• State
  • Locked Locked
  • Replies 10 replies
  • Subscribers 48 subscribers
  • Views 2803 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.

THS4522: Offset from Wheatstone bridge

Maria Medina
Prodigy 110 points

Part Number: THS4522
Other Parts Discussed in Thread: ADS8354, TINA-TI

I am using a THS4522 to amplify the signals coming from two full wheatstone bridges. The signals go through the THS4522 and then are converted from analog to digital using the ADS8354 chip. Is there a way to remove the initial offset voltage when no strain is applied? I am currently doing it in the code, after the signal is converted, but the offset does not let me use the entire's converter range to measure the change in strain applied.

Can the THS4522 be controlled to adjust the initial offset to zero? I do not want to use potentiometers, as I am planning on making several of these and they will all have different offset values.

  • 0
    Guru 140200 points
    Hi Maria,

    how to proceed depends on your circuit. Can you show us a schematic?

    Kai
  • 0 in reply to kai klaas69
    Prodigy 110 points

    I have something very similar to the below diagram, just with a gain of about 800, and using the THS4522 which has two channels. The diagram is only for one channel.

    AVDD = 5[V]

    DVDD = 3.3 [V]

  • 0 in reply to Maria Medina
    TI__Genius 15975 points
    Hi Maria,

    Unfortunately, there is no easy way of calibrating the THS4522 input offset voltage other than doing it through ADC. If you cannot use potentiometers, then the only other way is by using a high precision DAC. You can use the ADC codes to calculate the offset and then generate the respective DAC output that will counter-act this offset at the THS4522 input. As I mentioned earlier, the circuitry would be a bit complicated than doing it through regular potentiometer.

    Best Regards,
    Rohit
  • 0 in reply to Rohit Bhat
    Guru 140200 points
    Hi Maria,

    how much is the offset you must adjust?

    Kai
  • 0 in reply to kai klaas69
    Prodigy 110 points

    One of them is 3.1 [mV], and the second one is 0.5 [mV].

  • 0 in reply to Maria Medina
    Guru 140200 points
    Hi Maria,

    you wrote that the offset does not let you use the entire's converter range to measure the change in strain applied. Is it because the offset voltage can be negative? Can you handle a negative input voltage at the ADC input? As far as I know your ADC has a bipolar input voltage range. So, you could handle negative input voltages.

    The reason why I ask is, usually a software adjust of offset voltage is the preferred way. You could add a configuration routine to your software which measures the offset voltage for each setup one-time and takes it into calculation for all later measurements.

    Kai
  • 0 in reply to kai klaas69
    Prodigy 110 points
    Yes. I am currently adjusting the offset in my program. The problem is that I am applying a gain of about 650 using the THS4522. That makes the input difference of my adc about 1.95 [V] for the first Wheatstone bridge, and an input difference of 0.325 [V] for the second one. If my reference voltage is 2 [V] that only leaves me with .05 [V] left to use, and I won't be using the entire's converter range, not even half of it.
  • 0 in reply to Maria Medina
    Guru 140200 points
    Hi Maria,

    now I understand! Will think about it...

    Kai
  • 0 in reply to Maria Medina
    Guru 140200 points

    Hi Maria,

    I have made some simulations to see how the offset adjust should be carried out. This is a bit tricky for a differential amplifier. Have a look at the results:

    What are we seeing? The THS4522 is supplied by a +5V single supply. Vocm = 2.5V. I have assumed 80k and 100R feedback resistances to set a gain of 800. R2 and R6 inject currents into the feedback paths to provide the offset adjust. (10k in parallel 101R01 gives 100R.) There are two sorts of offset voltages: Vocm shows an offset voltage of up to +/-9mV and isn't affected by the gain. As the TINA-TI model comes with a pre-installed offset voltage for Vocm of 6.24mV I have adjusted it by the help of V5 for the sake of simplicity. Then there is the input offset voltage which is heavily affected by the gain and which you want to be removed. As the TINA-TI model comes with a pre-installed input offset voltage of 0.24mV I had to adapt the simulated offset voltage set by V4. The maximum input offset voltage is +/-3.5mV. Taking into account the pre-installed offset voltage of 0.24mV results in V4 = 3.26mV and V4 = -3.74mV.

    The first and third simulation show the non-adjusted offset voltages and the second and fourth simulation have the offset voltages adjusted. You will notice that according to the polarity of input offset voltage either V6 or V7 has to be set to the proper calibration voltage. The other calibration voltage must be set to 0V. Of course, if the input offset voltage is smaller than +/-3.5mV the according calibration voltage must also be set smaller. V6 and V7 should allow to be varied in a range between 0...350mV.

    Do you need further assistance?

    Kai