• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 48 subscribers
  • Views 1972 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.

LMP7717 as unity gain buffer at high frequency and 1.8V single supply...

DONGJIN KWAK
Prodigy 40 points
Other Parts Discussed in Thread: LMP7717

Hi.

I need a unity gain buffer when input signal is 50MHz,  max 1.8V min 0V square wave. and I need to use single voltage supply 1.8V... 

So I surf in TI products, I found LMP7717. I followed the application circuit in datasheet.

figure58 circuit from datasheet, Lead-Lag compensation for non-inverting configuration for unity gain as shown below.

and transient simulation result is...

I changed input signal freq 1MHz, and capacitor 1pF, I got better result.. but this signal is not what i want...

I can see there is bandwidth violation at suggested configuration(C=10nF, Rc=330ohm, input signal freq 50MHz) from datasheet. 

I try another configuration gain +2 non-inverting configuration but almost same result at 50MHz.

I think this happen because of op amp's own bandwidth limitation.

Why this happen? datasheet says unity gain BW is 88MHz...

Should i use another opamp? Is there any other unity gain scheme using LMP7717 for 50MHz signal, single supply 1.8V?

Thanks you for reading!

  • 0
    TI__Genius 13935 points

    Hello Dongjin,


    I think there are several things you need to reconsider.  First, if you want to reproduce a 0V to 1.8V signal you need a supply voltage that is slightly larger than your signal.  Can you use a 3.3V supply instead of 1.8V?  Can you use a split supply? 


    Second, with a square wave signal you need to figure the slew rate and ultimate bandwidth required to create the signal fidelity you desire.  An ideal square-wave has infinite bandwidth, so you need to determine the rise time and fall time that you need and then find an op amp that has that slew rate and bandwidth.  Since your signal is fairly large you cannot look at small signal bandwidth, you need to look at the op amp large signal bandwidth.

    Generally speaking you need at least 5th order harmonics to make a reasonable approximation of a square wave.  For a 50MHz square wave the 5th harmonic is 250 MHz.  The slew rate you need is around 1000V/us.  If you find an op amp with 250MHz large signal bandwidth and 1000V/us and use a 3.3V supply I think you'll have better results. 

    [EDIT] I originally calculated bandwidth and slew rate for the 3rd harmonic, but third harmonic only makes a very sloppy square wave. 

    Regards,

    Loren