap of the ADC).  

    Best Regards,

    Rohit