 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 : visual effect of capacitors
  Page 19 of 25   |
|||||
| Author | Message | ||||
| carl1 Regular Member  Joined: 16/04/2007 Location: AustraliaPosts: 79 |
Hi to everybody re: cheep caps and schottkys diodes: this might have been posted before: we just had a kerbsite garbage collection for the last 2 weeks. Plenty of discarded computers on the roadsite and each Power supply has a pair of 200 to 250 volt 330uF to 470uf caps and schottkys 20 to 30amps 45 volt in it. cheers Harald  |
||||
GWatPE Senior Member  Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Just an update on capacitors. Have just modified Bolty's windmill rotor blade setup again. The light weight wooden blades have been attached directly to the std rotor hub without the extenders. What an amazing transformation. The capacitor doublers had to be removed, as they loaded down the output now at low rpm. The operating tsr of the blades, without the extenders is higher. The rotor spins up to alternator cutin quickly without the extra inertia. The rotor quickly tracks the changing windspeed. It seems that the high system inertia before was preventing the rotor from following changing windspeeds. This setup will now be monitored to determine what experimentation will be required next. Gordon. become more energy aware |
||||
| kenssurplus Newbie  Joined: 22/03/2009 Location: United StatesPosts: 4 |
Hi Gordon, New member here, but have been flying windmills for a few years now. Last year I built an 10 ft diameter axial flux design mill (12 volt stator). I didn't have the funds for the correct wire sizes, so I used litz wire in the coils, sized for the equivalent disign wire. (turned out bad) To be brief, the mill really had to be spinning to reach battery cut in, so I looked on the net for voltage boost solutions. I found the cockcroft / walton voltage boost circuit using capacitors and diodes in a similar fashion to what you have done. I tried it out using three 2500 uf 450 volt polarized capacitors to see if the effect would work and it does. I achived 2 - 5 amp increase in output just after cut in, and found cut in to be at lower rpms as well. Obviously my use of these capacitors was not the best choice, but did prove well worth the effort for me. I was concerned about blowing up the capacitors because of missaplication so I removed th C.W. boost circuit off of the mill last fall. I now would like to reinstall it back on as I am missing the extra power now. My question to you all is, I have a good pile of 41000 uf 15 volt polarized capacitors on hand. If I place them back to back and happily live with half capacitance, will the mill voltage blow those capacitors or do I need to place them in a different configuration? I don't have the spare funds to source the correct oil filled run caps, so I would like to use these big cans if I could. Any help appreciated, Ken |
||||
| davef Guru  Joined: 14/05/2006 Location: New ZealandPosts: 499 |
As I recall you need to put a diode across each "back to back" unipolar capacitor so that reverse voltage can not be developed across it. Search for "making bipolar capacitors out of unipolar or electrolytic capacitors". So, to get the voltage breakdown rating, for each 2500uF/450V cap, you need to series 30 of the 41000uF/15V caps to get 450Volt rating but only 41000/30 = 1366uF. So, you are going to need 60 caps to replace each 2500uF/450V. Something odd about the value of these 15V caps, 41000uF is not a standard value, do you mean 47000uF? Guess I am surprised that that 2500uF/450V capacitors didn't blow up if they have AC across them and/or appreciable ripple current. |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Ken, the Cockroft-Walton [C-W] arrangement is a series, relatively high impedance cct. All the active caps are in series with the loading. In the full XXX arrangement, the DC caps are essentially connected back to back as well, but on each end of a winding. In the arrangement I posted on p15, this is a parallel, relatively low impedance arrangement. I have a good number of 10000uF 16V caps that will not be used in any series or back to back combo. Hi dave, the diode issue has been discussed, and seems unnecessary, as std N.P. electros don't have any internal diodes. Gordon. become more energy aware |
||||
| kenssurplus Newbie Joined: 22/03/2009 Location: United StatesPosts: 4 |
Hi Dave, Yes the correct value is 44,000 uf 15 vdc (mentally mixed these up with a few 51,000 uf 50vdc caps I was looking at). Hi Gordon, as a side note, I built your circuit a week or two ago using twelve 100 uf 450 vdc 'lytics. With a wide range of windspeeds now, I can report no apreciable current going through this boost circuit with these cap values on a lousy 12 volt stator. I will have to set up the 44,000 uf caps back to back tomorrow using both iterations of boost circuit to see which I like best. Thanks for your comments, they have helped me understand more fully. Ken |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
A search for the terms I indicated found the abstract for this article "A novel large value ac capacitor using two dc capacitors", 7th hit on the first page. here The Goggle description tells you more than the article description on the web page. You need to register and/or pay for the whole article. It does mention the use of diodes and surprisingly some switches as well. Also, here I am sorry if I missed this discussion. I have not followed all of this rather long thread. I do not understand that because "bipolar (std NP) caps do not incorporate a diode" that it follows that when you try to make a bipolar cap out two unipolar caps that you also do not need a diode. Also, Wikipedia Special capacitors designed for AC operation are available, usually referred to as "non-polar" or "NP" types. In these, full-thickness oxide layers are formed on both the aluminium foil strips prior to assembly. On the alternate halves of the AC cycles, one or the other of the foil strips acts as a blocking diode, preventing reverse current from damaging the electrolyte of the other one. Essentially, a 10 microfarad AC capacitor behaves like two 20 microfarad DC capacitors in inverse series. Could you tell we roughly where this discussion took place on this thread? Thanks, davef |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
hi Dave, discussions were around p26. Hi Ken, you have a dual rotor AxFx windmill. The output frequency is quite a bit lower than a F&P for the same rpm. I have 66uF cap voltage doubler and 230uF series caps on my 24V battery loading and, 275uF series caps on the 48V grid inverter loading. I have temporarily removed the 340uF cap voltage doubler from the 48V loading, until I finish making a 48V diversion regulator. I have been comparing iron cored and ironless cored windmill performance. The lack of iron drag allows the rotor to achieve a much higher tsr in the same winds, unloaded. No energy is lost in iron hysterysis. In my testing the ironless alternator achieves a similar result without the need for additional external voltage converters. The magnetic iron drag in the iron cored machine slows the rotor and lowers the output emf. The external cap voltage multiliers allow power extraction at the lower rpm. The matching process with capacitor multipliers works best with an alternator that was wound for a high cutin rpm. The original cct you made for the AxFx mill used 2700uF caps. I would expect the current in the 100uF cct to be 1/50 of your original just from the cap size differences. This is not really a valid comparison test. I don't expect the multiplier to pass more than 10-20% of the alternator power at maximum output power. The proportion increases at lower power levels. If you are expecting more, a rewire would be a better option. The ccts I have are to purely better match the alternator to the wind energy curve. I will be testing my new dual rotor AxFx mill with caps as well, when it is finished and flying. Gordon. become more energy aware |
||||
| kenssurplus Newbie Joined: 22/03/2009 Location: United StatesPosts: 4 |
Hi Gordon, I ended up using your circuit and placing the twelve 44,000 uf 15 dvc caps in there. The results are very pleasing: At cut in an extra 2 - 5 amps with the boost ckt installed. At mid range rpm 20 amps through the main rectifier and 10 amps through the boost ckt. At higher rpms 30 + amps through the main recifier and 15 amps through the boost ckt. It even brought one phase that was seriously sagging in volts up and producing well with the other two. All in all, I am very happy with the performance so far. I have not tested the C-W ckt setup with these values of caps yet (probably leave it this way as I have really no reason to be dissatisfied with it). Many thanks and have a breezy day! Ken |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Ken, As a matter of interest. This is a 12V system. I would say that your windmill would be very late cutin. What maximum power has the windmill produced in the past?, without the cap ccts. The voltage doubler is producing results a little higher proportion to the main rectifiers than I would consider optimum. This would be a result of the probable high cutin. I would say the caps are relatively high ESR. If the blades do not appear to stall at low rpm, then this is a workable setup. It is reassuring to have positive results published. Maybe other readers will have success like you. Gordon. become more energy aware |
||||
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Ken... Are you able to detect any sign of worrying heat build up in the caps at those high currents in the 12v system. I am surprised you haven't stalled the mill with such a powerful boost value.... you must have had plenty of shaft power in reserve. As Gordon says, it is nice to see some successful testing being done on this.... .........oztules Village idiot...or... just another hack out of his depth |
||||
| kenssurplus Newbie Joined: 22/03/2009 Location: United StatesPosts: 4 |
Hi Gordon, Absolute peak maximum power that I have noted from this mill was around 45 amps for 2 phases, and somewhat less for the third phase (without caps). I am not sure about the stalling of the blades. It has never gone into turbo boost runaway so perhaps the mill has always been in stall mode??? This is an area where I am a little unclear how to identify stall, or the absence of it. Hi Oztules, After reading your post, I checked the caps during a windy section of the day yesterday, and although the amps produced were lower than the first day I reported, I could'nt detect any heating in the caps by feel. Thanks for the replys all. Ken |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Ken, and other interested readers. It is relatively easy to get an feel for how a windmill performs. I monitor visually, how the rotor responds to changing winds, first without a load, in light to moderate winds. This gives an indication of blade performance and bearing/iron/inertia affects at low rpm. The important aspect is the rate at which the rotor tracks the windspeed changes, and the rpm tracking accuracy to the changing windspeed. This rotor response behaviour becomes the benchmark for other visual comparisons. The mill is then loaded and the rotor behavior is noted with the changing windspeeds. A windmill that is overloaded will not track the windspeed well, particularly at low windspeeds. As a general rule, if site location winds are reasonably constant, then rotor inertia will be less significant with affecting output performance. If the windmill is subjected to wind turbulence from trees and buildings, or terrain, then it is desirable to have minimum rotor inertia. A windmill that is overloaded from the alternator at low rpm will behave similarly to one with high rotor inertia. The rotor will respond slowly to changes in windspeed. The alternator overloaded windmill will struggle to increase rpm with increasing windspeed, and slow down quickly with decreasing windspeed. The windmill with high inertia will have similar rpm response to increasing or decreasing windspeeds. If more discussion is needed, then perhaps a new thread should be created. Gordon. BTW Ken, I would not expect an AxFx mill with adequate loading would runaway. There would be an external problem, too big a rotor, or inadequate furling or inadequate diversion loading. These alternators do not generally suffer from inductance limiting like some iron cored units. PS edit : this was edited at 10:10AM on 26/03/09. Looks like thebackshed home server has had to skip the country. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I am not sure that the impact that capacitor voltage multiplyers has on a windmill is fully understood by many readers. I raised the topic recently to some windmill enthusiasts, not familiar with discussions on this forum. They were intrigued by the simplicity and benefits that would be gained. The most revealing aspect was the more efficient use of the wire in the coil windings that would result. For the simplest voltage doubler arrangement, the effective halving of the wire resistance required for the same cutin voltage, occupying the same physical space would effectively halve the overall copper losses. They agreed that some matching of the load to the alternator was required, and that capacitors were not a universal black box solution fix it. The lack of electronic black magic and switching devices was of interest to those in the bush, where reliability was high up the wish list and reliance on radio communication where RF-interference could not be tollerated was important. Full AC coupling of a windmill will require careful sizing of capacitors, but the more universal low output power, voltage multiplying aspect can be incorporated on most windmills utilizing a PM alternator. Personal feedback from readers indicates, many are now benefiting from additional power from their windmills with lower cutin rpm. Once the engineers have the opportunity to fully optimize the use in new designs, the increase in efficiency offered for the same windmill alternator sizing will becomne available in an off the shelf unit. Sources of cheap caps and bridge rectifiers suitable for voltage multipliers come up often on the internet, so making units from new components should be realtively inexpensive. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Just some follow up on series caps arrangements. I dismantled my high voltage doubler and 48V series cap system on my F&P mill. I have noticed a slight leakage of electrolyte from one cap. This was from a block of 6 that was passing over 500W. I am recommending that series caps, where possibly the full output power of the windmill may pass through the caps will require careful cap sizing. The voltage multiplier arrangements place a much reduced stress on the capacitors and these show no signs of leakage. Have any other readers noticed similar signs of possible long term failure of caps in their own systems. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
It would seem that the AC frequency aspect of capacitor sizing is not critical. The same capacitors used on a F&P 56magnet rotor do just as good a job as those on a 12magnet rotor AxFx at similar rpm. I will look into this further when I get time. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I have prepared a graph of the relative contributions to the output power from Caps in a voltage doubler arrangement. 
I have plotted the total power, series caps, cap voltage doubler and relative cap doubler contribution. The scales are 1.3 times the actual power. At low power levels, the significant power comes from the cap doubler. As the power levels increases to 200, the relative contributions are the same. As the power levels increase further, the doubler plays less a role. This is a seamless process without any switches. The graph is showing that I can increase the series caps still some more, as the loading is still not quite enough at higher power levels. Gordon. become more energy aware |
||||
| Ali1 Newbie Joined: 22/12/2008 Location: RomaniaPosts: 3 |
Hi, Gordon. Please tell me what are the best value for the caps and what schematics to use, a voltage doubler or a series caps for a LG direct drive motor wich is corespondent of F&P and with a chinese Ax mill. Both are used with 12v configuration and need some solution to put some Amps in the batteries at 4~5 m/s wind speed. Thx. Alin |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
12V systems require large caps and the diode losses negate effectiveness of the capacitor voltage doubler. My systems are 24V & 48V loading. You need to provide operating AC Hz at operating rpm, and windmill power outputs. All of my cap systems are empirically determined. Chances are that you will need to do the same. I can only give an estimate, given that I don't have an LG windmill or an Ax, whatever that is. Most small windmills at 12V system voltage are usually not good candidates for significant power boosting. They are usually overrated in output performance. It is easy to stall a small windmill with a cap voltage doubler. If you have a wind distribution graph for your location, then this will assist in deciding the amount of power available at the windspeeds, and if there is benefit to making changes. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I have been informed that capacitor voltage doublers also work on a F&P windmill, wired for 12V operation. The system worked with 100uF 400V DC caps wired back to back in a voltage doubler arrangement. Battery charging current was obtained in conditions of low windspeeds where previously the windmill turned, but rpm was insufficient to reach cutin. The additional power outputs are in the order of 1-2A, and this is in winds that would normally give 0A. The reasons for not adding some form of capacitor voltage multiplier or loading matching system is diminishing. Gordon. become more energy aware |
||||
| Page 19 of 25 |
|||||
 Print this page |
