• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 480 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.

PGA900: T ADC Offset

user5886507
Prodigy 30 points
Part Number: PGA900

Hi all

I try to measure a single ended Temperature voltage with the T Gain input. The measurement works finde, bot there is a offset in the PGA Measurement of about 2000 bits.

My Configurations are following:

I am working with the PGA900EVM

AMUX_CTRL:

TEST_MUX_T_EN = 1 (Tgain to TADC)

TSEM_N = 0 (Single ended)

T_GAIN_SELECT:

T_GAIN = 11 (20x Amplification)

The input is configured as 16Bit.

If i measure my signal at the VINTP input with a DMM i measure 0.056V

Calculation: 0.056V * 20 = 1.12V
1.12V / 5V *65536 = 14680 Bits

If I debug my software and read out the TADC_DATA the value is: 16770

Can someone explayn where this offset can come from?

Thanks

  • 0
    TI__Mastermind 31655 points

    Hello,

    Have you tried different gain settings? How does the offset change if you select a different gain? There shouldn't be such a large offset due to the gain stage, but I'd like to see how it's effected.


    What kind of sensor are you connecting to the Tgain inputs? Do you see the same offset if you supply a fixed voltage from a bench supply?


    Regards,

  • 0 in reply to Scott Cummins
    Prodigy 30 points

    Hi Scott

    At the moment i only measure the voltage over a resister so its kinda of a fixed voltage supply. I also tested the internal temperature sensor which needs a gain of 5. Even with the internal temperature sensor i see a large offset.

    I will now try to set the gain to 2 and measure a fixed voltage from a bench supply. I will also try out the differential measurement and see if the offset stays the same.

    By now i tested two different PGA900 chips with about the same results.

    Regards

    Tobias

  • 0 in reply to user5886507
    TI__Mastermind 31655 points

    Hi Tobias,

    Please let me know the results of your tests with the differential measurement and different gain. The internal temperature sensor itself can have quite a large offset, so it is possible that is the source of the offset when measuring in that mode. The offset when measuring from the input pins should be very minimal however.


    Regards,

  • 0 in reply to Scott Cummins
    Prodigy 30 points

    Hi Scott

    I have tried the measurements with gain = 2 differential and single ended:

    The differential results are much better than the single ended results.
    If I connect the same Voltage to VINTP and VINTN i get an offset of -38 Bits which stays the same over all Gains.

    If i measure single ended with a fixed voltage, i get the same error as bevore (10 times smaller because of the smaller gain 20 --> 2).
    Interresting is that if I connect gnd to the VINTP pin in single ended measurement the adc displays exactly the offset i have in my measurements.

    It seems like the single ended measurement ins't measured correctly respect to gnd.

    Regards

    Tobias

  • 0 in reply to user5886507
    TI__Mastermind 31655 points

    Hi Tobias,

    It seems like there may be something internally contributing to this offset in the single-ended mode. I will have to discuss with a designer to see.

    Can you use differential mode in your application, or are you able to accept the offset from the single-ended mode? When performing the calibration procedure, a digital TADC offset correction is applied which should take care of this.

    Regards,