Abstract[edit | edit source]
The purpose of this project is to present logged data from the CCAT house solar panels at Humboldt State University onto a public-access website in real-time. To access the live data click here: http://www.pvoutput.org/intraday.jsp?id=31570&sid=28922
Background[edit | edit source]
The Campus Center for Appropriate Technologies (CCAT) house at Humboldt State University in Arcata, California has a variety of solar-powered projects on site. Our Engr305 Appropriate Technology project for the Spring 2014 semester is to provide CCAT with a solar energy monitoring system, which will upload the voltage, watts, kWh's, and display this information on the CCAT website. CCAT has asked that we create a real-time energy monitoring system that can store logged energy generation data from the PV panels, and can also graphically present the data on the world-wide-web in an easy-to--read manner. The information that has been tracked by these PV panels can be used by science or engineering students at HSU to analyze, and also to show the world how CCAT is making use of 10+ year-old solar panels, and the efficiency of the appropriate technologies.
Problem statement[edit | edit source]
The objective of this project is to develop a data monitoring system to capture the electricity generation from the photovoltaic panels located on the rooftop of the CCAT house at Humboldt State University and display the data online in kWh's. This will be accomplished by researching and determining the best software and hardware to install, allowing CCAT officials to store the solar power data and access it at any time. Ideally, with open-source software and affordable hardware that will allow any one to access the information online. The data will be displayed on the CCAT websitein easy-to-read graphs and will continuously log the data throughout the lifetime of the system using a raspberry pi mini-computer hooked up to a Magnetek Aurora inverter, and using pvoutput.org to store the data on their servers, free of charge.
Criteria[edit | edit source]
The goal of this project is to install PV monitoring hardware and software at the CCAT house located at Humboldt State University. The data must be uploaded to the CCAT website and be available 24/7 to anyone interested.
|Accessibility||Solar data from CCAT must be streamed online in kWh, and must be accessible to any one online|
|Education Factor||Historically logged data must be easily accessible and formatted in such a way that students can use the data for future projects, assessments, etc.|
|Low maintenance||Solar controls must automatically upload information to the CCAT or Appropedia website to display logged solar data|
|Budget||Must not exceed budget of $1,000|
|Real time data||Data must be graphically represented on the CCAT or Appropedia website, and updated at least twice per day, and at most, in real-time|
|Ease of use||Graphical representations of data must be easy to follow, and software must be easy for CCAT co-directors to navigate, if necessary|
|Aesthetics||Graphs must be pleasing to the eye, look professional, and be easy to read|
Literature Review[edit | edit source]
Many Residential Solar Data Collection and Monitoring design options were reviewed during the research and development process.
Electronic Data Logging[edit | edit source]
This general term is defined as an electronic device that records data over time. It can be built-in to a system to receive readings off of a sensor or directly from the data source. Small processors that collect the data, utilize computer software, and display the data with either a direct LCD screen or transfer the data to an alternative storage device, such as the world-wide-web. One of the primary benefits of using a data logger is the ability to collect live data automatically in real-time, continuously throughout the day. Once activated, data loggers do not require human maintenance; they can be left unattended and continue to successfully log information. 
Solar Photovoltaic Panels[edit | edit source]
Solar cells, also called photovaoltaic (PV) cells, take light photon energy and convert it into voltage, creating electrical current. PV cells are made out of a semi-conductive material, mainly Silicon, and capture excited electrons, allowing them to flow through conductive materials. The typical solar PV panel contains about 40 cells, and each home contains about 8-15 PV panels to meet the electricity demand of the average residential home. 
Energy[edit | edit source]
Energy, in this case, is the rate of fuel, electricity, or sunlight that is used or created over a period of time.
Energy Units[edit | edit source]
- Watt (W) -A Watt is a unit of power, equal to energy generated or used. It is equal to one Joule per second.
- KiloWatt (kW) -A kW is a unit of power, equal to 1,000 Watts. This is the scale of power that a residential PV system will generate. If you think of this in terms of traveling in your car, a kW would be your instantaneous speed at that specific time, or the average speed of your car while driving to work.
- KiloWatt Hour (kWh) -A kWh is a unit of energy. It is equivalent to 1,000 Watts used for a 1-hour period of time. This measurement is usually found on home energy bills. If you think of miles traveled in a car, energy (kWh) is equivalent to the miles per hour, or distance over time. In terms of PV systems, energy is the amount of sunlight hitting a solar panel over a period of time. 
Raspberry Pi[edit | edit source]
The Raspberry Pi is a wallet-sized computer that was developed in the UK as a tool for children to learn computer programming. It is compatible with many operating systems such as Python, Raspbian, Linux, and many more for the beginner to advanced level computer programmer. You can run word processing and spreadsheets just like a laptop computer, as well as watch High Definintion television through its' HDMI port and audio outputs. It is not limited by its physical components, as you can solder other parts to the device, and connect through its' 26 dedicated GPIO pins, which can be infinitely expanded upon. 
Operating System[edit | edit source]
When using a device such as a Raspberry Pi, it is necessary to assign it an operating system. An operating system is the most vital program that runs on a computer. It is responsible for managing tasks and programs, which your computer relies on to function. Raspberry Pi computers rely on Operating System Image Software, and they can operate on a wide variety of images.
Linux[edit | edit source]
Linux is a combination of Graphical User Interface (GUI) and the traditional Command Line Interface. Running Linux allows you to directly send commands from the keyboard to the operating system through a terminal. It was originally developed as, and is, and open-source software.
NOOBS[edit | edit source]
NOOBS stands for the New Out Of Box Software. It is a package that allows you to pick from a variety of operating systems and allows you to edit their configuration settings. It allows you to switch from different operating systems, and upon inital startup, presents a windows-like desktop experience. NOOBS allows new users to easily navigate throughout the Start menu, Operating System Images, and Internet to program the Raspberry Pi for any application you like.
Open-Source Software[edit | edit source]
Open-source software is computer software that can be written, shared, distributed and freely edited by any one with no restrictions. Open-source software relies on individual creation and user input for a variety of computer programs and applications. An excellent resource for open-source energy monitoring projects is: Openenergymonitor.com.
Current PV System[edit | edit source]
Currently, the CCAT house has 8 PV panels on the rooftop of the main house. These panels are the same ones that were generously donated by ASE Americas back in 2001, and remounted onto the rooftop when CCAT moved locations in 2008. The Inverter Currently Installed at CCAT is a Magnetek Aurora 3600-US-OUT. The company Magnetek has since been bought by Power-One. The Power-One Aurora 3600-US-OUT installation manual can be found here. The Wiring diagram to the right is from 2014, and is an accurate record of the current system.
The 8 solar panels on the rooftop are wired together in 2 strings of 4, mainly because the panels were previously hooked up to two separate inverters. Since there is only one inverter, both strings of the 4 solar panels had to be wired together in series, and hooked up to the one updated inverter through the "PV Input 1" port on the Magnetek Aurora inverter. Instead of re-wiring the panels on the rooftop when replacing the two old inverters with the most current Magnetek Aurora inverter, Solar Roger, the local solar electrician/installer, decided to just splice these wires together. This way, the 2 strings of 4 solar panels wired separately on the rooftop are merged together in series, into one continuous string, connected to the input of the one Magnetek Aurora inverter.
The current CCAT Photovoltaic system is shown below through images of the DC disconnect box, the box of wires below the inverter, and the Magnetek Aurora inverter.
This image shows the PV systems' external wire box in the downstairs storage closet of the CCAT house: the red and white tape, joining the black and white wires are the location where these 2 strings of 4 panels, wired in parallel, became one string of 8 solar panels, wired in series.
Construction[edit | edit source]
Time Table of Tasks[edit | edit source]
|Research||Week 07, Friday, 03/07/14|
|Shopping||Week 08, Friday, 03/14/14|
|Learning Raspberry Pi||Week 10, Friday, 03/28/14|
|Programming microSD card for Raspberry Pi||Week 13, Friday, 04/18/14|
|Installation and Testing||Week 15, Friday, 05/02/14|
|Create YouTube Video||Week 16, Friday, 05/09/14|
|Finalize Appropedia page||Week 16, Friday, 05/09/14|
|Showcase final product to CCAT and ENGR 305 class||Week 17, Monday, 05/12/14|
|Project Completed||Friday, May 16, 2014|
Costs[edit | edit source]
Table of costs:
|Quantity||Material||Source||Cost ($)||Total ($)|
|1||Raspberry Pi||Amazon.Com- Raspberry Pi (free shipping)||39.00||39.00|
|1||Raspberry Pi Case||Amazon.com- Raspberry Pi Case (free shipping)||9.00||9.00|
|1||RS485 to USB cable||Mouser.com||42.50 +shipping||49.49|
|1||Micro USB to USB power cable||RadioShack||21.64||21.64|
|1||5V USB Power Adapter||ebay.com- USB power adapter||5.00||Donated|
|1||Ethernet cable 25'||HSU Information Technology Services||20.00||Donated|
|1||SD Card 8GB||Best Buy||10.00||Donated|
|1||HDMI to HDMI cable (used for set-up)||Amazon.com||10.00||Donated|
|1||HDMI compatible monitor (used for set-up)||Newegg.com||100.00||Donated|
|1||Computer Mouse (used for set-up)||CDW.com||10.00||Donated|
|1||Keyboard (used for set-up)||Monoprice.com||10.00||Donated|
Operation[edit | edit source]
Maintenance[edit | edit source]
The brilliant part about the Raspberry Pi is that once it is programmed and set up initially by this project team, there is no need for any future tinkering with the device or its' components. CCAT Co-Directors can simply log onto the internet and view the live-stream energy data from the website. The Raspberry Pi itself is stored in the downstairs storage closet next to the inverter. As long as the power is plugged in, the device will need no maintenance.
Project Design[edit | edit source]
The cheapest, most versatile option for this design is to connect directly to the inverter through its' RS485 connection. Currently, the Magnetek Aurora inverter has data logging capabilities and can translate that data to Aurora Monitoring software. That monitoring software presents the electricity generation data graphically in real-time, but it is bound to a local computer only. Using a Raspberry Pi, this data can be sent to the internet using open-source codes.
Components[edit | edit source]
This design incorporates a Raspberry Pi, a USB to RS485 cable, an ethernet cable, microUSB power cord, and a MicroSD card.
The RS485 to USB cable transfers the logged data from the Magnetek Aurora inverter to the Raspberry Pi. The Raspberry Pi is a small computer that requires an SD card encoded with an operating system; we used Raspbian with an 8GB SD card. The SD card that runs Raspbian is also encoded with Aurora open source forge software that displays the inverters' logged solar power data online. It is sent through the Raspberry Pi network plug to the Local Area Network (LAN) connection located in the CCAT house. Eventually, we will have HSU's Telecommunications crew install a LAN connection in the downstairs storage closet where the inverter and Raspberry Pi will reside.
Once the data from the inverter is sent to the internet via LAN, open-source software translates the Aurora monitoring data onto a website called PVoutput.org. This website graphically shows how much electricity is being generated by CCAT's PV system in 5-minute intervals.
This design option costs no more than $100 to implement, and is an excellent demonstration of the abilities of open-source software to provide a community with PV data monitoring.
Replication[edit | edit source]
For those of you who wish to replicate this project:
This hardware is ideal for a multitude of projects, ranging from programming robots, to data collection. This software is specifically meant for this Aurora inverter, and will not benefit any other projects, unless they are running Aurora Monitoring software directly through an Aurora inverter. For replications of this project with other inverters, research will be necessary to find the right open-source software for your needs.
Project goals[edit | edit source]
- Learn about the CCAT PV system
- Research the best monitoring design for this system
- Learn about the Raspberry Pi and the Raspbian operating system
- Research and determine the best open-source software for the Raspberry Pi
- Learn how to program the Raspberry Pi with open-source software
- Get project idea approved by Information Technology Services (ITS) at Humboldt State University
- Get project approved by CCAT Co-Directors, CCAT Key Advisor, and HSU's Director of Sustainability
- Get the kWh data from the Aurora Inverter to the world-wide-web
- Embed the Aurora Monitoring Program code onto the CCAT website
Instructions[edit | edit source]
We used a Mac Computer with an OS X Operating system for all the following steps.
Part 1: Installing NOOBS Software onto SD Card Software[edit | edit source]
If you are starting from scratch, follow these instructions to install NOOBS onto your 8GB or larger SD card.
However, to save 40 minutes of your time, it is highly recommended to purchase a pre-loaded SD card from a trusted online distributor, such as Amazon.com.
If you are starting with a pre-loaded NOOBS SD Card, skip to Part 2 below.
How to Format your SD Card and Install NOOBS:
Part 2: Installing SourceForge.net Aurora Monitor to the SD Card[edit | edit source]
How to Install Aurora Monitoring Open-Source Software Onto SD Card:
Part 3: Raspberry Pi Hardware Setup[edit | edit source]
How do I set-up the Raspberry Pi Hardware?:
Part 4: Formatting Raspberry Pi[edit | edit source]
hit ENTER on the keyboard to complete each command.
How to format your Raspberry Pi with fully loaded SD card:
Part 5: Configure Aurora Monitor Settings[edit | edit source]
How to Finalize Raspberry Pi Aurora Monitor Software:
Conclusion[edit | edit source]
Testing Results[edit | edit source]
The testing was all done after our programming was initially proven successful. Therefore, the results of our physical testing were successful. Programming the Raspberry Pi was broken up into 3 separate working sessions. The first part of testing went well up to the point of programming the SD card with the Raspbian image and Aurora Monitor sourceforge.net software.
The second part required minor adjustments to the open-source instructions to best fit our project needs. We received help from Roger Tuan, a former CCAT Co-Director, to make sure we were on the right track with changing the codes. The code changes included: setting up an IP address for the Raspberry Pi as well as filtering out a couple type-o's in the open-source code.
The third round of testing was a dry run hooking up the inverter to the Raspberry Pi and the LAN connection. We had to work with Telecom to get a MAC address of the Raspberry Pi approved before being able to connect to the Humboldt State University network. We also had to set a temporary IP address for internet connection to the Raspberry Pi because at this time it does not have its' own LAN connection in the CCAT house, it is just borrowing a LAN connection from the CCAT library computer.
We drafted a test page for the CCAT website, with the help of Roger Tuan, embedding the pvoutput.org graphs onto a /PV link on ccathsu.com.
After all testing was completed, the project is deemed successful. It was initially tested in April, 2014. Since then, the CCAT LAN line has been permanently connected for the Raspberry Pi to permanently link to the internet. As of December, 2014, the results are continuously streamed live on the pvoutput.org website: pvoutput.org user id=31570 and the ccathsu.com/pv website.
The world wide web now streams real-time data displaying the Watts, Volts, Amps, kW, and kWh data. The data is updated onto the pvoutput.org website from the inverter every 5 minutes.
The image below is a screenshot of the PVoutput.org website, up and runnin in April, 2014.
Discussion[edit | edit source]
With 3 design changes, we learned a lot about the University requirements, the CCAT PV system, and how many different software programs and components work together to log energy generation data and monitor that data online.
Lessons Learned[edit | edit source]
- Do your own research, and reach out as much as possible to multiple people from different social layers within the community.
- It would have been nice to purchase an SD card that was pre-installed with the NOOBS software, as it was the same cost of a normal 8GB SD card, and it saved a lot of time and effort.
- To avoid dependency on corporations, go directly to open-source software in the beginning stages of a project such as this. Open forums are surprisingly informative, and the writers and users are eager to help, especially if it means talking about their own projects!
- ITS at HSU requires approval for any wireless router to be installed on campus. If the frequency of the wireless signal is 2.4 Ghz or between 5.0 to 5.7 Ghz, the request will automatically be denied, as it will interfere with the IT system on campus. Most of the wireless routers we found to transfer local data information to a web server were around 433 Mhz, like the Efergy Energy Monitor for example, and are "good to go", according to ITS.
Troubleshooting[edit | edit source]
- This website was used to download the Aurora Monitor software that is displayed through the Raspberry Pi to a local monitor on-site.
- This website was used to follow instructions for the bulk of our project. We did change a few details such as: omitted most of Part 2 and this project used a wired connection instead of Wi-Fi.
- Contact Us!
December, 2014 Updates[edit | edit source]
The pvoutput.org website was not receiving data from the inverter through the Raspberry Pi since May, 2014. To troubleshoot, we did the following:
If the data is not updating to the website, KEEP EVERYTHING PLUGGED INTO THE RASPBERRY PI, especially the SD card.
1. Plug a USB hub into the one open USB slot on the Raspberry Pi. This will be used to connect a keyboard and a mouse to the Raspberry Pi.
2. Plug in a monitor to the open HDMI slot on the Raspberry Pi, using a male-to-male HDMI cable.
3. Turn on the monitor, and the Aurora program should appear automatically at start-up.
4. On the Aurora screen, navigate to SETTINGS -> PVOUTPUT.ORG, and click to open.
5. Check that the API and System ID are filled in. (In our case, they were blank, so we had to fill them in again.)
6. On a separate computer, go to the www.pvoutput.org website, login to the ccathsu account (login info is written on the Raspberry Pi project box in the CCAT downstairs closet plastic bin, labeled "Ribbons")
7. Navigate to SETTINGS
8. At the bottom of the page, you will see a section titled API SETTINGS
9. Copy the API KEY and the SYSTEM ID, and input them into the Aurora PVOUTPUT.ORG screen from Step 4 in the respective fields.
10. Additionally, under the API SETTINGS section from Step 8, click on the HELP button to the right of API ACCESS
11. Scroll down the HELP page for various SERVICE URL's to input into the Aurora PVOUTPUT.ORG screen from Step 4. We used pvoutput.org/service/r2
12. Wait until the sun is shining, and check the pvoutput.org website from a computer to check the data is uploading correctly.
Update October 2016[edit | edit source]
As of October 2016, the last upload from the Raspberry Pi data logger to the pvoutput.org website occurred on March 14, 2016. This is due to technical difficulties with the ethernet port connection. The ethernet port being used did not meet operating standards within the IT department at HSU, and the data logger had to be unplugged. CCAT staff are currently working to have a new, fully functioning ethernet port installed in the CCAT main house, in order to get the data logger up and running again. Up until having to be unplugged, the data logger had been operating successfully and restoring the system will be as simple as plugging the Raspberry Pi device back in to its respective ports.
References[edit | edit source]
- "Data Logger". Wikipedia. 2014. Retrieved 2014-05-10.
- "Learning About Renewable Energy". NREL. 2012. Retrieved 2014-05-10.
- "kW and kWh Explained". Bizee. 2014. Retrieved 2014-05-10.
- "Raspberry Pi". Raspberry Pi Foundation. Retrieved 2014-05-10.
- "What is "The Shell"". Linux. 2014. Retrieved 2014-05-10.
- "Raspberry Pi". Raspberry Pi Foundation, Liz Upton. 2013. Retrieved 2014-05-10.