CMT on K64: why add a glitch on every zero transmission?

[MarcoGiammarini]

Dear all,

I am trying to use CMT on K64. The modulation works fine, but I have a problem with every 0 transmissions. As you can see on the following image, where the second trace is the output of CMT_IRO pin, there is a glitch on every zero.

I really can't understand why!

Can you help me?

Thanks

Marco

[kerryzhou]

Hi MarcoGiammarini,

   1. Do you can reproduce the issues on the FRDM-K64 and it's SDK cmt project?

   SDK_2.8.2_FRDM-K64F\boards\frdmk64f\driver_examples\cmt

   Please help to check, whether it also have the glitch or not?

  2. About the testing tool

    Could you please also use the oscilloscope to test it, and remove your logic analyzer, just to check the wave directly with the oscilloscope, I want to see the original wave, the glitch pulse width, and the detail situation, after you test it, you can share me your wave picture.

Best Regards,

Kerry

[MarcoGiammarini]

Dear @kerryzhou 

any news for me?

Regards

Marco

[kerryzhou]

Hi MarcoGiammarini,

   Do you mean the test result like this:

And in each 0, it contains one pulse, right?

And you don't want that one pulse , right?

From the readme of the cmt project:

SDK_2.8.2_FRDM-K64F\boards\frdmk64f\driver_examples\cmt

This result is correct.

You can check the cmd example readme:

Note

The example modulates 11 bit data is modulated and output as: 10001100011.

The output waveform is captured on the oscilloscope like:
- data order: 1 -> 0 -> 0 -> 0 -> 1 -> 1 -> 0 -> 0 -> 0 -> 1 -> 1
- the period of the basic carrier generator signal is 25 microsecond
- bit 1 is about four normal carrier signal period.
- bit 0 is about one normal carrier signal period and three space carrier signal period.

So, please give me updated information, do you want bit0 just the space carrier signal period and no high period?

Best Regards,

Kerry

[MarcoGiammarini]

Hi @kerryzhou 

Yes, I mean the result that you show in the picture. Yes, I don't want that one pulse during zero periods.

I confirm that I want bit0 just the space carrier signal period and no high period.

Best regards

Marco

[kerryzhou]

Hi MarcoGiammarini,

Seems this is related to the following content:

About more details, I will do more checking.

But due to the Chinese Spring Festival, I will reply you until I am back on Feb. 18th.

Best Regards,

Kerry

[MarcoGiammarini]

Hi Kerry,

Yes, I tried the example into SDK and reproduce the same issue.

In attach you can find the acquisition done with an oscilloscope. The only thing that I have changed is the carrier frequency: is the same as my project that is 36000Hz.

Regards

Marco

[MarcoGiammarini]

I think the problem is: the mark period will not be never 0 because the formula is (markCount + 1) / (Fcmt/8).

But is very strange...