A moving platform test system for VAWTs

[oneirondreamer]

Hello All,

I'm just ironing out the final kinks and hoping that soon I'll be in a position where I could use some advice, on how to wind a stator that matches this turbines output.   I've got some idea's around alternators, and I'm the best I've ever met at putting things together or seeing how things work, but I'm not great with electrical engineering.

Here's a video that I made mostly thinking of you folks, so that it's clear I'm really making stuff, because I know there's a lot of people who come here "looking for help", with no real intent to do anything and just want to chinwag.

I'm hopeful, but I also understand that VAWT's, especially of this type (lower speed, lots of embedded material) may rarely be economical, even if they are efficient, especially small versions.   I continue to put the energy into my work that I do, not really because of the wind turbine market, but because of the kinetic hydro market.     I do hope that it'll be efficient enough to be useful in some wind locations, but I'm not holding my breath.

[MagnetJuice]

I have a good understanding about designing and building alternators. We could collaborate and should be able to put together a nice system that gets those resistor coils smoking. That is, if the turbine can put out the power.

Ed

[Adriaan Kragten]

It's a good idea to test your rotor on top of a car because then you have an expanding wake and the wind speed is the driving speed if the real wind speed is zero.

On my website www.kdwindturbines.nl you find different ideas about PM-generators at report KD 341. In the reference of this report you find reports with detailed information about a certain option.

[Bruce S]

Even if it's not as efficient as a HAWT or even as high a power unit as you hope for it sure is nice to look at!!

Keep up the great work!

[oneirondreamer]

Thanks Bruce, Adriaan, and MagnetJuice,

I have read Adriaan's report KD 341, more than once, and while I definitely learned some things, a lot of it is over my head.   

I agree with what I think is Adriaan's general analysis, that low TSR, high solidity VAWT's have some significant disadvantages which are not always recognized: the blade takes more material than the same swept area HAWT, the lower rotation speed means higher alternator cost (or additional gearing costs), and as there is no furling option, the bearings and mount must be stronger (and more costly) than the equivalent HAWT.   These things make me see that to be successful a new VAWT must be made inexpensively from tough long lasting materials (galvanized steel sheet?), and it will need the least expensive alternator and control system that can be made. 

Living in the mountains of BC, there is a substantial wind resource but it is stranded at the tops of mountains, and as I understand it, largely inaccessible to modern HAWT's due to the rapid wind shifts and turbulence, and so, if I can get over the first two hurdles I am hopeful that it may find some utility locally.   It's my understanding that the mountains are largely inaccessible to HAWT's of any size, due to wind shift and turbulence issues.   

I'd like to know more about the venerable “Windside”/GUS and the alternators they use.  They have been in business for a long time.   Their original claims were that their rotors reached Cp's of 0.21, however they have erased those claims and it's my understanding that purchasers found them to produce much less than 50% of expected power.   Their alternators are custom made, and it's my recollection that they used to sell them for use as very low speed high torque alternators.   They've gone through a lot of changes, and their website no longer lists them for sale.   They claim to be low or zero cogging, but seem to be fairly conventional radial flux machines, however they are very heavy!!  38kg for a .5 m tall and .3 m diameter unit, with a Fiberglas blade!.    The blade should be light, less than 5kg, so around 30kg of alternator and bearings!   I think they are using ceramic magnets, but I'm curious as to the geometry that gives them low cogging.    I have a bunch of 1”x1”x6” ceramic magnets, and it's tempting to try to build a radial flux machine,,,, but I'd better keep on track for now.   

[Bruce S]

It could be as you said Ceramic magnets.
However Zubbly, Windstuffnow, Jerry, etc were known for skewing their Neo magnets to offset the cogging issues.

Other have done pretty good right-ups about skewing the mags,
As soon as I get back to HQ I'll do some searching.

Cheers

Bruce S

[SparWeb]

That sparweb guy did a  write up on magnets....  :-)
But a much better one came from Peter Dingemans 
https://www.fieldlines.com/index.php?action=profile;area=showposts;u=3538

In time I can probably find the specific post...  he was prolific.

[SparWeb]

Drew,
On watching your vid, some suggestions.
If there is a breeze when you test, drive the vehicle in opposite directions into and away from the breeze,  and average the results.
If the breeze adds a crosswind on both tracks, then averaging isn't enough.
Then use trig to take the component vector out of the average.

[oneirondreamer]

Hello All,

Tomorrow I hope to publish a draft report on my testing today but I'm having trouble figuring out how to get a low enough resistance load to keep my alternator from dragging down the turbine speed below ideal load.   I outline my resistance problem below, and hope for some advice.

  I'm still sorting out some test procedures, and apparatus.   I'll do a number of draft reports that I hope to use feedback from to improve.

A preview of my results for tonight, based on the load cell reading torque, optical gate reading turbine speed in 36 pulses per rev, and factory calibrated wind sensor is that it appears that my unloaded tip speed is above 1.3, however at higher vehicle speeds the unloaded turbine develops some vibration that I avoided, when lightly loaded this seems to disappear, and heavily loaded it doesn't occur.   My load doesn't have enough resolution for me to determine ideal tip speed (which I expect around .9).   With the load I have now maximum turbine output (shaft Watts) occurs at TSR 1.1-1.2, around 55km/hr, at around 650rpm.   As with my previous turbine, this turbine exhibits two different operational speeds.   A low TSR (.2-.3) mode, where the turbine has been dragged below most useful speed, and a > .9 mode where the turbine operates more effectively. 

My load bank problem.    First problem is that my alternators output voltage at low speed barely makes it past the recitifier.   I assume the rectifier adds some resistance to the circuit, that's what eats the 1.4 V?   Next problem is that attaching the load bank I have is too much, it is too much drag.   It seems like lower resistance is the way to go, but it's already so low, I'm not sure it's the right thing.   Maybe I'm missing something?

What I've got is about 50 wire coil resistors I made from stainless hard wire a few years ago.   Each one measures Around  2.25  ohm but I am not confident in that reading as my multimeter is cheap, and it ranges around and never really totally settles, generally keeps slowly declining, I'm guessing as the battery gets killed by trying to load the coil?    I have them in banks of 8 parallel right now.    If I measure 1 coil it measures around 2.25ohm.   If I measure 4 coils in series I get 7.5 ohm, and 4 in parallel I get 0.8 ohm.   

Clearly this is beyond my cheap multimeter.    I do have 0-30 voltage probe and 10A current probes, and a low wattage/voltage power supply and my data logger.   I guess I could test them with that?   But I'm not sure that I need to know what the resistance of this bank is, other than knowing what I'd like is something less than 1/4 it's current resistance is my guess, which is 48 of these coils in parallel.   If I broke each coil into 4 coils, and wired them in series, would this in theory get me what I need?    I think I'll just get some heavier stainless wire, but I'm curious about this.   Does anyone have any suggestions for a very low resistance load that can handle as much as 500W, but more likely 100-200W?

[MagnetJuice]

What you need is to get your turbine spinning to the RPM that you want, and then switch on the load remotely from inside your truck.

For that, you need a heavy-duty resistor (Dryer Heating Element). This element total resistance is about 10 ohms and can take 5 Kw. With this setup you will be able to set you load to the exact resistance that you want using the battery clamp.

Here is all that you need:

Dryer Heating Element.
A 12 Volt solenoid
A battery clamp
A toggle switch

You can order all from Amazon Canada.
Here are the links:

https://www.amazon.ca/Heating-Element-Replacement-Whirlpool-Kenmore/dp/B07R62Z9N6/ref=sr_1_42?crid=2AP676Q9USSEG&keywords=dryer+element&qid=1566544069&s=gateway&sprefix=dryer+ele%2Caps%2C228&sr=8-42

https://www.amazon.ca/Oregon-33-334-Solenoid-Mower-Solenoids/dp/B000ALAUAE/ref=sr_1_8?crid=194DFG5NMANWR&keywords=solenoid+12v&qid=1566543869&s=gateway&sprefix=solenoid%2Caps%2C676&sr=8-8

https://www.amazon.ca/Vehicle-Battery-Alligator-Crocodile-Clamp/dp/B071W272KM/ref=sr_1_46?keywords=battery+clamp&qid=1566544332&s=gateway&sr=8-46

https://www.amazon.ca/Baomain-Toggle-Switch-waterproof-cover/dp/B01IYD8SHQ/ref=sr_1_14?crid=35V7ZUTJU2G49&keywords=toggle+switch+12v&qid=1566545032&s=gateway&sprefix=toggle+sw%2Caps%2C228&sr=8-14

I will draw a wiring diagram for you tomorrow. It is late now; I can hear my pillow calling me. 

Ed

[Adriaan Kragten]

Drew,
On watching your vid, some suggestions.
If there is a breeze when you test, drive the vehicle in opposite directions into and away from the breeze,  and average the results.

I don't agree with this advice. The generated power is proportional to the cube of the wind speed. The positive effect of driving against the wind is therefore much higher than the negative effect of driving in the direction of the wind. So you should wait untill there is absolutely no wind if you want to have reliable measuring results.

Another point is that real wind is always fluctuating so this fluctation gives fluctuation of your measurements even if you drive with a constan speed.

[DamonHD]

First problem is that my alternators output voltage at low speed barely makes it past the recitifier.   I assume the rectifier adds some resistance to the circuit, that's what eats the 1.4 V?

Diodes do have a forward voltage drop, which is not just resistance.  It is possible to reduce that in come circumstances (with schottky diodes, or active 'perfect' diodes) though each has its own issues.  And the voltage drop will be higher under high load (over 1V per diode, so over 2V for the bridge).

(Other people please chip in if I'm talking nonsense; this should be easy but I'm suddenly unsure!)

Rgds

Damon

[oneirondreamer]

What you need is to get your turbine spinning to the RPM that you want, and then switch on the load remotely from inside your truck.

For that, you need a heavy-duty resistor (Dryer Heating Element). This element total resistance is about 10 ohms and can take 5 Kw. With this setup you will be able to set you load to the exact resistance that you want using the battery clamp.

Here is all that you need:

Dryer Heating Element.
A 12 Volt solenoid
A battery clamp
A toggle switch

You can order all from Amazon Canada.
Here are the links:

https://www.amazon.ca/Heating-Element-Replacement-Whirlpool-Kenmore/dp/B07R62Z9N6/ref=sr_1_42?crid=2AP676Q9USSEG&keywords=dryer+element&qid=1566544069&s=gateway&sprefix=dryer+ele%2Caps%2C228&sr=8-42

https://www.amazon.ca/Oregon-33-334-Solenoid-Mower-Solenoids/dp/B000ALAUAE/ref=sr_1_8?crid=194DFG5NMANWR&keywords=solenoid+12v&qid=1566543869&s=gateway&sprefix=solenoid%2Caps%2C676&sr=8-8

https://www.amazon.ca/Vehicle-Battery-Alligator-Crocodile-Clamp/dp/B071W272KM/ref=sr_1_46?keywords=battery+clamp&qid=1566544332&s=gateway&sr=8-46

https://www.amazon.ca/Baomain-Toggle-Switch-waterproof-cover/dp/B01IYD8SHQ/ref=sr_1_14?crid=35V7ZUTJU2G49&keywords=toggle+switch+12v&qid=1566545032&s=gateway&sprefix=toggle+sw%2Caps%2C228&sr=8-14

I will draw a wiring diagram for you tomorrow. It is late now; I can hear my pillow calling me. 

Ed

Thanks Ed,  I think you are right about needing to switch on the load when the turbine is at speed, though I am concerned that the sudden torque may twist the stator on it's mount.   I may need to beef that up.   What I did on my last moving platform was have a band brake, with wooden friction blocks.   I controlled the band tightness (load) with a little 9v gear motor, and you are correct would start with low load and progress up.   I had a cohort who would run that while I drove.   

The beauty of that setup was infinite number of load's.   The downside is that the blocks roasted and the friction co-efficient would change, as well as the reality that it was difficult change the load accurately or quickly. 

I like the idea of using a dryer element, in fact I changed one here a couple years ago, and kept a few bits of element.   I've got some stainless tie wire that might be close to equivilent.    I also have a couple of old heavy welding contractors I think, but I'm not sure what voltage the solinod runs at.     I'm pretty sure that I can wire that up without a diagram, the larger question is how do I vary the load?    If I made an 8 ohm resistor that was a series of smaller resistors, then I could use a jumper wire to short around some of the resistors maybe.   So test could go solenoid off, drive to speed, solenoid on, wait, solenoid off, change load w jumper, solinid on.     

I don't want to sound melodramatic, but I have no money to spend on this right now, every penny has to be borrowed, but I have a lot of stuff to work with.   My old truck could die at any time as well, so I'm trying to get the most out of every test.

I am worried that perhaps my calibration of the new toothed photo gate wheel, from 12 to 36 tooth, could have thrown off my numbers.   I'll be checking that today.   As well I've got another wind sensor, and I may do a test run without the turbine, but instead the second wind sensor in it's place, to ensure that I'm measuring the input accurately.    I also recall that I can add a gpx file if I can remember how, which would give me real speed over the ground, as well as wind speed.

Thanks for the tip, it's been a long time since I did my previous tests, and I was hoping that I could just use a “fixed” load, and hadn't thought that through.    I wonder how I could/if I could somehow soft start a load and where I could salvage some high heat terminal blocks.....   

[MagnetJuice]

Here is a diagram for a circuit that will allow you to load your alternator incrementally. If you want to have more options of resistance, just add more solenoids and a rotary switch with more contacts.

Make sure that the rotary switch is 'make before break' to keep the alternator loaded during switch-over.

Since that is pure resistance, you don't have to worry about the polarity of the alternator.

The wires to power the solenoid coils could be 18 Ga. and wires coming from alternator should be 8 Ga.

[SparWeb]

Drew,
On watching your vid, some suggestions.
If there is a breeze when you test, drive the vehicle in opposite directions into and away from the breeze,  and average the results.

I don't agree with this advice. The generated power is proportional to the cube of the wind speed. The positive effect of driving against the wind is therefore much higher than the negative effect of driving in the direction of the wind. So you should wait untill there is absolutely no wind if you want to have reliable measuring results.
Another point is that real wind is always fluctuating so this fluctation gives fluctuation of your measurements even if you drive with a constan speed.

Very good point, AK.  Now that I check my with pocket calculator...
This would become inaccurate if the wind speed were more than about 15% of the vehicle speed when driving into or out of the wind.
And when driving with a cross-wind, a 10% cross-wind will exaggerate the power by 33%.  Driving on the reverse direction wouldn't cancel it out because the cross-wind vector is still additive.  For any vehicle tests to be accurate, the wind needs to be very still.

Fluctuations are inherent in all natural air flows.  Depending on what you want your test result to represent, natural fluctuations may be beneficial.

[SparWeb]

Drew,

Rather than trying to trust your multi-meter on the low-ohm end of the scale, there's another way.  To measure resistance of low-R resistors, try applying a fixed DC power supply to the resistor and measure the voltage and current.  Use Ohm's Law to work out the R. 

Watch out for applying so much voltage that the resistor heats up significantly, which will affect the results.   Not that that's bad - it's supposed to heat up - but do it carefully so that you know how much, and don't set your workbench on fire. *

You have to measure battery voltage and current simultaneously on the ends of the resistor wires.

* voice of experience