• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 918 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.

OPA388: noise issue

Yong Xu
Prodigy 20 points
Part Number: OPA388


Hi,

I built the following circuit to measure the noise floor of OPA388.  R2=200k ohm, R1=100 ohm (precision metal resistors). The op amp is power by + - 3 V lithium batteries.   I then recorded the output using 24-bit NI DAQ board with a sampling rate of 1 kHz.  I calculated PSD using Matlab periodogram function, and finally divided sqrt (PSD) by the gain (1+R2/1R1=2001), referring the noise back to the input.

The resulting noise PSD (V/sqrt Hz) is shown below. I also included the noise plot from OPA388 data sheet for a comparison.  Above 1 Hz, the noise is 3-4 times larger than the value reported in the data sheet (7 nV/sqrt Hz).  What concerns me more is the 1/f region.  OPA 388 should have zero 1/f noise according to the data sheet.  Does anybody know why?  I also measured the noise floor of NI DAQ board by short circuiting its input.  It is much smaller, especially at low frequency region.  So I know it is not caused by DAQ.

  • 0
    TI__Guru* 93105 points
    Hello Yong,

    I am unable to open your attached images. Can you send them again?

    Please note that precisely measuring the spectral noise density of low-noise operational amplifiers and obtaining accurate results requires specialized methods that has required much development on our part. The ability to produce the results such as those seen in the graph of OPA388 datasheet Figure 15 are a result of the effort.

    Once we see your circuit and results we may be able to offer some general suggestions, but we can't reveal any proprietary information about out measurement methods.

    Regards, Thomas
    Precision Amplifiers Applications Engineering
  • 0 in reply to Thomas Kuehl
    Prodigy 20 points

    I encountered some difficulty of inserting figures/images.  I tried a few times but still couldn't insert images.   Anyway, the circuit is simply a non-inverting configuration (R2=200k, R1=100) with a again of 2001.   the input signal is replaced by short circuit, so the output will be noise voltage.

    the resulting noise PSD has very significant 1/f components.  Can you kindly advise what could be the source of this 1/f noise?

  • 0 in reply to Yong Xu
    TI__Guru* 93105 points

    Hi Yong,

    Since the OPA388 is a true chopper op amp it should not exhibit the 1/f noise region characteristic. The 1/f noise normally observed with non-chopper op amps is aliased by a chopper amp to the chopping frequency region where it is filtered by the internal sync filter (see the EDN article link listed below). There really shouldn't be any evidence of 1/f noise in your measured noise vs. frequency results.

    Check to make sure that you obtained your OPA388 from an authorized TI distributor. If by chance it was obtained elsewhere, there is the possibility that it could be a conventional, non-chopper op amp marked as an OPA388. In that case, it would exhibit a 1/f noise region.

    Regards, Thomas

    Precision Amplifiers Applications Engineering