Hello Michael ,

Thanks for your reply. What you say is right. In my first software , I had not written a NACK subroutine and just sent a STOP subroutine, so each time I powered on and read the EEPROM I used to get junk data.After writing a NACK this problem was sorted out.
Nevertheless I think it might be a good idea to reset the devices and then begin operations to avoid bad data due to a hardware glitch.
My case is if I have 4 EEPROMS on the same bus , and a random one is floating due to a hardware glitch , how would it be possible to detect and reset the same?
Or is this possibility so remote ,that I can ignore it?

Thomas