 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Windmills : - NEW ALUMINIUM BLADES -
  Page 5 of 7   |
|||||
| Author | Message | ||||
oztules Guru  Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Pete, you don't see pvc pipe doing 3kw. The Al needs support to protect it from fatigue. It can help stiffen the steel rod to some extent, and maybe damp some oscillations. I still think solid from start to finish is inside the safety limits.... regardless of the extra weight it brings. That only figures in the centrifugal force... and it is easily strong enough. Solid steel (mild) can see off 60000 psi to break, and 40000psi to deform. The centrifugal force in this case (with extra weight all the way) should be less than 12000lbs. so room to spare. The bending should still be the 265 lbs that Perry has calculated, that does not change with the weight... and we now have extra strength... all good. I have never been able to bend 1" steel rod with only .875mm leverage, and I don't expect the mill to either. It will flex, but I don't see it going over 40000 psi at the hub to get to deformation stage. The extra weight will slow the yaw down as well, helping to keep it tamer in the gusts and yaw conditions. The same turning (yaw) force into a bigger load will mean slower response as well....... I think anyway. Maybe Perry will wish to comment otherwise.... I can't begin to work out how to deal with the yaw calculations on a blade, but the oscillating nature of the beast will be a killer for aluminium or any other material that cannot stand fatigue well. ............oztules  Village idiot...or... just another hack out of his depth |
||||
Perry Senior Member  Joined: 19/11/2009 Location: Posts: 190 |
Hello Oztules, I think you may actually have a little overkill with a solid rod all the way through. Actually when you are dealing with these loads and geometries, the center of the rod contributes very little besides weight. A thick walled tube is just as strong. What if we used a 1" sched 40 pipe that runs the length of the blade with 125mm sticking out of the blade for attachment. This gets rid of all the plug welding and stress concentration areas which are problem areas as well. The pipe is readily available and cheaper than rod. I ran through the calc's as I outlined in previous posts and achieved the following results. 1. The centrifugal and bending moment forces are highest right where the pipe attaches to the hub. 2. centrifugal forces here = 194 MPa (28.1 ksi) 3. Bending moment from thrust forces here = 73.8 MPa (10.7ksi) 4. The highest combined force is 267 MPa (38.7 ksi) So there you have it. By this model a plain old schedule 40 pipe (I used SS304 for the calc's)that runs the length of the blade incurs stresses that are well within the capabilities of many materials. No need for the rod, thin walled tubing, plug welding, stress risers, etc. And simpler is better. I would probably recommend the pvc extrusions for this particular design. Without knowing I would assume it is lighter than the aluminum. You would have no galvanic problems and you have some good attachment options. SEE THE FOLLOW UP POST BELOW |
||||
SparWeb Senior Member Joined: 17/04/2008 Location: CanadaPosts: 196 |
Thank you Perry, For your detailed explanation and thoroughness. The numbers bring reality into focus, and provides a rational basis for making choices. I expect safer machines will be put into service following your work. (Somebody owes you a debt for your professional services.) [quote]Perry: P.S. please check all this over! [/quote] Yep. I do my calculations in a different order, but I get nearly exactly the same thing. I used the same assumptions regarding Betz limit and delta-V=1/3 in that condition. The one thing I would do differently is apply the thrust at 3/4 of the blade's span, not 1/2. The reason for this is the greater speed of the blade at the tip makes the thrust force ramp up to the tip. With the thrust applied farther out the blade, the bending moment is, in my opinion, worse than even what you've shown. Your result proves your point more than adequately, so a detailed re-calculation isn't necessary. Thank you for offering a solution, too. Nobody should feel sh** upon when you use your work to validate the solution, too. Armed with more detailed information about the stresses in the blade, we can work out what is necessary to attach the blade to the tube. I've seen various riveting schemes, which seem suitable.... The tensile load ramps up to about 5000 pounds at the hub, but in normal service it will be more like 2000 pounds or so. A rivet driven through plastic will only support a load equal the the force required to crush the plastic. I would expect a 1/8" rivet to hold only about 100 pounds or so before the hole starts to stretch. Distribute that 2000 pounds among rivets that can handle 100 pounds each, then you need at least 20 rivets. To satisfy the runaway condition you might use up to 50. With one rivet row on the front and one on the back, each face of the blade would have up to 25 rivets. The blade is around 60 inches long, therefore one rivet every 3 inches would give you 20 per side and that would be plenty for all but the worst conditions. The aluminum blades will offer a better rivet joint strength than plastic, so you could get away with one every 4-5 inches or so. [quote]Oztules: I have never been able to bend 1" steel rod with only .875mm leverage, and I don't expect the mill to either. It will flex, but I don't see it going over 40000 psi at the hub to get to deformation stage. [/quote] Bear in mind that the hub itself isn't perfecly rigid. The blades will bend and stretch whatever hub plate you use. When you say that the bar/tube attaching the blade to the hub has so much bending moment and tensile force on it, that's all transferred into the hub. The stress then carries through into hub plate or disk, which has to hold it all together. The bending that comes from thrust and gyroscopic action is even more insidious, because it also loads up the fitting on the shaft, to the shaft itself, then the bearings, and so on. All of this analysis should put the use of a "4000 pounds trailer hub" in context. Steven T. Fahey |
||||
| Perry Senior Member Joined: 19/11/2009 Location: Posts: 190 |
Hmmmm, I was double checking my work on the problem above where I proposed the sched 40 pipe and I think I messed up the bending stresses at the hub. Steven, can you help me out here. Given: thrust force on the blade is 1175 Newtons applied at the 3/4 of the blade point as suggested. So d = 1.25 m from the center of the hub (assumed mounting point) So at that point the bending moment is; M = Force x distance M = (1175 N)(1.25m) = 1469 N-m For the round mounting tube the stress is given by; bending stress = M*c /I with c being the radius of the tube For a sched 80 pipe the moment of inertia is I = 1.603x10-8 so stress equals; bending stress = (1469 N-m)(.0133m)/(1.603x10-8 m^4) bending stress = 1218 GPa = 176 ksi It has long since failed. My question from this very elementary mechanics equation is; 'Are these blades mountable by any means to survive a 65 mph 1000 rpm event?' Perry |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
Oz, I dont disagree 100% with you, but the thought of 3 crowbars whirling around up there is a bit scary. And god forbid if one came off, it would take out a low flying aircraft or harpoon a whale a mile away. I do understand your point on strength and no welded sections etc and agree on these points 100%. Perry, I also agree with a heavy wall tubing and had thought of a seamless tube/pipe similar to used in hydraulic systems. Where as nominal bore pipe do have variations in wall thickness and could effect strength and balancing to a small degree. Im not convinced on sch-40 as a high enough wall thickness and would aim for at least 6mm wall or greater. The other advantage to pipe it allows for rivets to be used easier compared to solid rod. Is the design for a 3kw mill at 1000rpm a tad overkill for the average F&P mill considering it may only produce 0.5 kw and at 1000rpm it would be out of control and likely to fail else where anyways. The other point i was trying to raise was a discussion on alternative mounting options outside the present method. What is the strength of the blade over 1.2 mtr without any centre support.? What is the weight of a metre of the profile? There is a lot of clever minds here and surely we can come up with a better solution than 3 crow bars, and if not at least some thoughts have been put forward and worked through. Pete. Sometimes it just works |
||||
KarlJ Guru Joined: 19/05/2008 Location: AustraliaPosts: 1178 |
I like the comparison to the PVC pipe type blades.. the moral of this story to me is a 1000rpm event is not only rare but few if any solutions exist to combat it. I think the math on the loading is in error based on this Its fully furled ie turned out of the wind 75degrees (a realistic furling as tail may not be able to go the full 90deg to the wind, due to risk of blade contact) Thus to me in this case your 1000rpm will not actually see the betz limit load by a fair margin ie perhaps 20% of the limit thus reducing the thrust load considerably. AND... as for the SS tubes..... Luck favours the well prepared |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Perry, I don't mean to make a bad case worse, but 1" schedule 40 pipe would have a 1.3" diameter.... did you do the calcs for 3/4"schedule 40 (1.05 " outer diameter) or the 1" schedule 40 (1.3" diameter). the hole in the aluminum is only 1" I think.... so that would be 3/4 pipe? Solid. I know won't set the world on fire from a bending perspective, but the extra weight won't harm either. It does help in the breaking strain caused from centrifugal force, and I still think it will slow down the yaw considerably.... which should help the figures for yawing.... interested to know your thoughts on this part. With your new figures from your last post, not knowing the moment of inertia for 1" solid, I can't know how the solid would cope. I know they won't leave the scene, but will they deform?.....I'll stick to wood I think. Downwind... they won't break, and if mounted sensibly, won't leave the scene. From Perry's last post, they may bend though.... will wait on his response for that.... however.... I agree with Karl, that to do 1000 rpm, it would be an F&P, an axial will have furled and hold the rpm down or if unable to furl... probably stop with a burnt up shorted stator at that rpm and stop the mill.... In furl the F&P thrust forces will not be anywhere near as great as the worst case scenario being calculated for... it is only capable of small power dissipation.. and will free wheel after the armature reaction becomes significant... so low thrust in any scenario. An axfx will soak up significant thrust though.... but for other reasons will not let it happen. In short, to get to 1000rpm, thrust will be low. centrifugal will be high. Perhaps, someone (Sparweb springs to mind) can calculate at what speed these blades would stall under free wheel... they do have a decent chord. .............oztules Village idiot...or... just another hack out of his depth |
||||
| Perry Senior Member Joined: 19/11/2009 Location: Posts: 190 |
OK, I have analyzed this problem long enough. As you can see, I have put some work in. I am not associated with Phill. I have never owned these blades. I did try to buy some but the e-mail form never worked. I am interested in them because the concept seems really cool. An easily customizable blade that you could work with and never have to carve an ounce of wood. Plus all the cool pitch control stuff you could work out. I really want them to work. Unfortunately I think that the math proves that in no configuration can they be mounted to absorb a 65 mph 1000 rpm event. As you saw in my models I started with the centripetal forces. Then moved on to the bending stress calc's. The bending stress calc's are not dependent on blade mass so it doesn't matter whether the blade weighs as much as a feather or a SUV, carbon fiber or stainless steel. Those equations show that under 65 mph/1000 rpm conditions the mounting tubes/rods will see over 753 MPa or 109 ksi of tensile stress where they mount to the hub. None of these materials come close Simply stated, 1 inch diameter solid rod is too weak to absorb even the bending loads at these conditions. Sorry to have to say it but the math is painfully obvious no matter how hard I would like them to work. I will probably take some flack for this but I am just trying to help people design safe turbines So, where does this leave us? Is it all over for the extruded blades? I think that for these conditions, yes. So maybe we should re-examine the boundary conditions of 65mph and 1000 rpm. In fact you can throw out the centrifugal forces as they are not as high as the bending moment forces. You would have to drop the spec on the highest survivable wind speed. Maybe it turns into 30 mph. Is this good enough. Probably not. Perry |
||||
fillm Guru Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Hi Perry , Once again , THANKS , for all your time and effort you have put into these caculations up to this point , and I am sure that I am speaking for all who are reading this thread . As I had previously said , I just plucked numbers from what I thought would be a severe senero , in actuall fact the highest rpms I have seen is 770 on the dual stator F&P , its dia is 3.15m , blade length of 1.4m , in that situation it was out of control because saturation is reached at about 500-600Rpm and then the torque drops of as the blades speed up with almost no extra output . The same set of blades on the Ax Fx reach a max output of 3.2Kw @ 550rpm in 50-60klm wind , which points out the inefficeinties in the F&P not being able to controll a set of blades of that size , 2.8 - 3m would be the max for a F&P dual depending on your wind zone . As I have said previously , these blades are only 100mm less in dia than the failed set and the s/s tubes are on 150mm steel solids and have copped more than the set that failed , but do not have plug welds on the transition point . I will not build again on such short solids as the leverage ratio is to high . Getting back to weights and materials , the aluminium section is slightly lighter than the PVC , and in my way of thinking and with what you have said , keeping the rotating mass to as light as safely possible will give the best survival margins in strong winds as well as ensuring that furling works at the desired wind speed and output. My ideas of going down the path of having no welding on the loaded areas, this reduces the points where a failure can start and is something I feel strongly about after the recent failure due to that fact and the loads exerted in that area . I feel that using S/S tube has its benifits in the fact that it does have a higher tensile strength than mild steel tube and when combined with the aluminium section with well fitted rivets forms a rigid structure in the middle of the blade skin which is also a convex and concave structure of aluminium that somewhere would be figured into the equasion . Galvanic corrision has been mentioned a few times ,in what time frame would this start to occur and what conditions accelerate this process , I dont see this as being of a major threat as perodic inspections should be carried out as you would with any blades . By keeping the inner end caps of and allowing moisture to be dried by some air flow and heat and centrifugal force might also be a solution . I will be back at home in a few days and will be straight into building the new blade tubes and mounts , I will also be able to give some more accurate weight figures , material specs and lengths as well as do some sheer tests on the aluminium opposed to PVC . Hopefully out of all this we will be able to have a design that is not to expensive to build and sorced out of readily available materials , when combined with proper design of a furling system and correctly sized for the load it is intended to power , will be able to give many years of safe reliable sevice . PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
| Perry Senior Member Joined: 19/11/2009 Location: Posts: 190 |
For the original calc I used 3/4" as you pointed out, 1.05" OD. Later calc's I just used 1" OD solid rod. This is true. My model does not take furling into account. All this is based on a non furling turbine, facing into the wind and converting the wind power to reactant forces. In other words, pointing into the wind under power. I think this is the condition we are worried about, right? Perry Perry |
||||
Greenbelt Guru Joined: 11/01/2009 Location: United StatesPosts: 566 |
The Chain Saw is likely the best tool in the shed. A wood blade properly dried and sealed with resin or Urethane can hold its own with metal. Fom a man with experience. http://www.pipercubforum.com/woodprop.htm I realize this is off subject but related. Most of you know that a wood blade will serve well when cared for, It has the qualities to stand up under repeated abuse. A tree after all is conditioned to the fury of nature and evolved to withstand the twisting and bending. It can fail but not fatigue. My post is directed to the wind power Newbies. Time has proven that I am blind to the Obvious, some of the above may be True? |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Perry, You have done very well. The only flaw is in the assumptions. The boundary conditions exist, and you have shown that the blades properly mounted (shedule 40 will do I suspect) will survive..... what am I saying??? Ok, now the dust is settled, no-one I know has an alternator capable of absorbing the 40000 watts that your scenario would require. So furled or otherwise, about 6kw is the maximum that the blades would be required to deliver as a load.... so maybe 120 kg for the whole blades set.... 40kg per blade thrust. This would be reached about 30mph or so. After that, without furling, the axfx would burn out to zero resistance and stop the thing, and the F&P would simply free wheel as it would have done from about 25mph or so at about 40-50kg per blade set.... only 13 or more kg per blade. This brings the things into fairly easy territory to deal with, and back in from the cold. After 30 mph, the things will be trying to furl and it is it is only the centrifugal forces that count as iron cored machines run away (the high RPM Phil saw with the dual F&P... but low power output), the thrust will not increase substantively, as there is no-one using the torque.....At 65 mph, there will be 35000 watts not being utilised. I don't know how much thrust is used up beating the drag for an unloaded prop at 65mph for an unloaded, unfurled machine.... but not 35kw, just the drag of the blades trying to get away from the wind.... and some wicked centrifugal forces. Job well done. ..............oztules Village idiot...or... just another hack out of his depth |
||||
| Perry Senior Member Joined: 19/11/2009 Location: Posts: 190 |
Thanks Oz, It has been fun and I'd like to help find a solution to use these blades. I will take a look at your thrust force assumptions in the morning and see what the numbers say. What you are saying is that I should back off from the Betz limit and lower the wind speed to when furling begins? I think the original boundary conditions are excessively harsh. I don't know what blade can survive 1000 rpm in 65 mph wind with no furling. More to come, Perry |
||||
| Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
I found this lurking in the depths of my computer. It an article on calculating blade efficiency and may offer some ideas. I dont know the source i got it from or the creditability of the information. Perry, you are what i mean by the "clever minds" here. Thanks for the time and information you have given. 2009-11-30_191311_Factors_to_consider_when_evalutating_small _wind_generators….pdf No matter how well you design something, the wind is a powerfull beast. There is always the unexpected. WHOOPS! 
Pete. Sometimes it just works |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Perry, This story from Phil shows that Betz is a long way from the truth. This was a dual stator, and even then only put out 735 watts at 750 rpm.... it would still get 700 watts or thereabouts at 500 odd rpm. If it were single stator, then increase the rpm, and halve the watts. If we are not using the power, then the blades must not see it.... at 1000 rpm, it will still be only about 750W, or 370w for a single stator perhaps. So the 1000rpm is very real possibility for the F&P, but not more than 500w for a single stator. Thrust must be correspondingly small too I suspect. The multi tens of kilowatts that you have calculated the thrust for are only a dream for most of us I'm afraid. ...........oztules Village idiot...or... just another hack out of his depth |
||||
| Perry Senior Member Joined: 19/11/2009 Location: Posts: 190 |
sorry, all my calc's in this post were wrong so I edited it all out. Perry |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Oh dear, the engineer v's the idiot... don't look good for me. However, the power in only a 31 mph column of air with a cp of .3 and a diameter of 3.2 comes out at around 4000 w, and for a 3.6m set (which I think I was using the calcs on ) is about 5000watts. thrust has been calculated as 86kg and 108kg respectively. At twice the wind speed we have 8 times the power... so 32kw and 40 kw respectively and about 4 times the thrust on the blades. This is for a system 1/2 the CP you are using. This is true to life as well I can muster about 5000w into a load in a roughly 33 mph wind with my 4m set. Phil was able to get over 3000 watts with a wind speed of only 57kmh with only 3.1m blades... here If we double the windspeed we get 8 times the power in the wind, and although I can't measure it, one must assume that if I could use the power, it would be at least 8 x 5000w in my case... so 40000 watts... assuming alternator and blade inefficiencies remain the same... the thrust would be 4 times as well or 155kgs would climb to 600kgs????? 66mph. My losses would amount to well over 50%. I would think. I can't explain why we are different, but a visit to this page here will give you an indication of my thinking on this. It appears to come true in practice as well. We are measuring only the output power into battery banks, and loads, so our inefficiencies are way way above your ideal CP=.6 with no losses. One can only guess at what they calculate out too without the alternator losses included. (edit: for a 3.2m @ 57kph (approx 16m/s) and a cp=.6 we get over 12000 watts... phils actual was just over 3000w) What am I missing? are we talking different things here? If we talk only of thrust, or pressure of the air in the column, we are not talking power, only force, so I'm a little lost as to what power you have calculated. It seems to bear no resemblance to what I am talking about..... I seem to be not getting it 
........oztules Edit, I'm getting an uneasy feeling that the thrust may to be independent of the cp.... and the power output. I don't fathom this at all ... action-reaction seems not to matter?????? I will be interested in my upcoming lesson. Village idiot...or... just another hack out of his depth |
||||
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi oztules, the commercial size windmills >1MW quote efficiencies of approx 50% wind energy harvest. Gordon. become more energy aware |
||||
| Perry Senior Member Joined: 19/11/2009 Location: Posts: 190 |
Oz, Sorry you had to write that long reply to my last post. All the numbers were crap because I squared the velocity instead of cubing it. Nobodies perfect. The Cp does indeed influence the thrust force. If your rotor captures a certain percentage of the wind there has to be a reactionary force behind that. Energy has to be conserved. I found this thrust force calculator http://www.thebackshed.com/windmill/Docs/Furling.asp My numbers were really close to this. I think if we put realistic boundary conditions as discussed we will get better answers. Perry |
||||
| carl1 Regular Member  Joined: 16/04/2007 Location: AustraliaPosts: 79 |
Hi I did some efficiency test a while ago on my lathe: mechanical power input versus electrical power output F&P 80 12v delta: @ 250rpm 69% going slowly down to 64% @ 950rpm cheers Harald |
||||
| Page 5 of 7 |
|||||
 Print this page |
 |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2025 |
