LIS3DSH question and documentation request

[Andrei Chichak]

Posted on March 31, 2017 at 07:49

I've been able to get the LIS3DSH working for single values by doing this:

transmitData[0] = (uint8_t) (LIS3DSH_READ | LIS3DSH_X_L);

SPISend( transmitData, receiveData, 2); xl = receiveData[1];

But if I wanted to read out multiple values in one transaction, I'd expect that I could do something like:

transmitData[0] = (uint8_t) (LIS3DSH_READ | LIS3DSH_MULTIPLE | LIS3DSH_X_L);

transmitData[1] = (uint8_t) (LIS3DSH_READ | LIS3DSH_X_L); SPISend( transmitData, receiveData, 3); xl = receiveData[1]; xh = receiveData[2];

Or have byte 2 just be 0xFF because the auto-increment would do what I expected.

But alas, it doesn't work. (yes address increment is turned on in register 6)

Can this be done? What am I missing?

Now for documentation, the temperature register just isn't documented in the data sheet. It took a while to figure out that it is Celcius and biased by -25C. Also, the M/S bit isn't documented either. I found AN3393 to be very useful, the data sheet, not so much.

Thanks,

Andrei from The Great White North.

[Miroslav BATEK]

Posted on March 31, 2017 at 10:33

I assume you are using 4-wire SPI interface, than you approach for multi read operation is correct. But I don't know what exactly you function SPISend does.

It would be good to check the signal with scope or logic analyzer to see if it match with the protocol.

[Andrei Chichak]

Posted on March 31, 2017 at 16:05

Sorry, yes you are correct, I'm using 4 wire SPI.

My SPISend routine is:

void

SPISend(

uint8_t

* transmit,

uint8_t

* receive,

uint16_t

length) {

HAL_StatusTypeDef

status;

HAL_GPIO_WritePin( CS_I2C_SPI_GPIO_Port, CS_I2C_SPI_Pin,

GPIO_PIN_RESET

);status = HAL_SPI_TransmitReceive( &hspi1, transmit, receive, length, 100);HAL_GPIO_WritePin( CS_I2C_SPI_GPIO_Port, CS_I2C_SPI_Pin,

GPIO_PIN_SET

);

if

(status !=

HAL_OK

) {

printf

(

''there is a problem with the transmit %d\r\n''

, status);

}

}

When I send two characters the pattern is:

So the signaling form is okay. At 5.25 MBaud I'm not buzzing the chip. The values that I'm getting out make sense.

I checked the wave forms on the three byte version (read | multiple | xlow, read | xhigh, dummy) and it looked good, nothing violated, but the data did not change.

Andrei

[Andrei Chichak]

Posted on April 25, 2017 at 16:37

Okay, I figured it out. 

According to the timing diagram in the data sheet, bit 0x80 is the R/W bit, and 0x40 is the M/S bit, with bits 0x3F being the address bits.

In real life, there is no M/S bit. The diagram is a cut/paste from a previous generation of some  other chip. 

To read multiple bytes, you set the address increment bit in control 6 and just keep sending clocks.

If you set the M bit 0x40, suddenly you are just reading other registers. This device has registers in the 0x4x-0x7x range, and you just end up reading them.

So, this snippet works:

for

(

uint8_t

i = 0; i < 7; i++) {

   transmitData[i] = 0 ;

}

transmitData[0] = (

uint8_t

) (LIS3DSH_READ |LIS3DSH_X_L);

SPISend( transmitData, receiveData, 7);

x = (

int16_t

) (receiveData[2] << 8) | (

int16_t

) receiveData[1];

y = (

int16_t

) (receiveData[4] << 8) | (

int16_t

) receiveData[3];

z = (

int16_t

) (receiveData[6] << 8) | (

int16_t

) receiveData[5];

So, ST, please correct your documentation.  The same problem exists in the LSM6DSL documentation, and I expect many more as well.

Andrei from The Great White North

As seen on the embedded.fm podcast and blog.

[Andrei Chichak]

Posted on April 25, 2017 at 17:23

Okay, in DocID022405 Rev 2, table 15 shows the default values for register 6 being address-increment being off and watermark interrupt being on.

Section 7.7 says nothing about the default for address-increment, but has watermark interrupt being off.

If I do not set the address-increment bit, the address still increments.

I think there is something wrong with the documentation.

[Miroslav BATEK]

Posted on May 30, 2017 at 17:19

I think you are right the Figures 7, 8, 9, 10, 11 in

DocID022405 Rev 2 (Datasheet for LIS3DSH) 

are not correct because the bit 1 is not the M/S but is MSB of the register address. I asked sensor specialists to confirm this and fix the datasheet. 

I don't see this issue in the datasheet for LSM6DSL

Concerning the default value of CTRL_REG6, it is not mentioned in the table 15. I think you mistake the default value with binary representation of the register address. But you are right in the section 7.7 the default value of ADD_INC is not mentioned. This bit is set to 1 by default.

I'm sorry for the inconvenience.