MSP432P401M: UART TX and RX error

Dear Champs,

Other problem is how to calculate UCBRSx? Like Table 22-4. UCBRSx Settings for Fractional Portion of N = fBRCLK/Baud Rate
It seems like not similar with previous method like processors.wiki.ti.com/.../USCI_UART_Baud_Rate_Gen_Mode_Selection
Could you kindly share how to calculate UCBRSx?
Thanks a lot.

Dear Evan,

Yes, I have studied all.

After studied it and hope to test the calculation, take an example as red block.

BRCLK = 32768, Baud rate = 1200. N = INT (32768/1200) = 27.

So that, OS16 = 1, UCBRx = INT (27/16) = 1, UCBRFx = INT([(27/16) – INT(27/16)] × 16) = 11. But my problem is UCBRSx.

For UCBRSx, 32768/1200 = 27.3066667.

As 0.3066667, I try to use Table 22-4, but how do I select in? 

If using  USCI UART Baud Rate Gen Mode Selection, http://processors.wiki.ti.com/index.php/USCI_UART_Baud_Rate_Gen_Mode_Selection

UCBRCSx = Round (0.3066667*8) = 2. 

Why UCBRCSx is 0x25hex (37dec)?

Could user only need to use Table 22-4 to get UCBRCSx or exist any UCBRCSx calculation method? 

Second, customer's demand is BRCLK = 24MHz, Baud rate = 115200, could you kindly provide TX Error (neg/pos) and RX Error (neg/pos)? 

If you have any concern, please feel free to let me know.

Thanks a lot.

Janet,

First point, to the link you provided, the "USCI UART Buad Gen Mode" link, is built to calculate the settings for a different class of devices. USCI and eUSCI (which is what is on the MSP432) are two different things and the calculations will not come out to be the same so please do not use that link for this effort. Please trust the UCBRSx value you received from the datasheet.

In the instance for your fractional portion, I would round to the closest setting for UCBRSx as you see was done for your first example.

I'll look into the TX/RX error calculation for 24mhz and see what I can find.

Dear Evan,

Thanks for your explanation.
So, as my example, |0.3066667-0.3000|<|0.3066667-0.3335|, so that select 0x25.
It means user just select the closest number as UCBRSx, right?

And for 24MHz BRCLK, please kindly provide the TX/RX error rate.

I will also calculate it and double check this with you.
Thanks a lot.

Dear Evan,

Thanks for your sharing.

After I use this method to calculate the TX/RX error.

I found one question on below condition: 

BRCLK = 24MHz, baud rare = 115200, so that 240000000/115200 = 4000000/19200 = 208.333333, but its UCBRSx = 0x84. 

But I can't find 0x84 in Table 22-4.

So when the fractional Portion of N is 0.3333, is UCBRSx "0x84" or "0x25" ? 

If you have any suggestion or idea, please feel free to let me know.

Thanks a lot.

Dear Evan,

Understood, I have based this calculated method to build up an equation in excel.

But just meet some little problems as above, like 4000000/19200 = 208.333333, but its UCBRSx = 0x84, Why not 0x25?

Could you please kindly help me check this little confusion? If it is TRM error, could you please kindly update it? 

Thanks a lot.

Janet,

I've recognized the same thing you have and we are investigating it. It could be as simple as a typo and we will update if so. In the meantime, it's best to refer to Table 22-4.

Janet, 

How did the advice work for the customer?