Hi everyone, I am using Jetpack 4.6.2 now. I found that I used the serial port assistant on the PC to connect to the board /dev/ttyTHS0 and send commands "0123456789abcdef"to the xavier nx At this time, XAVIER nx will send data to PC ‘’345678’, which is a little bit of crosstalk. I adjusted to connect with the board /dev/ttyTHS1, and this phenomenon does not occur. Is there a way to solve it?
I don’t know if this is what happens in your case, but one of the first things I’d try, if there is any significant length to the cable, is shielded cable (shielded internally so TX and RX are isolated). An example of cable with this is Cat 7 ethernet cable (pairs are in one shield, you’d use both halves of a pair for TX, and another pair wired together for the RX; if CTS/RTS are used, then similar). Here are some pictures of Cat 7 cable:
https://www.google.com/search?q=cat+7+ethernet+cable&tbm=isch&source=hp&biw=1858&bih=824&ei=VDMjY6XZC97h0PEPlduy-Aw&iflsig=AJiK0e8AAAAAYyNBZLgQDfXn8P7Y7UGpyQYVCpd58PT4&oq=cat+7+&gs_lcp=CgNpbWcQAxgBMgUIABCABDIFCAAQgAQyBQgAEIAEMgUIABCABDIFCAAQgAQyBQgAEIAEMgUIABCABDIFCAAQgAQyBQgAEIAEMgUIABCABDoICAAQgAQQsQM6CAgAELEDEIMBOgsIABCABBCxAxCDAVAAWMoGYLMYaABwAHgAgAFHiAGPA5IBATaYAQCgAQGqAQtnd3Mtd2l6LWltZw&sclient=img
I almost forgot: You might also test in loopback mode where on the Jetson the TX and RX of the UART are wired directly together, and you use your terminal to echo back to itself. If there is cross talk at the given settings while in loopback, then you know the problem is in the Jetson, but if the cross talk only occurs going over the cable, then you know the problem is external. Slight differences in capacitance or placement of cable can make otherwise working hardware behave differently.
OK thanks
One more detail about loopback: You can also use the cable itself in loopback. Instead of connecting the device which is remote to the Jetson, just take the far end of the cable and jumper TX and RX there instead of right at the Jetson (also CTS/RTS if you use that). If you get corruption without cable, then then issue is at the Jetson. If you get corruption only at the far end of the cable, but not on the Jetson, then you know it is cable cross talk. This is of course not definitive since attaching a device can change capacitance and inductance and thus cross talk, but it would be a very big clue as to whether or not this cross talk occurs in loopback.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.