Help with power on reset

[galapogos]

Hi,
I have a chip that has a reset input that is currently connected with a
simple RC circuit, according to reference design. It is an active high
reset signal, i.e. when power is applied, the chip resets. However, I
wish to control the actual reset of the chip with an external signal
via a MCU's reset pin. I have tried to connect a MCU's GPIO pin to the
chips reset, and it works, but I'm wondering if it's a good idea to do
that with the existing RC circuit in place? Since my MCU will be
holding the reset signal low until it decides to reset the chip,
wouldn't I be better off with a pulldown instead of a pullup for the RC
circuit(pulldown because on system startup the MCU's IO pin is high,
and only a split second later does my firmware run and set it to low,
so for a split second the chip reset could be high instead of low,
triggering a reset when I don't want it to happen)?

I'm sorry if this is a stupid question. I don't have any experience
with reset circuits. Thanks!

 

[ehoosee.tianya.cn]

I do not understand why use a GPIO to control reset?

If you really want to do like this,RC is no need.

"galapogos дµÀ£º
"

 

[[email protected]]

Hi,
I have a chip that has a reset input that is currently connected with a
simple RC circuit, according to reference design. It is an active high
reset signal, i.e. when power is applied, the chip resets. However, I
wish to control the actual reset of the chip with an external signal
via a MCU's reset pin. I have tried to connect a MCU's GPIO pin to the
chips reset, and it works, but I'm wondering if it's a good idea to do
that with the existing RC circuit in place? Since my MCU will be
holding the reset signal low until it decides to reset the chip,
wouldn't I be better off with a pulldown instead of a pullup for the RC
circuit(pulldown because on system startup the MCU's IO pin is high,
and only a split second later does my firmware run and set it to low,
so for a split second the chip reset could be high instead of low,
triggering a reset when I don't want it to happen)?

I'm sorry if this is a stupid question. I don't have any experience
with reset circuits. Thanks!

I can't follow this since at times you indicate reset is high, and
other times reset is low. If reset is low, you would call it reset bar.

 

[mkaras]

Hi,
I have a chip that has a reset input that is currently connected with a
simple RC circuit, according to reference design. It is an active high
reset signal, i.e. when power is applied, the chip resets. However, I
wish to control the actual reset of the chip with an external signal
via a MCU's reset pin. I have tried to connect a MCU's GPIO pin to the
chips reset, and it works, but I'm wondering if it's a good idea to do
that with the existing RC circuit in place? Since my MCU will be
holding the reset signal low until it decides to reset the chip,
wouldn't I be better off with a pulldown instead of a pullup for the RC
circuit(pulldown because on system startup the MCU's IO pin is high,
and only a split second later does my firmware run and set it to low,
so for a split second the chip reset could be high instead of low,
triggering a reset when I don't want it to happen)?

I'm sorry if this is a stupid question. I don't have any experience
with reset circuits. Thanks!

If the RESET to your microcontroller is high active then that is what
it is and it will require a high level signal to assert the reset.
Whatever pulldown/pullup you try to apply will have to take this into
account. If you intend to have the MCU able to reset itself from the
use of an I/O pin that pin will need to be set high as an output to
cause the reset. You do have some issues to consider however.

First off you will still need to provide for a valid initial reset at
power up. Whatever you do for this will have to somehow combine
logically with the port pin logic you want to use and whatever other
reset sources you may have such as a manual reset switch. The app note
reset circuit consisting of the resistor and capacitor may be suitable
for this initial power-on reset or it may not. Many designers opt to
use one of the more predictable reset generator chips (also known as
microprocessor supervisory chips) that are made by the likes of TI,
Maxim, National, Philips or Analog Devices.

Secondly you need to consider what happens to the microcontroller when
its firmware programs an I/O pin to output and drives the reset line to
the same microcontroller to an active assertion. This will almost
instantly stop the program that is running and return the
microcontroller back to almost the intial start up conditions. This of
course will include returning the I/O pin back to its default state
which is most likely as a high impedance input. This subsequently has
the effect of causing the firmware asserted reset signal to be taken
away. For a great many microcontrollers the time the reset signal stays
asserted in this senario is shorter than the minimum reset assertion
time that the MCU manufacturer recommends for their device. So you may
very well need to arrange the circuitry so that the I/O pin triggers a
reset assertion through some type of pulse extension scheme via a
resistor/capacitor circuit or through the use of the "reset switch
input" that can be found on many MCU supervisory chips.

- mkaras

 

[[email protected]]

Hi,
I have a chip that has a reset input that is currently connected with a
simple RC circuit, according to reference design. It is an active high
reset signal, i.e. when power is applied, the chip resets. However, I
wish to control the actual reset of the chip with an external signal
via a MCU's reset pin. I have tried to connect a MCU's GPIO pin to the
chips reset, and it works, but I'm wondering if it's a good idea to do
that with the existing RC circuit in place? Since my MCU will be
holding the reset signal low until it decides to reset the chip,
wouldn't I be better off with a pulldown instead of a pullup for the RC
circuit(pulldown because on system startup the MCU's IO pin is high,
and only a split second later does my firmware run and set it to low,
so for a split second the chip reset could be high instead of low,
triggering a reset when I don't want it to happen)?

I'm sorry if this is a stupid question. I don't have any experience
with reset circuits. Thanks!

Hi Here your capacitor will take of that short unwanted glitch. Pull up
or pull down may not matter much. your RC timing should be designed in
such a way that the unwanted spike or glith pulse width is less than
your RC timing

NSR