LIS3DH X-axle value is stuck in maximum value

[Jukka Lamminm ki]

Hi,

I've been testing LIS3DH sensor as an alternative to the LIS3DSH that we are using.

In testing with two different devices I've seen that with range values 2 ... 8g the X-axle value is stuck in the near maximum value 0x7FF, while two other axles are working fine.

I've using the chip in high resolution mode, the configured registers and read values are as follows:

range selected is 16

configured registers (hex): 23B8 2100 2200 2400 2502 B100 1E90 2057 A600

OUTX-register raw values(x,y,z): E02D 2000 2005

Interpreted values: x: 8601 y: 23 z: 960

range selected is 8

configured registers (hex): 23A8 2100 2200 2400 2502 B100 1E90 2057 A600

OUTX-register raw values(x,y,z): 807F 60FC 400F

Interpreted values: x: 7781 y: -221 z: 930

range selected is 4

configured registers (hex): 2398 2100 2200 2400 2502 B100 1E90 2057 A600

OUTX-register raw values(x,y,z): 807F 20F9 701E

Interpreted values: x: 3890 y: -209 z: 928

range selected is 2

configured registers (hex):2388 2100 2200 2400 2502 B100 1E90 2057 A600

OUTX-register raw values(x,y,z): 807F40F2D03D

Interpreted values: x: 1945 y: -209 z: 943

Other axles works fine when the device is tilted, X-axle is with 2 ... 8g range always fixed.

With 6g range the X-axle value changes as expected around ~8g offset.

The code used for reading is same as with LIS3DSH sensor.

Could it be that the sensors have been damaged in soldering process?

Best Regards, Jukka Lamminmäki

[Eleon BORLINI]

Hi Jukka @Jukka Lamminmäki​ ,

It looks like that -for some reason- the X axis is not well responding to the stimulus at high FS (but also at lower FS it is not well behaving, right? The offset is very high)... and well, the a soldering process damage might be the root cause.

Which is the failure rate you are facing? 2 of 2 devices? Could you share your reflow profile? The suggested surface mounting guidelines are shown in the TN1198 document (soldering profile as per JEDEC J-STD-020).

To check if the X axis is damaged, I would suggest you to perform the Self test procedure, described in the LIS3DH AN3308 application note, p.57, and coded in the lis3dh_self_test.c Github C example. In this way you will check the sensor functionality without moving it. When the self-test is enabled, an actuation force is applied to the sensor, simulating a definite input acceleration. In this case the sensor outputs should exhibit a change in their DC levels which are related to the selected full scale through the device sensitivity, especially in case of high FS.

-Eleon

[Jukka Lamminm ki]

Hi Eleon,

Thank you for the fast response.

Out of 9 devices only one (device #6 in the following list) has got somewhat acceptable offset value in X-axle. The following list is measured using 8g range:

Device 1: x: 7781 y: -22 z: -938

Device 2: x: 7682 y: -22 z: -961

Device 3: x: -6550 y: 23 z: -960

Device 4: x: 2882 y: 11 z: -996

Device 5: x: -2578 y: -19 z: -930

Device 6: x: -377 y: 7 z: -953

Device 7: x: 5488 y: 0 z: -968

Device 8: x: -2269 y: -7 z: -930

Device 9: x: -2910 y: 7 z: -961

Our HW-department says that soldering mask below the chip has NOT been removed, copper layer is removed though. Also there's no screws near the sensor. However we have manufactured tens of thousands of devices using this same layout with LIS3DSH sensor with no problems whatsoever.

I'll run the sensor self test tomorrow and let you know about the results. I also send the reflow profile as soon as I get it from our subcontractor.

Rgrds, Jukka

[Jukka Lamminm ki]

Hi Eleon,

Just got results from the self-test procedure:

SELFTEST RESULTS (mg, normal:selftest):

X 2040.0 : 2040.0

Y 19.2000 : 428.8000

Z 1000.8000 : 1430.4000

Axis[0]: lmt min 68.0 mg - lmt max 1440.0 mg - val 0.0 - FAIL

Axis[1]: lmt min 68.0 mg - lmt max 1440.0 mg - val 409.5999 - PASS

Axis[2]: lmt min 68.0 mg - lmt max 1440.0 mg - val 429.6000 - PASS

I think that this shows that the X-axis is damaged. Is this the definite proof that it has been damaged in soldering process?

BR: Jukka

[Eleon BORLINI]

Hi @Jukka Lamminmäki​ ,

>> I think that this shows that the X-axis is damaged. Is this the definite proof that it has been damaged in soldering process?

Ya, unfortunately it could be... can you try to change the soldering process?

-Eleon