IIS3DHH accelerometer temperature drift

[GBra]

The datasheet reports an offset drift TCOff of max. 0,4mg/°C.

We made temperature tests with 3 samples and found the offset drift beeing 0,17 ... 0,58 mg/°C so higher than allowed. Does this meet your expectations?

Can you share your measurements?

Thank You

[GBra]

We tested -20 ... +50°C

[Eleon BORLINI]

Hi @GBra​ ,

Some questions for toy: Is the OFF_TCOMP_EN in CTRL_REG4 (23h) set to '1' on your devices (as reported in the datasheet p.3), right? And did you keep the device in a steady state condition and in a mechanically controlled environment during the test, in order to avoid the concurrence of external variables on that thermal gain? Did you normalized the gain with respect to 25°C? And is your Vdd = 2.8V as in the datasheet, or higher?

I cannot unfortunately share you characterization results so far, but please let me check, in case...

-Eleon

[GBra]

Hi Eleon,

Yes we have set the OFF_TCOMP_EN Bit.

Yes we have a good mechanically controlled environment and we did the test 2x.

Our powersupply is 3,3V.

Here our meassurement. It shows the deviation in arcsecs over the temperature in °C.

[GBra]

[GBra]

The blue curve is X-axis; the red curve is Y-axis.

[Eleon BORLINI]

Hi @GBra​ ,

thank you for sharing the measures.

So you are plotting the LSB on y axis? Can you please convert them in mg according to the 0.076 mg/LSB sensitivity?

If I well understood, you found 1 on 3 devices with a TCOff of 0,58 mg/°C... this might be correlated also to the Vdd = 3.3V, which is different from the 2.8V of the datasheet and could increase internal heating.

-Eleon

[GBra]

Hi Eleon,

find attached the plotted drift with mg on y-axis.

Gerhard

[Eleon BORLINI]

Hi @GBra​ ,

thank you for the data. Are you sure it is a IIS3DHHC device (with the final "C")?

By the way, what if you calculate the drift (TCOff) of the overall acceleration vector, i.e. sqrt(ax^2+ay^2+az^2)?

I have to check internally, but the TCOff parameter on the datasheet might be referred to this joint value...

-Eleon

[GBra]

Hi @Eleon BORLINI​ ,

we use evaluationboards. See attached picture.

How do you determine the revision "C"?

Can't find any information in the datasheet.

Gerhard