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.

ADS1256: Power Drain during channel switching

Part Number: ADS1256
Other Parts Discussed in Thread: MSP-EXP432E401Y, TPS63020

Hi,

This is in regards with ADS1256.

History: The ADS1256 VDD is connected to 3.3VDC rather than 5VDC, this is an design error. Since we did not notice any abnormal operation on the ADC functionality and clarification from this forum we have not modified the circuit and continued.

We are using all the 8 Channel of the ADC for measurement of Temperature (1 Channel), pH (1 Channel), conductivity (4 Channels) and Turbidity (2 Channels). The ADC reference voltage is set to 1.6VDC

Test setup: 3.3V to the system is provided from the Launchpad MSP-EXP432E401Y.

System operation: On completion of AIN0 sampling and conversion, the channel is changed to the next channel until AIN7 is complete. The sampling rate for each channel is different at mentioned below.

AIN0 Sample rate 2.5SPS with 10 Samples

AIN1 Sample Rate 2.5SPS with 20 Samples

AIN2 Sample Rate 25SPS with 10 Samples

AIN3 Sample Rate 25SPS with 10 Samples

AIN4 Sample Rate 25SPS with 10 Samples

AIN5 Sample Rate 25SPS with 10 Samples

AIN6 Sample Rate 3750SPS with 10 samples

AIN7 Sample Rate 3750SPS with 10 samples

Observation 1: With this setup we the observed the ADC reading for AIN0 or AINx are different from the reading taken when sampling only one channel (collect 10 sample do required calculation and continue to collect samples in a loop). Basically if the sequence is followed we get 'x' value from the ADC for AIN0 and if only AIN0 is sampled continuously without channel change we get 'y'. The difference (x-y) is linear for all inputs on AIN0. We would like to understand why there should be a difference in the conversion data if the channel is changed. We are debugging the code to check if there is a firmware bug.

Observation 2: When the system supply is provided using TPS63020 (BUCK-Boost) with a battery input, we see that there is power drain (Voltage drops linearly 3.3V to 2.8V and rises linearly to 3.3V) for a period of 3ms on the 3.3V rail which is resetting the micro-controller (MSP432E4) when we are changing the channels of the ADC, it is quite random sometimes it is on the first channel change or second channel change.Which is not observed when 3.3V is provided from the Launchpad.

We have tested the TPS63020 separately by applying with sudden load upto 1A, the drop noted is 0.1V, but no micro reset (micro operating voltage is 2.9 to 3.5). The drop is from 3ms, we tried adding 1000uF cap to the 3.3V rail, but no change. TPS63020 power save mode is disabled by connecting it to VINA as recommended. We changed the inductor for high current capability, still there is no change in the power drain. Based on the testing we have done. We are confident the problem noticed is not related to the TPS63020.

When we insert a new battery, the power drain on the 3.3V is noted when the battery open voltage is 2.7V. We are testing the TPS63020 functionality and not observed any issues with it till now, no load or with load or when load is applied suddenly up-to battery voltage of 2.3V. (Load applied on 3.3V is 250mA)

Since the problem is noted only during the channel changes. We would like to know if there is any specific sequence to be followed before changing the channel. In debugging mode we are unable to note exact point of reset. We have only LED indication to provide the info on micro-reset.

Regards,

Muruga

  • Hi Muruga,

    Did you have additional questions regarding the AVDD voltage? I provided some guidance in a previous post (see: e2e.ti.com/.../751792), but perhaps I should have clarified that running AVDD at 3.3V is not recommended as it is outside of the datasheet specifications.

    Do you know what the typical input voltage are on each channel? Perhaps there is some relationship with the reduced AVDD voltage, but it is also possible that when switching channels the PGA inputs could be over-ranged (causing the PGA outputs to saturate) and drawing additional current. It might be a good practice to program the ADCON register (to reduce the PGA gain) first, and then update the MUX register when switching channels to avoid this possibility.

  • Hi Chris,

    Thank you for the response.

    We changed the AVDD of ADS1256 to 5V from 3.3VDC. All channel sampling is done with PGA set to 1. As the AVDD is set to 5V, the input voltage AINx should not be a problem. We have replaced the ADS1256 with new IC before testing with 5V AVDD.

    We could share the complete schematics if required for your reference.

    Following are testing done and observations:

    1. Changed 3.3V AVDD to 5V AVDD.
      1. Still the problem is observed on the 3.3V rail, the voltage drop to nearly 2.9V, resetting the micro.
    2. Changed battery from Duracell to Lithium Ion 4.2V as the input to TPS63020,
      1. Still the problem is observed on the 3.3V rail, the voltage drop to nearly 2.9V. Comparatively on the fully charged battery it is less.
    3. Changed battery nominal voltage from 2 batteries in series to 3 batteries in series
      1. Still the problem is observed on the 3.3V rail, the voltage drop to nearly 2.9V, resetting the micro.
    4. The analog section (except the ADC) of the board is powered individually. e.g. temperature measurement circuit is powered and readings are taken and powered down, pH measurement is powered and reading are taken and powered down.
      1. The problem is not noted during the power on of the section or power down of the section of the circuit. It is specifically noted when the channel change is done.Please find the attached screen shot of the oscilloscope capture. The Small drop noted at 3ms scale is the power ON of the particular section, the waveform signifies that the load more capacitive. The Second drop noted at 3.5ms is when the channel settings are done, the waveform signifies a Inductive load which we do not have on the board. We are currently writing in to 4 registers (Buffer Register, Channel Register, PGA Register and Data Rate Register). We understand there is a drop in the battery voltage hence the drop in output from the waveform. What we do not get is why the sudden inrush current, since the ADC is already powered up.
      2. Please note the drop is the 3.3VDC is not captured when the board is powered from Launchpad. Only difference is the TPS63020 is bypassed.
    With Regards,
    Muruga
  • Hi Muruga,

    If you would share your schematic with me it would be very helpful. You can either send me a private message on E2E with an attachment or email it to pa_deltasigma_apps@ti.com.

    My initial impression is that you might need to wait a bit longer before communicating with the ADS1256. If you are able, I would recommend waiting for a /DRDY interrupt as an indication that the device is powered and actively converting.

    Also, which o-scope channel is which? These are the 3.3V (DVDD) and 5V (AVDD) supplies, correct?
  • Hi Muruga,

    I haven't heard back from you on this issue. Where you able to resolve it?
  • Hi Chris,

    Sorry for the delayed response.

    The scope capture was battery voltage and 3.3V rail

    We had probed the 5V rail and did not see the dip on the rail. Hence it is not the ADS1256 which is creating the problem or might be (because it happens when there is a channel change).

    Now we are looking at TPS63020 to check why the behavior is so.

    Currently have isolated the supply to Main board and TPS63020 by a diode, we are testing the configuration to check the performance.


    Regards,
    Muruga