• State Resolved
  • Locked Locked
  • Replies 16 replies
  • Answers 1 answer
  • Subscribers 50 subscribers
  • Views 3259 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.

TINA/Spice/LMH6629: 111

Ali Haddadpour
Prodigy 80 points
Other Parts Discussed in Thread: LMH6629, TINA-TI, BUF602, LMH6559, OPA656, OPA300, OPA857

 

  • 0
    TI__Expert 7875 points

    HI Ali,

    Your problem is likely with the buffer configured amplifier that you are using. The LMH6629 requires a minimum gain of 10 V/V in its default configuration for stability. The WSON-8 package includes a compensation pin that can be driven to reduce the minimum required gain to 4 V/V at the loss of some bandwidth, which would work for your last stage.

    It's a bit challenging to comment on the stability of your transimpedance stage because it looks like you are trying to "bootstrap" the capacitance of the diode which will have some effect on the amplifier stability. It may be in your best interest to try and use a more stable amplifier if you can take the hit in bandwidth.

    Regards,  

  • 0 in reply to Ali Haddadpour
    TI__Expert 7875 points
    Hi Ali,

    When you say 110nF capacitor, do you mean the 10nF series capacitor you have connected for the bootstrap? The extra noise is an unfortunate side effect of the transimpedance configuration. The only way to reduce the noise peaking is to move the noise gain pole cause by the input capacitance by reducing the capacitor. Can you possibly use a lower capacitance source (I'm assume a photodiode?)

    Regards,
  • 0 in reply to Ali Haddadpour
    TI__Expert 7875 points
    Hi Ali,

    I understand what you mean now. When you say you are seeing noise at the output. Do you mean from a noise simulation or do you mean you are seeing voltage disturbances on the output with a transient simulation? Remove the large input capacitor should have actually reduced the total noise of the circuit, but based on the configuration of the bootstrap, I would expect it to still oscillate.

    Regards,
  • 0 in reply to Ali Haddadpour
    TI__Expert 7875 points
    Hi Ali,

    Are the voltage disturbances you see periodic? I think that they may actually be an oscillation occurring most likely from the bootstrap amplifier. Are you able to share an image of what you are seeing?

    Regards,
  • 0 in reply to Ali Haddadpour
    TI__Expert 7875 points
    HI,

    That definitely may be an oscillation occurring. I noticed your scope is only 200 MHz of bandwidth so it may not actually be able to capture the oscillation because the LMH6629 has much higher bandwidth than that. Do you have a high frequency spectrum analyzer that you could use to look at the frequency content of the signal? Preferable several GHz of bandwidth.

    Regards,
  • 0 in reply to Ali Haddadpour
    TI__Genius 15975 points

    Hi Ali,

    Sorry for the delay in reply.

    1. It seems that the first stage LMH6629 TIA is a bit undercompensated which is why you might be seeing oscillations in the output waveform. I would recommend you to increase the feedback capacitor in the first stage to 500fF (or higher to 1.2pF) and see if the oscillations go away.

    2. Most high speed op-amps (LMH6629 is no different) tend to oscillate when the feedback resistor is high unless they are compensated otherwise using a feedback cap. I noticed that on the 2nd stage the Rf value is 9.5k, which might be the second source of oscillation. I would recommend to make that as 500-ohms and subsequently the Rg as 50-ohms.

    3. I would recommend the C4 cap to be reduced from 100pF to 10pF in-order to get more BW out of the cascaded stages.

    The AC response below for the above suggestions in TINA-TI look to be good with no apparent peaking, and should indicate the system is stable. I have attached the TINA-TI model for the same.

    Best Regards,

    Rohit

    LMH6629_NoBootstrapTIA.TSC

  • 0 in reply to Ali Haddadpour
    TI__Genius 15975 points

    Hi Ali,

    I don't think I have the bootstrap in my model, because I was assuming that you were seeing oscillations without the bootstrap?

    Also, for the bootstrap to work properly, you would need the BW in the bootstrap amplifier to be at-least 10x that of the first stage TIA BW. I don't think you have correctly configured the LMH6629 bootstrap op-amp in the figures I am seeing, because the device cannot support unity gain configuration. I think it would better if you switch to OPA656 which can support unity gain operation and has FET inputs. Unity gain buffer amplifiers such as the LMH6559 or BUF602 are also good choices for bootstrapping across the photo-diode cap and LMH6629 TIA gain stage inverting input.

    Best Regards,
    Rohit

  • 0 in reply to Ali Haddadpour
    TI__Genius 15975 points

    Hi Ali,

    What is the issue you are seeing with the bootstrap? Is it oscillations? You could try to increase the value of Rf and Cf and see if the TIA is stable as shown below. 

    Also, is there a reason as to why you absolutely need the bootstrap? I understand that you want to reduce the photodiode (PD) capacitance, but it comes with the added risk of instability across the TIA loop as well as the bootstrap loop. From the OPA300 datasheet page 8, you can see the output impedance for G = 1 starts to increase at fairly low frequencies ~100kHz with 20dB/decade. It is possible that the issue you are seeing is because of this increase in output impedance where the bootstrap loop is unable to keep the PD cap low at high frequencies. That's the reason for my recommendation of a faster buffer such as the LMH6559 or the BUF602. 

    The other thing is that the problem you are trying to solve can be achieved without using a bootstrap as well. I have attached a schematic below of the LMH6629 achieving 100MHz BW without using bootstrap. If you looking for even higher transimpedance gain than 1.5k-ohms, then I would suggest looking at the OPA857 which should give ~105MHz BW with 20k TIA gain. 

    Best Regards,

    Rohit

    1. LMH6629 with bootstrap:

    2. LMH6629 without bootstrap (100MHz BW):

  • 0 in reply to Ali Haddadpour
    TI__Expert 7875 points

    Hi,

    My apologies for the delayed reply. I would remove the 100pF capacitor on the non-inverting input as that is likely causing the oscillation on the second stage.

    Regards,