Super sketchy, totally DIY, low budget wind generation experiment

harrisonpatm

Member
Joined
Jan 5, 2022
Messages
392
I got on the forum for all of it's awesome resources having to do with using secondhand lithium cells, and I know that nearly all of the readers are solar users. Solar is great. And yet, building wind generation just had such an appeal to me for the last few months. I've learned enough (still so much more to learn) to know that at some point in the next couple of years, I'm sure I'll be breaking down and finding some secondhand solar panels to stick on my house and get some use out of these thousands of 18650's you folks are helping me process. In the meantime... behold the jankiest homemade wind power setup you've ever seen!

Goals for the project were to spend as little money as possible and do as much construction from scratch as I could. While the ultimate goal of having a few wind turbines down the road, with enough output to fully supply power to my house, would be awesome, it's less than likely. Main goal was to learn, secondary goal was to have a small setup to charge my phone, kid's toys, tablet, small devices, some lighting, ect. There are also local restrictions in place in my area from installing anything significant (including solar). And I'm not quite there yet! But I have made a lot of self-taught progress and I would love some feedback and suggestions as I move forward.

Turbine #1
IMG_20220313_133004.jpg

This is a 5-bladed Lenz style turbine using a modified hoverboard wheel as an alternator. A scrap bicycle wheel is mounted centered on top of the alternator, with the 5 "blades" mounted evenly on the rim. Blades are made of 4in PVC, halved, mounted askew using aluminum L-angle, and flushed out with scrap house siding to catch the wind on the drag side. Galvanized wire on top to keep them from wobbling. At 20mph winds, I get about 80-120 rpm. Benefits is that I don't need a yaw mechanism, it catches the wind from any direction. There's also a relatively significant amount of weight from the vanes on the exterior of the wheel, which gives it a nice flywheel effect. It's efficient at catching even small breezes, and once it gets going, the flywheel keeps it moving. Great torque. Downsides are that vertical axis turbines are not as efficient at catching windpower, and despite having great torque, it never gets terribly fast even in high winds, so therefore low voltage. On my to do list is to remount it totally differently; I want the turbine's spin to gear up to the alternator, instead of direct drive, then I should be able to get more voltage. Even 1:2 or 1:3 would be awesome.

Turbine #2. This one is still in prototype testing stage, lots of improvements, but it's generating as is, so I'm including it
IMG_20220313_133304.jpg

Another hoverboard wheel (so cheap and easy to find, if not free), this time a more traditional horizontal axis. Blades are once again PVC, this time 1/3 lengths, shaped to give a leading and trailing edge. The mounting plate and bearing is repurposed from an old industrial fan that I found. Yaw mechanism is a thrust needle bearing, no slip ring yet, cuz I'm still just testing. Which is also why it's not mounted higher up yet. The tail, as I've learned, also needs to be bigger, and on a longer boom. So most of it's time it's not actually facing wind, I either move it myself or the wind is strong enough to push it around, until I get around to remaking the tail fin and boom. But oh boy, when it does get going, it barely needs any wind at all to get ripping. This is in 12-15mph winds:
IMG_20220311_181946.jpg
It's no wonder why 90% of all the giant 300ft wind turbines all look the same. It's a very effective design, even on a small scale with little precision.

And now for some good laughs: in moderate 12-20mph winds, turbine #1 ouputs about .5-1amp at 3-6 volts. Turbine #2, 1-1.5amps at 8-12 volts. Barely anything! Wohoo! So what's the point, amiright?

Point is it was all junk, scrap, used parts, spare time, and a ton of learning experience. I've spent less than $200 on everything all together, including the electronics introduced next, and I'm getting a little bit of usable power out of it. The learning can easily be applied to larger control systems that I can purchase later with solar or higher-output turbines. Plus, I'm simultaneously collecting, processing, testing used 18650's, on which I'm spending exactly $0, for an eventual home powerwall. These projects are giving me something to do with my substandard cells, and I'm practicing using them, so I can learn from my current mistakes before I build the big powerwall.

Onto the charge control and storage. Since I'm only getting like 5-15 watts at a time, no Amazon or Ebay charge controller is able to stoop that low to actually get me to store any of this energy. I have to build it myself using off the shelf components, doing things with them they are very much not designed to do. Since I don't have a lot of electronics experience, this has been very fun learning experience, and the setup is always changing when I want to try new ideas. If you like nice tidy organized wires, please scroll down and skip the next picture.
IMG_20220313_133830.jpg

Battery is a 1S24P made of my mediocre cells, 1800-2000mAh. All in parallel, I get about 45-48 usable amp hours. Since I'm just using this for charging 5V devices for now, this a relatively huge bank. This is so that if I have multiple non-windy days in a row, the battery is never actually in much danger of draining dry. I have a small voltmeter connected to it for SOC, which I can turn on/off with a switch so that it's not a parasitic drain. It doesn't have it's own BMS, persay, but the output is done through a TP4056 (introduced below), and I don't have any series connections to worry about getting out of balance. Plus I'm playing with the setup and monitoring it constantly every day, so I'm not worried about it getting under- or over-voltage.

The "charge controller" is on the small wooden board on the right side of the picture. Since turbine #1 never gets high voltage, what I've done is rectified the 3-phase output and connected that to a small boost converter, MT3608, calibrated to 5.0V output. When the turbine spins up to about 1.7V, this provides enough input voltage to boost to 5V. The 5V output goes to a TP4056, which goes to the battery and charges it with whatever scavenged milliamps it was able to get from the wind, The TP is rated for 1 amp, and I can guarantee the turbine rarely actually gives it that much. For output, on the same "circuitboard," I have a cheap USB boost module taking the battery voltage and boosting it to 5V to charge phones, devices, ect. I have a 2A dual USB charging module on order, and when it arrives, I can put everything in a decent box to tidy up the wiry mess.

Turbine #2 was much easier to get power into my 1S pack, it spins a lot faster in relatively low winds and easily gets up to 8-10V. Yet still not high enough for even the cheapest Amazon wind charge controller to handle, which usually require 12-24v and would be a $30 waste of money. Instead, I've connected the rectified output to a buck module, a cheap 5A board for $1.52, and set the output to 4.18V. I've tested this in a controlled setting with a power supply to charge 18650's normally, and can confirm that as long as it's set correctly, there's never a chance of battery overvoltage. So what happens is that turbine #2 spins up, and once it gets above the battery's current voltage, any current generated goes into the battery, regulated to 4.18V. 8, 10, 12V coming out of the turbine, only 4.18V into the battery. High turbine voltage is essentially getting turned into higher amps to maintain 4.18V (or at least that's as much as I've been able to learn. Someone please correct me if I'm worng!) The buck converter is rated for about 30V input, and this current turbine stays way under that. I can also limit current with this board, and have set it to 2A, but again, I never get that high.

Exact power generation, in watt hours, is hard to calculate, mostly because cheap power meters have a minimum input of 6V, but at the same time, I don't particularly care. I get red LED's when charging is occuring, I charge mine and my wife's phone and my kids' toys every day, and the battery gets replenished 24hrs a day, sporadically. It moves between 3.5 and 4V. So the system is working. It's horribly inefficient, and a mess, and a constant project, but it's also fun and incredibly educational, and all my materiels, excluding cheap electronics, are used, scrap, secondhand or junk, extremely low cost. I also know that this method of charging is essentially trickle charging, and not particularly good for the health of the battery, but it's my low quality salvaged secondhand cells that were about to be thrown away by a computer repair shop anyway, so that also doesn't really matter in this situation. I'm saving the good cells for bigger projects.

So that's about it for now. I would love some feedback. It's a totally unique project, so it's been quite hard to find suggestions on the internet, because who in their right mind would spend this much time and effort on something that only gives you a few watts? This is going to be a constant work in progress. I have 2 more turbines planned. I'm always changing my mind or trying out a new setup with the charge controller or the blades or the placement in my yard. And solar is definitely a couple years down the line.

Thanks for reading!
 

Attachments

  • IMG_20220306_092428.jpg
    IMG_20220306_092428.jpg
    2 MB · Views: 105
You might like reading up at this site:
& particularly this project:
with this type of alternator:


note his comments about using any gearing = losses due to friction.
direct drive into a permanent magnet type alternator will be most efficient.

re electronics to get the most power out of your windmill, maybe have a look at "maximiser" circuits.
The idea is they don't load the turbine until it's producing some useful voltage. This way, it can come up to an efficient speed.
Eg modify this circuit to lower voltages:
Instead of a motor output, use the FET or 311 output to control an enable line (or similar) for a charger circuit downstream.
Maybe use a TL431 instead of a 6.2V zener + adjust other resistors accordingly.

Maybe use high power schottky type diodes for less rectifier losses. At these low power levels, 3A ones should be fine, eg SB350, 3A, 50V, ~$1 each.

Sounds like you'll need speed limiters soon on #2! Calming your neighbors after bits fly through a window might be interesting!
 
You might like reading up at this site:
& particularly this project:
with this type of alternator:


note his comments about using any gearing = losses due to friction.
direct drive into a permanent magnet type alternator will be most efficient.

re electronics to get the most power out of your windmill, maybe have a look at "maximiser" circuits.
The idea is they don't load the turbine until it's producing some useful voltage. This way, it can come up to an efficient speed.
Eg modify this circuit to lower voltages:
Instead of a motor output, use the FET or 311 output to control an enable line (or similar) for a charger circuit downstream.
Maybe use a TL431 instead of a 6.2V zener + adjust other resistors accordingly.

Maybe use high power schottky type diodes for less rectifier losses. At these low power levels, 3A ones should be fine, eg SB350, 3A, 50V, ~$1 each.

Sounds like you'll need speed limiters soon on #2! Calming your neighbors after bits fly through a window might be interesting!
Thanks for the suggestions!

That website you linked was already a great resource for me on my build. Lots of great information.

Thanks for the suggestion on maximizer circuits, I haven't seen those yet and will do more. As for building my own... not there yet! Still out of my depth, but I'm learning a ton and I'll get there.

As a general observation, using buck conversion instead of boost to regulate power seems to be more efficient. I can see that with a buck, load isn't actually applied to the motor until it gets above the set voltage. So it's already kinda doing what a maximizer would do. Trouble is catching the wind to keep it above that speed! Better turbines, bigger blades.
 
Based on your pictures, you are in a quite turbulent area. The VAWT would definitely be the better one to go with over the HAWT.
If you can get it a little further away from obstructions, maybe even raise it up a bit higher, that would help quite a bit.
VAWT's really do well in turbulent/random wind. But it still needs to be flowing 'horizontally' across the ground, not coming down off the tree line or buildings.
 
Based on your pictures, you are in a quite turbulent area. The VAWT would definitely be the better one to go with over the HAWT.
If you can get it a little further away from obstructions, maybe even raise it up a bit higher, that would help quite a bit.
VAWT's really do well in turbulent/random wind. But it still needs to be flowing 'horizontally' across the ground, not coming down off the tree line or buildings.
Indeed. Which is why I'm aiming to gear up the VAWT at some point. Despite increased torque requirements, even 2:1 is gonna give me a nice bit of constant supply.

Just got back from my mother in law, who has a 30ft steel antenna tower that she's been looking to offload. Dibs!...as long as I can clear it with the city.
 
Just got back from my mother in law, who has a 30ft steel antenna tower that she's been looking to offload. Dibs!
Before asking them straight up, find out what the ceiling is first. You can probably ask them anonymously, aka w/o giving them your address/phone#. Basically it's so that, let's say the ceiling is 50Ft, the tower is 30ft. Technically it's within the limits. However, there's probably bound to be someone from code enforcement wanting to stop by and see what else you may have done that they claim is not in accordance.

Despite increased torque requirements, even 2:1
Personally, I would recommend redoing the generator. I'm not sure exactly what you're using as the generator (you have quite a bit of text to read through and I don't remember from the initial read :p ), but I'm going to assume it is an existing motor of some sort.
It's very easy to build, I think it's called, a flywheel generator. Instead of a classic design where the Rotor spins inside or around the Stator, they are pancake designed, like this:
1647294840021.png


The magnets have a longer time across the coils creating more current flow instead of only being on the edge of the coil. The magnets are usually arranged with 1 less the number of coils (if i remember correctly; maybe it's 2 less), and they spin on top of the coils. The trick is to get the space between the magnets and coils as small as possible.

This would allow the VAWT to have low resistance from weight on the Rotor and allow them to spin faster. You could probably also redesign your blades to make them lighter, as well. If you have access to a 3d printer, you could probably print the ends to save weight
1647295089392.png

Those actually don't have to be very heavy duty as they don't do much other than to make sure the air flows around the foil instead of creating a vortex on the ends causing drag.
 
Before asking them straight up, find out what the ceiling is first. You can probably ask them anonymously, aka w/o giving them your address/phone#. Basically it's so that, let's say the ceiling is 50Ft, the tower is 30ft. Technically it's within the limits. However, there's probably bound to be someone from code enforcement wanting to stop by and see what else you may have done that they claim is not in accordance.
Excellent idea: how much can I get away with under the law!

Personally, I would recommend redoing the generator. I'm not sure exactly what you're using as the generator (you have quite a bit of text to read through and I don't remember from the initial read :p ),
Hoverboard motor
It's very easy to build, I think it's called, a flywheel generator.
Also called a magnetic flux generator.
The magnets have a longer time across the coils creating more current flow instead of only being on the edge of the coil. The magnets are usually arranged with 1 less the number of coils (if i remember correctly; maybe it's 2 less), and they spin on top of the coils. The trick is to get the space between the magnets and coils as small as possible.
In this case, you'll need magnets both above and below the coils to extend the magnetic flux. Definitely doable, and in fact this design of motor is something I have in progress for a flywheel-driven bike generator. Another option, if I go with gearing up the drive, is to simply use a much smaller motor, more in the 50-100w range; it'll have significantly less resistance.
You could probably also redesign your blades to make them lighter, as well. If you have access to a 3d printer, you could probably print the ends to save weight
What I'm currently using in that spot is aluminium flashing, incredibly lightweight, I doubt it's heavier than plastic. Unfortunately I don't have access to a 3d printer, and certainly not on my budget.

I'm torn on reducing the weight of the vanes themselves. Want to do it for sure, on my next one, which is going to be 3-bladed Lenz instead of 5. And yet, I like the flywheel effect the weight around the sides is having. Less responsive to varied gusts, but keeps good momentum once it starts.

As I'm learning it's all a give and take in what you're able to get out of the wind. Omni Calculator has a great tool that calculates available windspeed given the dimensions of a turbine and windspeed, so it's nice to be able to have a "goal" of how many watts to shoot for.
 
With aluminium parts, metal fatigue won't be your friend!
With the hover board generator, those seem to like higher rpms. You might like to look at similar devices with larger rotors, eg the f & p washing machine motor. These are also wound for higher voltage (mains) so likely to get you useful output sooner?
 
With aluminium parts, metal fatigue won't be your friend!
With the hover board generator, those seem to like higher rpms. You might like to look at similar devices with larger rotors, eg the f & p washing machine motor. These are also wound for higher voltage (mains) so likely to get you useful output sooner?
Heard on metal fatigue! Looking forward to seeing how long it lasts.

And I appreciate the advice on the washing machine motors. Im on the lookout for one, Looking forward to trying it out. The hoverboard motors definitely want the high rpm. But their benefit is that they're a dime a dozen, so I have a ton to play with and try different ideas until I get a better alternator. Thank you!
 
As a 1980's builder of small wind turbines a couple of tips. 600+ machines.

#1 your horizontal axis turbine is dangerous, very dangerous you must get it up high enough so that no little kid can get killed.
#2 If the horizontal axis unit is making any sort of howling noise it is being generated by the turbulence at the trailing edge, if you make it sharp it will become much quieter.

#3 The VAT blades would work much better if you carved them out of 2X4's I used a NACA 0010 foil split in half with a smooth or slightly convex upwind surface and 0010 curve on the downwind side, again but not certain the sharper the trailing edge the quieter the machine.
 
Thanks for the reply! I appreciate your experience. The big HAWT is already down for remodel. Anyway, the picture you saw was it on the temporary test stand I use for a new setup, it's never the final position. More recently, I'm trying a smaller blade setup of 22in, on a much smaller motor. I've learned more about better matching blade size to generator output.

Can I ask, with your experience. I've only recently learned about the standardization of airfoils and their designations. Can you recommend a resource for picking an airfoil and transferring onto raw material for shaping? Without CNC?
 
Sure, glad you have the blade up higher than anyone can reach.
A tip if you need to stop a runaway HAT blade take a loose bundle of rope 1/2" or 3/8" maybe 20' long toss the whole loose bundle into the blades, the blades will stop with no damage to rope or blade, dunno why I thought of that but it may be an age thing.

I figured out the power needed for the generators and the rpm to get there, then calculated the angles and length (diameter) of blade required,
Then calculated what windspeed it would take to "fill the circle" ie have the blade spinning sufficiently fast that in theory every molecule of wind would impact the blade.
Just don't ask me to do the math anymore, and that was before we had computers.

As I recall decide on the corner of the 2x4 that will be the leading edge, and leave it straight right out to the tip, it will be on the oppsite edge of 2x4 on the upwind side, on the other half of the blade, once you have the leading edge, begin removing the wood from the upwind side at the root of the blade 4" away from the generator to the opposite side of the 2x4 that will become the trailing edge, then draw a line from the trailing edge at the hub along the 1 1/2 inch side out to the tip opposite the tip that will be the leading edge, mark a line on the end of the board at about 1.5 degrees and remove all the material from the upwind side working between the line on the trailing edge and the edge that is the leading edge. then fllip the blade over and make it look like the top of an aircraft wing, that will get you a blade that will spin and give you a jumping off point to refine the design. I started out using an inclinometer Cheap at home depot and a 6" angle grinder, at the end was making airplane props and had a crew of 14 workers.
 
Like it....
For your charge controller... just use a 1200W boost unit like this one https://www.aliexpress.com/item/1005002760452503.html
They allow you to set the input voltage level and this is key for wind. Manual adjustment but they allow you to take a low input level and boost it up. I seem to remember they do have a limit as to how much the voltage multiple can be boosted, but not sure off hand and I have slept a lot since then....
I used one of the 1800W units on a 700W turbine which is about the same size as your turbine 2...
The fixed voltage level can work quite well if your not expecting any storms,
The 700W unit I had I managed to get a peak just over 2kW in a storm.... when I thought the turbine was going to get destroyed.
Unless your desprate for every last potential watt, I'd just use a simple boost unit for the moment and continue learning :)
 
Like it....
For your charge controller... just use a 1200W boost unit like this one https://www.aliexpress.com/item/1005002760452503.html
They allow you to set the input voltage level and this is key for wind. Manual adjustment but they allow you to take a low input level and boost it up. I seem to remember they do have a limit as to how much the voltage multiple can be boosted, but not sure off hand and I have slept a lot since then....
I used one of the 1800W units on a 700W turbine which is about the same size as your turbine 2...
The fixed voltage level can work quite well if your not expecting any storms,
The 700W unit I had I managed to get a peak just over 2kW in a storm.... when I thought the turbine was going to get destroyed.
Unless your desprate for every last potential watt, I'd just use a simple boost unit for the moment and continue learni
I have used this unit and two other boost units already, in both real world wind and in my test setup where I can spin the generator via a pulley system to simulate different wind speeds. I am hopeful they will work in the future, but so far they have not done what expected. Let me share what I learned so I can be corrected if I am doing it wrong. Maybe you can help.

These boost converters are designed to work with a relatively stable power source to produce a higher voltage. My goal is 24V. So let's say I set my boost converter to 24v using a 12v power supply. All good, i input 12v and get 24 out, excellent. Now I move the module to my test generator, which I spin at various speeds to simulate changing winds. I can get anywhere from 5 to 30 volts on my generator. This is a problem with a boost converter. Using this example of a module set to 24v via a 12v power supply: at 5v input I get 10V out. At 10V input I get 20V out. At 18V input I get 28V out. The boost converter is doing it's job, boosting voltage, but it's not designed to handle such varying voltages. I have tried it with 3 different modules at different ratings: the one you linked, a 10A module, and a 5A module. Similar results all around. Am I doing it wrong? And like you mentioned, what about storms when wind speeds rise unexpectedly? The boost module cannot buck and it could damage batteries.

This is why, for now, I am focusing on buck converters. Most cheap units input all the way up to 35V, and I have bench tested several: if I set it or 12V, it never goes above 12V. Any excess wind speed, the buck turns it into extra amps at 12V. Which, I know, is what a charge controller is supposed to do. But it's nice to know that even the cheap units still do their job, albeit less efficiently.
 
M not familiar with these motors, so will ask are they brushed or brushless, if brushed you should not need a controller, just a blocking diode to stop the motor spinning in no wind situations.
Shortly after leaving the turbine business, had an acquaintance who was very proud of his $700 three bladed turbine mounted on his RV trailer.
When we camped nearby, used to drive him nuts, would put mine up and bypass the diode, so it spun even if you could not feel any wind.
 
Shoulda mentioned that measuring open circuit voltage from your generator is meaningless, you need to know the wattage, if you are charging lead acid batteries the voltage input tops off at the level needed to charge the battery and the amps will increase or decrease depending on rpm of the generator. Used to rate my systems at 200 watts, because that is the power they put our in 20kts wind speed. We recommended taking them down if it was more windy.
 
...
It's horribly inefficient, and a mess, and a constant project, but it's also fun and incredibly educational, and all my materiels, excluding cheap electronics, are used, scrap, secondhand or junk, extremely low cost.
I find it simply lovely! Do keep in mind safety, especially if they're reachable by children. And seen as I think you did a great job by hand-making everything, I'd just give a cleanup to the wires, get rid of the alligators connecting your DC step-up. Keep it up!
 
I find it simply lovely! Do keep in mind safety, especially if they're reachable by children. And seen as I think you did a great job by hand-making everything, I'd just give a cleanup to the wires, get rid of the alligators connecting your DC step-up. Keep it up!
Lesson learned already on alligator clips. They were great at the time to easily swap different connections and configurations. Then one day two of them touched and immediately exploded the IC of a buck converter and sent it flying across the room. Insulated connectors on everything now.
 
Sure, glad you have the blade up higher than anyone can reach.
A tip if you need to stop a runaway HAT blade take a loose bundle of rope 1/2" or 3/8" maybe 20' long toss the whole loose bundle into the blades, the blades will stop with no damage to rope or blade, dunno why I thought of that but it may be an age thing.

I figured out the power needed for the generators and the rpm to get there, then calculated the angles and length (diameter) of blade required,
Then calculated what windspeed it would take to "fill the circle" ie have the blade spinning sufficiently fast that in theory every molecule of wind would impact the blade.
Just don't ask me to do the math anymore, and that was before we had computers.

As I recall decide on the corner of the 2x4 that will be the leading edge, and leave it straight right out to the tip, it will be on the oppsite edge of 2x4 on the upwind side, on the other half of the blade, once you have the leading edge, begin removing the wood from the upwind side at the root of the blade 4" away from the generator to the opposite side of the 2x4 that will become the trailing edge, then draw a line from the trailing edge at the hub along the 1 1/2 inch side out to the tip opposite the tip that will be the leading edge, mark a line on the end of the board at about 1.5 degrees and remove all the material from the upwind side working between the line on the trailing edge and the edge that is the leading edge. then fllip the blade over and make it look like the top of an aircraft wing, that will get you a blade that will spin and give you a jumping off point to refine the design. I started out using an inclinometer Cheap at home depot and a 6" angle grinder, at the end was making airplane props and had a crew of 14 workers.
Thank you for your advice. Despite having zero experience carving wood, these turned out great on first attempt, so hopefully they will get even better.
IMG_20220410_111933.jpg
IMG_20220410_165720.jpg
 
Back
Top