Weather Station Project #13 - Thinking of a Different Way to Send Snow Related Weather Data

I am sitting in the Detroit airport waiting for my flight and it occurred to me that I might be able to complement my weather station with some additional data. What I would do is get the additional data, such as snow accumulations and do a surface extrapolation to find approximately what it should be near me. Then, I could use some laser scanning techniques to extract a free fall snow drift. I would need to shelter the area where I would determine the size from wind (so that it doesn't drift from other places, giving a false reading). By using a laser scanning technique, I could measure the height of the snow pile and thus the amount of snowfall. Before I send the information out in my normal weather stream, I could use the extrapolated point to determine if my findings are within the "ballpark" of a correct amount. This would give me some confidence in my findings. Again, this could be transmitted via a wifi link to my network to be combined with other weather information.

I should think about how I am going to interject the other three sensors with this type of thing.

Maybe I should think about opening up a blog on the Weather Underground to talk about DIY and Hobbyist Built Weather Stations.

Rearranging the Network for Fun

Well how annoying.  I haven't been blogging to this page for a while so I might want to catch up with what has been going on in the network at home.  I recently added a Personal Weather Station (PWS) and a Cisco router to the mix.  I find that once again I am not really trusting of the Verizon folk and I want to keep separate from them as much as is possible.

I did some wiring changes to the 4th switch so that the router only needs 2 wires through the wall; one from the switch in the side room (connected to the other 3 switches, with trunk lines) and one directly from the Verizon FIOS ActionTec router (as the WAN input).  The PWS is connected from a Raspberry Pi to a 3rd wire into the side room and into the 4th switch where it goes into a port with a "patch" VLAN that keeps the flow of information outside of my home network.  This "patch" VLAN then goes from switch #1 to the ActionTec.  Thus, I only have two connections now to the ActionTec.  I have effectively isolated the traffic from the PWS from everything else (as long as the switch on the ActionTec router acts like a switch).

I am also in the process of standing up a Ceton InfiniTV 6 to replace my HD HomeRun Prime.  The difference is 6 tuners instead of 3, and this is a pooled device; meaning I just grab one of the tuners from whatever is connected to it.  Since I really don't believe that my ZTec will be able to keep up with 6 tuners, I am going to stand up a Windows VM in the Mac Mini to host the Windows Media Center to the two Ceton Echo WMC extenders.

I might also use the 1 to 1 NAT capability of the Cisco router so that I can access all media devices from the subnet on the ActionTec.  This will simplify the use of iPhone/iPad apps to control the media devices.

I have some time this weekend to start making all of this work together.  Hopefully, I can get it into some form that will be acceptable to the wife and we can start getting rid of the FIOS Stb and VCRs to save some rental fees ($33 each month).

Stupid RPi Tip #5 - Conversion of SD Cards

I was able to convert an SD card from an RPi B (my work MediaPi) to a micro-SD card for an RPi B+.  I had gone through the trouble of putting some drivers on the regular 8GB SD card and I didn't want to lose the configurations that had already been done when converting over to an RPi B+.  The key to all of this was the use of Win32 Disk Imager and raspi-config.  The normal use of the Win32 Disk Imager is explained below:

Using Win32 Disk Imager
-----------------------------------
*** extracting an image file from an SD card ***
-- launch Win32 Disk Imager
-- load SD card
-- switch dropdown on right to the SD card that was loaded
-- select the file folder icon
--- go to the directory that you want and set a file name <filename>.img
--- click OK
-- you should see the location for the image file filled in, and Device being SD card
-- click on the Read button
-- when the file is finished, exit the program
-- you now have an image file of the SD card

-----------------------------------
*** pushing an image file to an SD card ***
-- launch Win32 Disk Imager
-- load SD card
-- switch dropdown on right to the SD card that was loaded
-- select the file folder icon
--- go to the directory with the image file, select it
--- click OK
-- you should see the location for the image file filled in, and Device being SD card
-- click on the Write button
-- when the file is finished writing to the SD card, exit the program
-- you now have an SD card with the image written on it

I followed the directions above marked "extracting an image file from an SD card" with the 8GB SD card to make an image.  I then followed the "pushing an image file to an SD card" with the 16GB micro-SD card to copy that image.  Once the RPi B+ system booted up, I then used raspi-config to expand the image to fill the entire micro-SD.  Problem solved - now back to troubleshooting.  I think I am now going to work on an SSH tunnel back to my house.

Weather Station Project #12 - Weather Station Still Holding Up

Maybe I'm paranoid but I do check on the weather station every once in a while. This involves looking at apps on my iPad to make sure that everything is working and the station is updating. It just seems that I have missed something because the data stream is running so well.

I still have to get around to making the sensors work and integrating it into the Weather Underground data stream. Now that it is winter and around Christmas time, the development will have to wait until I am able to have time. Right now I have a ton of things going on at work and I have family responsibilities.

I do wish everyone who comes across this little blog a fond Merry Christmas!

Got WiFi adapters TL-Link WN725N to Work

After getting the Netgear wifi adapters to work with my bring-to-work RPi box, I took them out and put them on the RPi at the back of my new ASUS touchscreen.  I then went to MicroCenter and got a couple more, this time they were TL-Link WN725N.  Well, when I booted the RPi box up it didn't work.  Simple explanation is that the native drivers were not usable.  So I got one of the Netgear wifi adapters and put it on the RPi box, to at least get WLAN0 working and connected through my DIR-505L wireless router.  After an update of the software on the RPi, I then started looking for a solution and found it at http://raspberrypi.stackexchange.com/questions/23338/tl-wn725n-v2-driver-needed-for-linux-3-12-28.   The solution was found on Gordon's site and the following worked great:

wget gordon.d4rc.net/8188eu_3.12.28.zip
unzip 8188eu_3.12.28.zip
sudo cp rtl8188eufw.bin /lib/firmware/rtlwifi
sudo install -p -m 644 8188eu.ko /lib/modules/3.12.28+/kernel/drivers/net/wireless
sudo insmod /lib/modules/3.12.28+/kernel/drivers/net/wireless/8188eu.ko
sudo depmod -a
sudo reboot

Thanks Gordon for the tip!  Although the GordonSys site is no longer there.

Working with a GuertDino

Ok, I now have a GertDuino in hand. I just happened to see it when I was in Micro Center. right now, I am planning on using it with the optical parts and the pollen parts of the weather station. It's Black Friday weekend and I think I will have some time to work on this. I have been facing a problem with how to transport the information from the sensors to my home network without having a physical connection, or just one cable. I have thought about using PoE injector/extractor pairs but that doesn't seem to be the correct way to do the job. The GertDuino might be a way that I can offload the sensor sampling from a RPi - not that I need to do such a thing; only that it is a neat use of technology to have a cross-platform sensor rig. Oh well, maybe I might use it for something else entirely. But a neat idea anyway.




-- LW

Weather Station Project #11 - Working on Code for Other Sensors

I had some time this weekend to work on the other sensors that I will be using in the weather station.  I decided to go ahead and test out the TSL2561, the SI1145, and the Sharp  Dust Sensor.  The Arduino breadboard that I am using for the other sensors in the test suite is the Parallax Board of Education.  This breadboard plugs right into the Arduino Uno (actually on top of an Ethernet shield and then into the Uno) and provides breakout of the analog and digital lines for prototyping as well as allowing other shields to be added to the mix.  I wanted to get the software working first followed by putting the sensors on a Arduino prototyping shield.  My thought was to use the Arduino to do the sensor interface and push a serial output down the USB line to a Raspberry Pi.  Alternatively, I have a couple of Wireless Transciever modules to do the same thing.  Haven't thought that far ahead yet.  The whole breadboarded setup is shown in the image below:




The visual sensors, the TSL2561 and the SI1145, are both cantilevered over the edge of the breadboard so that I can get to the pins below for the jumper wires.




The Dust Sensor in this photo is simply sitting on top of the Board of Education since there are no electrical conductive parts on the bottom.  I simply plugged the wires into their respective locations according to prototype setups that I found online.



The wiring setup for the Sharp Dust Sensor is according to what is recorded at standalone-sharp-dust-sensor.  The only difference is that I used A3 for the analog instead of A6 which is not available on the UNO.  The connections for the SI1145 is according to the instructions found at Adafruit-si1145-breakout-board-uv-ir-visible-sensor.  The connections for the TSL2561 is according to instructions found at Adafruit - TSL2561 Luminosity Sensor. Since the I2C addresses are different between the two visual sensors, I connected the SDA and SCL connections in parallel.  The two visual sensors make use of the Adafruit sensor library.

I tested each of the sensors independently from one another using test programs that were provided.  Now I am in the process of combining the test program pieces from each of the sensors into the same sketch.  After that, I will run a test to gather data over the period of a couple of days to try and compare to known values from other weather stations.  Once that is accomplished, I will endeavor to solder the sensors into the Arduino shield and place the whole lot into the specified fin enclosure.

-- LW