Customer inquiry;
I've noted considerable variation in the DCO clock frequency when using the internal resistor. Could this explain why some parts do not start up without an external clock?
Comments....
-Tom-
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
No.
DCO is inprecise by definition. This is why some MSPs have calibration data for one or more standard frequencies.
However, DCO frequency has nothing to do with startup. And the DCO is not so unprecise that the power-on-frequency will exceed the allowed core frequency on minimum supply voltage.
If an MSP doesn't start up, it is in most cases caused by a slowly rising and instable power supply. It may cause the MSP to begin working, which increases teh current drawn, whcih in turn decreases the still unstable supply and causes the voltage to drop below the allowed range again.
To prevent this from happening, put a pullup resistor of 47k between RST and VCC and a pulldown capacitor of 2.2nF between RST and VSS. This way, RST will rise mor eslowly than VCC does and RST is considered low until VCC has settled and RST can catch up.
Higher values for the capacitor increase stability if the power supply is really slow, but this willa ffect JTAG access, especially with the SBW protocol. (4-wire JTAG is more robust, and 10nF or even 100nF can be used)
Jens-Michael Gross said:... If an MSP doesn't start up, it is in most cases caused by a slowly rising and instable power supply. It may cause the MSP to begin working, which increases teh current drawn, whcih in turn decreases the still unstable supply and causes the voltage to drop below the allowed range again.
To prevent this from happening, put a pullup resistor of 47k between RST and VCC and a pulldown capacitor of 2.2nF between RST and VSS. This way, RST will rise mor eslowly than VCC does and RST is considered low until VCC has settled and RST can catch up.
Higher values for the capacitor increase stability if the power supply is really slow, but this willa ffect JTAG access, especially with the SBW protocol. (4-wire JTAG is more robust, and 10nF or even 100nF can be used)
I agree. However, 47k*2.2nF is about 0.1msec, 47k*100nF is about 5msec. A more effective and flexible way is to have a delay in the start-up code before the MCLK is increased and other power hungry features are activated.
A software delay won't help if the MSp already crashes when trying to start wunning with default speed.old_cow_yellow said:A more effective and flexible way is to have a delay in the start-up code before the MCLK is increased and other power hungry features are activated.
Of course, having a delay (or better: a check for sufficient VCC) before switching to higher speeds is mandatory. But the resistor/capacitor on RST prevents the chip from crashing long before the user code is executed at all.
**Attention** This is a public forum