• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 48 subscribers
  • Views 47 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

PGA308: determine an individual error

Sven Kalms
Prodigy 10 points
Part Number: PGA308
Other Parts Discussed in Thread: INA333, PGA309

Tool/software:

Hello TI Support Team,

We are currently experimenting with the PGA308 as a replacement for an INA333. The goal is to change things like offset and gain without additional soldering.

If I understand correctly, each PGA308 has individual inaccuracies. Therefore, each PGA must be calibrated with the system?

All the calibration procedures I have found so far (PGA309 / PGA308) use the actual stimuli of the sensors. Here, the minimum and maximum values of the sensor must be applied to the component in real life. This is not possible in our application (strain gauges).

It becomes even more problematic when our system is used by customers. Here, only the range in which the measuring bridge is detuned is given.

Therefore, I experimented with the formulas from the application note. However, when I load the calculated register values into the PGA, the output is always a few mV off. I think this is due to the manufacturing tolerances of the ICs mentioned above.

Is there a way to measure the PGA in advance with a bridge simulator (0.25 mV/V, 0.5mV/V, 1mV/V, 2mV/V, 4mV/V, 5mV/V, 8mV/V, 10mV/V, 25mV/V, 50mV/V) and determine an individual error from it?

This error could then be included in the calculation to obtain a more accurate output value for real settings, e.g., 2.38mV/V. Is there perhaps already a corresponding set of formulas including such a correction value?

Best regards,
Sven

  • 0
    TI__Mastermind 19307 points

    Sven,

    You are correct that the inaccuracies in the PGA308 are normally eliminated by calibrating the PGA with the other components as a system.  If you could apply test signals to your system during the manufacturing process, you could create an offset and gain error lookup table to get good accuracy.  The lookup table would be stored in your microcontrollers memory. This would require a factory calibration where precision test signals are applied to the PGA308.  The formula would be a simple straight-line fit to the gain and offset of the PGA308 (Vout = Gain x Vin + offset).  In this formula the nominal gain/offset and the measured/calibrated gain and offset would be different.  The measured/calibrated gain/offset would come from the lookup table.

    Best regards, Art