• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 142 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.

AM263P4-Q1: AM263P4-ADC sequential and simultaneous sampling

Jayakeerthana P
Intellectual 405 points

Part Number: AM263P4-Q1
Other Parts Discussed in Thread: AM263P4

Hi Team,
For real-time monitoring of motor three phase voltage and current, we are using multiple channels of the same ADC module (ADC4) in the AM263P4. Since the channels within a single ADC module are sampled sequentially, there will be a time difference between the voltage and current samples.

We are observing sudden transient/noise spikes during monitoring. Is there any recommended ADC configuration or sampling method to minimize or eliminate the sampling delay while using the same ADC4 module? Please suggest the recommended SOC/trigger configuration or any other method to ensure the voltage and current measurements are as closely synchronized as possible.

  • 0
    TI__Prodigy 460 points


    Hi,
    •Sequential sampling within a single ADC module cannot be eliminated by configuration, since each module has one sample-and-hold. Recommendations:
    Use multiple ADC modules instead of only ADC4. AM263P4 has 5 ADCs + 2 ADC-Rs. Assign phase currents to different ADC modules and trigger all of them from the same EPWM SOC event. This gives true simultaneous sampling. Voltage channels can tolerate the small skew.
    •If you must stay on one module: place the most timing-critical channels in the earliest SOCs, use the minimum acquisition window your source impedance allows, and run the ADC at maximum clock. The remaining skew (a few hundred ns per conversion) is deterministic and can be compensated in software if needed.


    Regarding the transient/noise spikes: these are usually not caused by sampling skew. Please check:


    -Trigger placement: sample away from PWM switching edges (e.g. trigger at counter zero/period for low-side shunt sensing). Later SOCs in a long sequence may drift into a switching instant.
    -Acquisition time vs. source impedance of the driving circuit, and channel-to-channel settling.
    -RC filter at the ADC input pins and grounding/layout.


    If you can share where your SOC trigger sits within the PWM period and what drives the ADC inputs, we can look further.


    Regards,
    K.B.

  • 0 in reply to Kudra Baruti
    Intellectual 405 points

    Hi,
    We are observing this periodic noise of 12s on every signal  we are probing,
       whether this can eliminate in ADC reading

  • 0 in reply to Jayakeerthana P
    TI__Prodigy 460 points

    Can you please provide a bit more information on this. I do not have enough to go on.