0
\$\begingroup\$

analog circuit

We have a circuit like the one I attached to this post.

The idea behind it is to generate a pulse every time the switch is closed.

We have two constraints for this circuit:

a. The pulse should be at least 50 micro-second

b. The max current of this circuit when the switch is on should be less than 250 nA while the switch is closed (only the analog, without the MCU part)

However, This circuit suffers from unwanted pulses in the MCU IO pin because the pull-down resistor is too big.

How can we solve it?

Or design another circuit with these constraints (and also the pulses problem...)

Any ideas?

\$\endgroup\$
3
  • \$\begingroup\$ Why such a low ON current of 250 nA? \$\endgroup\$ Commented Jan 6, 2020 at 15:39
  • \$\begingroup\$ How often is the switch switched (ie, frequency)? What is the switch duty cycle? How long does it stay on/off? \$\endgroup\$
    – bobflux
    Commented Jan 6, 2020 at 16:33
  • \$\begingroup\$ What are the input characteristics of the GPIO pin? Maximum input leakage current and pin capacitance? What is the parasitic capacitance of the wiring? \$\endgroup\$ Commented Jan 6, 2020 at 17:32

1 Answer 1

1
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

  • the bounce time must be less than T, otherwise increase T

Rev 1

  • Added buffer hinted by Elliot. (chose Schmitt Nand)
  • corrected polarity for negative logic resets.
  • added Power on Reset (POR), low leakage Si diode.
  • added Watch Dog Timer reset input (WDT) (useful)

For more wisdom, read what microchip say...

http://ww1.microchip.com/downloads/en/DeviceDoc/31003a.pdf

\$\endgroup\$
5
  • \$\begingroup\$ I don't think this will be reliable if it is connected to the input pin of a microcontroller, where the input capacitance may be 10pF and the input leakage current is up to +/- 1 microampere. \$\endgroup\$ Commented Jan 6, 2020 at 17:31
  • \$\begingroup\$ Yes then a Schmitt buffer is preferred. Leakage is worst at max temp. \$\endgroup\$ Commented Jan 6, 2020 at 17:33
  • \$\begingroup\$ I am concerned from the leakage current. Does not the Schmitt itself consume more than 1 uA? \$\endgroup\$
    – Yonatan
    Commented Jan 7, 2020 at 12:05
  • \$\begingroup\$ @TonyStewartSunnyskyguyEE75 thanks. Still have some questions: I don't understand all the POR, WDT, diode and uC reset stuff. The pulse is not for reset is for counting. What about the schmitt? I understand it is good for leakage current - but the schmitt has its consumtion, is not it? In addition, the 33pF is enough for the unwanted pulses I get? \$\endgroup\$
    – Yonatan
    Commented Jan 7, 2020 at 13:24
  • \$\begingroup\$ Define uC input used for wake up. I assumed it was reset.. like Power on Reset POR. My fault.. \$\endgroup\$ Commented Jan 7, 2020 at 17:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.