What config to use for owl node, like to monitor inputs and determine time between events?


(Lo Ra Westerhoven) #1

Hi,

I need some advise on which configuration I should use to get my project working.
What do I want to do: register the activity in nesting boxes for owls.
These boxes are at remote places so I need a battery operated system. I devised an arrangement with two micro switches (and one with two reed switches) that register an owl getting in or out of the box. Currently I am using a LoPy unit but that device cannot keep track of time in deepsleep mode. I need this functionality to keep track on how long an owl is inside the box.

So I am looking for another configuration that can provide the following functionality:

  1. Two digital inputs for the switches
  2. Keep track of elapsed time in between the activation of the switches while in deep-sleep mode.
  3. A “powerful” LoRa unit. As the boxes are at remote area I need a lot of range.
  4. As we inspect the boxes at least once a year a battery life of 2 years should be good enough.

Any advise is appreciated!


#2

interesting project.

So you detect if the owl is in or out by a microswitch.
How does that work ? (by weight?)

Keeping track of time can be done by an I2C RTC module, just read the time on situation A and situation B
The battery life depends on how many times the owl is getting in/out of his nesting box.


(Lo Ra Westerhoven) #3

I am using a microswitch to detect if an owl is passing through the lock. this lock is to prevent other animals to enter the box. The owl needs to pass this lock and will trip one of the micro switches.
Unfortunately your answer does not help me too much. can you advice on which devices to use e.g. a certain arduino/wmos board in combination with which LoRa unit?

image

20171115_225210


#4

Oh I’m sorry… do you have any experience in electronics / lorawan ?
If yes search this forum, you’ll find many ‘node’ projects, if not buy a complete LoRaWAN node.

do you have a gateway setup to receive the signals ?
how do you process the received data ?
what is 'a lot of range ’ ?


#5

For the lora bit, You could use 2 of these for “point to point” comms, They can be programmed via the Arduino IDE

Nice project - thanks for pics - please post back with what you decide in the end (even if it isn’t Lorawan )


(Lo Ra Westerhoven) #6

Yes I do have some experience in electronics and manage to program a LoPy in python (although I like old-school Pascal :grinning:) Recently setup a IMST based gateway and testing range with the LoPy GPS shield.
have no data processing in place yet but probably go for a simple google docs integration.
my shortcomming is that I recently started to do this kind of project and am not up to speed with things like arduino or attiny. I will need to learn but after LoPy I’d like to avoid un-necessary platform switches.


(Arjan) #7

So (after some debouncing) it clearly triggers one of the switches for each direction? (If not, then you would only know if the entrance was used, not if the owl entered or left.)

Also, are owls solitary birds?

I guess you already determined you cannot send a message for each event? (If you could, you could use the server time. But that depends on how often the state changes?)

I also guess you could not somehow detect if there’s some animals inside, as then it would always detect live once eggs have hatched? (If you could, you could measure at some fixed interval and only transmit, say, every 10 measurements. You would implicitly know some time then.)


(Lo Ra Westerhoven) #8

Yes I already thought about this. One switch is for incomming the other for outgoing. Keeping track of the order of switching I can count ( or maybe better estimate) how many owls are inside the box. E…g. Two times in and one time out means one owl in the box. Registering the time makes it possible to estimate how long 1,2 or 3 owls are inside the box but not which one. I add all the 1, 2 etc resident time up so once every 30 passages I send a message containing the number of passages and total residence times of 1,2 or more owls inside the box.
If there are less then 30 passages in an hour the node will send an alive message.
So besides keeping track of time I also need to store some data which could be done using a sd card or non-volatile memory.

I do not know what to expect but hope to find a pattern related to e.g. the presence of eggs or young birds.