cancel
Showing results for 
Search instead for 
Did you mean: 

ST please document properly revisions of development boards (this time, STM32F3DISCOVERY)

The STM32F3DISCOVERY board's main perk are the MEMS gyro and MEMS accelerometer/magnetometer combo.

In different revisions of the board, these chips were changed (whatever was the reason). This of course has impact on working of software, including examples.

However, while UM1570 rev. 6 mentions, in chapters ST MEMS E-compass (ST MEMS LSM303DLHC or LSM303AGR) and ST MEMS Gyroscope (ST MEMS L3GD20 or I3G4250D), that these are different parts, it fails to explain, how to identify, on which revision of the board is which part combination. Chapter 7.2 Board revision history fails to mention these changes entirely.

ST, please take documentation seriously.

JW

1 ACCEPTED SOLUTION

Accepted Solutions

More.

#define I_AM_L3GD20                 ((uint8_t)0xD4)
#define I_AM_L3GD20_TR              ((uint8_t)0xD5)

The Cube code checks against both these values. Yet I can't find any L3GD20TR which wouldn't be just the tape-and-reel packaging option of L3GD20, with the same 0xD4 WHO_AM_I value than the non-TR L3GD20. Or at least can't find it in public pdfs.

But given the code, it obviously exists in the wild otherwise the Cube coders wouldn't be bothered; and what's worse, it exists on the F3 DISCO (and who knows on what other devboard, too).

ST, any explanation?

I made this an idea, too, so ST can ignore more of them.

JW

View solution in original post

5 REPLIES 5

And TEST example code for functionality. The ability to identify and self-report components and versions would be helpful.​

Tips, Buy me a coffee, or three.. PayPal Venmo
Up vote any posts that you find helpful, it shows what's working..

> And TEST example code for functionality.

Even if not functional for all revisions of board, document that. Here, in STM32F3-Discovery_FW_V1.1.0

Although, of course it would be nice to see it properly updated, including binaries.

JW

(Rant stemmed from this thread)

Posted also as an idea.

JW

More.

#define I_AM_L3GD20                 ((uint8_t)0xD4)
#define I_AM_L3GD20_TR              ((uint8_t)0xD5)

The Cube code checks against both these values. Yet I can't find any L3GD20TR which wouldn't be just the tape-and-reel packaging option of L3GD20, with the same 0xD4 WHO_AM_I value than the non-TR L3GD20. Or at least can't find it in public pdfs.

But given the code, it obviously exists in the wild otherwise the Cube coders wouldn't be bothered; and what's worse, it exists on the F3 DISCO (and who knows on what other devboard, too).

ST, any explanation?

I made this an idea, too, so ST can ignore more of them.

JW

Sco
ST Employee

Hello,

I checked the document, and below information from hardware side FYI.

From version D to version E, the changes for the product are:

  1. U3 changed from LSM303DLHC to LSM303AGR
  2. U5 changed from L3GD20 to I3G4250D
  3. C7,C9 changed from 20pF to 33pF

Best Regards,