 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 13 of 25   |
|||||
| Author | Message | ||||
fillm Guru  Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Glenn, Sounds like you have the thinking "cap" on , you may have read the dual piclog on electronics is up and going so if you need your piclog back for any testing with this I can send it via mail , drop me an email if you do ...Phill.. PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
| Ali1 Newbie  Joined: 22/12/2008 Location: RomaniaPosts: 3 |
Hi all, i watched all discusion on this thread and i have some question to you all. First i want to try to build my F&P mill but because in Europe we don't have those motors i try with an LG motors equivalent i think on the 80 series. for now i work on the mechanic system and in 3weeks i hope to be able to run some tests. Also i have a small chinese mill model Aeolus3oo 
but like other else chinese mills it's spinning hard to put something in my 12volts battery bank. It's a 3 phase mill at 12v rated 300w and maximum 500w ( in practice only in hard wind it's make approx 8 amps) and in body have a regulator that cut the output at 14,5~15v under load condition. I want to try these series cap but my probs are the mill is at 12 meter high and is very hard to rise and down the tower to make those tests. Is it possible to get down on slipring with alternative and to renounce at ground wire (slipring has only 3 outputs)and put caps on the tower base or better at 40 meters distance in my garage where are the batteries? My wires are 16mm2 copper. Also what setup to chose it's another question for you, maybe some schematics with paralell like Bolty will help me. Sorry for my english and i wish you all Merry Christmas |
||||
| Jarbar Senior Member  Joined: 03/02/2008 Location: AustraliaPosts: 224 |
Hello Ali1,welcome to this forum.Although I can't answer your question your location reminded me of another Romanian who was working with an F&P mill.If you look up the name Vasi in the members list you may be able to track him down through his webpage.I am sure others here can also help.Merry Christmas to you. Anthony.  "Creativity is detirmined by the way you hold your tounge".My Father "Your generation will have to correct the problems made by mine".My Grandfather. |
||||
Don B Senior Member Joined: 27/09/2008 Location: AustraliaPosts: 190 |
Hi Glenn, I don't think that a capacitive voltage divider is practical for 12V (or indeed any power) applications. My only experience with capacitive voltage dividers is for (very) high voltage 50 Hz potential transformers, where the load is minimal. Even these also incorporate a transformer. The only random thought that crossed my mind for an LV application is series transformers, with, say, a 10:1 or higher ratio, with the higher number of turns (primary?) in series with the mill phase leads. If a capacitor is then connected across the secondary winding, there should be be an impedance transformation of the capacitor impedance equal to the square of the turns ratio (ie X 100 in the example given). This would, theoretically, transform a 100 MF capacitor to give the effect of a 10,000 MF capacitor equivalent on the primary side. This thought completely ignores the no doubt hugely adverse effects of the transformer winding resistances, impedance, and losses, plus the fact that you would be expecting the capacitor to carry significantly more current at a lower operating voltage than if it were directly in series with the phase leads. Its equivalent series resistance would also be transformed up by the square of the turns ratio, and I imagine that the whole concept is probably just plain silly. If you were going for any sort of transformers, then you might as well use them in the conventional way to transform the voltage back down again from an F&P reconnected to output higher voltage, and incorporating series caps. Better yet would be to store your DC at a higher voltage (ie 24V or higher) direct from the alternator and caps (as Gordon does), then drop down via a switchmode DC-DC converter for your 12V loads. This, in turn, would only be worth the bother if the gains from the series capacitor coupling exceeded the losses from the DC-DC converter, although these can be pretty efficient these days. Then there is the question of the cost of the equipment additions and changes, although these may well be justifiable over time. Regards Don B |
||||
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi kiwijohn, I have 4 pics, that tell some interesting stories. The rpm was not constant, and the capacitors reduce the loading, and the rpm increases. This is what happens on the windmill as well, and is why I started to investigate with a doubler, to regain this power. pic 1. current and voltage with no caps across bridge rectifier, single independent winding output to 24V nom battery load. This was 0.87ADC to battery. current was indicated by the voltage developed across an 0.8ohm series resistor, for a 6ohm winding resistance. The voltage was the AC to the input of the bridge rectifier. 
pic 2. current and voltage as above with 50uF series caps with the winding. 0.81ADC to battery. The voltage was the AC to the input of the bridge rectifier. 
pic 3. current and voltage as above, with 25uF series caps with the windings. 0.72ADC to the battery. The voltage was the AC to the input of the bridge rectifier. 
pic 4. current as above, with 25uF series caps, but voltage measured on the coil side of the caps. 
This last pic, shows the current to the battery is not in phase with the emf from the coil, when capacitors are added. The coil produces a more sinisoidal output voltage, unlike the clamped type output when directly connected to the battery. There is approx 160Vpk-pk compared to the 60Vpk-pk without the caps. I am sure someone will have some explanations. Gordon. become more energy aware |
||||
Bryan1 Guru Joined: 22/02/2006 Location: AustraliaPosts: 1344 |
Hi Gordon, Above is some of the things we talked about today in which I'm still in the dark, like those scope images. One thing that was said to really test this cap theory is REAL wind conditions. I will be making a new wind genny site for my 2hp conversion and I will be happy to hold a cap testing session in SA at Kanmantoo on my farm. As every tower building guy would know it takes a couple of weeks to let da concrete set. So if you guys are really interested in seeing caps work ontop of my hill where I'm sure Murphy wont pay a visit in real world conditions I'll be happy to accomadate. Lots of step will need to be made but as I really started this I'll be happy to try and prove to all the naysayers Caps really work in REAL world conditions. All that is needed is a few chairs, an esky and maybe some shade for the equipment. Oh and and as my place is off da grid maybe a 12 volt to 240 inverter to power the logging equipment which can be powered from a batterybank placed onsite. This is only a concept at this stage but I am happy to do this just to prove the concept. Cheers Bryan  |
||||
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Hi Gordon Scope Images interesting . Unfortunitly no conclusions can be drawn as the measurements are only showing the characteristics of the load accepting current from the available voltage. Maybe if you added a refeance winding to sense the induced emf rather than the supplied emf . Caps as expexted show a lead in the last trace . I can not from this info confirm or as i suspect kill the idea of leakage reactance. The generator is merely doing what any generater does supplying the load placed on it. My theroy tells me that the caps supply added excitation to compensate for the effective bucking of the mmf of the rotor as the load current increases.Not entirely unrelated to how a induction motor can be self excited. The rotor moves back and forward as load pf changes get a strobe old car timming light . Bryan IMHO no testing can be conclusive on live testing you are merely changing the characteristics of the generator to suit the driving force it works but is a stab in the dark. I admit I have used caps for years to tune hydros but once i find the operating point I adjust the windings to suit rather than leave the caps in, in the mill yes we have varying speeds and caps will have benifit left in. Only testing on a bench will be able to determine and understand the ultimite setup. Herb ps just looking again at last trace see 6 divisions one cycle 360 Deg therefore approx phase shift 1.5 div gives 90 deg there fore as we also have resistive load there cannot be much inductance effect from the windings. |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi herb, Therer must be something with terminology you are using. I see the last trace as showing that the current is achieving a maximum after the voltage has already achieved a maximum. The current into the loading after the capacitor is lagging the voltage that is driving into the other end of the capacitor. In laymans terms, I see the voltage as similar to the motion of a swing being pushed. As the swing is pushed harder, the motion increases in amplitude. the capacitors are like a bungy cord attached to the swing. a load is attached to the other end of the bungy, like the rectifier and battery to the capacitor. The swing pulls on the bungy and exerts a varying force on the load depending on the amplitude of the swing. The position of the load attached to the bungy is changing depending on the stored energy in the bungy and the position of the swing in the cycle. The maximum amplitude of the load occurs at a time after the swing has attained a maximum and all the stored energy in the bungy has been utilised. The swing transfers energy to the bungy and then the bungy transfers this energy to the load. The capacitors perform a similar function. This is how I interpret the pic 4 traces. As the frequency increases, the capacitor reduces impedance, so the load more directly affects the driving force. BTW, I feel the emf of a sense winding on the same poles will closely follow the loaded emf, possibly without the rounded top of the waveform. I can only measure 2 channels without isolation. To accurately show the winding emf and current in relation to the load voltage and current will require 4 channels with electrically isolated inputs, which I don't have. The current sensing resistance contributes a small proportion of the loading. This would be 0.8 ohms. The other resistance quoted was that of the winding at approx 6ohms. even so the total 6.8 ohms would present only about 5V of the 28V to the battery. There is another 25 or so ohms that is not pure resistive. Gordon. become more energy aware |
||||
| herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Hi Gordon Its true Caps Inductance resistance all have their mechanical equivalents but to think in these terms is verylimiting in fact mechanical values are often converted to electrical values to allow analysis. I am suprised that you you visualise this way. To comunicate effectively I need to follow the tried and proven AC therory terms that all the values of impedance formulas are based on. The emf on a sensing coil would only be the same if the winding didnt posscess any inductance in other words using the phase relation between this emf and the current would have some meaning still only measuring one emf one current Herb |
||||
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Pic4.... is the current measured after the caps and before the diodes. Is the voltage measured after the coils and before the capacitor and diodes? .........oztules Village idiot...or... just another hack out of his depth |
||||
| herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Hi Yea something is wrong current should lead voltage in a this arrangement I suspect one trace invertered Herb |
||||
| andrewf Newbie Joined: 24/12/2007 Location: New ZealandPosts: 15 |
just to chuck out something completely different, but quite the same, have you considered using an active PFC ( Power Factor Correction ) circuit inbetween the Mill and the load? The caps certainly will help in a given situation.. ( ie, rpms, load etc etc ), but theres way too many variables in a windmill changing to put a single "solution" in. PFC technology is pretty common place, theres small PFC's in just about any modern wall wart, and larger ones about the place as well. As a silly question, is is possible to get the F&P's output voltage in the range of about 80-250V? If so, it might just be possible to use an appropriately sized switch mode power supply?? EVERYTHING SHOULD BE AS SIMPLE AS IT CAN, NOT SIMPLER - EINSTEIN |
||||
| KiwiJohn Guru Joined: 01/12/2005 Location: New ZealandPosts: 691 |
Andrew, an unmodified F&P would produce voltages in that range. Something that agitates my gray matter somewhat is the fact (?) that the F&P (and I guess any other alternator) can really only deliver anything to the load for that very narrow part of each cycle where the voltage is greater than the battery voltage. IIRC someone has already talked about a square wave output alternator though I dont recall how he (proposed) to do that. |
||||
| andrewf Newbie Joined: 24/12/2007 Location: New ZealandPosts: 15 |
What i should have said, was a "off the shelf powersupply".. A PFC could be built for pretty much any range. Using one that was designed for universal voltage ranges would just make life easy. EVERYTHING SHOULD BE AS SIMPLE AS IT CAN, NOT SIMPLER - EINSTEIN |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi herb, Yes, one trace was effectively inverted. I rewired the measuring to this arrangement. 
The DSO gnd was connectred to the Gnd point and then the current was measured across point A. The voltage was measured across B and then C, before and after caps. I was unable to reproduce the smooth voltage sinewave as before. 
I hope this is a little clearer. On another point: The real point of the exercise, that has been lost lately, was to better match the windmill and battery loading to the wind energy curve. This is not about maximizing the efficiency of an alternator. A PFC correcting device connected to an F&P, or other alternator will not necessarily match the loading to the wind energy curve. There will be some data and graphs from Phills' test rig on "Cap testing" that will make the objective of capacitors a little clearer for windmill owners. There may be spinoffs for other applications, but this is primarily aimed at windmills. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi kiwijohn, I missed the square wave output note. My AxFx unit had a close square wave output. I used 4phase, with overlapping coils, and a constant magnetic field, as was shown by a mf simulation provided by "dinges". I won't elaborate more here though. Gordon. become more energy aware |
||||
| andrewf Newbie Joined: 24/12/2007 Location: New ZealandPosts: 15 |
Oh ok, well, then time to move to a different thread about maximising efficency in the overall system. EVERYTHING SHOULD BE AS SIMPLE AS IT CAN, NOT SIMPLER - EINSTEIN |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
andrewf For hydro and engine driven alternators, it is efficiency that is the goal.... for wind it is of no big importance. So much power is being wasted by mismatch between the blades and the alternator and the battery, that alternator efficiency is only important in regards to heating. With wind, it is load matching. If you get it right somehow, you can easily achieve a 1000% increase in power into the batteries over a poorly matched system. In some cases as easily as changing the battery voltage from say 24 to 48v (mine will go from 200w to 2000w just by doing that in the same wind)..... so efficiency is only about small changes... a few percent here a few percent there. These are the changes Herb looks for in his hydro schemes... every little bit counts. With wind, it is of no consequence to loose substantially in efficiency, if it will improve the loading dramitically. We do this in some cases with large resistors.. totally wasteful, dissappating hundereds of watts sometimes.... to get a vast improvement into the batteries.... by being less efficient. Silly isn't it. A 100% efficient alternator is unable to charge batteries at all with wind. The moment it reaches cut in speed, is the moment it stalls at this rpm. It cannot increase beyond this point. (batteries behave as a zero impedance load, you can figure it from there) This thread of Gordons is to explore if and how caps can be used to help to match the load to the prop. This has shown an increase in system efficiency.... not necessarily in alternator efficiency.... in fact lowering the alts effective efficiency with series caps for part of the power curve and then loading it up when the frequency increases (rpm and XC) seems to pay big dividends. I agree with Kiwijohn, that the bench testing is limited in scope, you need your particular blades and wind to see if anything useful will happen.... thats where the payoff is. If we can get the alternator to "roll with the punches" (perhaps by using caps in differing combinations or circuit position), then we will/maybe do better. Success will depend on the matching before you start. A well matched system requires nothing. A poor one needs all the help it can get.... and sometimes it was so badly matched, that the improvements are dramatic. However, you do seem to need to be in the same post code with regards to impedance of the stator. I don't presume to get much relief from a 60s stator into 12v.... no matter what caps you use. The impedance is too high to help.... a transformer is the only hope or a rewind. It's not magic. ..........oztules Village idiot...or... just another hack out of his depth |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi andrewf, I did not wish to offend you personally. You may wish to start a new thread on PFC in alternators in the electronics section. This thread has served a useful purpose, extending early work by Dennis and Bryan. I suspect more information will continue to come on the Cap testing thread. This is not done yet. I believe "oztules" has summed things up quite well, thanks for that! I do think that bench testing in the manner that is set up at Phills', can be translated to blade behavior in the wind. The torque measurements can be translated to blade torques. Blades work effectively across a range of tsr. There is an optimum of course and any method of automatically adjusting the loading to the wind energy can result in big gains. My real world testing of windmills shows that better matching at low power levels is more useful than just attaining the maximum power attainable, even though people take note of maximums. Get it right for the winds that occur 80% of the time. Gordon. become more energy aware |
||||
| KiwiJohn Guru Joined: 01/12/2005 Location: New ZealandPosts: 691 |
Lots of good comment here! Much healthy brain exercising stuff too! So a few more thoughts of my own. A single phase producing a sine wave voltage will only deliver to the battery for that part of the cycle where the voltage is above the battery (plus rectifier) voltage and I presume any increase in RPM will raise the voltage and increase duration of the current each cycle and the load will come on pretty quickly as RPM increases. Suppose this current period is only 10% of the time, that would mean the stator current during that period would be 10 times what we would see as charge current on an analogue meter! But as the RPM increases so does the voltage and the current too but the ratio of stator peak current to charge average current is much less so the 'R' losses in the stator are decreasing as a percentage of the total losses. Is this logical thinking? If so it gives me the idea that higher voltages would be better and might even more than compensate for whatever is used (switch mode power supply?) to get down to battery charging voltage. We can simulate a battery load (I think) by using something that would act like a humungous Zenor to ground, something that once a particular voltage is reached produces a very low impedance path to ground. That might be a convenient alternative to a stack of batteries when doing something like Phill's lathe tests. But how can we simulate the characteristics of the aerodynamic part of the mill? Can anyone summarise the characteristics of wind turbine? I am thinking that low RPM means low torque but increasing RPM brings an increase in torque up to some optimum speed from where torque increases more slowly? Is that how it works? If we can better simulate the battery load and better simulate the action of the wind turbine in our lab (i.e. Phill's nice clean workshop) we might even be able to eventually formulate some rules for building the 'best' wind turbines we can, we sure do seem to have enough eager minds on the project! 
|
||||
| Page 13 of 25 |
|||||
 Print this page |
