Underlying Ideas:
The initial idea is to power a our internet receiving radio, a small computer that runs our internet connection, and the wifi access point that gets the internet from the roof to our rooms. Because our internet is so expensive (about $250/mo for a connection twice the speed of dial-up) we need to make good use of it. One of the ways we can do that is by having the connection download all of our virus updates and big files at night when no one is using it so that we can have the whole 128kb connection available for people to use during the day. However, since the generator shuts off at 11pm, we need a way to keep the equipment running through the night.

The initial plan of powering the internet equipment will serve as a test case. If it's successfull, we can quickly think of many many other uses for the technology where we are. Friends of ours who work out in the bush where even fuel for generators is hard to come by, for example, would benefit greatly from even a small amount of power to run lights and such. Remote villages could use the power to help with communications, irrigation, schools, etc. The possibilities are endless!

So, the use for such a system is great, but I still need to get from knowing nothing to building a system that can be carried over in airline luggage, assembled in Africa with only rudimentary hand tools, and then be so rugged that it will work with little (or better, no) maintenance for a long time, in hot sun and pouring rain. Sounds like a really fun challenge, huh?

On to the prototypes!

I've focused my initial reading on a design called a "Vertical axis wind turbine" which is abbreviated VAWT. The names may sound odd, but when you remember that a normal windmill is actually a "horizontal axis wind turbine" device, then it makes a little more sense. So a normal windmill spins around perpendicular to (into) the wind, but a VAWT spins on the same plane as the wind.

VAWT designs are interesting to my application in Africa for a couple of reasons:

  • They can start spinning in slower wind - this is the most important concern because the wind we have is very gentle.
  • They can be used lower to the ground since the length of their blades spreads out across the ground instead of toward it like a normal windmill. For us, this means we don't need giant, expensive towers.
  • They don't care what direction the wind is coming from. A traditional windmill can be made to rotate with changing wind directions, but that's one more moving part that can break.
  • They're simpler and hopefully a little more resilient to inclement weather.
  • We don't care about their main disadvantage, that they can't generate as much energy. We don't need the Megawatts of power that commercial giant traditional windmills can generate.

With the above points in mind, I built the first small model of a design I found here. Go to the small cardboard Lenz VAWT model page to see info on that. Also, don't forget to check out information about the follow up production sized model once I get that posted!

“Slatmill” VAWT - 18" Model
The “slatmill” was constructed after the cardboard Lenz model. The original slatmill was three feet tall and had a diameter of about 18". It was built from thin Schedule 30 PVC pipe cut in quarters length-wise for the slats, fiberglass reinforced plastic sheeting for the top and bottom section, and hardwood/aluminum to attach the top and bottom.

The original proof of concept model was in operation about two months. It demonstrated very good low wind performance, the design shows promise! It flew apart during a strong wind gust while we were away for training in Denver. The model's week point was that the slats were attached by hot glue to thin wooden dowels. The wind blew the slats right off. The slats weren't attached in a very strong way because for the model I wanted to be able to adjust their angle. Now that I have a good idea of the best angle, I'll attach the blades on the bigger one much better.

The model never generated electricity, but I think that this was related more to a bad drive system than the model's aerodynamics. The model was attached to a 5/8” shaft on a pair of bearings, and the shaft was using pulleys and a belt to drive the generator. I have since decided that the action of bending the belt (a V belt, like what you'd find on a lawn mower) around the pulley in order to turn it was wasting too much energy because the belt is so stiff. I'm switching to chain and sprockets for the next model and hopefully that will be smoother so more of the power can make it to the generator. (You might wonder why the windmill isn't attached directly to the generator, it's because the speed is so slow it won't generate any electricity – I need to take advantage of the extra torque the models are generating by gearing them up to trade that extra torque for more RPM's!)

The “full” size slatmill is now under construction. It will clock it at about six feet tall and three feet in diameter. It should be able to grab some serious wind!

PVC Pipe HAWT Turbine - 30" Model (Coming Soon) - 10' Model (Coming Soon)
While I was scavenging for pipe for the slatmill project, I found a scrap piece of big 6” diameter pipe. I decided to give a traditional horizontal wind turbine a try to see what it could do, and also to have a reference to compare the other types against. So far, the traditional kind is looking pretty good! (Of course, I'm testing them in windy Michigan, not slow winded West Africa!)

Before trying to build a bigger PVC turbine, I decided to use 3" Schedule 30 pipe to make a smaller one. The plans I found for making blades from PVC pipes were something I hadn't tried before, so I figured a small version would be an easier start. That turned out to be a good idea, because the big one was finicky enough that I might have given up on the idea without the success of the small one to push me on!

The basic idea behind this project is that there are many ways to make propeller blades, but most of them (fiberglass molds or hand-carved wood) are pretty labor intensive. The goal with this project was to produce a simple, quick set of basic blades that might not be the most efficient, but would be easy to make. The process it simply to cut a length of pipe lengthwise in quarters and then cut a diagonal off to make the blade thinner on the tip than the base.

The first model used thin 3” Schedule 30 pipe and the blades were about 15” long. I cut a wooden hub out of plywood and hardwood, and friction fit it to the pump motor that I've been using to test the different turbines (for info on why this is a really bad way to test turbines, see the Axial Alternator section below). I held it up by hand in a pretty strong wind and it started spinning like crazy! This was the first wind project that yielded measurable electricity! Because of the way a pump motor is wound, it only gave about a quarter of a volt, but I was able to measure it.

I hadn't planned to spend much more time on regular propeller type turbines, but after how simple and effective the small one was, I had to make a bigger one! I had a six foot long section of heavy Schedule 40 6” diameter pipe from the trash pit in back and decided to make it my victim. I followed the same procedure, painted them with spraypaint, and bolted them to the steel hub I'd been using for the slatmill project. It wouldn't spin even in strong wind! The blades were made of so much heavier plastic and were so much longer than the small model, that their imbalance meant the heavy blade just fell to the bottom. I spun it by hand and it would do about two turns before the heavy blade settled back down to the bottom.

I pulled the blades back off of the hub, weighed them, and found one to be a few ounces heavier (they are each several pounds) which seemed like an insignificant amount! I ground a little off of a couple of them to get the same weight, remounted them to the hub, put the turbine back on the test stand, and still nothing! The reweighting helped a little, now they would spin a four or five revolutions if turned by hand before settling in to the annoying heavy-blade-on-bottom stillness. In the end, I decided that the angle of one of the blades was slightly different because of variance in the drill pattern in my homemade hub. That little angle difference made a big weight difference in blades that are so long!

I decided that to take all of these factors into account, the most accurate way to balance the assembly was to bolt the blades onto the hub, and then attach a string to the center of the hub. By holding up the string and watching which blade fell to the ground and which rose, I could find the heavy side of the whole assembly. I did, ground a little more off of the heavy one, tested again, ground a little more, etc. After a few hours, the assembly was pretty well balanced.

I reinstalled the turbine and it's been turning happily for several weeks! I'm pretty impressed, and it's a lot of fun (for me) to watch. I think it's one of those things where if it wasn't your creation you wouldn't get it, but for me it's pretty cool to see!

Since this is a decent sized turbine (10' in diameter), I thought it was time to find out if it could generate some electricity! I attached a pulley to the shaft and used a belt to drive the same (horrible) pump motor. This is an aweful way to arrange a wind turbine. See the section below about the Axial Alternator for an explanation of why that is and what I think the solution is going to be.

All in all, the PVC blades project is a great success! I think if my generation system were better and the turbine was on a rotating windmilll platform at a proper height, it could generate a good amount of power, given my target application!

Logging Anemometer - Soup Spoon Version - Weatherproof "Production" Version (Coming Soon) - Electronics
The more I read about wind turbines and alternators, the more I realized how important wind speed is. Duh... but really how much of a difference it makes. (O-B-V-I-O-U-S-!) No really, clearly you need wind to make a windmill turn, but did you know that the power you can get from a windmill is not linear with the wind speed? A 20mph wind doesn't give you twice as much power as a 10mph wind, in most designs it's more on the order of 5x! So obviously if we're going to install windmills efficiently in Africa, it's going to be necessary to be precise in assessing the wind at given sites to choose the best locations - and not just the wind at a given time, but the cumulative amount of wind at a site over time.

Of course my first thought after googling “wind speed detector” to find that it was called an anemometer was “build it!” Then (as I normally try to do) I thought “this is going to take a long time, how much would it cost to buy it?” The answer is, the “no work” model is several hundred dollars. Davis Instruments makes a (I'm sure awesome) unit that measures wind speed and direction. It's only about $120, but the problem is that in order to get a reading from it, you have to hook it up to their weather station electronics, which are a few hundred more. That's at the outside of what I would be willing to spend, but is way outside when I remembered that I probably need three or four of these to be able to test several sites at the same time.

I looked around the renewable energy websites and saw some very clever designs which require a little bit of work. This one is particularly clever because it uses a bicycle speedometer which you can pick up from Walmart for $15-20. The part that actually catches the wind, the cups, can be bought for another $20 from otherpower.com. It's a nice, inexpensive solution that works great for knowing the current windspeed. The bike speedometer even has the ability to automatically remember the high and low speed. Clever!

However, neither of these solutions are suitable to do what I want them to do. Since I don't have a whole month to sit and watch the speed on the gauge to evaluate a site, and in most cases won't be within ten miles of the site during the time I want to evaluate it anyway, I needed a solution that can store the information for me. The bicycle speedometer does have the ability to calculate the average speed, which would be a good starting point except that it ignores the times when there is zero wind. Even if it was smart enough to compensate for that, an average would be of only marginal use anyway. As we remembered earlier, when comparing two sites, one with a constant 10mph wind will give less power than another with a 20mph wind half of the time.

So, I embarked on a bit of research to figure out what I needed to do in order to make a “logging” anemometer, one which would record wind speeds for a time interval of several days so that I can use the data later to analyze the site. Check the links above for the project status, but at this point I'm very optimistic that I'll have a bulletproof solution that will log wind speeds six times a minute for a whole month, for around $100 per device. This information will hopefully allow me to correlate data from a windmill design (so many watts at 5mph, so many at 10mph, etc) with a site and be able to calculate the exact amount of power that design would have given on that site during that month. Armed with that information, we'll be able to make the most efficient decision for a given project whether solar or wind would be cheapest for the amount of power the project needs.

I'm geeked.

Axial Alternator

The homebuilt axial alternator pioneered by a legend in the field of homemade alternative energy projects, Hugh Piggott, has been very popular as a generator for windmills for years. As my ideas have continued to evolve, his design is looking more and more useful. I'll be building one when we get home to Michigan in February, so stay tuned!