• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 48 subscribers
  • Views 465 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.

ADS7828: Apparent HW latch-up

Joe SixPack
Prodigy 20 points
Part Number: ADS7828

Hello, I'm troubleshooting a design that uses the ADS7828 ADC and gives bizarre behavior.  The ADC will convert correctly for extended periods of time (hours) and then suddenly start reporting values that are offset from correct by approximately 1/4th of full scale.  The offset is constant once it shows up, e.g. 1.25V will read 0x1FF for a while and then suddenly jump to 0x0FF; if I increase the input voltage to 2.5V the reading goes to 0x2FF instead of the full-scale 0x3FF.  Cycling power on the board restores normal operation, consistently.  I assume this is caused by noise on my circuit board, but would like to understand how it's possible for noise to affect the IC this way, so I can add some kind of protection.  The analog inputs are all sourced with low-impedance and the reference voltage appears stable; is it possible for noise on SCA or SCL to cause this kind of behavior?  Thanks.

  • 0
    TI__Expert 5485 points
    Hi Joe

    There can be couple of reasons for such behavior

    1. Please monitor device VDD/ REFIN. If there is glitch on these lines which is forcing any of these pin voltage less that POR voltage of device that can cause abnormal behavior. When you recycle VDD power that resets the device and device starts working normally

    2. Please check that power rails monitoring VDD and REF are not noisy. You can try adding a small series resistor and a decoupling capacitor (RC filter) on VDD pin. When you say there is a noise in your circuit, can you elaborate more on this.
    Do you mean to say that you are hardwiring the devices to get the prototype working or you already have proto boards / hardware in place.

    Thanks & Regards
    Abhijeet
  • 0 in reply to Abhijeet Godbole
    Prodigy 20 points

    Hi Abhijeet,

    Thanks for your reply.  We have already fabricated boards, so this is not hand-built or wire-wrap.  This IC is being powered from 3.3VDC, and there are capacitors connected across its supply pins, so I think that situation is okay.  We are using an external 2.5V reference and it measures stable with an oscilloscope.

    I searched the data sheet for POR voltage but didn't find any.  I guess you might be referring to the minimum VDD requirement of 2.7V?

    Also, do you mean to say that there is a POR level for REFIN?  That seems strange since the data sheet says REFIN can be "as low as 50mV."

    The noise in our circuit comes from an external device that is similar to an ESD generator.  It isn't connected directly to this circuit, but when it operates we can see spikes on an oscilloscope without the probe connected to anything.

    I wonder if it's possible for the problem to be caused by the digital input (I2C) lines exceeding their allowed input range?  It's not simple to clamp the lines within 0.3V of the supply range, so I think these are the least well-protected pins.

    Thanks again,

    Joe