LPS25HB questions

John L · ‎2017-02-14

Posted on February 14, 2017 at 23:24

We are evaluating the LPS25HB BP sensor, and have a couple of questions. I notice that several forum members have asked similar questions, but without receiving a reply. I would really appreciate an answer, as this would indicate that the LPS25HB is not an 'orphan', unsupported product.

1) The datasheet mentions a soldering effect on BP, and says that a register is available to store a one-point calibration. There are no other details at all about this in the datasheet. We have found substantial (~0.5 kPa+) pressure inaccuracy in samples we have tested. Is this to be expected? Exactly how are we to use the calibration register? How is it calculated? Again, these are important points for any user, and the datasheet gives no meaningful information.

2) We notice a very substantial temperature effect on BP. It appears that no temperature compensation is being done internally to the chip, although the datasheet mentions quadratic temperature compensation, and claims that a 20-60C change will cause only a 20 Pa (0.2 hPa) change. I see that magnitude of change or greater from a ~5C ambient temperature change with no significant change in actual ambient BP. Does internal temperature compensation need to be turned on? Or is it up to the user to determine an empirical temperature calibration?

Answers to these questions would be greatly appreciated and will help us to evaluate the LPS25HB relative to competitive sensors made by other manufacturers.

#lps25hb

Miroslav BATEK · ‎2017-02-15

Posted on February 15, 2017 at 15:54

One-point calibration can by applied by filling in RPDS_L and RPDS_H registers (see chapter 8.22 and 8.23 in datasheet). It is basically offset which is subtracted from the output value. It is 16-bit value with weight of 16 most significant bits of output value.

Pressure inaccuracy 0.5kPa = 5hPa can be caused by soldering and/or cleaning process. It can be compensated by the one-point calibration mentioned above.
Internal temperature compensation is always on, you don't have to switch it on. Absolute pressure
accuracy over temperature mentioned in the datasheet is typical value. Max/Min value is not guaranteed.

John L · ‎2017-02-15

Posted on February 15, 2017 at 16:19

Thank you! I am sure that other customers and viewers of this forum will also find this information useful.

I have one more question, based on your explanation of the correction register. Usually, calibration errors are corrected by a multiplicative factor, rather than a fixed constant to be subtracted from or added to the value that the sensor gives. You mentioned that in the case of this sensor, the calibration register stores a fixed offset, i.e. a constant. This would correct values very close to the initial calibration point, but I am wondering if this correction would hold true for values removed from that point? In other words, should the correction be multiplicative rather than additive? Do you have any empirical data to back up the use of a fixed constant, which in our experience would be unusual?

Thank you in advance for your answer. I am sure of that other customers will be wondering the same thing.

Miroslav BATEK · ‎2017-02-16

Posted on February 16, 2017 at 11:00

From our experience soldering process can create an offset but the sensitivity is not influenced. The offset is the same for whole range.

From that reason only additive correction is needed and available in the sensor.