The CS pin of SPI

I recently used the SPI on Xavier. I extracted the corresponding pins of the SPI on the 40 pin plug, and then connected it to the STM32 through the digital isolator MAX14851.At present, it has been determined that the use of the digital isolator is no problem, but it is found that the CS pin of the SPI is problematic when using the xfer function in the spidev library to transmit data on the Xavier as the Raspberry Pi uses the SPI.
The CS pin of the SPI is an open-drain output, followed by a pull-down resistor of one megaohm. I tried to add a pull-up resistor, and tried the output of the CS0 pin and the CS1 pin is not normal. When the pull-up resistor is small, the oscillation is large. When the pull-up resistor is too large, the CS pin cannot be pulled high. I can only send SPI data by using other pins as chip selections.
I want to know how to use the CS0 pin or CS1 pin to control the SPI selection.

Hi zc15262950139,

Are you using the devkit or custom board for AGX Xavier?
What’s your Jetpack version in use?

Have you referred to the design from the devkit?

Hi, please also refer to the Jetson Nano Developer Kit 40-Pin Expansion Header GPIO Usage Considerations Applications Note for the rerquirement of 40-pin header pins.

i used the agx xavier developkit. The Jetpack viersion is 5.0.2

At first, I did not add a pull resistance to the chip selection pin of SPI, and then found that the chip selection signal could not be pulled high. Then refer to the ’ Jetson _ AGX _ Xavier _ Developer _ Kit _ Carrier _ Board _ Design _ Files _ ’ on the download center, and add a pull-up resistance of 20K, but the low level of oscillation is very large. When it is increased to more than 50K, there is the same situation as without pull resistance. Now I 'm checking the output mode of the CS pin selection and the effect of TXB0108 with reference to Trumany 's advice. I will give feedback if I enconter problems.

1 Like

Thanks for your guidance, I read this document carefully and checked the circuit again with the oscilloscope. First, the CS of SPI is a push-pull output, which was mistaken for an open-drain output before, and then I removed the previously added pull-up resistor. Later, it was judged that the bidirectional channel in the MAX14851 isolation chip would affect the TXB device in XAVIER, while the unidirectional channel would not. MAX22164 is being considered to replace MAX14851.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.