About SPI slave on 51422

RE: About SPI slave on 51422

tetiana — Thu, 19 Jun 2014 16:14:42 GMT

The problem is solved - I misconfigured my master so that slave select didn't work the proper way

RE: About SPI slave on 51422

Håkon Alseth — Thu, 19 Jun 2014 11:31:36 GMT

Are you setting the buffers after the completed transaction? By calling "spi_slave_buffers_set" after a finished transaction, you will acquire the semaphore again and ready the SPIS for a new transaction.

RE: About SPI slave on 51422

tetiana — Wed, 18 Jun 2014 15:42:42 GMT

P.S. I figured out that SPI1_TWI1_IRQHandler is never being called for the second time, and all the NRF_SPIS1 registers are in the same state before and after the transaction (checked it in the main slave example loop).

RE: About SPI slave on 51422

tetiana — Wed, 18 Jun 2014 15:27:10 GMT

I have the same issue with nrf51822 board: 1. Yes 2. Yes 3. Yes. I'm also using STM32 discovery as a SPI master. Are there any updates for the issue? I don't have any other ideas what might be wrong :(

RE: About SPI slave on 51422

Juliane — Tue, 06 May 2014 03:43:55 GMT

hi Elliot , I use STM32 MCU for SPI master side and Nordic for SPI slave side. How can I transmit data from slave to master?

RE: About SPI slave on 51422

Elliot.Chao — Thu, 20 Mar 2014 06:58:53 GMT

I resolve this issue, because I set CSN pin that have some problem.

Thank you.

RE: About SPI slave on 51422

Elliot.Chao — Wed, 19 Mar 2014 17:03:24 GMT

I always get 0xAA from Nordic.

RE: About SPI slave on 51422

Elliot.Chao — Wed, 19 Mar 2014 17:01:56 GMT

I use STM32 MCU for SPI master side and Nordic for SPI slave side. How can I mark sure that I send out the clock and data is consistent with Nordic? How can I know which mode I want to set it?

Thank you.

RE: About SPI slave on 51422

Håkon Alseth — Wed, 19 Mar 2014 15:43:08 GMT

Up to 2MHz is within spec and should be working. Yes, just make sure that both SPI-Master and Slave has the same SPI mode set. It might be that the SPIM clocks out data too fast after CSN is asserted, and adding a delay can work around this.

-H

RE: About SPI slave on 51422

Elliot.Chao — Wed, 19 Mar 2014 15:38:02 GMT

Do you mean mode0~3 on 51422? I Set mode 0 OK,I will try to delay time.

But, I use 300KHZ clock from master, is it ok?

RE: About SPI slave on 51422

Håkon Alseth — Wed, 19 Mar 2014 15:03:15 GMT

Have you set the same SPI mode on both sides? Could you try adding a delay of some microseconds after asserting the CSN line until you clock out data on MOSI?

BR Håkon

RE: About SPI slave on 51422

Elliot.Chao — Wed, 19 Mar 2014 13:37:23 GMT

1.Our master clock is 300kHZ 2.Yes 3.you mean "Serial Peripheral Interface Slave (SPIS) specifications" - SPIS timing diagram, one byte transmission, SPI Mode 0

RE: About SPI slave on 51422

Håkon Alseth — Wed, 19 Mar 2014 13:14:00 GMT

Hi,

That indicates that an ignored transaction has occurred (DEF register is set to '0xaa' in the SDK example).

Have you tried reducing the frequency on the SPIMaster side?
Do you always see this issue?
Are you sure that your SPIM is within the timing requirements stated for the SPIS module (PS chapter 8.3)?

Best regards Håkon