Author Topic: Mailbox Notifier project question  (Read 1526 times)

labusr

  • Newbie
  • *
  • Posts: 34
Mailbox Notifier project question
« on: March 17, 2014, 03:32:52 PM »
I want to recreate the mailbox notifier project.  But I have a few questions:

1)  Is there some other type of sensor that consumes less current than a hall effect sensor?  I thought magnetic sensors were like a limit switch on a small spring and the magnet pulls the limit switch into the contact...but that only exists in my head apparently because I couldn't find it on ebay.  Maybe a physical limit switch would consume less power, but that's hard to mount and have it go off consistently.

2)  Sleep Mode:  which method is more energy efficient?

a)  Sleep the Arduino for a second, and wake up just long enough to power the hall effect sensor via digital output pin and sense whether mailbox door is opened.  Assume mailman keeps the door opened for greater than 1 second.

b)  Put the Arduino in deep sleep but power the hall effect sensor all the time.  The hall effect sensor wakes up the Arduino.

Felix

  • Administrator
  • Hero Member
  • *****
  • Posts: 6022
  • Country: us
    • LowPowerLab
Re: Mailbox Notifier project question
« Reply #1 on: March 17, 2014, 03:41:27 PM »
You mean a reed switch?
You could use that but typically it's a bit bulky, whereas this TINY hall sensor is vritually non existent next to the Moteino which itself is very small.
You could also use a small tilt switch where the internal mercury or metal ball moves when there is movement causing a contact closure. But those are mechanical so prone to error sometimes.
I am experimenting with a PIR sensor with very good results. I may release a kit based on this method. The PIR consumes only 60uA while non active and about 260uA when motion detected, causing an interrupt which can wake the Moteino on pin D3.
So yeah I would recommend trying PIR. That one does not depend on a magnet or another part that it has to align with. It will always detect the motion when the box opens.

The hall effect uses 2.5mA according to DS: http://www.melexis.com/prodfiles/0004822_US5781_rev014.pdf
So leaving it ON is a huge energy loss.
Also 1 second mailbox open is about the max I would allow for security purposes, someone could peek in your mailbox for less than 1 second, so you may want to catch that action. With the PIR you don't have to worry about that. Any motion will trigger it.
« Last Edit: March 17, 2014, 03:43:56 PM by Felix »

labusr

  • Newbie
  • *
  • Posts: 34
Re: Mailbox Notifier project question
« Reply #2 on: March 17, 2014, 04:42:25 PM »
The PIR consumes only 60uA while non active and about 260uA when motion detected, causing an interrupt which can wake the Moteino on pin D3.
So yeah I would recommend trying PIR. That one does not depend on a magnet or another part that it has to align with. It will always detect the motion when the box opens.

Are you looking at one of these PIR?

https://www.sparkfun.com/products/8630

I couldn't find the current consumption of the PIR sensors.  Their data sheet isn't very good.  Also, for the PIR, are you bypassing the voltage regulator on the PIR's circuit board?  I read somewhere that the PIR can run on 3.3V if you bypass the voltage regulator.

Felix

  • Administrator
  • Hero Member
  • *****
  • Posts: 6022
  • Country: us
    • LowPowerLab
Re: Mailbox Notifier project question
« Reply #3 on: March 17, 2014, 07:42:18 PM »
I have these in my store but currently out of stock: http://lowpowerlab.com/shop/index.php?route=product/product&product_id=113
I have verified the consumption on these to be around 50-60uA including the Moteino in sleep mode.