Differences between F411RE and F767ZI SPI interfaces?

[RH38]

Hi,
I am accessing a DRV8873s motor driver (SPI version) on two different NUCLEO boards. My usage is limited to reading data from 2 registries.
I got it to work on a NUCLEO-STM32F411RE using SPI2, with the following settings:
Motorola format, 8 bits, MSB first, prescaler 32 to get a 1.5625 Mbits/s rate, CPOL = low and CPHA = 2Edge.Now I try to do the same with a NUCLEO-STM32F767ZI and the same code and configuration (apart from the baud rate, set to 1.6875 MHz) don't work anymore.
I have tried SPI4 and SPI5 on this board with the same result: the data read by HAL_SPI_TransmitReceive is not correct.
Is there a difference in the way STM32F4 and STM32F7 perform SPI communication, or in the electronics?

 

[Andrew Neil]

Welcome to the forum.

Please see How to write your question to maximize your chances to find a solution for best results.

 

---

@RH38 wrote:

I got it to work on a NUCLEO-STM32F411RE using SPI2 ... with a NUCLEO-STM32F767ZI and the same code and configuration (apart from the baud rate, set to 1.6875 MHz) don't work anymore.

---

Have you used an oscilloscope and/or logic analyser to see what (if anything) is happening on the wires?

Have you tried a basic SPI example for the F7 board?

[RH38]

Thanks for the reply

first point: I have no instrument to see if the signals are correct or not. If the default persists I'll have to spy these signals with another card (I have several)

second point: I am currently studying the F7 example for SPI in polling mode, and I can't find any difference with what I (think I) do. I am going to try a basic communication between two cards for a start.

I keep on trying!

 

[TDK]

The peripherals are different but the abstracted HAL_SPI_TransmitReceive call is the same for both. Show your code. Probably something else going on.

[RH38]

Thanks for the reply

Yes, this is why I prefer using HAL functions; btw the same code runs on a G474RE board as well. As my project is rather complex, it is probably a good idea that I restart from scratch with just this SPI comm to eliminate side effects

 

[RH38]

Hi,

I rebuilt my circuit from scratch using a NUCLEO-F411RE and a NUCLEO-F767ZI with the same symptoms until... I discovered that one of the Dupont connectors I was using with the latter was faulty!
Shame on me!
So both NUCLEO boards DO work identically when I use the HAL layer.
False alarm, sorry about that...

 

[Andrew Neil]

Great - so please mark your post as the solution;

Instructions here: https://community.st.com/t5/community-guidelines/help-others-to-solve-their-issues/ta-p/575256

 

---

@RH38 wrote:

I discovered that one of the Dupont connectors I was using with the latter was faulty!

---

This is why it's so important to look at the hardware as well as the software!

And, thus, why it's so important to have hardware testgear - an oscilloscope would quickly have spotted this.

 

#NeedAScope #HarwareMatters