Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
02/11/07 16:39
Read: times


 		 #132527 - What resets?
Responding to: ???'s previous message
Daniel said:
- I tried to correct the spikes I mentioned earlier, produce by LM2575 switching regulator. The simple method of increasing filter capacitor, obviously produces a longer Vcc rise time.

You mean something like that?




A better way to reduce the ripple (than to increase the capacitance) is to take care that the electrolytic capacitors are low ohmic! I often use this one in switchers:

http://docs-europe.electrocomponents.com/webdocs/0808/0900766b80808c61.pdf


Daniel said:
- Once the AT89S8253 starts with a rise time it does not like, no matter how you try to reset it again (forcing +Vcc in pin 9, debounced, for example) and/or how long. Result: the internal clock goes half the speed (checking ALE, or any external signal the software generates).

Have you tried the reset circuitry shown here:



Daniel said:
- In POR and BOD (figure 7-1 page 11 of Atmel manual) description, seems that Vcc has to be >2.7V before Internal RESET signal. OK, this is logic. But external reset (pin 9 in DIP40) is unable to reset/correct ALL the "wrong things" produced by an internal Reset happened when Vcc<2.7V. Weir, at least.

Don't use the internal resets. Try the reset circuitry shown above.

Kai


List of 83 messages in thread
TopicAuthorDate
Comments about AT89S8253            01/01/70 00:00      
   are you surprised, it's Atmel            01/01/70 00:00      
      are you surprised, it's Erik            01/01/70 00:00      
   Not neccessarily bugs...            01/01/70 00:00      
      AT89S8253 ALE pulses            01/01/70 00:00      
         Yes, a bit imprecise            01/01/70 00:00      
            And Vcc rise time...            01/01/70 00:00      
               What kind of "external reset"?            01/01/70 00:00      
      not that fast, Kai            01/01/70 00:00      
         Not so fast, Jan...            01/01/70 00:00      
            speaking from experience...            01/01/70 00:00      
               ...as engineer?            01/01/70 00:00      
               make that darn datasheet 5000 pages            01/01/70 00:00      
                  Exactly!            01/01/70 00:00      
                     I'm not sure what you mean, Kai            01/01/70 00:00      
                        That's what I meant            01/01/70 00:00      
                  Serial hung.            01/01/70 00:00      
            migration            01/01/70 00:00      
               Please quote carefully...            01/01/70 00:00      
         is it possible            01/01/70 00:00      
            I don't think so...            01/01/70 00:00      
   Any reset            01/01/70 00:00      
      I've wondered about this ...            01/01/70 00:00      
         Indeed!            01/01/70 00:00      
            Perhaps they're part of the problem ...            01/01/70 00:00      
               Do you think...            01/01/70 00:00      
                  That's not clear to me.            01/01/70 00:00      
               not here            01/01/70 00:00      
                  These are all COTS SMPS\'            01/01/70 00:00      
                     startup            01/01/70 00:00      
            Is it not instead a problem of letting reset go be            01/01/70 00:00      
               staring problem            01/01/70 00:00      
                  have you never stared at a problem :)            01/01/70 00:00      
                     Ah, so you meant....            01/01/70 00:00      
      What resets?            01/01/70 00:00      
         Any reset!            01/01/70 00:00      
            I got what you mean            01/01/70 00:00      
               Circuit proposed            01/01/70 00:00      
                  Interesting!            01/01/70 00:00      
         SMPS and LC filter            01/01/70 00:00      
            Switchers & low ESR            01/01/70 00:00      
               you missed the point            01/01/70 00:00      
                  What 100nF?            01/01/70 00:00      
            Rob is right            01/01/70 00:00      
               Ah            01/01/70 00:00      
                  Erik, you do use switchers frequently...            01/01/70 00:00      
                     nope, all I know is that they work and work and wo            01/01/70 00:00      
                        Thanks            01/01/70 00:00      
                           so ... Kai ... how would YOU build a power switch?            01/01/70 00:00      
                              there is no difference between "no power" and "in            01/01/70 00:00      
                                 There is a difference ... no power means no clock            01/01/70 00:00      
                                    ah, an investigatory measure            01/01/70 00:00      
                                       The Max1232 gets pretty confused ...            01/01/70 00:00      
                                          what decoupling do you have across the supervisor            01/01/70 00:00      
                                             This was a Philips ...            01/01/70 00:00      
                                                Too fast?            01/01/70 00:00      
                                                   The risetime isn\'t all that fast.            01/01/70 00:00      
                                          More details, please?            01/01/70 00:00      
                                             So far, it's just the reset that's been broken            01/01/70 00:00      
                                                Throw away the DS1232!!            01/01/70 00:00      
                                                   I\'ll have to send for a couple ...            01/01/70 00:00      
                                                      I've seen that one and it did not work            01/01/70 00:00      
                                                         Especially difficult to design            01/01/70 00:00      
                                                      Datasheets            01/01/70 00:00      
                                                      610msec reset active time of MAX1232 not enough?            01/01/70 00:00      
                                                         Let's get to the bottom of it ...            01/01/70 00:00      
                                                            Some ideas...            01/01/70 00:00      
                                                               each flash/eeprom HAS a charge pump...            01/01/70 00:00      
                                                                  Exactly, Jan!            01/01/70 00:00      
                                                               Yes, but ...            01/01/70 00:00      
                                                                  Why there are reset issues            01/01/70 00:00      
                                                                     so you're saying I should pull down ALE and nPSEN            01/01/70 00:00      
                                                                        Real Reset            01/01/70 00:00      
                                                                     be careful...            01/01/70 00:00      
                                                                     never heard of that one, can you elaborate            01/01/70 00:00      
                                                                        Elaborations on Reset            01/01/70 00:00      
                                                                           Mayabe that's worth some study ...            01/01/70 00:00      
                                                                              Which chip are you using?            01/01/70 00:00      
                                                                                 I can try several, but for now, Philips            01/01/70 00:00      
                                                   DS1232?            01/01/70 00:00      
                                                      it means Dallas Semiconductor            01/01/70 00:00      
                                                   re 1232            01/01/70 00:00      
   Sorry.            01/01/70 00:00      

Back to Subject List