| ??? 12/03/12 14:05 Read: times |
#188953 - Some datasheets show numbers... Responding to: ???'s previous message |
Daniel said:
Hi Kai, Richard, clearly is NOT a problem with Vcc rise/fall time. Seems connected with the moment the clock stops. Yes, dear Daniel. But in the past we have discussed this topic a lot and in those cases improper power-on-reset and power-on slope rates were an issue. Richard said:
If you can provide a link to a datasheet or other document that specifies details of Vcc rise and fall, I would be very grateful. You can find these data in more and more modern microcontrollers, because the manufacturers have realized in the mean time, that there's a problem with isp capable flash micros and improper power-on rise times in combination with the internal POR and BOR circuitry. Some data can be found here, e.g.: http://ww1.microchip.com/downloads...39755a.pdf http://www.atmel.com/Images/doc8161.pdf http://www.nxp.com/documents...PC92X1.pdf Now you can say, it's only the POR and BOR which need the controlled Vcc rise times. But I think, everything else inside the micro, for instance the auxiliary oscillators and charge pumps for the flash programming high voltage, etc., is adjusted to that specified power-on slope rate window as well. I have made good experiences by controlling the power-on slope rate. I never had a flash crash so far. It's not too tricky to control the slope rate of Vcc. If you power the micro by a switcher, you can use the soft-start control. If you power it by a LM317, you can choose a suited cap at the adjust pin. In battery powered applications a high side switch with a PMOS and a suited gate cap can be helpful. Even a simple RC-element can do the job. Kai Klaas |
