 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 : OEM mills
Page 1 of 3   |
|||||
| Author | Message | ||||
norcold Guru  Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Purchased a couple from film, have them up and running . Have them on a 48v system with Cap-treblers. They cut in at 5km/hr (approx) with output at 1amp at 10k/hr. Great low wind performance. They have 51 fields but only 46 magnets, believe their performance could be improved by updating magnet numbers to 4 over any 3 field windings at any time as is "norm" for 3phase. However my experience here is limited to theory, film has a little trouble getting his head around this unusual arrangement too. Note also there is no space between magnets. Just wondering if the voltage would increase if this magnet arrangement were to be increased to the 4 over 3 rule, at same revs. Bob, have been playing with tower height from 10m to 12 m at my site output goes close to doubling. No logging as yet so this is very much an observation at random intervals but do know whilst I have heaps of room ( on farm) have a very turbulent wind site. Once I have logging will be able to give meaningfull figures. Believe these mills with the Al blades have much more potential, feel the magnet arrangement is holding them back. Am just not sure enough to warrant the expense of magnets, thus hoping for some insights. We come from the land downunder. Vic |
||||
| Xmaswiz Regular Member  Joined: 14/04/2011 Location: United StatesPosts: 69 |
look forward to updates! Santa Maria, CA. Noel |
||||
Gizmo Admin Group  Joined: 05/06/2004 Location: AustraliaPosts: 5078 |
I would leave the OEM motor as is. Yeah its an unusual layout, but there would be a reason behind it, like to make best use of the available magnets and space, and it does work very well both as a alternator and motor. I think if you change anything you'll end up with less, not more. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
fillm Guru Joined: 10/02/2007 Location: AustraliaPosts: 730 |
I would agree with Glenn , these motors are used in all kind of small EV aplications and like 99% of the stuff built in China , they are built to be mass produced as cheaply as possible for that aplication, would more than likley be the reason. I think trying to modify windings etc, would end up in disaster and the effort and cost may be better spent elsware. To date I have not seen to much development with these electric bike motors as wind generators, apart from what OEM is doing, which is basicly a few mods to put blades on them, being no different than a washing machine motor with blades except they have Neo magnets and can certinaly exceed their rated power to the point of destruction unlike a F&P which tops @350w max. Even in 48V I have logged peaks of 1300W in 40klm wind. I would also think that getting 1A @ 48V ( 54V on Batt Float ) = 54W in between 5 to 10Klm ( 2 to 3 M/S ) is not to bad . The amount of power available in low wind is very low and care has to be taken not to overload and stall the blades in low wind as this is detremental when the wind speed increases . As you know I have been playing with mine in 48V, with the Cap/Tripler & Quaddrupler in paralel with the doubler where I can switch them in and out, with the 1600uF caps it over loads to early for the 2.8m blades and for the small amount that is picked up in light wind is quickly lost in med wind. I will be trialing 680uF on the trippler so as not to load the alternator. Bob uses a different method with the voltage boosters, he could probably enlighten you more with them . I purchased a couple of 130w 10-32V-IN to 38-60V-OUT of ebay a while back to give them a try . I have not tried them on the mill but I did connect them to a 12v batt bank to dump power into the Grid Feed as a test, the amount of heat generated was well over 100deg C which I would think is a lot of wasted power with only 20-30W being feed into the grid. It would be good to see some pics of your set up now that it is all up and running . I am a firm beliver in the Cap/Doubler - Trippler - Quaddrupler for applications like these alternators with high cut-in ( low resistance ) windings , there is virtually no heat loss and the mill can be tuned with differing uF caps to get the optium blade loading. PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Phil I am still playing with mine and am totally impressed with its performance so much so that as soon as funds allow a second one will join it. I am now using two boost modules one that cuts in at 4 volts and conducts 3 amps to 27 volt charge, this happens around 2 MS. The second module cuts in at 11 volts and by 20 volts it is doing 9 amps this is around 4MS . At these power levels you need a lot more heat sink on the bigger module. The cap doubler is still hooked up at the moment and contributes 2 amps at 16 volts input but doesn't go higher than that possibly because the caps are a bit small. As the wind speed comes up the main rectifier starts to conduct at 27 volts, the other two then neutralize and drop to zero this happens at around 5.5 MS and by 12 MS it is doing 50 amps at 30 volts, this is where it is furling if it goes any higher. In gusty weather before I got the furling better it burned out a 50 amp meter, cooked the composite shunt, now replaced with a proper metal shunt. There is no sign of blade stall because it is only loading from 4 volts so the blades are turning well at that stage, I think it is also because of being in Delta. I have increased the blade diameter to 3 meters but that didn't make any difference I could note. All the best Bob The blade now starts just over 1 MS and speeds up to catch the slightest puff of breeze. Foolin Around |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
As in first post they are great low wind performers but believe that can be improved on . I am comparing them with a Jaycar 200w mill I`ve had up some 4 years or so. On 48v with a cap trebler or doubler the 200w mills performance with it`s 1.8m blades are equal to or better than the 2.74m & 3m OEM`s( same conditions except 1 OEM(2.74m) is 2m higher where it clearly shows height is King). Thus you can see why I am dissapointed with the OEM`s I expect more from the additional blade dia, and believe they are being held back by either magnets or windings. My system is a working system, while I like to tinker and am prepared to do more, I require wind turbines to produce usable power with the wind I have( I am not on the grid). I did not go in blind the 4years of the 200w mill has shown the limitations of my wind site The 200w I estimate allowing for my poor wind site to be a 100w-120w mill. I do not factor in high wind just the average usable wind. The OEM`s I expected to double that in the 250w to 300w range, they do that effortlessly on a 12v system(tested) but not on my 48v system. They are held back by low voltage output which can be compensated to some extent by cap-rectifier combination. If I can get those 51 windings to output the voltage on 3phase they should be, these mills will also fly on 48v and thus add to my system if not than they are not suited to a 48v system. We come from the land downunder. Vic |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi norcold I would suggest that you match the blade loading on the two mills so they perform more or less the same and then series the two to get higher voltage, if possible wire them to delta as this will give a efficiency increase of nearly 30%. 
The photo is both mills in the same wind 4 MS OEM producing 300 watts and the 500 watt version of yours producing 90 watts. 
I found that it is better to go to delta as in star the internal resistance of the windings is double but the voltage out under load for star is 1.5 times delta. This i believe is caused by two factors 1. the increased internal resistance in star. 2. The phase angle losses between the two windings in star. The alternator is about ideal for 24 volt systems but at the top of its range on 48 volts, as I posted earlier the machine will drive good power at lower volts and thats what I pick up with the boost converter. It is not so easy to get a boost converter off the shelf for 48 volts however it is possible to put one 24 volt one on each mill and series the outputs to 48 volts, that could be worth a try, as I believe based on the performance of mine that you would get 300 watts on the boost module before it goes to the main rectifier in higher winds. Mine does 600 watts on the boost module before going to the main rectifier, so I gave a conservative output on yours. As your test showed they are happier at lower volts on lower wind speed. When they get rated wind speed they go wild and produce much more power than rated 8.5 MS for 500 watts I believe, can't find the destruction book to check that out at the moment, the wife has been spring cleaning and has snuck into my office as well. Not sure which is worse wife in clean mode or grand kids in wild mode, at least I can bash them a bit, and only get called Grump par. All the best Bob Foolin Around |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5078 |
If you were really keen, you could rewire the stator using more turns. I think there were 5 or 7 wires wound together. Running several wires together increases the current capacity of the alternator, while keeping the cost of copper down and also making it easier to wind, as a single thick wire is harder to bend than several smaller wires. You could unwind all the windings on the stator, taking care to note down exactly how it was wound, ie pole 1, 3 turns clockwise, pole 3, 3 turns anticlockwise, etc. Like I sad before, its an unusal winding system they have used, but it works perfectly, and changing it will break it. Then buy yourself a length of the same wire as you took off, but instead of winding 5 or 7 strands together, use 3 or 4 strands, and almost double the number of turns, ie if it was 3 turns originally, try 5 turns. This will give you a alternator with a higher voltage, lower current capacity. Same watts, and better suited to low winds at 48v. Or you could use cap doublers, etc as suggested above. Either way will help you with low winds. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| fillm Guru Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Hi Vic, Below is a table I made up showing what output can be expected from different Dia/Radius blades , this was done using the Blade Caculator at the warlock site , there are other caculators "Altons" that I think Bob has used in the past. 
It shows 3 blade sets R.9m R1.4M & R1.5m and what is the predicted output in wind from 1m/s to 10m/s , the tsr was 6 and the blade eff .3 for the R.9 and EFF of .4 for the other 2 which are the GOE222,it is debatable as it goes against Betz's Law. One thing the table points to is that size does matter with wind turbines. If you go to the PICLOG page HERE where my 48V Mill is logging and put in 24/5/11 and 25/5/11 into the Dates filter and hit apply , it will bring up 2 days where I had reasonable wind . If you then click on the WIND V POWER , it will bring up a scatter graph and by running your mouse over the dots it will show the output for a given wind speed . Because my annemometer goes up in 3 klm increments it will give a number of outputs at a given W/S , I usually go about the middle.The annemometer is a Davis and is calibrated to my Davis Weather station so it is fairly accurate. It pretty much traces the output of the graph for a 2.8m blade set give or take some wild readings that are in the middle of nowhere, and taking into consideration that that logging will have a percentage for error. For the application we use the loggers for it can give you an accurate enough idea of how a mill is performing which can not be done by watching an amp meter and assuming windspeed or even trying to watch a cheap annenometer that updates every 120sec, which is what I tried a few years back. Getting some form of logging would definatly shed some light on what is happening. With regards to your 200w Jarcar mill with a 1.8m blade set outperforming the 2 OEM mills with 2.8m and 3m blade sets seems to go against the maths of windpower, but there could be a logicial explanation. Have you checked what AOA the blades are set at ? The GOE222 need to be @ 8~10degAOA which is +3deg when setting them accross the face, if even one blade is not correct it will mean a power drop in low wind. I recall you were trialing a Doubler/trippler configuration with just series connecting the DC out of the rectifiers and paralell connecting the AC with no Caps , can you take some pictures of the trippler/doubler set up ? To get low wind power in 48V these have to run with a Cap/Doubler What is the cable run & size from the mills to rectifiers and then the DC run and size? Did you upgrade the wiring and connections through the slip rings? PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Phil Just did a backwards exercise on the power calculator and to get the power I am getting at the moment it shows the geo 222 blades at 60% efficiency. This is very similar to the Altons calculation as well. Bob PS Just confirmed this wit the blades on the 500 watt mill and they came out a 32 %, to be fair to these they are high speed blades and pick up efficiency as speed goes up. Foolin Around |
||||
| fillm Guru Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Hi Bob, To be on the money , I think you would need to be accurate with the "Output V Windspeed", I would be exstatic to be able to claim that the Blades I sell beat betz's law by 100%..... Mabe I should factor in a Price Increase 
We should be all logging our windmills, but the sad truth for me is the laptop probably consumes more power than the mills make most days in my location. This is the link to those 60V dc dc Boosters I bought - HERE Mabe you could do a mud map on How You connect them in to the system. With the Delta and 48V I think the actual cut-in going up to over 500RPM would be to high for even boosters and Capacitors to work. PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Phil The readings was with the anometer radio thing in my hand and it updates every 5 seconds, standing watching the amp meter' and volt meter on the mill so While it isn't logged every fraction of a second it is really a good indication. Those blades are better than you give them credit for, as they have a CL of 2.4 as compared to altons at 1.3. I was talking about putting the mills in series and using delta with two boosters on the output also in series to get the voltage up while maintaining efficiency. I will do a mud map and post it tonight as at the moment i am trying to get some chink junk working. I suppose I should go the the trouble of getting a logger but I can't bring myself to work to try to get one going a but fiddly for this old fool, and to be any good for what I want to do it would need to monitor six mills at the same time. I will just settle to a comfortable chair and some analogue meters. All the best Bob Foolin Around |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Guys Sorry it is a bit rough, but I got the shakes a bit. 
First one is my mill with two boost circuits on it 1 a 4 volt to 35 volt at 3 amps 2 a 11 volt to 35 volt at 9 amps each has its own input 3 phase rec and a bridge on the output for isolation. Suggested for norcold two mills wired in series and delta connected with a pair of boost modules also wired in series and isolated bu bridges on the input and the output. All the best Bob Foolin Around |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
The output from my mills at any given wind rate are the same as yours fillm. We have covered that in our private emails and I think we should not go there in interest of the 4M. My critism of the mills is portrayed to find a solution, I believe they are even more capable than the excellant low wind performance they currently show. Rectifiers connected in series on DC side and parallel on AC side will act as voltage multipliers without the caps. The caps basically only ease some of the load, filter etc. Thus the importance of cap capacity. My issue is not this. My understanding of 3Phase PMA that relates to this is that for efficient top performance 3 fields should pass over 4 magnets at any time. I was hoping for some feedback along this line. Perhaps my understanding is incorrect here. Bob`s idea to couple the rectified output in series from each mill thus giving higher voltage is an idea I`m going to trial. The unloaded voltage(AC) from the OEM`s and the 200w Jaycar will show why it competes with the OEM`s on 48v. As stated it is left behind by the OEM`s on 12v. Am also considering this may have something to do with the smaller dia allowing higher revs under the same wind. I will get these unloaded voltages after this current frosty spell we are going through. The old adage "mild winters follow large wets" has been blown away this year. Naturally frosty days have no wind. Although currently the 12m OEM is outputting 0.2 amps the 10m OEM is not turning neither is the 200w Jaycar, the leaves on the trees are not even rustling. That is low wind performance probably all the graphs and data would never agree with, my eyes may be getting on but this is fact. We come from the land downunder. Vic |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Norcold Just a little snippet to show the extra efficiency of a delta connection. Bob There are basically two ways to wire a 3 phase alternator, star ( or Wye) and Delta. With Delta you get lower voltage but more amps. In star you get higher voltage but less amps. You can calculate these by using the square root of 3 ( or 1.732 ). Each coil set is a "phase" of the alternator so when you measure voltage,ohms or current to test one phase of the alternator you would measure the "phase". Once you know what the output will be from one phase you can calculate the "line" output of either delta or star. The line voltage would be measured from any 2 of the 3 outputs. If one phase measured 22 volts in your test and 10 amps then the star configuration would produce 38 volts and 10 amps ( 22 x 1.732 ). The amps remain the same as the phase measurement because the star is basically series'd to another phase. In Delta you would get 22 volts at 17.32 amps (10 amps x 1.73 ). If you calculate this out 22 volts x 17.32 = 381 watts and 38 x 10 = 380 watts... so what is the advantage? Typically the resistance in Delta is 1/3 the resistance of star. If the resistance of star was 1.5 ohms we could calculate the output ( see formula section ). Lets assume the test was at 600 rpm, we achieved 38 volts in star ( about 16 rpm per volt ) so at 1000 rpm we would get 62.5 volts less battery voltage of 12.6 = 49.9 volts / 1.5 ohms = 33.26 amps * 12.6 = 419 watts... not to bad. Now in delta we had 22 volts at the same rpm ( about 27 rpm per volt ). So at the same 1000 rpm we get 37 volts - 12.6 battery = 24.4 volts / .5 ohms = 48.8 amps * 12.6 = 614 watts. Almost a 200 watt gain !!! The advantage of star is the higher voltage at lower rpm which means our unit would have to make 201 rpm to start charging at 12.6V where the Delta would require 340 rpm to start charging. Disregard the rpm figures as this is a different style alternator. All the best Bob Foolin Around |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Must be our age,Bob,see you`ve been up early. Think I understand what you are saying about delta to an extent but believe I must get the voltage up, delta will not achieve this without further reliance on cap-rect or some other voltage multiplier. As is would be good to utilize trebler at low winds than as wind picks up direct to single rectifier, however getting those revs consistently may be beyond the blade-pma-multiplier combination as is.We know the low wind performance is tops, what other mills putting power into the battery at 5km/hr? But that`s the voltage multiplier and those Al blades at work. If I can will try the 1.8 Jaycar blades on OEM and Al blades on Jaycar mill, I believe that may confirm my view. Have a Altronic-SC data logger kit sitting here ready to be constructed, that`ll consume any spare time I`ve got for awhile. But first I`m off on holidays to catch a few barras and see the g-kids week after next. We come from the land downunder. Vic |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Have found a formula for the coil-magnet relationship. C=Nx(M/2) Where C=Coils N=Phases M=Magnets Thus Mx2=C/3 We have 51 coils divide by 3=17 multiply by 2=34 magnets required this formula apparently blows away my original 3 to 4 formula. Different source different formula. But it appears from reading there is a definite relationship to voltage- coil number and magnet number. I say appears because there is a lot of conflicting info out there. If the OEM coils(standard) are definately wired in Wye than instead of increasing the magnet number I should be decreasing from 46 to 34. It also appears from my investigating that the magnets should not be touching. As both Glen & Phill have pointed out I probably will not gain but lose by changing. If Phill you have a spare rotor or can obtain one for me I would be prepared to give it a go. Rare earth magnets are not cheap thus would have one go at this. What grade of magnet is advisable to use? Now the above thinking and calculations may not be correct and if you feel that is so please correct me. Just have to cut the number of rectifiers down to stop the wastefull loading, to do that must increase the voltage at same revs. We come from the land downunder. Vic |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
http://www.stanford.edu/~hydrobay/lookat/pmg.html Forgot to put the link in my post. Bob, on reading and trying to understand, delta probably is the way to go. Probably than the loading up caused by multiple rectifiers will be offset to a certain extent. I wonder if boosting the voltage on the AC side is possible also,sort of a transformer for each phase before the rectifier. The 48v DC booster Phill has found has possibilities here to, but some energy would be lost to heat.Head is full of ideas but without a technical background in this direction, wonder if I`m chasing my tail. We come from the land downunder. Vic |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi norcold There are always losses in conversion, but I look at it this way, before I put in the boost module I had no output at all, after I put it on I had 10 amps over the two modules and a nice little hand warmer for the winter, so In reality I gained something, not to maximum efficiency for sure but power was there where there was none before. The running in delta and picking up 27 % efficiency there more than compensated for down stream losses. The cap doubler gave only 2 amps but didn't cut in till 15 volts dc, whereas the two boost modules started at 5 volts. With the 3 meter blade it handled this without stalling or excessive slowing down according to the frequency meter on my DMM. Also when the cap doubler is working there is a bit of heat produced there so it isn't without losses. The transformer situation is also a bit lossy, something like 10% at 50 HZ, the problem is that the transformers loose efficiency at lower frequency, just where we want pick it up, and loose efficiency when the frequency goes high due to iron losses. If the output frequency is between 40 HZ and 70 HZ it will work OK, so it depends on the frequency that you are trying to transform as to the through efficiency. I believe toroidal transformers would be better but haven't tried them. On one mill I am playing with at the moment it is an unmodified LG motor fitted with neo magnets, I expecting 300 vac + out of it and it will be connected to 3 transformers 240 to 36 volt, this is then connected through a star delta relay to give voltage buck boost at the output. The cut in voltage is around 40 HZ to full power at 80 HZ so will be a bit of loss at bottom and top end, but thats life. I hope the simple non electronic component approach will outweigh the complicated electronic solutions I have been looking at. You may be able to use 2 to 1 ratio transformers wound on toroidal to get what you want, I think that the delta winding set up with voltage doubler transformer and a boost module would be the simplest and least expensive with the only weak link the boost module, it needs extra heat sink to dissipate the heat, and even if the smoke gets out only costs $10 / 15 to replace. I have blown several but only for my stupidity, so they are fairly reliable. I have ordered several of the sixty volt units to play with for my EV project so that will be interesting. All the best Bob Foolin Around |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Bob, Notice in your sketch of your setup, you use 3x3Phase rectifiers. Would it be possible to run the boost modules in parallel off the one rectifier DC outlet? If so believe this would remove some loading. Going further should not just one rectifier be enough with the two boost modules and the direct output to the battery? Have ordered 4x60v boost modules as it seems my magnet alteration idea has little support and thus probably "hare-brained", and will have to trial other ways to get these mills working on 48v. Glens rewiring would work but I`m not that keen. Leaning towards trialling the boosters first than rewiring one mill to delta. Have given your idea of of connecting in series the DC of the 2 mills, but have noticed although my 3 mills are all within 50m of each other, one or two or three can be producing singly or all and not all tracking the same wind direction. Tried the Jaycar at 15m height briefly but let it down to 10m for safety reasons. Must build better towers to get that kind of height. We come from the land downunder. Vic |
||||
| Page 1 of 3 |
|||||
 Print this page |
