LP-AM243: MCSPI Waveform abnormal

Hi Hang,

• Is the observation above causing the example to fail?
• Can you also share the measurement setup details (scope model, acquisition bandwidth, trigger setup, analog vs digital probes and their model numbers)

Best,

Daniel

Hi Daniel,

1. the example is successful, all test is passed.
2. I am using Scope model MSO6054A with 500 Mhz bandwidth, analog probe p6100 with 100 Mhz bandwidth

Since the loopback is successful, I suspected that it was a measurement issue, so I drop the CLK frequency to 50kHz to make it easy for the scope. Since the measured waveform of SPI CLK are sharp square waves, I think there should be no measurement problem remaining. Moreover, my customer got the same measurement using both scope and digital analyzer. This problem should be easy to reproduce, one just need to simply probe D0 with the example running.

Regards,

Hang.

Hello Hang,

I apologize for the delay on this, other items took most of my bandwidth this week. I will take a closer look at this problem next week. 

Could you please tell me which version of the SDK are you using and also if you are running the LP in DEVBOOT mode after running the initialization script? I want to make sure I replicate your setup in this case

Best,

Daniel

Hi Daniel,

Below are the sdk versions.

I have not checked the development boot mode yet, as I am not very familiar with this device. Perhaps you can check it from the photo above or instruct me how to check it. I have run the initialization script before running the example.

Thanks,

Hang.

I can see that you are working on QSPI bootmode. You can check this information in the LP user guide: AM243x LaunchPad Development Kit User's Guide (Rev. E) (ti.com).

I see you are also working on the latest versions of the SDK. I will take a closer look at this so please allow me a couple of days to run some tests. I'll update you by Thursday with more information

Best,

Daniel

Hi Hang,

I have been running into some issues getting this test to work on my launchpad. I will sync with other experts and get back with more information tomorrow

Best,

Daniel

Hang, 

I am taking over this thread, and will have time to investigate this tomorrow.

Regards,

Brennan

Hi Brennan,

Thank you for replicating the issue. I can see the from your experiments that CLK is normal and there is data on D0, (which is similar to my test results too), but I am not sure if the data you send matches the waveforms.

In my tests, I can see similar CLK and D0, but the waveform does not match the data I send (from ascii 'a' to 'e').

Could you please help checking the data and waveforms?

Thanks,

Hang.

Hang, 

I have added the SDK Example owner, Anand, to this thread so we can investigate further.  He has greater SW expertise on this issue than I do.

Hang,

Upon investigating the issue, I was able to obtain the desired waveform successfully. I recommend trying a different oscilloscope probe and retesting. For your convenience, I’ve attached a screenshot for reference.

Thanks,

Akash Soni

Hi Akash,

Thank you for the efforts! Could you please share the CCS project with me so that I can tested it too?

Thanks,

Hang.

Hi Hang,
 
please find the attached .out and appimage for your reference.
 
Thanks,
Akash Sonimcspi_performace_am243x_lp.zip