A new solar panel is born – More Power!

As we learned from our feedback, solar cells must all be of the same size/power capacity to hook them in series properly. In the original solar panel we had snapped our third row of cells in half to get them to fit down the remaining glass and make it easy to solder together. That essentially limits the current output of the entire panel because the full size cells are choked by the limited current that can pass through the snapped cells. So we built another panel using all complete cells. It made it take a little longer wiring the cells together since we couldn’t just run down the straight line of them all, but we found that using tabbing wire for everything worked good. If you bend or fold the tabbing wire to make 90 degree angles it made it pretty easy to solder to the next cell.

Laying out the new solar panel

Wiring up the solar panel

You can see we bent and folded most of the tabbing wires for the last row. It went surprisingly well and the solar panel tested good on the first attempt

Another up close shot of the 3rd row wiring using the tabbing wire

Ready for Epoxy

And finally the solar panel is ready for full testing

Results:

The results are definitive!  The new panel using cells in full (without breaking them in half) generates almost double the power! We tested it Jan 21st 2012 in full sun on a 25F degree day when the sun was almost mid sky.

First Solar Panel Attempt
22 Full & 11 half cells – essentially 27.5 cells worth of surface area
35 Watts (20.7V x 1.7A)

Second Solar Panel Attempt
28 Full Cells
59 Watts! (16.7V x 3.55A)

So now that we know how to build these DIY Solar Panels properly we just repeat the process. Let us know what you think by leaving a comment below.

Ben

Solar has really come down in price since the last time I looked. We wondered if we could start powering our shop off of solar. A good project would be to see if we can build a small scale solar array which would generate power back to the shop and keep it under $1/watt. This would have to include everything from the solar cells to getting usable power out the other end. So we started searching around.

Solar Cell Selection

We went to our favorite buying spot (eBay) and did some looking around to see how much we could find some solar cells for. We found that there are production error cells out there that have some blemishes or chipped edges and you can pick these up for dirt cheap….about 25 cents a watt. So we grabbed 2KG worth (they sell them by weight) which will allow us to make about a 500 watt array. The cells are 3.25″x6″ and are VERY FRAGILE! Who knew these things would break if you breathe on them too hard. They are 0.5 VDC with a maximum 3.6 Ampere output meaning they have a max output of 1.8 watts. Since some of them are chipped or blemished we need to figure we will get less than perfect output from them.

Solar Panel Size and Voltage Choice

We had 9 of the same old windows laying around the shop and they would work really good for about a 50 watt panel each. Wired together they should make a 450 watt array. Really any window panel size will do. Since we are making this on a budget, free windows will definitely help. Making common voltage solar panels helps in wiring them together so they are all working equally. You will lose some voltage from the diode that has to be installed on each panel, but shoot for 12, 18, 24, or 48 volt output panels or arrays.

Electrical Grid Tie Inverter Selection

What exactly is a Grid Tie Inverter and why is that what we want to use? It connects to your existing household AC outlet on on side, and your wind turbine on the other. It converts the DC wind power to an AC pure sine-wave matching the phase of the grid. Since the electricity is being generated on the house side of the power meter, the generated power will actually slow down your power meter or bring it to a stop. But if you are lucky enough to have an old dial turning style power meter AND are generating more power than you are using, it is possible to spin your electrical meter BACKWARDS! I didn’t believe it either until we tried it and it worked.

Grid tie inverters have come down in price and gotten pretty good recently. We expected to pay about 25 cents per watt in this part of the build. They are usually stackable too, which means as your system grows, you can just add another grid tie inverter and they will all just work in tandem.

This is the grid tie inverter we got. It was $138 (27.6 cents per watt) and again we sourced this from eBay.

Assembly and Testing

Here you can see we wired 11 cells in series for each row (5.5VDC). The cells vary voltage from top to bottom and vary current side to side. Meaning if you broke a cell in half 3×3 you would still measure full voltage and have half the amp output. We made 2 full rows, and a third row of cells we broke in half like we just mentioned. Edit: As pointed out by Hack a Day commenter Mark, we effectively limited the current of the entire panel by doing this. This can also lead to overheating in the half panels. At the time, our main concern was getting the voltage high enough for the inverter. For the next set of panels we will be using only full cells. So 3 rows of 5.5VDC = 16.5VDC, then minus the diode voltage drop and we should see around 15-17 Volts Output in operation. We plan on putting 3 panels in series and each series in parallel making a 45-51 Volt (or 48V standard) array.

It makes it easier if you pre-wire a lot of cells before you start. Here you can see we started making piles of them. It really goes quick and is pretty easy to get the hang of once you do a few. The trick is to get a soldering iron that can get hot enough to make it so you don’t have to spend a lot of time on any part of the panel for too long. It took us a few hours to build a 50 watt panel. That time involved will play a part in how big of a system you want to start with.

The cells get wired together easily by running “tabbing” wire down them after applying a flux pen. This make soldering to the solar cells painfully easy. Here you can see all 33 cells wired in series. With each full 3×6 cell theoretically capable of producing 1.8 watts and the halves at 0.9 watts this would be a maximum of 49.5 watts. We’ll round and call this a 50 watt panel.

Once all the cells are wired in place, we tape everything down and give it one last electrical test. This will temporarily hold it in place while we pour a two part polyester resin epoxy over it all to lock it in place.

Here is the epoxy sitting to harden up. When we made this we used old epoxy in cool temps and it came back to bite us. Ours peeled up and ended up taking so long to set up that it leaked through our masking tape and got under some of our panels. We figure we now need fresh epoxy and use about 1/10th of a gallon or less for this size panel. A little goes a long way and really this is just to give it a thin seal and lock everything in place.

Results and Thoughts

Here is the breakdown of our costs:

$125 Solar 3×6 cells creating 500 watts worth
$0 Windows
$50 Gallon of Epoxy
$25~$50 Tabbing wire and diodes
$138 Grid Tie Inverter
————-
So for roughly $375 we have built a Solar Array capable of producing 35-50 watts in it’s current state, with a maximum of 450 watts if we can get full sun. If we did this again we could get the price down even lower, but for about 75 cents per watt it would only take a few years for us to pay this system off.

The next step on this project is to finish the rest of the panels and tie them all in together and see some long term power generation to get an average for a year. Our first panel puts out about 35 watts in full sun, so we have some improvements to do for the next one.

Let the government pay for 30% of your alternative energy generation project!

As we were digging around with this project I want to make you aware that there are some really good government grants (AS IN FREE MONEY!!!) for those that take advantage of them. As always consult with a professional before deciding to spend any money and never take my word for it. But for what it is worth this website has a lot of good information about getting financial assistance in building alternative energy sources. Check out http://www.dsireusa.org/ to find out if your state has any local policies. As of the writing of this, as long as you START your build before the end of 2011 the Feds will pay for 30% of your project if it is under a commercial business. There are LOTS of grants and loan programs out there to help pay for the costs of installing these systems and now seems like a great time to do it. Earth will love you for it too.

Ben

For this build we wanted to make a low cost wind turbine that could give us some good usable power at a low dollar-per-watt budget. We started out designing a 7′ 6″ diameter turbine and used a motor that has a max output of 1,100 Watts. Our goal is to make a wind turbine that costs less than $1 per watt to build and operate. This includes the tower, the wind turbine, and the inverter to tie it in to our electrical system. We were pressed for time on this build because we had very high winds forcasted just two days after we started the build. We needed to come up with something quick to get this thing in the air and see what it could do.

Motor Selection

We chose to use a Weslo 1.5 HP DC electric treadmill motor (model# 104593) which we picked up for $35 off eBay. 1.5 HP equates to a little over 1,100 watts of which, if we can utilize 500 watts I would consider this a successful first build. This particular motor rates at about 50 RPM per Volt which we calculated by spinning the motor and measuring the output. We measured about 40 volts at 2,000 RPM by spinning it in our lathe. We tested different speeds and the output seemed to be pretty linear. That means we would need to spin this motor at approximately 6,000 RPM to reach its max voltage of 115 Volts. 50 RPM-per-volt is a little on the high side and means we will have to spin our motor at least 1,000 RPM to generate the 20 Volts minimum to match our Grid Tie Inverter DC input side. We also need to be aware that we shouldn’t spin it faster than 3,000 RPM to stay under our 60 VDC max input for our Grid-Tie inverter. I would recommend finding a motor with a lower RPM per volt as it will be easier to match the blade speed to the motor shaft speed. We will need to gear this motor up from the blade shaft speed to generate any usable power. For a first build it should suit fine for testing.

Motor shaft speed 1,000 RPM to 3,000 RPM = 20VDC to 60VDC

Electrical Grid Tie Inverter Selection

What exactly is a Grid Tie Inverter and why is that what we want to use? It connects to your existing household AC outlet on on side, and your wind turbine on the other. It converts the DC wind power to an AC pure sine-wave matching the phase of the grid. Since the electricity is being generated on the house side of the power meter, the generated power will actually slow down your power meter or bring it to a stop. But if you are lucky enough to have an old dial turning style power meter AND are generating more power than you are using, it is possible to spin your electrical meter BACKWARDS! I didn’t believe it either until we tried it and it worked.

Grid tie inverters have come down in price and gotten pretty good recently. We expected to pay about 25 cents per watt in this part of the build. They are usually stackable too, which means as your system grows, you can just add another grid tie inverter and they will all just work in tandem.

This is the grid tie inverter we got. It was $138 (27.6 cents per watt) and again we sourced this from eBay.

Blade Design and Manufacturing

A lot of thought went into the blade design at first. We learned more than I thought there was to know about a theories, shapes, and efficiency of a turbine blade. The two most important things you will need to know are are that we can only extract a maximum of 59% of the energy of the wind because of Betz’s Law, and that blade design is realistically the least important thing. All different shapes of blades will work just fine and all be relatively close to extracting the same amount of power. Basically we found that you can spend a lot of time in designing the perfect shaped blades, but in practicality they all work about as good as each other. The diameter of the overall blade size is the most direct and easiest way to extract more power from the wind.

I must thank Hugh Piggott for a wealth of information found at http://www.scoraigwind.co.uk and specifically the blade design notes in PDF format found here: Wind Turbine Blade Design

We looked around at 3 different diameter PVC Schedule 40 pipe. Lowes carried 4″ and 6″, but after doing some research we thought that the diameter was just too small for the scale of turbine we were building. We found a company Ferguson Plumbing that sold us a piece of 8″ diameter Schedule 40 PVC 10′ Long for $90. We cut this 43″ long. Then we sliced the PVC into thirds. Each third makes 2 blades leaving us with 6 blades as shown in the picture above. The blades were positioned and bolted 3 bolts per blade onto a 10″ diameter hub around a 4″ diameter inner circle. That left us with 2 x 43″ long (2 blades) and 4″ across the hub = 90″ (7′ 6″) diameter turbine.

Why 6 blades? With 6 blades we have a lower startup or cut in speed and the turbine doesn’t have to spin as fast because the blades are in more places at once catching the wind versus having only 3 blades. We are hoping that this will give us a more consistent rotational speed. With only 3 blades we would need to have thinner blades with a faster tip speed ratio and the rotor would turn faster in higher winds. So we opted for a thicker blade at the tip which spun slower. There is LOTS of theory and like I said before it all boils down to the same basic idea. You can dismiss a lot of the technical blade jargon and still have good working turbine if you just build the damn thing.

Tower Design and Selection

We found an old welded steel frame with a 2″ Pipe attached to it that worked our perfectly for our “tower”. At first we will just be using this basic tower for testing, but ultimately we need to go higher. How High? Lot’s of people told us lots of different things, and from the research we did it seems to all come down to go high, and build far away from trees. The rule of thumb we liked the best was be at least twice as far away from the trees as they are tall AND be at least 30 feet above the top of the tree canopy for best results. This would be pretty tall for our build so we will use what we have and prove out the mechanical of the turbine itself before we invest in going higher. Once we feel we have a solid working turbine we will get some guy wire and extend the height of our tower to see some real world long term testing.

Assembly and Testing

Most of the parts were made from 6061 aluminum in the CNC part of our shop as can be seen above where we cut out a hub for the blades to attach to. We chose 3 bolts 3/8″ in diameter for each blade to attach to the hub. The tail fin is a piece of thin stainless sheet steel we had and the tail body is 1″ aluminum square tubing that we cut to length.

Since we know that the average Joe doesn’t have access to a machine shop, we may decide to offer some of these parts for sale once we finalize a design that we like. Email us crew@ttjcrew.com if you are interested.

The blade shaft spins inside 2 snowmobile idler wheel bearings. Another 2 of these bearings are the pivot point on the tower. This was a first test of holding it together and it seemed to work for a little while. We are currently developing a more secure way to hold everything together with aluminum on the CNC machine. For a first run down test of a wind turbine we just wanted to prove that it is possible, and see how much wind we could capture. Once all these things are figured we can make a more stable and secure turbine which will be ready for taller heights.

On our first tests with the motor’s shaft connected directly to the blade hub we found that in very high wind (20+MPH) we were only spinning about 150 RPM and there was no problem spinning the motor even when it’s wires were shorted creating the most load to the blades. So we should easily be able to gear this up to be able to spin the motor faster. For now we will be testing the turbine in high winds with a straight 1:1 gear ratio.

Results and Thoughts

Here is the breakdown of our costs:

~$50 for Aluminum material
$35 Motor
~$35 PVC for Blades (75 cents per inch and we used 43 inches of the tube)
$10 Paint
$0 Old snowmobile idler wheels and bearings
$0 Stainless sheet steel for tail fin
$0 old tower
$138 Grid Tie Inverter
————-
So for roughly $275 we have built a wind turbine capable of producing about 10 watts in it’s current state, a maximum of 1,100 watts if we can spin the motor at it’s max RPM, and about 500 watts if we can spin it at a reasonable amount.

The next logical step for this turbine is to gear this motor up and get this turbine higher up in the air. That will be coming in part 2 of this build, but for now we wanted to show you the basics of what we did and hopefully learn something in the process. Enjoy and oh teah….don’t touch the blades while they are spinning, they have an incredible amount of built up energy and they will usually win.

Ben

In this episode the Tech Junkies tackle cheap alternative energy systems. Ben and Eric guide you through the process of assembling your own solar panels for around 50 cents per watt and how to construct a low cost wind turbine. This is only the beginning…more to come in part two!

More details here:
Wind Turbine Build
Solar Panel Build

Almost all of this information and work was put together by Micah Elizabeth Scott over at scanlime.org. We simply experimented with the technology and found what works best for us.

Hand Wrapped Coil Antenna and ATtiny85

At this point, if you haven’t already watched the RFID episode of TTJ, you’ll want to check that out now.

As shown in the video, using the code provided at scanlime.org, you can easily create a clone of an EM4102 (Parallax) RFID tag or a standard HID 125kHz tag.

Although it is shown on scanlime.org that you can clone an RFID tag simply using an AVR ATtiny85 and an inductor, we had a lot of trouble getting this to work well. Our first attempt at this used only the ATtiny85 and the inductor and we were only able to get Parallax readers at very close proximity to detect the spoofed tag. It definitely works, but we wanted to build something that would work from a bit greater range.

For some experimentation, we went to our local RadioShack and found this pack of magnet wire for $7.39. Our plan was to hand wrap a coil:

Magnet Wire from RadioShack

After making a few different coils, we found that the 30 gauge wire and approximately 100-125 wraps gave us the best results. For actually wrapping the coils, we dug out an old perf board that another project had been built on. We found an empty space on the board that was 3″ x 1.5″ and then placed four screws into the board. This then allowed us to easily wrap the coil and simply remove the screws after wrapping. We went with this size because it was close to the existing RFID tag and we were curious how well it would work.

Old Perf Board

Wrapping the Coil

As for programming the ATtiny85, you’ll need some sort of programming cable and something that can act as an ISP programmer (Arduino can do this). For our purposes and in line with the previous RFID post, we used a USBtinyISP. We found a small perf board and wired up the 6 pin header to an 8 pin socket. Below are the pinouts for the ISP programming header and the ATtiny85:

ISP Header (USBtinyISP connects through this)

ATtiny85 Pinout

Alright, time to get down to programming. If you haven’t already “acquired” your target tag that you’ll be cloning, you’ll need to do this now. As stated in the video, you can get a Parallax EM4102 tag key by using a Parallax serial RFID reader and anything that can talk TTL serial. For our example, we used an FTDI cable. The easiest way to get data from the FTDI cable is to simply load up the Arduino IDE and select serial monitor. Set your baud rate to 2400 and wave the tag in front of the reader. Write down this value and save it for when we load up the code.

As for getting an HID tag key, you’ll need an HID reader (we prefer the Wiegand version of the MiniProx). We then used an Arduino to connect to our HID MiniProx and get the tag value. Below is a download of our Arduino sketch that explains how to hook up the reader and get the usable site code and unique key that you’ll be entering into the program to be loaded onto the ATtiny85.

HIDReader_v01.zip

The rest of the code loading process is covered in detail in the video. The zip file below contains the source code from scanlime.org, the Makefile, and a text file with the avrdude commands you’ll use to flash the code to the chip. Keep in mind, you will not be able to reprogram the chip again after executing these commands unless you have an external clock source.

rfid_cloning_v01.zip

If you don’t already have the AVR toolchain for compiling and writing code to AVRs, make sure to grab the software package for your OS:

OS X: CrossPack
Windows: WinAVR + USBtinyISP Driver
Linux: avr-gcc/avrdude from your package manager

Once you’ve loaded up your ATtiny85 with code, pop the AVR out of the socket and wire up your coil to the clock pins (PB3 and PB4 on the Pinout above). For our purposes, it doesn’t matter which wire is connected to which pin. At this point you may wish to neatly pack your ATtiny85 in with the coil and wrap it in electrical tape or use some sort of rubberized coating. Now go try out your new spoofed tag!

Wrapped RFID Tag Clone

RFID Lock PCB

Prototype RFID Lock

In the first half of the RFID TTJ episode, I walk through the process of building this design and loading the board up with code. I’d highly recommend watching the video as it will give a good explanation of how everything is put together and goes further in detail than this post.

Now then, let’s get onto the details. The first thing you’re going to need for this build is the actual circuit board that everything gets connected and soldered to. I put together the minimal board design using Eagle.

You can download all of the Eagle files and generated board files here:

rfid_lock_eagle_v01.zip

For actually having the board manufactured and sent to you, I highly recommend BatchPCB. Although you may have to wait upwards of a month to get the finished board in the mail, it’ll only cost you $10-$15 including shipping. I’m hoping to put together kits in the near future to make this step a bit easier.

As explained in the video, you have two options for RFID readers. You can either use an HID reader (common among universities and corporate environments) or a Parallax reader that uses the EM4102 card standard. As I was issued an HID card to gain access to the various buildings on campus, I decided to go with an HID reader for my purposes. The main issue is that HID readers are significantly more expensive (~$150 retail). You can find them cheaper on eBay, so if you really want to be able to read HID cards then I would suggest that route. If you don’t already have any RFID cards, the Parallax reader (~$40 retail) is most likely the way to go. This circuit board will work with both options – you’ll just be connecting the readers to different pins and loading different code.

As for all of the parts you’ll need for this build, I’ve compiled the list below:

1. (1) Custom Circuit Board (~$15 ordered from BatchPCB)
2. (2) 47uF Capacitors [Mouser.com Part #: 647-UHE1E470MDD] ($0.30)
3. (1) 5V Regulator [Mouser.com Part #: 511-L4931CZ50-AP] ($1.15)
4. (1) 20 Pin Socket [Mouser.com Part #: 571-1-390261-6] ($0.21)
5. (1) ATtiny2313 [Mouser.com Part #: 556-ATTINY2313-20PU] ($1.91)
6. (1) Header Pins [Mouser.com Part #: 649-68000-416HLF] ($0.40)
7. (3) 3 conductor screw terminals [Sparkfun.com Part #: PRT-08235] ($4.50)
8. (1) 2 conductor screw terminals [Sparkfun.com Part #: PRT-08084] ($1.25)
9. (1) Magnet [Sparkfun.com Part #: COM-08914] ($1.00)
10. (1) Reed switch [Sparkfun.com Part #: COM-10601] ($2.00)
11. (1) Dome Button [Sparkfun.com Part #: COM-09181] ($10.00)
12. (1) 6V Power Supply [http://www.radioshack.com/product/index.jsp?productId=2049709] ($1.50)
13. (1) Hitec HS-311 Servo [http://www.amazon.com/Hitec-RCD-Inc-31311-Standard/dp/B0006O3WVE] ($10.00)
14. (1) Parallax RFID Reader + Sampler Kit
[http://www.parallax.com/StoreSearchResults/tabid/768/List/0/SortField/4/ProductID/441/Default.aspx] ($43.00)
15. Door mounting hardware (foam, double sided tape, zip ties) [Home Depot] ($5.00)

Total Cost: ~$100

As far as assembly of the board goes, it’s pretty straight forward. Solder in the 20 pin socket (make sure to line up the notch), snap the screw terminals together and solder them in place, solder in the 2 capacitors, the voltage regulator, and then break apart the header strip to populate the programming header and servo header. Pop your ATtiny2313 microcontroller into the socket (making sure to match up the notch) and you’re ready to program.

Assembled Circuit Board

At this point you’ll need something that is capable of writing code to an AVR microcontroller. If you have an Arduino, it’s possible to use that. For this build I ended up using a USBtinyISP Kit. You can pick one up from that link for $22 and it’ll give you a simple interface between USB and the ISP programming header (which will also come in handy for the second half of the episode). In the video I demonstrate how to write the code to the RFID board. The 3 software packages I mention correspond to which operating system you’re using:

OS X: CrossPack
Windows: WinAVR + USBtinyISP Driver
Linux: avr-gcc/avrdude from your package manager

You can download the source code for either a Parallax reader or HID reader from here:

parallax_rfid_lock_v01.zip
hid_rfid_lock_v01.zip

As an additional step that wasn’t included in the video, you’ll need to get the code off your HID tag if you’re using an HID reader. I’ve put together an extremely simply Arduino sketch included in the HID source code zip file above that will allow you to get your tag code. All of the explanation should be included in the comments at the top of the file. If you’ve still got questions after checking out the sketch, feel free to post them in the comments.

Now that your board is assembled and the code is loaded up, all you have to do is screw in your connections and mount the reader to your door. Once again I’m going to prod you to watch the video because there’s a pretty good explanation of how all the components wire together (with the Parallax reader). I also talk about how the HID MiniProx reader gets connected. In summary, there are 4 wires you’ll need to connect. You can wire the Vcc and Gnd directly to the 6v adapter input and then just connect your Data 0 and Data 1 wires to the corresponding IN0 and IN1 connections on the PCB.

Fully Wired Board

As far as mounting goes, each door is different. For the most part I’ve been able to get away with using zip ties to hold the servo onto the actual deadbolt handle and then use lightweight foam and double sided tape to hold the servo in place. Stick your magnet onto the door frame and mount your reed switch on the edge of the door so it detects the door opening/closing. Everything else can be mounted using double sided tape.

Servo Lock Mounting

Foam used for mounting the Servo

There you have it! Your very own RFID door lock that requires no modifications to the door. I know there was a lot covered in this post, so I will continue to update it with information I feel is useful. Although we love using RFID for a lot of our projects, make sure to check out our post on spoofing RFID tags to understand how they can be exploited. Hit up the comments below if you’ve got any questions and I’ll do my best to answer them and get your RFID door lock up and running!

In this episode of TTJ we show you how to build your very own RFID door lock using our open source plans. We also take it to the next step and show you how RFID tags can be read and duplicated.

Make sure to get all the details and source code in the episode writeups:

RFID Door Lock
RFID Tag Cloning

In this shorter half-episode, we recap some of the cool things featured at Maker Faire Detroit 2011 and a bit more video of the Mantis robot driving around.

Drive for Innovation – All hail Mantis, the open-source robot

We’ve got more videos/projects on the way, so stay tuned!