Post Go back to editing

Reference: Figure 6 of, ADXRS646-EP part datasheet.

We understand, Null Bias is ratio metric to Vratio input.

For Vratio of 6V, Null will be 3V(Typical). But in the graph (Vratio is mentioned as 5V) and Null bias is still 3V.

As per our understanding it should have been 2.5V

(Or)

Vratio=5V is a TYPO in datasheet.

## Top Replies

Parents
• Hi,

If NULL is ratiometric to Vratio, why there is huge variation in NULL over temperature (even at 25C over sample to sample)? How can be compensate this error?

• Hi,

If NULL is ratiometric to Vratio, why there is huge variation in NULL over temperature (even at 25C over sample to sample)? How can be compensate this error?

Children
• Hi

it seems that there is a typo in the Figure 6, thanks for noticing. I will work with our editorial team to correct it.

From what I have been able to learn, through review of internal design documentation, there are process variation that causes null bias to shift from ideal, this is normal. At that time, lead customers were asking for improvement linear vibration rejection and nonlinear error and were willing to support this level of variation in the null bias, over normal process variation and temperature.

Digital (downstream) calibration has been the most common approach for customers that we have worked with, directly.  You can see an example of how one of our IMUs applies these calibrations, on page 17, of the ADIS16507 datasheet: https://www.analog.com/media/en/technical-documentation/data-sheets/ADIS16507.pdf#page=17

Regards,

Pablo.

• Hi Pablo,

On a similar note, is calibration possible for Sensitivity as well? (As we have variation in sensitivity also for sample to sample).

Data sheet specifies +/- 3% over temperature. However, is there any Temperature coefficient (PPM) to calculate for a given range? (if requirement is beyond the range specified in Note3)

Thanks & Regards

Raga

• Hi Raga,

Sorry, the test was performed measuring the limits shown in the note 3 only. So we cannot provide a temperature coefficient.

Regards,

Pablo.