• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 164 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.

DAC8811: Sudden Noise in Analog Output Circuit Using DAC8811

Conor
Guru 12235 points
Part Number: DAC8811
Other Parts Discussed in Thread: OPA2192

Tool/software:

Hi,

Currently, one of my customers is using a DAC8811I in an analog signal control circuit, and has noticed a phenomenon in which sudden noise (approximately 300nV) occurs intermittently at the output (Vout).

■ Circuit Configuration

The circuit is based on Figure 26 in the DAC8811 datasheet, with the following component values and specifications:

Item Part Number Specification
DAC DAC8811IBDGKT VDD = 3.3 V, GND = 0 V
Trans-impedance amplifier OPA2209AIDGKT V+ = +15 V, V− = −15 V
Gain & offset stage amplifier OPA2209AIDGKT V+ = +15 V, V− = −15 V
R_G1 - 2 kΩ
R_G2 - 2 kΩ
R_REF2 - 2 kΩ
Input (VREF) - Sine wave, Vout = ±5 V (1 V/div), f = 8.2 kHz

■ Description of the Noise Phenomenon

  • Phenomenon: The noise appears as a deviation between Vout and VREF, up to approximately 300 nV.

  • Waveform: Resembles either a rectangular or jittery noise burst, occurring at random intervals. In some cases, it disappears after some time.

  • Frequency dependency and triggering conditions remain unclear.

■ Our Internal Analysis (Possible Causes)

We have considered the following possibilities on our side:

  • Ground loop between digital and analog grounds

  • Glitch current from the DAC output (Iout) during code updates

  • External noise coupling (e.g., EMI, ESD)

  • Ripple or transient disturbances from ±15 V power supplies

  • Potential instability at the MDAC output load (A1 input)

Could you please advise us whether such a noise phenomenon could be expected in this configuration, and if so, what countermeasures or design adjustments you would recommend?

Best regards,

Conor

  • 0
    TI__Guru 53983 points

    Hi Conor, let me take a look at this.

  • 0 in reply to Paul_Frost
    TI__Mastermind 24618 points

    To all,

    I agree with all the potential causes of the noise shown in the diagrams.  However it is also possible that it is popcorn noise in the amplifier.  To try and understand if this is popcorn or not I have a few questions.

    1. Does the noise follow the device?  In other words, does the noisy device always stay noisy or does the issue come and go.  In general devices with popcorn noise will remain noisy, so if the problem comes and goes it probably is not popcorn.  
    2. Can you replace the 2kΩ resistors with 20kΩ, and remeasure the noise?  Popcorn noise is a bias current step change and will be proportionate to the feedback and source resistance.  If the resistance increases by a factor of 10, the noise will as well.  This test is not definitive because noise pickup will often also increase with increasing source resistance.  Note that the thermal noise (random looking noise) will also increase by the square root of resistance.
    3. Can you shield the circuit?  EMI and other noise sources can look like the image shown.  It would be interesting to see if the noise goes away when the circuit is shielded.  
    4. Note that in general CMOS devices will not show popcorn noise.  The OPA2192 is a precision low drift, low noise CMOS device.  If you substitute the op amp for OPA2192 do you ever see this noise?
    5. Final Note: popcorn noise is often significantly larger than the specified op amp noise.  In this case the noise that has the "pops" looks only slightly larger than the noise without the pops.  Do you see any that are significantly different in amplitude?

    I hope this helps with the troubleshooting process.

    Best regards, Art

  • 0 in reply to Art Kay
    Guru 12235 points

    Hi Art,

    Thank you for your previous support regarding the noise observed in the DAC8811-based analog output circuit.

    Based on your advice, we are now conducting additional evaluations on the actual hardware, including considerations of possible external noise sources (EMI).

    We would like to confirm one point regarding the interpretation of popcorn noise:

    If the noise is determined to be popcorn noise, is it correct to assume that the source of the noise is more likely the op amp in the output stage, rather than the DAC8811 itself?
    Is it also correct to understand that the DAC8811, being a CMOS structure, is generally unlikely to exhibit popcorn noise?

    We would appreciate your confirmation on this interpretation. Additionally, please find our responses to the previous diagnostic questions below:

    Art Kay said:
    Does the noise follow the device?  In other words, does the noisy device always stay noisy or does the issue come and go.  In general devices with popcorn noise will remain noisy, so if the problem comes and goes it probably is not popcorn.  

    We have observed the phenomenon across multiple units. Once the noise appears, it tends to persist for 20–30 minutes, and then spontaneously disappears. Given this behavior, would you say that popcorn noise is less likely in this case?

    Art Kay said:
    Can you replace the 2kΩ resistors with 20kΩ, and remeasure the noise?  Popcorn noise is a bias current step change and will be proportionate to the feedback and source resistance.  If the resistance increases by a factor of 10, the noise will as well.  This test is not definitive because noise pickup will often also increase with increasing source resistance.  Note that the thermal noise (random looking noise) will also increase by the square root of resistance.

     Yes, we plan to perform this test by replacing the resistors with 20kΩ and will observe the impact on the noise amplitude. 

    Art Kay said:
    Final Note: popcorn noise is often significantly larger than the specified op amp noise.  In this case the noise that has the "pops" looks only slightly larger than the noise without the pops.  Do you see any that are significantly different in amplitude?

    So far, the noise amplitude has remained in the range of 300–500 nV, with only slight deviations. No clearly distinguishable step-like bursts have been observed.

    Best regards,

    Conor

  • 0 in reply to Art Kay
    Guru 12235 points

    Hi Art,

    Do you have any updates on this matter? Any updates on my comments from 6/11?

    Best regards,

    Conor