• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 49 subscribers
  • Views 1086 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.

DVGA input/output external biasing

Abel woldemeskel
Prodigy 100 points
Other Parts Discussed in Thread: LMH6881, THS4500

Hello , 

I am using lmh6881 for a certain project , my question is i have another differential(THS4500) before lmh6881 that outputs a differential signal output with DC offset Voltage at 2.5V.  On datasheet(page 19) it states that it is possible to externally bias inputs to 2.5 , therefore will it make sense as a design point of view to directly input this differential signals to the DVGA (Lmh6881)? Also if i do that the datasheet also states that it is internally biased to 2.5V will that make DC offset at 5V and cause the DVGA to malfunction?  

I would greatly appreciate  any help on this  ,

Thank you. 

Regards 

  • 0
    TI__Genius 13935 points
    Hello Abel,
    The output of the THS4500 at 2.5V can be directly connected to the LMH6881 input which will aleady be self biased to 2.5V.
    The bias voltage is common mode, not differential. The LMH6881 will have a 2.5V common mode voltage by default, but it will have a 0V differential voltage. The THS4500 output should also be set with a common mode voltage of 2.5V. Common mode voltages do not add.
    The differential voltage of the THS4500 will become the input voltage for the LMH6881. Common mode voltages do not contribute to the output voltage.
    Regards,Loren
  • 0 in reply to Loren Siebert 1
    Prodigy 100 points
    Hi thanks for the reply ,

    I am still learning these material and please bare with me ,

    Also i have another question , right now i followed the datasheet for THS4500 and i am leaving the Vcm input pin of THS4500 floating which resulted in a differential signal centered at 2.5V which is as per the datasheet.( i have physically tested this and behaved as described) But as a design point of view is it better to bias the Vcm input pin to 2.5V or would it also be ok to leave them floating.

    Thank you so much for your time and help.
  • 0 in reply to Abel woldemeskel
    TI__Genius 13935 points
    Hello Abel,
    I looked at the THS4500 datasheet and it is OK to leave the Vcm input pin floating. In Figure 106 of the datasheet it shows that the Vcm pin voltage is set by internal resistors such that the voltage is the mid supply point.
    Regards,Loren
  • 0 in reply to Loren Siebert 1
    Prodigy 100 points

    Ok Great.. Missed that somehow on the datasheet.  I will physically  test these combination very soon.

    Thanks  for the help Loren.

  • 0 in reply to Abel woldemeskel
    Prodigy 100 points
    Hi Loren ,

    I tested the LMH6881 as a single ended input , but signal is very noisy and am not able to get results as expected. Also we like to drive our ADC using this part so we plan to connect VOCM pin to ADC Vcm which is 0.95 Volts. However in datasheet page 21 . it is recommended to use VOCM from 1V - 2V. Will that be detrimental to our design?

    Also we are grounding the digital pins instead of actual digital inputs right now could that be the reason the DVGA is malfunctioning ?

    I am using the recommended circuit of figure 45 on page 22.

    Thanks ,

    I really do appreciate your help.
  • 0 in reply to Abel woldemeskel
    TI__Genius 13935 points

    Hello Abel,

    Using 0.95V on the Vocm pin will be OK, that will give an output common mode of 1.9V which is only 100mV under the recommended voltage.  


    Have you confirmed that the DC voltage at the ADC input is correct?  Are you using the resistors to ground as shown in figure 45? 

    I think that figure 46 is more likely to be the circuit that you want, figure 45 does not have a full divide by 2 on the common mode, so the ADC input may not be biased correctly.  

    Regards,

    Loren