# Discrepancy between AN4519 and MPL3115A Datasheet?

cancel
Showing results for
Did you mean:
SOLVED

## Discrepancy between AN4519 and MPL3115A Datasheet?

947 Views
Contributor II

I'm having trouble deciding how I should convert the value stored in the fractional part (bits 4-5) of the OUT_P_LSB for CTRL_REG_1 & ALT == 0 (a.k.a. the output pressure should be in pascals). The datasheet states that the value of the OUT_P is in unsigned Q18.2 format which should mean that bit 5 of the OUT_P_LSB would represent 2^-1 (a.k.a. 0.5) place while bit 4 of OUT_P_LSB would represent the 2^-2 place (a.k.a. 0.25)? However, it seems that Table 3 in AN4519 shows the exact opposite.

Is this a problem with the table, or am I completely wrong in my understanding of Q number formats?

1 Solution
517 Views
NXP TechSupport

Hi,

You are right. In barometer mode (ALT bit = 0), bit 5 of the OUT_P_LSB register represents 0.5 Pa while bit 4 corresponds to 0.25 Pa. Table 3 indeed shows the exact opposite and I have already asked the responsible team for correction. Sorry for any inconvenience this may caused.

You might find useful my example code where I illustrate, among others, how to convert the 20-bit data from registers 0x01 – 0x03 to real values in Pascals.

I hope it helps.

Regards,

Tomas

4 Replies
518 Views
NXP TechSupport

Hi,

You are right. In barometer mode (ALT bit = 0), bit 5 of the OUT_P_LSB register represents 0.5 Pa while bit 4 corresponds to 0.25 Pa. Table 3 indeed shows the exact opposite and I have already asked the responsible team for correction. Sorry for any inconvenience this may caused.

You might find useful my example code where I illustrate, among others, how to convert the 20-bit data from registers 0x01 – 0x03 to real values in Pascals.

I hope it helps.

Regards,

Tomas

517 Views
Contributor II

Also, in section 7.1.3 of the datasheet, it states that the temperature is in signed Q12.4 format. However, there are only 8-bits specifying the non-fractional part in the OUT_T_MSB. Is this too a mistake? I think it should be signed Q8.4 if this is using Q number formats?

517 Views
NXP TechSupport

You are right again, thanks for pointing it out, it will be corrected in the next data sheet revision.

The 12-bit temperature measurement in degrees Celsius is comprised of a signed integer component and a fractional component. The signed 8-bit integer component is located in the OUT_T_MSB register (0x04) and the fractional component in bits 7-4 of the OUT_T_LSB register (0x05). It obviously corresponds to the Q8.4 format.

Regards,

Tomas

517 Views
Contributor II

Thanks for the link and quick response!