 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 : Yet another axial
  Page 6 of 8   |
|||||
| Author | Message | ||||
ChrisOlson Regular Member  Joined: 19/01/2010 Location: United StatesPosts: 60 |
Unless the efficiency is very poor (which would mean a lot of heat dissipated), 300-500 watts would not develop sufficient braking torque to keep a 3 meter rotor from over speeding. Air core generators do not have any reactance limiting, so the harder you spin them the higher the open voltage goes, and the more power they make. With a geared drive on a 3 meter machine it's possible you might be able to select the right gear ratio to get this generator up to about 1 kW in 24 volt delta configuration. Then you'd have sufficient braking torque for the rotor as long as you didn't run out of ampacity in the winding and have the efficiency drop below about 70%. However, with your 12 volt wye internal resistance at 1.2 ohm, I have a feeling you're going to find it hard to keep the power efficiency above 70% in 24 volt configuration. For a 3 meter 24 volt machine, you really need the internal resistance down around .5-.6 ohm to get decent performance with an air gap generator. -- Chris off-grid in Northern Wisconsin, USA |
||||
Gizmo Admin Group  Joined: 05/06/2004 Location: AustraliaPosts: 5040 |
I might do a few tests running the alternator in 24v configuration star, on a 12v battery. This will reduce the cut in, bad, but then the ceramics will slip compared to neo's, so maybe it wont stall the turbine. The phase resistance would be 4.8ohm. I see your point about the resistance Chris, I dont want this thing to cook, and dont expect anywhere like those power levels. It's only a small alternator, so I'm trying to see what I can get out of it. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| ChrisOlson Regular Member Joined: 19/01/2010 Location: United StatesPosts: 60 |
Hi Glenn, The ferrites might not "slip" as much as you think. I thought that too. But my test turbine with the ferrite generator topped out at 137.93 amps yesterday in a storm, when the gust front hit. It ran from 70-85 amps for about a half hour during the storm and generated 1.16 kWh during that 30 or so minutes. That's right on par with the neo generator for the same turbine. When I first tested my generator in single phase configuration, it acted like it "slipped" and let the turbine spin faster. But that was simply because the single phase was not powerful enough to provide enough braking torque. I thought it might just go up to a certain output and not go over that - not so, as I found out later!! This is not an iron core machine and these air core units will surprise you. You get more flux thru the coils with an air core axial with none of the losses of an iron core machine. I've found, just like the neo units, the faster you turn it the more power it will make and it never "slips". I suspect when you get enough power to the shaft on your test setup, you're going to discover the same thing. Edit to add some info: I should clarify this to say, Glenn, that eventually you'll run up against the wall of internal resistance in the winding. Internal resistance is the only limiting factor in an air core axial. The higher internal resistance will make it act like it "slips" but in reality all you're doing at that point is loading the winding under 50% power efficiency and you'll burn it up at any sustained power levels at that efficiency. You need to develop an ampacity rating for the winding for a particular voltage and winding configuration (wye, delta or IRP). Then test it up to that amapacity rating to develop a power curve for it. The power curve will flatten considerably after the peak capacity of the winding is reached. But at that point you're converting more of your shaft input power to heat than output, and the braking torque is still available to the rotor. How long it will take it depends on cooling ability and some other factors. -- Chris off-grid in Northern Wisconsin, USA |
||||
Tinker Guru  Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Glenn, when you're done with the bench testing and before you mount it up a pole do yourself a favor and paint the plywood as well. Clear epoxy coatings break down easily in the sun's UV light. It crumbles from underneath and exposes the raw plywood then. Unless you have a hood over the lot or paint on it. Klaus |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5040 |
Thanks for the tip Tinker. This weekend I hope to use a friends lathe to do a full set of rpm tests. Thanks for the offer to use yours Bob, but this one is a lot closer to home  , about 5 minutes away. Once done I can decide how I will use this alternator.
Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Glenn No Worries mate good to see you conserving fuel and helping the environment, offer is open any time you need it though. All the best Bob Foolin Around |
||||
scoraigwind Newbie  Joined: 23/09/2009 Location: United KingdomPosts: 21 |
Hi Glenn, I've been reading through the story with interest. I must say it is very beautiful but an awful lot of work. I suppose I am lazy but I like to find the easiest and cheapest way to do things. You can predict the cut in rpm with reasonable accuracy using equations like the ones in the back of my Recipe Book. In your case the flux density works out around 300mTesla for 170 rpm cut in at 12 volts. That's quite a high density for ferrites. Working flux density is always less than the 'Br' or 'remanent flux' quoted in the magnet specifications. I wonder how you plan to prevent those magnet stacks from flying off - is it with plywood really? thanks for sharing your project. Have fun Hugh Piggott, Scoraig Hugh Piggott |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5040 |
Hi Hugh, thanks for the comments. I chose those magnets because they had the highest "Pull" figures compared to several other similar priced magnets. Its a 22mm diameter 25mm long cylinder with a pull figure of 2kg, for $1.10. Similar sized magnets were a 25mm square by 10mm tall magnet with 1.3kg pull for $2, or a 30mm diameter 10mm tall cylinder with 1.15kg pull for $1.25. I figured the height was giving it its higher strength, the further apart the poles the better. At only 22mm diameter, I could fit more around the circumference, a 16/12 arangement instead of 12/9. That means a higher frequency, more coils cut per revolution. I went for the side by side arrangement to give a effective 44mm by 22mm pole, be it figure 8 shaped. The thinking is I could put another two magnets further out, to give a 88*22mm pole, doubling the power per rpm, for a alternator that was 88mm wider in diameter, but that was for the next build. The thing is, now that I'm starting to get some figures, I dont know what I should be comparing it to, or if its a success or not!? Putting aside all the other costs, which would be pretty standard be it neo magnets based or ceramic magnet based, I think there are two costs to take into account, copper and magnets. There is 1.5kg of copper, for $65, and $70 in magnets, but that was closer to $100 delivered. So magnets and copper cost about $170. It looks like I should get 150 watts at 500rpm, with a fast set of blades and good winds, but I need to really test at higher RPM's to be sure. I dont know how well the money spent in copper and magnets would compare to a neo based alternator with the same budget, or even if thats a fair comparison. I did try a whole bunch of ideas on this alternator, and I'm pretty happy with how they worked. The CNC mould does take more time, but the results are very good. If I used a jigsaw and some ply wood I could have made the mould in 10 minutes that would have worked just as well, but I had the CNC tool, so decided to try it out at mould making. I'm also happy with the alternator layout, its easier to assemble than the conventional layout, and I can adjust magnet spacing and stator position easily. So I think its been a success, but its hard to say. The plywood, yes I'm sure it will be fine. There is no weight in the magnets, the plywood has been painted with epoxy and its set hard, and each magnet is held in place with more epoxy, the whole thing is glued together, plus the outside spacing bolts keep it all in place if the epoxy to metal bond fails. Also the magnets cant tilt up, the holes in the plywood are tight. But I will be painting it before it sees any service out side. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
Rastus Guru Joined: 29/10/2010 Location: AustraliaPosts: 301 |
Hi Glenn, Harking back to earlier comments in the thread,the mill head is a great success! It provides a viable alternative to the more exspensive options,includes improved innovations,and for those who can't afford the Rolls Royce they can go for the rolls can'ardly figuratively speaking.Cheers Rastus see Rastus graduate advise generously |
||||
| scoraigwind Newbie Joined: 23/09/2009 Location: United KingdomPosts: 21 |
I suppose that's always going to end up a matter of opinion. Personally I like to build the alternator after the bearings are already mounted. I find it easier to adjust the stator position using external mounting studs (which are also less stressed) and I just use a identical stacks of nuts and washers to do the spacing of the rotors, which usually works first time, but it only takes a few minutes to jack the front rotor back off and add or remove a washer if I decide it's necessary. I am moving more and more toward allowing a larger clearance gap for better cooling and peace of mind. But maybe 1 mm is not so tight on a very small alternator like this. Overall this kind of project pays off much better as you go up in size (since the work is similar but results are many times higher power.) I would say that a single rectangular or round magnet is more efficient than a pair in terms of filling the coil with flux, but obviously commercial factors - what you can get at what price - do weigh in there.) With neos it's not really feasible to place them in pairs like that anyway. And I see the logic of the series/parallel trick (although I always avoid parallel connections so as to minimise parasitic currents). The fancy coil mould and the use of epoxy in the coils seems to add a fair bit of work but also improves the cooling I imagine. It otherwise seems like a solution to a problem that doesn't exist since stator castings are one thing I have never had an issue with. Ferrite magnets flying off was a much bigger issue for me. I solved that once and for all by embedding wire in the outside of the magnet rotor casting. Nowadays with neo magnets I submerge the magnets in resin and put glass on top, but wire around the periphery works nicely for big ferrites, as do the little steel spurs welded on that Chris uses, and the lovely stainless steel hoops at Otherpower. I had assumed that welding steel spurs onto the outside of the rotor would rob some flux but Chris has found that in practice it only robs flux that was leaking anyway, so it's a useful solution. The best thing about these discussion boards is the way we can see lovely photos of all the different ways to do things - and they mostly seem to work very well. Which way you choose will depend on whether you have a machine tool shop, CNC router, etc, and what you personally find attractive to do. Hugh Piggott |
||||
| ChrisOlson Regular Member Joined: 19/01/2010 Location: United StatesPosts: 60 |
I plugged Glenn's results into my spreadsheet and he's getting exceptional flux density in the air gap with his setup. I'm running the air gaps at 17mm on one stator and 20mm on the other one, and I'm getting only about 2/3's of what he's getting (around .2mT). I too have gone to running a wider air gap and either throwing more speed at it, or a few more turns of wire to prevent contact between the stator and rotating assembly in operation. I like Glenn's design but on a bigger and more powerful unit I would be worried about the mechanical stress on the stator mounting it in the middle. The torque on the mount is substantial, as it is, when you mount the stator on the outside. When you get to 2 kW of power output at the stator @ 70% efficiency the stator assembly is providing about 67 lb-ft of braking torque to the rotating assembly @ 300 rpm. I could work out the stresses on the stator casting and mount when it is mounted on the inside like that, but I suspect that the mechanical working limits of a fiberglass casting would be exceeded at that power level if it were mounted on the inside, unless some sort of reinforcement was added to it. -- Chris off-grid in Northern Wisconsin, USA |
||||
mac46 Guru Joined: 07/02/2008 Location: United StatesPosts: 412 |
Hello Glen, I can see that there are a few folks that are picking you'r project apart. Thats a good thing I guess,,,else we would'nt learn a thing. I for one and pretty proud of the creativity and craftsmanship you've demonstrated. I'm not up to speed with these mills and this project has really helped me to understand things better. Keep up the good work Glen. ....Mac46 I'm just a farmer |
||||
| ChrisOlson Regular Member Joined: 19/01/2010 Location: United StatesPosts: 60 |
Few folks are building generators with ferrite magnets so Glenn's project is quite interesting. I'd like to find out more about those magnets he's using because he's getting more flux density in the air gap than my gauss meter showed my ferrite blocks had at the surface of the magnet. I used C5 blocks and they're supposed to be about .35 mTesla in the core of the magnet, which is about 1/3 of what a N42 neo is. And that's about what I got in the air gap - 1/3 of what the neos are. I used a larger (50mm x 50mm x 25mm thick) block that has twice the surface area, and twice the thickness of the neo magnets I normally use, so I ended up not having to use three times the turns of wire. But Glenn's design is quite amazing using the two magnets per generator pole. -- Chris off-grid in Northern Wisconsin, USA |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5040 |
Hey Mac. I dont see it as picking the project apart at all, its all good discussion. Besides Chris and especially Hugh have got a lot of experience behind their comments. I measured the gauss of a single spare ceramic magnet Chris, it comes in at 1340 gauss. I intentionally spaced the magnets apart by about the same distance as the magnet gap across the stator. In my case, the magnets are about 11mm apart at their closest point. The way I see it, any flux that takes a path from one magnet to its neighbor is wasted, cause its not passing through the stator. So by spacing the magnets apart, I hope that most of a poles flux will pass through the stator to its opposite pole on the magnet on the other plate, and not the magnet next door. I know a lot of builders place the magnets next to each other, but its just my thinking that they should be apart. Regarding the stator mounting points Hugh. I do understand there will a lot of torque at this point. There is a 3mm steel plate on each side of the stator, bolted through the stator with 6 bolts. The stator is thicker in the center, 15mm, and filled with fibreglass mat. I dont think this area will cause me any problems. The three positioning threads could be a problem, but this is only a test alternator and I would normally have these a lot shorter and use 16mm threads. Time will tell, there are a few ways I could strengthen it. I made another test today, 145w at 550rpm 12v star. Ran it for a couple of minutes at that output then felt the coils. The outside loop was cold, but I could feel some warmth to wards the inside part of the loop where cooling would be less. Maybe 25c, ambient is about 15c. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5040 |
Link to the magnet I used. http://aussiemagnets.com.au/product/-22-x--25mm-Rod-%28Ferri te%29.html I see its rated at 1800 gauss. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| scoraigwind Newbie Joined: 23/09/2009 Location: United KingdomPosts: 21 |
The ones that I use are listed here http://aussiemagnets.com.au/product/-50mm--x--50mm--x--20mm- -Block--%28Ferrite%29-.html I pay 0.60 US (60 cents) for these. Measurements of flux density depend entirely on how they are set up. It varies with the magnetic circuit so it's a bit meaningless. A long magnet and a small gap does create a higher flux density. Hugh Piggott |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5040 |
http://aussiemagnets.com.au/product/-50mm--x--50mm--x--20mm- -Block--%28Ferrite%29-.html The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Hugh I notice you pay 60 cents for these magnets but we are charged $6 in the southland is this a typo or are you able to source at that much better price. All the best Bob Foolin Around |
||||
| ChrisOlson Regular Member Joined: 19/01/2010 Location: United StatesPosts: 60 |
Yes, I suspect you're right. I had to "scrunch" my magnets closer together to get 10 poles to fit on 280mm rotors and I may be losing a bit of magnetic performance at the inside corners where the magnets are about 8mm apart. See the attached photo of the center rotor to visualize what I mean. I know another fellow from Ireland that built a 5 meter machine using two 50mm x 25mm bars arranged in a "tee" configuration for each generator pole. He also got good results by using more than one magnet per pole. -- Chris 
off-grid in Northern Wisconsin, USA |
||||
| scoraigwind Newbie Joined: 23/09/2009 Location: United KingdomPosts: 21 |
No this was not a typo but I got that price in the context of a $1500 deal with a chinese company. They shipped the magnets by sea, which saved a lot of money and only took a month or so. Hugh Piggott |
||||
| Page 6 of 8 |
|||||
 Print this page |
