Mike Palmer

9S08GB32 reset?

Discussion created by Mike Palmer on May 4, 2006
Latest reply on Jan 30, 2007 by jorge courett
MC9S08GB32CFU silicon 3L31R
Project is having apparent problems with RESET. It's possible to get the device in a weird state via external power application where the processor appears dead. If I take a voltmeter or scope probe and just touch the RESET pin (pin 1), the processor will start normally.
For low current consumption, we currently have a 330K resistor pulling RESET to Vcc. However, in trying to debug this, we substituted a 5K6 resistor with no luck. We also tried connecting a 0.1uF capacitor between the RESET pin and ground, again with no luck.
When no external power is present a 3.6V battery powers the micro through a 3.0V linear regulator. In this mode, the micro firmware tidies up and then executes a STOP instruction putting it into a STOP3 sleep. Current consumption from the battery in this mode is about 12uA which is exactly what we expected. The system is awakened by pulling IRQ low which can happen due to several events including external power being applied. If the micro is woken from sleep with the external power and then the power is removed and then re-applied in short order, the problem appears (i.e. micro apparently dead; touch reset pin and it wakes up and runs.)
I've read the 3L31R errata sheet and note that there is a RESET-related issue tantalizingly close to what we're seeing. We can also, once in a blue moon, get the thing so messed up that not only does external power have to be removed but the battery as well: That is, simply touching the reset pin doesn't help; forcing the reset pin low doesn't help; only if the power is completely removed can we get it to start again. The low-value resistor and capacitor issue mentioned on that errata sheet do not seem to help with this "sensitive" RESET pin issue we're seeing.
Any ideas?