• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 517 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.

LMH5401: Vcm input

Ramin Fardi
Intellectual 280 points
Part Number: LMH5401
Other Parts Discussed in Thread: LMH6554, ADS54J54

Hi,

I’m working on a 4 channel FMC converter board design that strictly requires DC coupled signal path through out. The input signal range is +/-0.5V single ended, so I have to use a diff amp as input stage. Most of the high speed ADCs including ADS54j54 have a rather high Vcm around 2V but most of high speed diff amps like LMH5401 have rails limited to +/-2.5V. This makes connecting Vcm output of the ADC to Vcm input of the diff amp impractical at best because the output of the diff amp will go too close to the positive rail (even with G=0dB) resulting in unacceptable harmonic distortion. I was considering LMH6554 which can be run with +/-5V rails but the part has too much HD2 and HD3 at higher frequencies (above 100MHz) degrading ADC linearity as a result. Is there anyway round this problem? Any suggestion device-wise, diff amp or ADC that works better together?

  • 0
    Guru 48395 points

    Hey Ramin, 

    Now you understand, one of the magic things about FDA's is they can take a +/-Vin around ground and level shift it up to swing around Vcm. So a 2V Vcm running the LMH5401 on single 5V supply should work fine. The input pins get level shifted above ground also. The LMH5401 has a pretty poor Vcm offset drift, so if your ADC has a tight tolerance input Vcm, concern the apps circuit in the data sheet, This Vcm servo loop idea actually came from some earlier work I was doing here for an AC coupled output filter - the difficulty there was a lot of ADC's have a clock rate dependent Icm input current that would vary the DC drop through the Filter R's. Adding the servo loop as part of the diff to single sense path fixed all that. 

    https://www.renesas.com/us/en/www/doc/guide/an1837.pdf

  • 0 in reply to Michael Steffes
    Intellectual 280 points

    Hi Micheal

    That’s indeed very good news. I will certainly give try it as soon as I get the EMV for LMH5401. Our application is a very high precision pulse digitizer, any ideas about compensation (calibration) of DC offsets and gain variations?

    About the Vcm servo loop, it seems smart thing to do.

     

    Thanks for your clever advise

     

    Regards Ramin

  • 0 in reply to Ramin Fardi
    Guru 48395 points

    you bet Ramin, 

    There are quite a few sources of DC error in an FDA, I tried to work through those in this recent article - Aside from the basic FDA errors, Resistor matching helps a lot, 

    https://www.planetanalog.com/author.asp?section_id=3404&doc_id=565037

  • 0 in reply to Michael Steffes
    Guru 48395 points

    I was trying to refresh my memory on just how bad the Vocm offset is on the LMH5401, these spec lines confuse the heck out of me, 

    But there is a plot, looks like a typ drift, what is missing is worst case 25C range and tempco range. Should have had a histogram of both.