On Tue, Oct 08, 2002 at 08:57:50AM +1000, Harry Lemmens wrote:
> Another problem I have seen with the MSP430 is difficulty with Reset. I
> (now) deliberately load the MSP supply rail with a 1K resistor to drop the
> supply rail on powerdown, as otherwise it takes several seconds for the
> supply to drop enough to guarantee that a correct reset occurs. (A 2u2
> capacitor takes a long time to discharge through the powering down current
> of just the MSP, probably nano-amps below 1 volt...) The Data sheet spec for
> reset to be guaranteed is 0.4 to 0.0 volts, the supply can easily hover in
> this region for many seconds without some extra load.
Reset and brown out protection is a real problem with the MSP430.
Your 1k resistor solution is far away from a "low power" design ;-(. I have
no real working solution yet, but a voltage monitor could help to tie the
supply of the MSP430 low, if the input voltage is low. Unfortunately that
monitor chips only have one switching element. E.g. Zetex ZXCM209R is
switching to ground in a reset condition (open collector).
A good solution could be to connect the power supply of the MSP430 to
the output of a special reset circuit - that circuit should have a
high side switch and a low side switch.
The high side switch is on if the voltage is ok. The MSP gets the power
through the highside switch. It is off and the low side switch is on
at power failure to bring the voltage at the MSP430 down very fast.
I think there is no ready made chip for this and I have not yet a good
an reliable discrete solution. It is difficult to keep it working below 1 volt,
pn and zener voltages are temperature dependent, zeners need current ...
Maybe a combination of a voltage monitor chip and some external components
could be the solution, but hey I am ending now with 90% of my components on
the pcb used for the reset circuit ;-(.
Matthias
