• State
  • +1 person also asked this people also asked this
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 105 views
  • Users 0 members are here
Support feedback
Options
Options
Related

ADS8598S: Spikes in measured data

Raman Sridahran
Prodigy 150 points
Part Number: ADS8598S
Other Parts Discussed in Thread: DAC8718

Tool/software:

Hi,

I have been encountering an issue with the ADS8598S.  I am getting some readings (seemingly) correct, but several readings seem to read incorrect voltages.

Right now I am communicating the ADC from an FPGA, which is also driving a DAC; for now just measuring the values from this ADC and reading them out through the DAC.  I believe the DAC is working correctly; it is an DAC8718.  I can output known codes or patterns to the DAC and it outputs follow.  To confirm it is the ADC and not the DAC, I have 8 LEDs connected to the fpga and I output the first 8 bits of the ADC data to the LEDs. When feeding +3V to the ADC I can see the MSB-Led flicker; indicating negative readings (since ADC outputs two's compliment).  This matches the spikes I see in the DAC signal.

I have also probed DB15, and can see it is often high for the MSB (sign) of the output of my channel.

The other strange thing I have noticed is that FRSDATA, which is supposed to pulse high on the output of the first channel, seems to pulse high several times during output of data.  This certainly seems to be a problem.

AVDD: 5V

DVDD: 3.3V

VRef: Internal

FSample: 100kHz

CS/RD tied together, Parallel mode (reading bits 17:2 then 1:0 on consecutive pulses of CS/RD.

One other thing to note: I see that the internal VRef, as measured by the REFIN/REFOUT pin is 2.9V, whereas in the datasheet it should be 2.5V.  Could this be part of the problem?

Thanks, looking forward to advice.

  • 0
    TI__Guru**** 180941 points

    Hi Raman,

    Can you provide a schematic of your connections around the ADS8698S and perhaps a screen shot (o'scope or Logic Analyzer) showing CONVST, RD/CS and FRSTDATA?  For the reference, is that measured with a DMM or an o'scope?  If DMM, can you verify using an o'scope that the reference is a stable DC value and not oscillating?

  • 0 in reply to Tom Hendrick
    Prodigy 150 points

    Hi Tom,

    The reference was measured with a scope, and it is very stable DC value.

    Below is a screenshot from a logic analyzer, showing CONVST, CS, RD, BUSY, FRSTDATA, and DB[15]. 

    FRSTDATA seems incorrect; and as I showed 

    For my schematic, I have configured it as suggested in the manual; 

    Thanks

  • 0 in reply to Raman Sridahran
    TI__Guru**** 180941 points

    I'd still like to see your schematic and the actual raw data from the ADC.  If the reference is stable, it won't cause the spikes.  If you can get CONVSTx, CS/RD, BUSY and FRSTDATA captured with a 4 channel oscilloscope, that would also be helpful.

  • 0 in reply to Tom Hendrick
    Prodigy 150 points

    Hi Tom, please see the connections above.  I am measuring on AIN1 (Blue). 

    Unfortunately I dont have the scope output right now; I will post it soon. 

    But I do have another data point.  I switched to use SPI instead of Parallel data transfer.  In doing so, the spikes disappeared.  But, I have another strange issue in SPI mode; I get a weird jump in the signal when it crosses 0 - and it depends on the slope (positive to negative, or negative to positive):

    I guess my first question is to identify the issue with the parallel transfer, but I think this will be an issue as well.  Do you have any recommendations on what could cause this as well?

  • 0 in reply to Raman Sridahran
    TI__Guru**** 180941 points

    I will wait until you have your proper o'scope captures and can provide the raw data that I asked for.