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.

ADS5294: Gain variation with time and power cycling

Part Number: ADS5294
Other Parts Discussed in Thread: ADS5292

My application will use this 8 channel part to sample 8 signals from DC to roughly 20 MHz.    Our product undergoes a one time factory gain calibration for the entire system.   We rely on the gain not changing with time or power cycling for each of the channels.    I realize the reference could change which would also impact gain, as well as temperature (which is well specified in the data sheet), but would whatever gain each channel had during our initial calibration remain stable with time, power up cycles, etc with this part?   If not, what would be the expected variation from power cycle to power cycle?

  • Hi,

    Thank you for using ADS5294 device.

    When you are running your signal from DC (0 Hz)

    then you need to keep your input differential signals with common-mode voltage output pin (VCM)

    please don't treat them as floating pins.

    (please use the ADS5294 pin as abut 0.95V,

    please see the datasheet page 10.

    Also please refer to the datasheet on Page 30,

    Figure 44. Analog Input Circuit Model.

    For your power up cycle to cycle,

    Please refer to the datasheet page 17 Figure 2.

    also on page 15:

    MIN TYP MAX

    UNIT t1 Power-on delay Delay from power up of AVDD and LVDD to RESET pulse active

    Typical is about = 1 ms

    Thank you!

  • I understand completely about how to dc couple the signal.   My question is in regard to the gain of each of the 8 channels.   I realize there is an initial tolerance and it can change with temperature.   I also realize it changes with the reference voltage changing.  

    My question is will the gain of each channel, between the analog input and digital output, stay the same as it was initially, every time the device is powered up, presuming the reference is perfect and no temperature drift.    

  • Hi, Thanks for letting us know more details.

    For your question,

    please refer to the ADS5294 datasheet page 12.

    It has already showed clearly:

    If you are using internal reference mode (please refer this more detail from the datasheet page 8 from REFB and REFT pin functions),

    its Gain Error (for your question) +/- 2 %FS error.

    If this is higher than your expectation, then you need to use external reference mode (from datasheet page 8)

    also please refer to the ADS5294 datasheet 9.6.1.14 on page 59 (very important for your case).

    so you need to setup registers 0x1, 0xF0, and 0x42 settings (please following one by one on Table 23 as well) .

    Thank you!

  • I get that the gain can be off a bit but my real question is if I power the unit off and back on, configuring it exactly the same as it was and assuming the reference is exactly the same as it was and the temperature is exactly the same as it was will the gain be the same?   I ask this because some devices from other suppliers undergo an internal calibration that could impact gain on every power cycle.   We want to be certain that this device doesn’t do that.

  • Hi,

    For more detail questions,

    please let us know more details such as what is your power cycle on your setup?

    and what the temp changed on your setup?

    and what gain error you expected on your side?

    Please your questions must follow ADS5294 datasheet specs.

    if you had follow the datasheet specs but you did not receive the results from the datasheet specs.

    Please provide more details.

    Thank you!

  • Hi,

    Also for example when you power off the device

    and then have you re-setup the device according to ADS5292 datasheet page 27 figure 45,

    and then you power back up to on again?

    These are very important.

    Thank you!

  • Also please refer to ADS5294 device datasheet,

    page 17 and Figure 1.

    Thank you!