NV RAM

do you have a real supervisor in your circuit or a stupid RC reset.

Erik

Sorry but can you elaborate more I haven't implemented any supervisor and even I don't have reset button - just an RC combination

a RC does NOT reset on power down and the uC just run wild.

Erik

Hello Erik Thanks for your interest Here is how my circuit is Vcc | | | Capacitor(10uF,25V) | |----------------------(Reset pin of uC) | Resistor(8k1) | | GND

Can you please tell me if there is any problem with this circuit and my NVRAM What can be the problem with this circuit in general it used to work fine

Sorry for that diagram but while typing it was looking good but after preview it is fuzzy Please try to understand it

It seems like you where trying to draw this:

 Vcc
  |
  |
  |
Capacitor(10uF,25V)
  |
  |----------------------(Reset pin of uC)
  |
Resistor(8k1)
  |
  |
 GND

But a simple RC reset is not recommended for several reasons. The voltage over the capacitor is changing slowly, so the processor may be reset multiple times in case there is noise in the Vcc and the reset input doesn't have a wide enough hysterese.

Another problem is that the reset is assymetrical. You want the reset signal to be activated while waiting for the supply voltage to rise on startup. But also when the supply voltage is slowly dropping after you have disconnected the power - or the battery voltage has dropped below the safe working range.

It is recommended that you use a dedicated reset chip. That may have a open-collector or a totem output, but will give a distinct toggling beteen active and inactive, and they will react to any irregularities in the supply voltage.

Having the processor run at too low voltages, or to try to program EEPROM, flash or battery-backed memory below the minimum required write voltage for the memory, may result in destroyed memory contents and may potentially activate read-protect and/or write-protect features of the memory.

Thanks Westermark I guess the diagram should have been in code form Anyway can you suggest any reset supervisor IC or Reset chip One more thing I checked the voltage levels(had DMM connected for all the time) at NVRAM they were always at around 4.5V to 5.10V however not at the time of switch off (removing plug)

One more thing what I've found in my case- This is the 3rd chip I've damaged, In my program I've sent all the data from RAM to PC through serial (i.e. using hyperterminal) This one was working fine Till a moment - Once when I tried to switch off the circuit suddenly I got all the data(resetted i.e. every mem location was holding its own value) on serial and this is how I'm making them unrepairable. Second one was damaged in same way Now can anybody help me or can anybody explain whats happening? Thank you guys

Is there a sequence which should be followed while powering down like - 1. power down NVRAM then power down uC
otherwise keep RAM always powered up just uC is powered down
Thanks in advance

this is a valid question to ask:
are you using a MAX232(equivalent) or one of those BLASTED so called "RS232 interfaces without the chip"

Erik

I'm using MAX232 obviously connected to Rx and Tx of uC but it shouldn't be problem right? As I'm reading from RAM and putting it on serial.
Without chip serial communication doesn't happen with PC(sometimes)

I'm using MAX232
good
Without chip serial communication doesn't happen with PC(sometimes)
without WHICH chip?

Erik

Its MAX232 series IC like MAX232, HIN232, etc.

Can anybody suggest me proper circuit(proper way of Vcc/GND) to work with RTC because I'm working with breadboard and i think that would be the problem as every time i put 2-pin vcc/gnd in berg there used to be some weird data on hyperterminal

When I removed interrupt pin i.e. pin 19RTC -> pin3.2uC
the program/circuit is working fine without any problem
What should I do about it I want to use alarm interrupt One more thing when I'm reading RAM all the data is correct except alarm location is there any problem with reading alarm locations?