• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 46 subscribers
  • Views 1390 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

using ADC ADC3444

Shyam Daram
Expert 1285 points
Other Parts Discussed in Thread: OPA699, THS4541, ADC3444

Hello TI,


we have an application in which we are trying to measure the pulse amplitude for power calculation purpose, the pulse is very narrow that is 10ns, i am done in designing first stage, where i use an TIA which senses 10ns pulses and amplifies by factor 1K and gives me inverted output

following which i am using a OPA699 as inverter stage and also expanding the pulse up to 1us using a 1nF and 1K elements after OPA699 stage

now the problem is, the output of OPA699 is 0-2.5V p-p, i want to convert this input analog signal to digital equivalent, for which i need an ADC which has the reference voltage as high as 2.5V or atleast 2.2V, as the signal i want to convert to digital should be essential a peak value conversion, i need to capture the ADC value immediately so i use a high speed ADC atleast 50MSPS so that i can get atleast 10samples closer to peak if i work at clock of 400MHz


i dont want to collect all samples also, i just want to take few samples for power measurement of input signal


i have 3 similar stages so all three stages digital equivalent has to be taken, this ADC output goes to a FPGA


i tried to use a THS4541 to convert the single ended output to fully differntial to interface it to ADC3444, but i have seen huge DC being inserted at output of FDA


so please suggest me ways to calculate the amplitude or the peak value of pulse, the pulse is 10ns , i have expanded it to 1uS and it repeats at rate of 20uS

as FDA is causing me these problems i thought of eliminating it and directly giving single ended input to ADC and grounding other terminal, please let me know what are the implications if i use a fully differential ADC as single ended

  • 0
    TI__Genius 12965 points
    Have you looked at this TI Design? It gives some ideas on DC coupling to ADC using THS.

    www.ti.com/.../tidu500.pdf


    Ken
  • 0 in reply to Ken C
    Expert 1285 points
    Thanks for help, i will come with the difficulties i am facing, after i complete design of interfacing differential driver , for now i have following questions

    i need to place a pulse expansion circuit of simple RC at the FDA output,

    moreover, what is the simplest way to interface this ADC to an FPGA, for me it looks like the serial LVDS is simplest to use, as i dont want to put energy in understanding and implementing or purchasing a IP core of JESD204B for the FPGA, any ways my datarate is not exceeding 50MSPS so i feel serial LVDS should serve my job
  • 0 in reply to Shyam Daram
    Expert 1285 points

    first of all sorry that i am not able to use TINA in below designs, just becuase the other party doesn’t provide spice models, i searched for converters even but unsuccessful

    ok here below is the design which tried to expand pulses after single to fully differential conversion

    where exactly problem happens is, after the expansion the amplitude of signal drastically goes down, almost by 10 and pulse is also not expanded as desired, as i have said, i want to expand the pulse and feed it to ADC, so that i can use a slower ADC and slower clock, as its only power measurement only, do you think i have to try a different approach and unlearn this approach ?

    one more thing i have done is tried to expand before, but when i feed such signal to the FDA , i see it going unstable