LIS3DH current consumption

[bojan]

Posted on January 16, 2017 at 13:36

Hello guys,

I'm using LIS3DH accelerometer in my current design. I need it to detect the movement.

ODR is set to be 10Hz. According to

http://www.st.com/content/ccc/resource/technical/document/application_note/77/ed/e7/e1/28/5a/45/d6/CD00290pdf/files/CD00290pdf/jcr:content/translations/en.CD00290pdf

document (page 10), LIS3DH should consume about 4uA of the current in both Normal and Low power modes.

However, when I measure the current consumption, I get the value of about 78uA (@2.5V power supply). When I remove the accelerometer from the power rails and measure the current consumption of mcu in a sleep mode, I get the value of 3.2uA. This clearly implies that about 75uA of the current is consumed by the accelerometer.

The important thing to mention here is that I was supplying the mcu and accelerometer with 3.7V to 4.2V for some time. This is above the recommended supply range for LIS3DH (3.6V maximum). So, is it possible that this over-voltage permanently increased the current consumption? I have the desired functionality even when I supply the system with 3.7V to 4.2V.

Thank you very much for your time and effort.

Sincerely,

Bojan.

#lis3dh

Note: this post was migrated and contained many threaded conversations, some content may be missing.

[Miroslav BATEK]

Posted on January 16, 2017 at 15:32

Hello Bojan,

the overvolatge can have an impact on the current consumption, but it is hard to predict how much the current would increase.

Some current can be also induced by internal pull-up resistors, but it would be probably more than 70uA.

Anyway can you please share you schematic, at least the sensor part, I can look at it.

[bojan]

Posted on January 16, 2017 at 17:00

Hello Miroslav,

Thanks for your prompt reply. Of course I can share the schematic with you.

As you can see, I have two external pull-ups for I2C. I used 10Mohms to additionally reduce leakage current and it I2C is working properly with them. Between MCU and the accelerometer, there is only I2C interface.

Is there anything you can suggest me? How could I disable internal pull-ups?

Thanks in advance,

Bojan.

[Miroslav BATEK]

Posted on January 16, 2017 at 17:11

There will be a current from SDO/SA0 pin to ground in your circuit, because you connected the pin to ground. According to datasheet the current should be higher than 70uA, but anyway try to disable the internal pull-ups in CTRL_REG0 register. Please check chapter 8.6 in the datasheet.

The pull-up for SDA/SCL are very high, you could have a problem with the communication. The recommended value for I2C pull-ups is 10kohms.

[Giraffy Fi]

Posted on March 31, 2017 at 08:03

Hey Miroslav,

Is there an equivalent 'REG_CTRL0' for LIS3DE?  I am also in the process of hunting down an extra 160uA static current between the MCU and LIS3DE with 4 SPI and 2 Interrupts pin connected.

I tried the trick above for the LIS3DH, by reading and writing address 0x1E.  The responses I got from the register does match the (LIS3D'H') datasheet.  I got 0xA0 @ 0x1E.  Drove SCLK and MOSI high or low did not change the current reading.  

Any advice would be greatly appreciated.

[Miroslav BATEK]

Posted on April 04, 2017 at 18:14

I'm double checking your question with sensor specialists, but I assume if there is internal pull-up on SDA/SA pin then it is automatically disconnected if you use SPI bus, because the pin is driven by push/pull output (MISO signal).

[Giraffy Fi]

Posted on April 04, 2017 at 22:14

Hello Miroslav,

Thanks for checking in.  So from the MCU side, adding a pullup on the IO to the SA0 (MISO) pin saved ~150uA, note that before it was high impedance (no pull), I guess the SA0 needs a little help.  I could use LIS3DH or E interchangeably, the LIS3DE happen to be available so we bought a batch for a project build.  Would be nice if you can confirm the 'Reg0' for the LIS3DE, just to give us another knob to turn.  Thanks.  Keep me posted.

FI 

[Giraffy Fi]

Posted on April 04, 2017 at 22:17

Interesting, k'n'o'b (above shown as ****) is censored by the site?  Ha.

[Tobias N]

Posted on February 22, 2018 at 11:20

Hi!

Very useful information above!

I have about 130 uA@3V3 higher consumption than expected for LIS3DSH. I suppose it is because pin 7, SEL/SDO is connected to GND. Any possibility to turn off internal pull up on this pin? I can not find in datasheet.

Thanks in advance! 

[Miroslav BATEK]

Posted on March 22, 2018 at 13:48

LIS3DSH has internal pull-up on SEL/SDO pin.

It is possible to disconnect this internal pull-up by setting to 1 the bit 7 (MSB) of register at address 3Fh (do not modify the content of the other bits!). Note: this functionality is not guaranteed (and tested) by STMicroelectronics on LIS3DSH device.

[Tobias N]

Posted on March 22, 2018 at 16:51

Thanks Miroslav

Best regards

Tobias