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.

LAUNCHXL-F28069M: Noise in the ADC reading.

Part Number: LAUNCHXL-F28069M
Other Parts Discussed in Thread: AMC1200-Q1, LM1117, C2000WARE

Hi you all, 

Due to this post I have been question to my self how to achieve a clean signal into the ADC, a person in this post told us about the poor accuracy or stability? of the  LDO  in order to get a good clean reference voltage I have been working with the LAUNCHXL-F28069M with a little sinusoidal signal provided by the AMC1200-Q1 and also I saw the spikes, and I used this configuration here:

 So I'm wondering if there is a way to achieve more clean signals not affected by the LDO or the LM1117, or if there are some guidelines while designing voltages to the adc. Even if I do that design, I have to reference with gnd of the adc, What happen if the ground of the ADC is not ok?

  • Mike,
    I'd like for you to run the example included with C2000Ware here: C:\ti\c2000\C2000Ware_1_00_05_00\device_support\f2806x\examples\c28\adc_soc\

    In the main .c file there is a call to the function "AdcOffsetSelfCal", we will use this function to further debug your issue.
    Once called, this function will re-direct to the F2806x_Adc.c source file where I'd like for you to step into the function.

    What you will see is that there is an internal channel, that is connected to VREFLO(which on the LaunchPad is shorted to VSS). Let the conversions happen, and then see what the data set looks like. Let me know if you still see spikes/noise on this channel or not.

    If there is noise it means that there is noise on the ground plane, or at least getting onto the pin ADCVREFLO. Since this is the ground reference of the ADC any noise on this pin will show up in the conversion process as well.

    If there is not noise on this conversion, then it means we must be getting the noise spikes from the voltage input to the channel you are converting.

    I'd like to understand this better before I suggest more debug tactics.

    Best regards,
    Matthew
  • Matthew
    I've already used the AdcOfssetSelCal. Also I tested the LAUNCHXL-F28069M and I see there are some spikes, the spikes add to my chain signal...
    Miguel
  • The question is more about if there is some LDO or Power Supply referenc to work with the ADC and the MCU in order to get such a clean voltage reference to the ADC
  • Mike,

    Just for clarity sake my understanding of the problem is:

    Q1)Can TI suggest a clean voltage source for the ADC to sample to see the performance of the ADC.

    Assuming this is true, the TI data converter group has this EVM http://www.ti.com/tool/PSIEVM for sale on the TI store.  It supports single ended as well as differential outputs and we have used this to demostrate SNR performance on both our 12-bit and 16-bit ADCs on our MCUs.

    Let me know if this helps.

    Matt

  • Matt

    Thank you for the reference, I think I don't saying it in a proper way, I'm looking for a recommendation about a power supply, one power supply which  has a clean voltage and little deviation of output voltaga, right now I'm using the LM1117 3.3V, but I have some spikes which screw up my signal, I can work with it, but I think there proper ways to do it, or in fact, some guidelines to work with a power supply dedicated to the ADC.

    I will check the reference to see what can work to me 

    Thanks

    Mike