• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 32 subscribers
  • Views 1100 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.

TM4C1294KCPDT: Consequences of VREFA+ greater than VDDA by small amount

Oliver Douglas
Prodigy 180 points
Part Number: TM4C1294KCPDT

Hello,

If I use a 3.3V (+/- 2%) regulator for VDDA, will there be any adverse consequences to using a 3.3V (+/- 0.1%) VREFA+? In the worst-case scenario, I would have a difference of VREFA+ - VDDA = 3.3 * (1.001 - 0.98) = 69 mV. This is a small difference, but since the TM4C1294KCPDT datasheet lists the nominal and max value of VREFA+ as VDDA (pg 1859), I am curious if any negative impact on ADC performance would result.

I noticed that the design guide (pg 22) mentions using a 3.0V reference for VREFA+. However, I assume a larger VREFA+ is better since it slightly reduces the effect of noise and offset errors in the sensor and conditioning circuits.

For reference, here is what I am considering doing:

  • 0
    Guru 27665 points
    You could just go with a 3V (or 3V2) reference. That avoids the issue entirely. I doubt there's a significant advantage to increasing the reference voltage by 10%.

    There is a significant advantage in making sure VDDA is properly filtered so switching noise does not intrude.

    Also verify the current requirements of VREF, they could be quite high. A/Ds often have sharp impulse currents. That leads to having to pay particular attention to the capacitor requirements.

    Robert
  • 0 in reply to Robert Adsett
    Guru 117855 points

    This much "care/concern" directed to the (inherently) compromised MCU's ADC appears curious.    (perhaps prematurely optimized & in disregard of the superior results achieved via "purpose designed" dedicated ADCs.)

    Such stand alone ADC's may be physically located to minimize input signal trace length while avoiding fast switching signals - both proven to degrade "serious" analog measurements - and much present w/in "mixed signal" devices (i.e. any vendor's MCU)...

  • 0 in reply to cb1_mobile
    Guru 10570 points
    Agreed, if Oliver is expecting 12 bit accuracy out of the device a re-read of the specification is in order. And perhaps a re-read of the sensor specification as well.

    Robert
  • 0 in reply to Robert Adsett72
    Guru 117855 points

    Indeed my (always) "Pre-Mature Optimizing Avoiding" friend - indeed.    And (another) "great Robert point" re: Sensor spec, as well.    It is "overall" system accuracy - not just (one) element - which proves vital!

    It is expected that such (early) optimization attempts do NOT end w/the ADC alone...

    Visit to "Robert's bookshelf" - very much indicated.

    (lip-gloss - applied (even) w/exacting care (especially w/such care) - may not overcome "normal/customary" device flaws/weaknesses...)