Winter Outlook 2017/2018

 

It'll be colder than last year!

 

The squirrels have visited the cafe till 18th of Aug. 2017 which is a month longer compared to last year's sudden disappearance. Additional 4 weeks of food consumption is a first evidence of getting a colder winter this year.

 

However, I'll have a look in the nut consumption data of 2016 and 2017 to find similarities or differences in the squirrel's behavior. A challenge in analyzing the nut data is that I can't differentiate individual squirrels. Therefore it is impossible to assign nut consumption to distinct squirrels. Neither do I have a clue if there has been any change in individual members of our squirrel group.

 

There are some challenges left to improve data gathering for 2018.

 

 

 

 

Hackster.io and reed switch problems

On June 12th TheSquirrelCafe got featured on the Hackster.io's Blog: "The IoT Squirrel Cafe Is Nuts! 🐿".  Well, it blow me down!

However where there is light, there is also shadow:

The first iteration of TheSquirrelCafe had a mercury switch attached to the lid to detect lid openings. To simplify the wiring the mercury switch had been replaced by a reed switch and a magnet attached inside the lid.

I thought that this would be an easy and reliable solution to count how often the lid is opened, but this was wishful thinking. Unfortunately I've experienced that the reed switch sometimes sticks until tapped. Maybe the reed's contacts got magnetized?

 

Read More

Remodeling

The Squirrel Cafe has undergone an extensive remodel. Our valued guests appreciated the remodel and are now more likely to spend time at the The Squirrels Cafe than before.

 

The feeder has been replaced by a new one and sensor gadgetry has been updated as well.

 

Instead of a mercury switch a reed switch is used to get rid of the lid's wiring. The blinking LED of the old Cafe had been upgraded to a 7-Segment LED display to be more informative. The display shows how often the lid had been opened by our guests. The original Raspberry Pi Cam has been replaced by a USB Logitech C270 webcam for simpler positioning and a crisper image.

 

The homepage has been updated to reflect the changes. From now on the Twitter TL is showing the captured images instead of links only. The source code on GitHub has been updated as well.

 

Squirrels are back!

Have a happy new year 2017!

 

As you may read in the Lab Log, squirrels have been absent since July 17th 2016. I do still have no explanation why this had happened.

However out of the blue, at least one squirrel stopped by and grabbed 16 nuts from the squirrel feeder within 30 minutes. I hope that the squirrel will spread the word about the exquisite food being served at our cafe :-).

 

Today it is rather cold compared to previous weeks. We have temperatures below the freezing point right now. However we've had temperatures below 0°C for a couple of days in 2016 as well and no squirrel has been interested to grab a bite from the feeder.

Looking forward to see if the squirrel's reappearance is a one trick pony or if we are going to see this more often in the near future.

 

I'm happy to see that the software written to capture the squirrels activity works as designed even after having no guests at the cafe for over 5.5 months.

 

How cold has it been last month?

I'm not interested in short term temperature variations but would like to answer a question like "How cold has it been last month?". A histogram of temperature variation is one possibility to answer such a question.

 

What you will see on my updated temperature page are histograms displaying temperature variation over a month's time period. On Sep. 17th I've relocated the temperature sensor and used a special housing for the sensor itself therefore the first histogram covers a time period from Sep. 17th to Sep. 30th only.

 

September 2016

Sept./Oct. 2016

Read More