Wind Turbine Air Compressor

[F0ggy]

Hi all, first post here but I have been creeping around these forums for a few years now.

Thought I would share my current project of building a direct drive wind turbine air compressor. To get the initial efficiency comments out of the way... Yes, compressed air is the actual form of energy I need stored so this seems to be the best route with my location.

Bit of background: I live in a town where the average wind speed is South West 16-22kts from May till September. Spring and Autumn are neutral and then the wind switches to North East for the winter thou less consistent than the Summer. The Summer inflow begins around 11am, ramps up till 3:30pm, holds for a few hours and then backs off at 7pm. This incredibly predictable wind attracted Windsurfers in the 80's and 90's and then Kiteboarders in 2001 and on. The popularity in the summer here has jumped so much that crowding on the beach has become a safety issue for launching and landing the kites. One way to alleviate this issue was to provide compressed air to inflate the kites much more quickly then the regular hand pumps and get sailors off the beach quicker. This has been run by a gas compressor for the last two years and has proven a necessity for organization and safety.

Our air consumption changes drastically depending on user volume, day of the week, weather conditions, wind speeds (kite size) etc... So for simplicity the goal is to have this turbine operational with no CFM target in mind. Our storage tanks are 930L (2 x 420lb propane tanks) What ever the turbine can supplement to the gas compressor will be a great starting point.

Rotor diameter is somewhat capped at 6ft for the "social impact" at the site. Because of this I have been sizing a compressor to match the rotor. Settled on the Air-X Apollo blades off eBay so I could focus less on blade design and manufacturing and more on the mechanical structure.

http://www.ebay.ca/itm/6-x-62-Wind-Turbine-Generator-Blades-Hub-Nose-Cone-6-socket-fit-Air-X-Apollo-/361452233427?hash=item54283b62d3:m:m7XVhvn_Rp6sTpTfMh7q1OA

Cut-In speed of 20kph in the goal for now.

Initial tests with these blades showed that coupled with a SV-202 (2hp rated) single stage compressor was too large for the wind speed range needed. This combo had a cut-in speed of 40kph.

I've now tested with a single cylinder .75hp compressor which works well but has some initial mechanical friction (piston rings) to overcome before the blades spool up. Cut-In speed was 50kph but was able sustain 50PSI down to 25kph once spooled up.

I've integrated automatic and passive furling with a spring and pneumatic cylinder.

Ok, now for the photos!


Main yaw bearing is a 1 1/8" bicycle head tube. Easy to work with, easy to find parts.



Large brass valve is a 3 in 1 check valve, compressor unloader valve, and adjustable pilot valve. The pilot valve supplies pressure to the 1/8" port once the tank has reached an adjustable set pressure, this feeds the extension port of the pneumatic cylinder and furls the turbine. The small needle valve keeps the tail from cracking the turbine in half when it furls at 160PSI! The pilot valve is vented to atmosphere when not engaged which allows the tail to return under spring tension.





This is the passive furling/tail return spring. Adjustments at both ends of the spring for tuning. A more permanent bracket will be made once a ball park spring rate and dimension is figured out in testing. Amazingly, my rough guess of rate and length furled the turbine at 70kph! This will change with compressor back pressure thou.




More photos in next post...

[F0ggy]

Compressor cooling tube. I have one port reserved coming from the compressor output for an expansion chamber. This will be used if the turbine develops too much back pressure when attempting to spool up.











Broke the main air swivel during testing on a bumpy road. Time to beef it up.

[F0ggy]

[ Specified attachment is not available ]

[F0ggy]

The main issue I'm having now is the wind speed needed to overcome the mechanical friction. Initial tests show the blades, once spooled to their designed TSR have plenty of power to keep the compressor going. Concerns with swapping to a lighter duty compressor with less mechanical friction would be the longevity. Remember the issue is mechanical friction NOT compressor back pressure, back pressure will be handled with an expansion/loading tank IF needed.

I see three options so far...

1. A centrifugal clutch thou without much thought, I feel it may have clutch shoe burning issues as the wind slowly builds from 0-40kph. One of these clutches may be the answer as it has a spring that keeps the shoes disengaged until a set RPM.
http://www.altraliterature.com/pdfs/P-1692-TBW.pdf

2. CTV belt drive stolen from an ATV, snow mobile, 49cc scooter... This seems to be the best options thou is a bit bigger/bulkier than I would like.

3. Gear/Belt reduction. Am I correct that a 2:1 reduction would half my wind speed needed to begin spool up or is there some odd physics going on with the power and square of the wind speed?

Questions, comments, advice? All are welcome!

[F0ggy]

Active furl testing using the manual pilot port control. This is only for testing, the port works automatically based on tank pressure.
Active Furling Test

Passive furling test. Works well. Tested the active furl at 50kph with the pump pressurizing to 75PSI, also worked perfectly. (Sorry about the video quality.)
Passive Furling Test

[Mary B]

Maybe a more solid rotor like th eold water pumpers and gearing to bring the RPM's up... much more low end torque!

[Frank S]

Maybe a more solid rotor like th eold water pumpers and gearing to bring the RPM's up... much more low end torque!

Mary B is correct in this regard the fan style windmills have loads more start up torque. However since they are drag instead of lift the TSR rarely reaches much more than 3 or 4 due to the massive amount of surface area
 water pumping windmills of 7 ft diameter can lift a 2 inch column of water about 100 ft  in winds as low as 7 to 10 MPH. The down side is they are pretty much limited to 500 RPM or less even in high winds due to the drag on the blade surface area.
  On of the big differences in design though is water pump mills use a 6 to an 8 to 1 reduction gearing to increase torque at the expense of speed.
 No matter which blade design is used though the available power from wind is increased by a factor of 87 with each 10 MPH increase in speed 10 MPH yields .007 HP per sf of swept area .06 @ 20 MPH and .42 @ 40 MPH
  translates to a .059 HP output for a 6 ft diameter mill fan or turbine @ 30% efficiency in a 10 MPH wind
 or 3.79 HP in a 40 MPH wind .
 These are all approximate figures.
 the best case scenario for a wind driven compressor, would be a variable displacement  bent axis. This could have a very low start up torque since it would be mostly rotary with little to no reciprocal friction at start up then as the wind increased the swash plate would cause the pistons to have a longer stroke producing more compressed air volume as well as being able to harness more of the actual horsepower out put from the turbine.
 I would shy away from any CVT belt of chain gear reduction design due to the amount of frictional loss 

[SparWeb]

Hi
I see in your first posting that you have a lot of winds above 10 knots through most of the summer season.
Later, the discussion about start-up is mostly concerned with the mechanical friction, rather than the back pressure in the cylinder, and you have potential solutions in mind.

I just watched your truck-test drive video (fun!) and it spins up to what seems to be several hundred RPM (judging by the wagon-wheel spoke effect).  I think your baby is ready to be thrown in the air.  If you have a tower ready to go (not necessarily the full height but just tall enough to keep the blades out of your hair) then you'll know by springtime what the bugs are.  A small test tower is also useful for allowing you to quickly make adjustments to the turbine without complicated winches, cables etc.

So, in a nutshell, go for it!

I look forward to seeing how it does.
Thanks for posting the project, and welcome to Fieldlines!

[kitestrings]

Welcome F0ggy,

The quality of what you done so far is very good.  I suspect there's been some trial and error to get the furling to work as it does in the videos.  Nice.

I would be concerned with compressed air about dealing with moisture.  It looks like you've where there is snow, so I assume condensation and freezing might be issues you'll have to address.

Anyhow, good luck and thanks for sharing.  ~ks

[MattM]

Very interesting concepts.  If longevity proves reliable and it can stay up at least a year at a time between seal changes, maybe it could pan out.  Pneumatic energy is quite useful.  Just need a useful dump load that removes pressure off the tank when the cup overfloweth.

Is pneumatic to electricity energy conversion in the future?

[MattM]

One other idea crossed my mind, you said that the need for air compression is mostly during daytime hours.  Why not use solar energy?  Surely you could use the solar panels dual purpose, to pump air and to provide some necessary sun relief in the area.  Solar panels are much safer around people than machinery.

[dnix71]

We have an Ingersoll Rand rotary screw air compressor at work for a conveyor, foil stamping and other tasks. Rotary screw compressors don't have the same high startup torque requirements as piston compressors, but a "small" compressor by IR standards takes a 5HP motor.

Have you thought of using a centrifugal clutch to allow the windmill to spool up before enganging the compressor.

http://www.northerntool.com/shop/tools/product_200415722_200415722  You would need a pulley set up to increase the speed of the clutch end from the windmill cut-in to 1800 rpm.

[F0ggy]

Ok, big update time... Just going to dump the photos off my phone and add a caption to each, everyone just wants to see photos anyway!




Tail section cut out of this amazing material called Durolam. It's two thin layers of aluminum with plexi or poly pressed in between. Incredibly stiff for its thickness. Planned to cut it out on the CNC router but this proved just as easy.




I started looking at other options for compressors to help the startup wind speed. Tried a few options like a small 50cc automotive A/C compressor but that still had too much internal friction. Took a new view on the issue and decided to try modifying the existing compressor which was already sized quite well for the project. I removed the two compression rings but left the oil scrubbing ring to reduce friction. Testing showed I could still produce 50PSI @ 1000RPM so I decided to keep moving in this direction.




Milling nacelle mold half sections. The width of the mold was limited by the Z depth on the CNC so I had to mill in sections.




Two sections joined, sanded and waxed for fiberglass.




I hate fiberglass...




Pretty ugly so far.




Getting better...




Trimmed base plate. New main swivel for air the travel down the mast. Nacelle mounts.




Mock-up.







Still working on the spring furling system. I'll probably attack it again after the second test when I know more about the spring location.




Handheld tachometers are either total junk or too expensive so I built one.




This will be for testing only so I can graph the performance. I might add a pressure and speed sensor along with a data logger, then all I have to do is drive for the tests and not stare at the pressure gauge.




First test run without the tachometer. I recorded the audio from the the purge valve while calling out vehicle speed and used that to calculate the RPM... Next run with the digital tach should have some more accurate data.




More to come...

[F0ggy]

Video of the automated furling/shutdown when a pre-set PSI has been reached. This was simulated by a pressure regulator being turned up to 60 PSI which triggers the pilot valve and pressurizes the cylinder.

Automated Furling

[F0ggy]

Quick run through of the tachometer. Number of triggers per revolution is adjustable.

Tachometer

[F0ggy]

http://sendvid.com/1uxrl0kq

[SparWeb]

Hey, this is fun (I love numbers)

Did you move your tinypic tacho video?  Link's broken.

Your fiberglass looks better than any I've done.

My only concern is that the tail looks too small.  The arm isn't very long, but it would be fine only if the fin were much larger. 
With the number of blades, and the offset of the pump, I'd say your turbine would have trouble running true to the wind.
I do like the laminate material you found. 

[F0ggy]

Did you move your tinypic tacho video?  Link's broken.

Try this one. http://sendvid.com/1uxrl0kq

The tail was ever so slightly too small during tests but not so much to stress about. I can compensate for this with the tail angle which has adjustable stop points.

[SparWeb]

That one worked.
I can't quite tell what you're doing with your dials.  Do you scroll a menu with one, adjust the pulse/rev with the other?