another idea

Sounds good to me, I was just trying to simplify the explanation of what's really going on for those who didn't seem to be getting it.

My first experience with a CVT was in a snowmobile that used a wide v-belt between two variable width sheaves that adjusted themselves automatically by the use of springs and centrifugal weights based on engine RPM and torque. My understanding is that belts provide something like 98% efficiency so a similar CVT might work well in this application.

 

The argument smasher....

Rick, Ken (et al)

I posted (attached) the pictures of my model which I constructed to settle a dispute about going windward. For those familiar with the idea, no explanations are necessary, as the way it works is self explanatory. It has a small spring wire that keeps the steering centered, but allows the windvane to correct it to windward. The nature of the bet required I could demonstrate it totally windward at all times, and this is not what the final controls would try and emulate. Keep in mind when I made this thing I had no idea of others that were working on this stuff, so what you see is my intuitive understanding of the issue, and what a general solution would look like.

Michael

 

The transmission of turbine wind power to the marine screw.

Rick,

I tried to catch up on your older posts last night to see where you were going with your concept, and I have to pretty much agree with how you understand the matter. As to the transmission of the power to the driving prop I would have the following suggestion: Since wind is a variable resource in both the long term (windless days) and in the short term (gusts, puffs etc.) and strong and weak winds, it will be necessary to have a torque management system in play to keep the power to the vessel approximately constant. You mentioned a CVT type of trans, well have you considered using a Solomon type of drive? Let me describe how I would tackle the problem. First, the turbine needs to be mechanically coupled to the load with some kind of very efficient chain drive first, not via the indirect generator to motor electrical coupling. The reason for this is that the purely electrical syatem would have to be robust enough to handle the entire power of the turbine, so it would be unnecessarily large. The turbine shaft would drive the input sprocket through a sprag clutch that allows the output propeller to overrun the speed of the wind turbine. The turbine therefore only produces output if it can catch up to the rest of the drive train. However, since the no load speed of the turbine is always many times higher than the nominal operating speed, it is always running at the operating speed, but not always producing enough torque to keep everything running by itself. This power feeds into one of the inputs (ring gear) to the Solomon drive. Briefly, the Solomon drive is a three input planetary transmission with each shaft both an input (power in) and an output (power out/generator). One of the other outputs (the planetary ring) drives the output propeller, and the third shaft (the pinion) is connected to a control motor/generator. In tandem with the output propellor is a direct drive bi-directional electric motor/generator equal to the capacity of the wind turbine. This may sound complicated, but really isn't, because the Solomon trans acts as a CVT since it can feed torque into and out of the system in response to the actual input coming from the turbine. That way you take what you can get from the turbine, and the rest of the time you are either 'bucking' or 'boosting' the output to produce a smooth ride. During the 'bucking' you are generating electrical power and storing it, during the 'boost' you are using electrical power. It is possible to run such a system as a 'net zero' power consumption at the power level of the turbine at ambient relative wind, or the relative wind can be augmented by inputing power from the electrical system, which I believe will amplify the turbine output by artificially increasing the vessel velocity.

 

The Solomon Drive

Rick,

If you need to look at the Webpage of this thing, here it is:

http://www.solomontechnologies.com/wheel.htm

Michael

 

My turbine prop...

Rick,

This is the shot of my prop under construction for the bicycle experiment. It has blades that are 24" long. The swept circle has an area of about 1720 sq. in. Should be interesting.

Michael

 

Whoops... picture failed, here it is...

Rick,

Prop picture...

Michael

 

Michael
My Solar-Wind boat is a balance between solar and wind with a slight bias on solar. The motor/generator I have for the wind prop will have much higher power handling than the turbine/prop is capable of in most wind conditions and I have very large energy storage. My goal was to keep it reasonably simple, at least in concept. Electrics give me complete flexible for a modular set up.

Your little model looks nice. You will find the bigger one will be more inclined to self start. The blades look OK. How well did you determine the pitch angle?

Rick

 

a vertical axis turbine would solve a lot of the pointing problems, as well as simplifying the whole machine.

*draws feverishly*

 

Ducted Wind Turbine

Rick
Doing some thinking about the idea of windmill powered propulsion, and I had a few thoughts I wanted to share and get some feedback from you, since you have given this some considerable thought as well. I believe there are some significant differences between just generating power from a fixed land based location, and one that is aboard a moving vessel. The most important is the magnitude of forces that are generated in the axial direction, since these are directly contrary to the forces that are required for propulsion. In a land based arrangement these forces are irrelevant other than the mechanical requirements to handle the thrust loads and stresses on the blades etc., since the pedestal is firmly anchored into the earth, and offers infinite resistance to movement. In a moving vessel these forces not only need to be mechanically resolved, but the forward thrust required must first counter this negative axial load before any net forward motion can be achieved. Using turbines with high tip speed ratios works counter to this goal, because as the speed increases the pitch must flatten out, and this shifts a large portion of the lift component to the axial direction, effectively making further gains in generating output useless beyond the point where these negative axial components equal the vehicle drag. In addition, the actual drag on the blades increases with higher RPMs so that negative axial component of this also increases.
The obvious answer to this dilemma is to find an arrangement where the tip and root speeds are very similar, so that the amount of pitch change is small, and never exceeds around 45 degrees so that no large axial component actually develops, and most of the energy can be converted into torque. This will require that the RPM be limited so that no large pitch changes are needed. To get the required air velocity, it would be better to augment the flow in a venturi duct first, and draw the air out the back with a flared diffuser. I have attached a sketch of something like this for those interested.

 

There is an earlier thread on this topic. I created a posting there http://www.boatdesign.net/forums/showthread.php?t=14182&page=12&goto=#174 that dealt with the rough math of why it is feasible to head directly into the wind with this concept. On that thread, this was a recurring theme of disbelief. My simplistic treatment might suggest how to analyse the problem of forces on boat-based vs land-based systems. The simplest case of heading directly into the wind is dealt with in my posting which however ignores energy losses. I suspect the level of math will become serious ...

Regarding turbine blade design all that is needed in principle is a constant pitch design. However, that assumes that the turbine is permitted to revolve at a speed that is efficient which requires a variable speed transmission.

 

With regard to ducting I am near certain that the added drag of all that area exposed to the air stream will more than negate any benefit that could be gained in turbine efficiency.

What I have shown is that the system performance gets down to overall system efficiency.

This means the combined efficiency from taking power out of the air stream, through the mechanical losses and then the efficiency of applying the power to the water. The important variable is effective gearing between the air turbine and water propeller while the only performance parameter of interest is the efficiency of power transfer. The system equations are:

Power In = Turbine Thrust x Apparent Wind Velocity

Power Out = Propeller Thrust x Boat Velocity

Propeller Thrust = Turbine Thrust + Boat Drag

Apparent Wind Velocity = Wind Speed + Boat Velocity (when going directly to windward)

Boat Drag = Function(Boat Speed) (For slender hull approximately Constant x Speed^2)

Power Out = Power In x Air Turbine Efficiency x Water Turbine Efficiency X Mechanical Efficiency

These are the equations that you need to know to determine system performance. Efficiency is KING and gearing makes it work. With a really efficient system turbine pitch can be as little as 1.6X the propeller pitch x mech gear ratio.

I design high efficiency propellers (or turbines) and I find they have quite high tip speed to get the best performance. This is particularly the case with small air turbines. My blades look more like they came off an aeroplane than the typical wind farm. The latter is designed to collect energy by extracting as much power as possible. They are defined by their power coefficient not efficiency. Your choice of propeller for you little cart is close to ideal I suspect. (Was this a well calculated selection or just inspired good fortune.)

The force acting on the blades is most commonly resolved into lift and drag components for calculation purposes. I recognise that at high tip speed the apparent velocity at the tip will be acute to the direction of travel and likewise the lift component but the small component generating torque is moving at extremely high velocity so producing heaps of power. Also the drag component is quite small and is at an acute angle to the direction of travel.

I have not seen any examples of properly designed boat systems that show designers understood what they were doing. They are usually a combination of a standard wind turbine and a standard water propeller. Neither are the best for best overall performance.

My interest in propellers grew from designing and building high performance pedal boats. My best design gets 12kph with 150W input. (When the engine is small efficiency is paramount.) This link shows one of my prop designs:
http://www.adventuresofgreg.com/HPB/uploaded_images/P9010014-783297.JPG
This one achieves an efficiency of 86% at design conditions. I can get slightly higher efficiency with a decent size air turbine; up around 88% at windspeed 3m/s and above. So plug these numbers into the above equation and see what is possible with a hull drag of 38N at 12kph where boat drag equals Constant x speed^2.

Rick W.

 

MP
I looked at a 5-bladed turbine of 1.5m (5ft) diameter designed for 10m/s (22mph) apparent wind with 100mm maximum chord blades.

The best speed is around 800rpm for these conditions. You should be able to extract 630W from it at this speed at 79% efficiency. The pitch curve is attached.

The foil I based this on is a NACA4410-44. It would be harder to model your tube section but if it is set up properly it could work quite well.

The tip speed ratio is 6 so it is a little lower than I expected. The efficiency is better at lower speed but the power extraction gets to be very low at low speed.

The drag on the turbine is 105N at 10m/s apparent wind.

Some rough numbers for the system are:

1. Bike windage at 22mph is 60N (including trailer wheels and frame). So total drag is 165N assuming negligible rolling resistance.

2. The power to the trailer wheels is 630W less mech losses - say 570W.

3. The bike speed over the ground will be 570/165 = 3.4m/s (7.6mph).

4. The true windspeed for this condition will be 10m/s minus 3.4m/s equals 6.6m/s (15mph)

So in 15 to 20mph winds you could expect something like 8 to 10mph with your turbine.

The turbine I have described would need to be geared 2.7:1 to achieve this using 27"bike wheels.

The reason for the relatively poor performance is the high wind drag of a standard cycle. If the rider was in a recumbent style seating performance would be better.

Rick W

 

The efficiency of a duct.

Rick,

Your concern about the duct drag is a little pessimistic. I once made a soapbox car for my daughter years ago that had a monstrous venturi duct as the body, with a frontal area 8 sq. ft., and it performed as well as tiny little cars that had about 1 sq. ft. area. A properly designed venturi is a net zero device, where the forces of flow are balanced between the inlet and diffuser. In effect, the car I made was a giant Laval Nozzle similar to a rocket engine. I have added some diagrams a pictures to illustrate what this contraption looked like. Also, the flow plot shows the conditions in the duct at the point that the throat starts to go sonic. Notice that the external flow is almost completely undisturbed.

 

Preliminary bike data.

Rick,

Thanks for the preliminary estimate for the bike experiment. I will use it guage how predictions stack up to real world conditions. Everything is on track to take it out to the desert next weekend the 14th of Sept.

As I was playing around with the problem of high tip speeds and steep pitch angle causing a detrimental thrust component, I had a wacky idea come to mind, and would be curious to hear what you think of it. Let me try and explain my reasoning.... first, I agree that the amount of power produced by the tips is huge, however, if it creates a component of thrust opposing the motion of the vehicle to the support pylon, then it is wasted output. If you look at vector diagrams of the lift vectors, and how they resolve into tangential and axial components, it is clear that at some point they are equal, and further increases in pitch are then counterproductive for a wind powered turbine/propeller combination. Long before that point you are getting diminishing returns. Any axial component of thrust has to be overcome by the drivetrain, which has efficiency issues. The source of the problem is the inevitable consequence of creating power via a rotating device that has long blades that have points that must travel along different helical paths, which produce force components that oppose our goal of producing pure output torque. My idea is, why not nullify this useless component directly at the turbine by making the tip of the turbine, at the point where the pitch flattens out, into a propeller surface instead that works in opposition to axial thrust component. This would entail no mechanical losses, as it is part of the same airfoil structure as the turbine, and it is flying in the best possible attitude, namely almost in a plane perpendicular to the turbine axle, at high tip speeds. The only loss would be the viscous drag on the foil, which is small. To anticipate your response, yes, it will draw power from the output to drive this tip, but better to do it at the source of the problem, rather than through the drive system with all the increased capacity necessary to handle this useless piece of dead ended energy. I carved one out of my choice prototyping material (schedule 40 pipe), and have attached some pics for you.

Stop laughing, it's a serous idea!

 

Axial Forces

Rick,

Just wanted to clarify what I meant by the buildup of the axial forces in a wind turbine blade as the tip velocity increases. In my diagram I show a typical foil at various radial positions along the blade, and the pitch angle at each point. All sections are considered moving with a zero degree angle of attack to the relative wind for simplicity. To keep it simple, we resolve the basic lift and drag, which at zero degrees pitch align along the tangential and axial axes perfectly, into a resultant force. Higher up on the blade, at 30 degrees for instance, it is clear to see that the resultant, when projected on the tangential and axial axes (green vectors), is transfering force increasingly into the axial direction. At 60 degrees, the axial component has swamped the tangential, even though the magnitude of the tangential vector may be many times larger than at 30 degrees. In a stationary wind turbine this is of little to no consequence, other than requiring a very robust support pylon, since in that situation the entire earth is the anchor. However, in a wind powered vessel the anchor consists of the thrust created by the underwater propeller, which in turn is derived from the output of the wind turbine. This explains some of the disappointing performance of upwind runs of turbines, and needs to be addressed if they are to take advantage of the greater energy available directly windward.