Lv2 going to a wind tunnel test faucility

[windstuffnow]

  Well, mike (rotornuts) is going to get his wish.  I've made arrangements to build another Lenz turbine and stuff it in a wind tunnel for an accurate accounting of what it will/will not do.  

  I searched all over looking for a test faucility, spent many hours on the phone and what do you know... There is a test faucility not 10 miles from my house.  Apparently it was built and installed less than a year ago for WMU's Department of Mechanical and Aeronautical Engineering.

  I'll be building another unit just for this purpose.  From what I understand they can test it in many different ways to find the actual efficiency.  I plan to run it through every test they have and a few of my own ( one full hours worth ) at various wind speeds.  

  I'll take some pics and video of the process.  It's gonna be fun!

.

[BoneHead]

When do you think you'll be doing this?

That will be great. I for one, will certainly be looking for the info.

[windstuffnow]

  I will take a tour within the next week or so to get all the details of how it has to be designed, size - mounting etc and it will be another few days to design and build the unit.  I'm not sure about the wait time but it sounded like I could get in fairly soon.  I'm hoping within the next couple weeks or so.  

.

[electrondady1]

thats really great ed , lots of misteries will be solved. will they use smoke  like i see on tv?

[windstuffnow]

  It seems like to me, I could be wrong, that the smoke stream would only be useful showing the flow over a locked rotor in different positions.  If the rotor is moving at any speed its simply going to disrupt the stream.   It might be interesting to see the locked rotor stream in various positions, may even invite some new ideas.  Since the machine is meant to spin at certain rpm in a given wind and extract energy the locked rotor flow might not be a good indicator of what the machine is really going to do.  

  Anyway, I'll have a better idea of what they can and can't do later this week.

.  

[rotornuts]

This is very exciting news Ed. I can't wait till you get the results. I would agree that the locked rotor flow visualization would be fantastic (I'm still very curious what's happening at the 9 to 12 o'clock position with regards to the open backed blades, assuming 12 is at the front of the unit). I think a quick look at airflow across the 2/3 left and right of center area during operating speed would be nice as well. I've always been curious as well to get an idea if the flow is somewhat deflected around the machine at operating speeds. It would help to undetrstand the setting of blade angles and also what may be happening to wind speed if the flow takes a curved path. If it does then we could understand the tsr's better and better understand what portion of the work is being done by lift.

If sufficient "appearant" solidity occurs at speed then maybe we would be better able to take advantage of the increased wind speeds and appearent wind direction as the airflow deflects around the machine.

I'm excited to say the least.

Mike

[electrondady1]

well, i for one am curious as to just what is going on in the center of the machine when at speed , and how the heck one wing can pull it self into the wind from  a dead stop at the 12 through 9 and on to the 6 oclock position as per your demo video.

[DanB]

Super Ed - that's exciting stuff!  Can't wait to hear more about it...

[windstuffnow]

  I'm pretty excited myself... Not only is going to be interesting to see how my turbine performs, the whole experience in itself is going to be fun and learnative.   It's kind of scarry having a professional wind tunnel so close... I can see my project money being depleated on testing.   I should be-friend a few upcomming Engineering students... maybe they could build and test some of my idea's for their grade.

  The tunnel is in an old hanger that was used by the college to train students to fly Cessna's.  Its 220 ft long and can produce winds up to 170 mph.  I'm really only interested in data up to 30 mph although I may have them take it to 40 to see if it actually starts to fall off in power.  I expect it will because the wings are pretty dirty overall.   Typically, on an ultralight the wing design was only good for slow speed and when you poured the power to them they simply burned more fuel and maintained a moderately slow speed.  A bonus on a VAWT.

  Here is a question for all that are interested...

My main objective is to find out the efficiency of the turbine.  That is the power output at a given windspeed vs the input power.  As well to observe it while its running and look for possible improvements.

If you have and idea or something that you would like to see tested let me know.   I'm only going to run the actual test for 1 hour but if the test's can be done with little to no change over time then I'll try to get it all in and post the results.  No guarantees of course but I'll do what I can.

.

 

[windstuffnow]

  Mike,

    The way I'm seeing it is when the blade is at the 12 it has stopped in relation to the wind ( that is to say it is no longer moving toward the wind or away from the wind ), the speed then increases slowly where a large amount of force is a applied before the blade speeds up again.  At the 9 position its traveling at its fasted away from the wind where the torque falls off considerably.  Just before the torque fall off from the downwind blade the upwind is taking over.

    Thats just my visual on the unit from hours of play time.  I'd still like to try a group of wing variations... There is more to capture!  

    I'm looking forward to seeing your fat cambered blade in motion!  That's going to be very interesting.  Build it up as a 2ft x 2ft model and it will fit in the smaller tunnel.  Their smallest tunnel is a 12 x 12 and the largest can fit a car in it.   The one I'm using has a test area of 3 x 4 ft and is far less expensive than the big one.  I'm not sure testing a model of 6" square would yield any reasonable data for a scaled up version.  

    Let me know if you have ideas on the testing phase, that you would like to see and I'll try to squeeze it in with the other tests.

    One thing we talked about, for efficiency, was a program that they had that measured incomming pressure and pressure behind the machine... understandable, but by monitoring rpm's and pressures they can predict shaft output.   I understand the pressure differences but I don't understand how that relates to watts through pressure and rpm.   I'll bet I'm going to learn a bunch in a short period of time.

.

[BoneHead]

One thing that I would like to know (after testing), is if you think that a slightly concave camber on the lift side would be more efficient. I hope I stated it right, I mean the flat side of the blade located on the inside of the circle that is traveled by the blades.

I'm making an itty bitty lenz with a modified blade design. The front that is rounded on yours is pointed slightly on these, and the backside, solid area is curved slightly inward. I was thinking I may get more lift that way.

If I can get some sketches scanned, I'll post them incase I'm not explaining it right.

Sheesh! I shouldn't have started this project at 36 years old. I should have done it when I was 16... I knew everything then...LOL.

[MountainMan]

Ed,

Many VAWT designs (including some of your earlier experiments) have a solid top and bottom plate.  If it were practical to attach some temporary cardboard top and bottom plates to your model and check for efficiency differences, that would be cool.

thanks,

jp

[windstuffnow]

  That's a good idea.  I had some question about that when I went from my test model to the larger one.   It didn't seem to make any noticable difference but it's worthy of a test.  I'll make up some discs that can be installed reasonably quick.

Thanks for reminding me!

.

[RP]

Ed,

Can you put a genny on it and get some data on optimum TSR (load) for power output?  As I recall the savonius peaks at 0.3 and I wondered where your's would reach max.

[rotornuts]

Of course I have some Ideas as well. I think I confused things a bit with my last explanation so I've included some drawings this time.

One is a flow visualization at operating speeds in perhaps 15-30 mph winds. the other would be a quicky locked rotor flow visualization.

Next

Anyway I'm sure your going to push your time limit easily so I don't envy the hard choices you'll have to make as far as what you can fit in.

Good luck and I wish I could join you.

Mike

[SparWeb]

Ed,

I've done a few wind tunnel tests myself, so I'm at least familiar with the equipment they'll want to use, tho in my case the one tunnel was smaller & slower, and the other tunnel was a supersonic one, thus not very useful to you!  ;^)

Wind tunnels come equippped with equipment to measure a LOT of variables simultaneously.  If you mount your model on the balance, you can measure the forces (Lift, Drag, pitch, roll, yaw) at different wind speeds.  Also note that elaborate models have working control surfaces and comples WT's can allow the researcher to study dynamic responses to control motions.  All of this means that you can mount your turbine on the balance beam arms and measure "Drag", "Pitch" and "Yaw" on it (I assume "Lift" would be zero, but you never know).  If you're following me so far, you should realize that "Drag" times half the height of your turbine should equal "Pitch".  If it doesn't, you have an answer to whether a circular top plate above the blades is useful.

Your test crew will also be familiar with the "Standard atmosphere", and if they're really nice, they'll convert the data into those terms for you.  Since you're an ultralight pilot, you may either have, or know where to get (Sporty's; nearby pilot's shop), an E6B calculator, which does the same thing for you.  

Definitely put the generator on, and bring various loads that you can quickly switch.  During a run at constant wind speed, you can apply different loads, and measure RPM with a digital camera, or with a helicopter rotor blade tracker.

Wow, sounds like lots of fun.  I envy you.  My VAWT's are on a drastic growth curve this month, which precludes cramming them into anything but the most grandiose of tunnels.

Vicariously yours,

[henjulfox]

Probably not practical but it might be interesting to get a HAWT in there next to it to compare efficiencies in the cotrolled environment.

-Henry

[hvirtane]

 "If you have and idea or something that you would like to see tested let me know.   I'm only going to run the actual test for 1 hour but if the test's can be done with little to no change over time then I'll try to get it all in and post the results.  No guarantees of course but I'll do what I can."

You might try fixing the blades so that

the pitch angle could be changed easily

for the tests?

- Hannu

[Ungrounded Lightning Rod]

It seems like to me, I could be wrong, that the smoke stream would only be useful showing the flow over a locked rotor in different positions.  If the rotor is moving at any speed its simply going to disrupt the stream.

IMHO Only if it throws turbulence - in which case the WAY it disrupts the stream tells you about that.  I'd think it would just chop up the smoke streams and you should still be able to get the air path from it despite the spin.  (You'll need to take a movie and examine the frames.  Comparing the smoke pattern in several frames from different cycles with the blades in the same position will tell you where your flow is laminar or turbulent.)

I agree with you on locked rotors being unrealistic.  I noticed your early postings involved the forces on a single blade fixed at a given angle to the wind, while in the real machine the apparent wind will vary at different parts on each airfoil (due to its rotation) while blade/blade interaction - with the blades directing airstreams at or past each other - should be a major effect (just as it is in a savonius).  This should help sort it out.

[Ungrounded Lightning Rod]

You can also use the current delivered at equilibrium as a measure of torque and the alternator output frequency (if you have a frequency counter handy) to measure speed.

Several load points at each wind speed will let you plot a curve to interpolate between them, so you don't have to hit particular TSRs dead on - or even match them from one wind speed to another - to characterize your mill's performance.

Of course if they have a motor handy that can force it to different speeds and measure the torque that does the job even better.  B-)

[electrondady1]

 ed if you do all the tests we'd like you to do your gonna need your own wind tunnel!!! lol

[SparWeb]

A link to a similar test, therefore it may be very interesting to compare the results:

http://infoserve.library.sandia.gov/sand_doc/1976/760131.pdf

"Wind Tunnel Performance Data for Two- and Three-Bucket

Savonius Rotors"

B.F. Blackwell, R.E. Sheldahl, L.V. Felts

SAND76-0131

(Of course, you might already be aware of Sandia's research).