• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 47 subscribers
  • Views 77 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.

Original question:

LMV712-N: LMV712TL/NOPB output dithers but not in response to input signal

LMV712-N: follow-up to "LMV712-N: LMV712TL/NOPB output dithers but not in response to input signal"

James Dupre
Prodigy 90 points
Part Number: LMV712-N
Other Parts Discussed in Thread: LMV712

Tool/software:

Hi,

Can you please confirm that the 2.6 mVpp output noise includes the dithering that is *very* low frequency (on the order of seconds) or if it applies only to broadband or "white" noise? Because my experience with noise in analog circuits is that it's typically random in amplitude and much higher frequency unless there is a deterministic noise source such as a switching regulator that is coupling into the signal of interest (in this case, the output of the LMV712). It seems like the 2.6 mVpp noise is very low frequency because the flicker frequency in the analog engineer's calculator was set to 1 for voltage noise (1 = 1 Hz), but it's not clear that the noise has nearly discretized levels as opposed to being mostly random in amplitude. Also, can you elaborate on what causes the output to dither at such low frequency and with nearly discretized levels? Is it possible charge trapping affecting internal MOSFETs of the device? Thank you in advance.

Regards,
James

  • 0
    TI__Guru**** 153506 points

    James, 

    James Dupre said:
    discretized levels

    This type of noise could be popcorn noise. This noise is also called burst noise and random telegraph signals (RTS). This type of noise is somewhat rare to very rare. What percentage of samples have this profile?

  • 0
    Prodigy 90 points

    Hi Ron,


    Thanks for your reply. Unfortunately, I don't have specifics in terms of the percentage of devices that have manifested this behavior. We have assembled around 300 of the PCBAs that use the LMV712 (one device per PCBA) but only confirmed this behavior on 2 PCBAs with an oscilloscope. However, our testing of these PCBAs does not have the ability to detect the low levels of noise that are characteristic of this issue, so it's possible there are many more affected devices. But if I tried to assign a percentage, it would just be a guess.

    I did a quick search on burst noise and found that the waveform in this Wikipedia plot strongly resembles the waveforms I've observed with the oscilloscope: https://en.wikipedia.org/wiki/Burst_noise. If this behavior is truly burst noise, are there op amps designed to be immune to such noise? If so, how can I identify these devices based on datasheet parameters? Also, is susceptibility to burst noise dependent on process variations such that devices from one lot may be affected but not devices from other lots?

    Regards,
    James

  • 0 in reply to James Dupre
    TI__Guru**** 153506 points

    James,

    The reason that time was taken to test these two boards could be used to determine that rate. If the reason was high noise was suspected, then the rate would be 2/300 = 0.7%. Popcorn noise is either obvious, not present or not detectable because it is much smaller than random noises sources. 

    In general, this is a bipolar problem. It is much rarer on CMOS process like LMV712. I would expect it to be even less common on newer CMOS (pA or less IIB) op amps that use more modern processes.