• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 50 subscribers
  • Views 1684 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.

ADS1278: ADS1278: problems with buffer

Jaroslaw Pulka
Prodigy 20 points

Part Number: ADS1278
Other Parts Discussed in Thread: THS4521, , TINA-TI

Hi All,

I would appreciate any help with my problem.

I have a data acquisition system based on ADS1278 (clock 25MHz) and THS4521 buffers. Buffers are connected as in Figure 89 of ADC datasheet. Capacitor across ADC input is 100pF. 

When I used 5.6nF (later tested with 1nF) capactor in bufer input, its power consumption was 40mA (!). After removing those capacitors current consumption decreased to 2mA, but I observed large oscillations on ADC input - approx 4Vp-p @ 6.3MHz. Could you advise me please if this is a normal situation and how should I approach the problem? Increase ADC input capacitor value?

Best Regards

Jarek

  • 0
    TI__Guru 68276 points
    Hi Jaroslaw,

    This sounds more like an amplifier question, so I've moved the thread to the High-Speed Amplifiers Forum. We may need to review your THS4521 configuration and verify that it is stable before considering the dynamic load of the ADC input.

    Best Regards,
  • 0 in reply to Ryan Andrews
    TI__Genius 15975 points

    Hi Jaroslaw,

    Sorry for the delay in reply. Would it be possible for you to attach your schematic in-between the THS4521 and ADS1278? Do you have any series resistor at the THS4521 output while driving the 5.6nF capacitor? I don't think the large oscillations you are seeing are normal and could be due to the THS4521 oscillating. Something you can try to stabilize the THS4521 circuit is add an additional 1.5nF cap across the THS4521 inputs.

    Best Regards,
    Rohit

  • 0 in reply to Rohit Bhat
    Prodigy 30 points

    Hello Rohit,

    Thank you for reply. Please find below schematic of the buffer and ADC input as requested.

    AIP(N) inputs are connected directly to the board input connector. Capacitors across ADC inputs are changed to 2.2nF and C42 and C46 are removed now. At the moment oscillation across ADC is much smaller but buffers consume around 20mA each. I will try later to increase the C15 to suggested value.

     

    Best REgards

    Jarek

  • 0 in reply to Jaroslaw Pulka28
    Guru 140200 points
    Hi Jarek,

    you wrote: "AIP(N) inputs are connected directly to the board input connector."

    That is the mistake! You don't have termination resistors at the left side of R40 and R45! And the 100p caps even worsen the situation.

    Take care, the circuit you have shown is expected to be driven directly by low ohmic drivers, which must sit directly at the left side of R40 and R45. An alternative method is to drive the inputs by a remote driver by the help of 50R technique. This would need a termination resistor at the left side of R40 and R45, a 50R line to the remote rivers, which again have 50R termination resistors.

    Please look at section 8.3 to see what I mean.

    Kai
  • 0 in reply to Jaroslaw Pulka28
    TI__Genius 15975 points

    Hi Jarek,

    Have you tried to increase the input cap (C15) value from 100pF to 1.5nF or higher value, and seen whether the oscillation goes away? I think your circuit will oscillate as it is because of not enough phase margin in the loop gain unity gain frequency with 100pF cap.

    For a 100pF input cap, your PM goes to about -1.7' meaning no phase margin leading to sustained oscillations in the circuit. The same PM improves to about 80' using a 5.6nF cap at the input, indicating a stable circuit. You can use the below circuit in TINA-TI to simulate stability of the THS4521 circuit, where the loop gain is Vout as shown below.

    To understand more about stability of fully differential amplifier such as the THS4521, I would recommend you to refer the following video.

    Please let me know if you have any questions.

    Best Regards,

    Rohit 

  • 0 in reply to Rohit Bhat
    Prodigy 30 points

    Hello Rohit, Kai,

    Thank you for your answers and suggestions. Regarding my circuit - this is what happens when one forgets about basics.

    To describe my system a bit more - input of the THS4521 is connected to the output of the AD8476 on different board (low ohmic driver). Buffer output still oscillates with  C8 (your schematic) set to 2.2nF. Removal of the capacitor solves the problem completely.

    Best regards

    Jarek

  • 0 in reply to Jaroslaw Pulka28
    TI__Genius 15975 points
    Hi Jarek,

    Understood. When you have the C8 installed as 2.2nF, do you also have the C6 and C7 installed as shown in the image I posted. If the C6 and C7 are not installed, then you are bound to oscillate using only C8 because of the PM reducing significantly at the unity gain loop gain crossover frequency as explained in my previous post. I will recommend you to install C6, C7 as 5.6nF and C8 as 2.2nF to see if the THS4521 output stabilizes.

    Best Regards,
    Rohit