DCO wake up time around 1uS?
This topic contains 3 replies, has 2 voices, and was last updated by Brad_Smyth 5 years, 2 months ago.
You must be logged in to reply to this topic.
Make it to the Right and Larger Audience
This topic contains 3 replies, has 2 voices, and was last updated by Brad_Smyth 5 years, 2 months ago.
TI MSP430 datasheet says DCO wake up time is around 1uS. It is very short. How do they achieve it? Normally xtal wake up will take ms and PLL setting time is around tens of uS or more.
It is true external Xtal normally takes ms to warm up. But DCO doesn’t use external xtal. Instead it is based on internal RC oscillator. As for PLL setting time, I think maybe it first starts as open-loop and then operates as close-loop.
I see. Thanks. But RC oscillator accuracy is very bad, like 30-40%. So if we use DCO as MCLK, MCLK can be way off from expected?
You can take a look of “MSP430 Microcontroller Basics” I posted:
http://www.valpont.com/vlbgrefv/book-msp430-microcontroller-basics/
Page 167-169 is about DCO. You are right that a plain RC oscillator accuracy can be as bad as 25%. But with special design, TI achieves 1-2% accuracy. This is normally good enough for digital circuit. But obviously it is not good enough if you have a PLL to generate very high freq like GHz level using DCO as reference clock. In this case, an accurate xtal clock is normally needed. Another caveat is even for digital circuit, if you have a circuit which is timing sensitive like UART, you need to double check if it works properly with 1-2% accuracy. Note that the other side may be another MSP430 which also uses DCO and has 1-2% accuracy. So sum up can be 4% difference in the worst case.
Hope it helps.
You must be logged in to reply to this topic.
Thanks for helping us better serve the community. You can make a suggestion, report a bug, a misconduct, or any other issue. We'll get back to you using your private message ASAP.
©2021 ValPont.com