 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 10 of 25   |
|||||
| Author | Message | ||||
fillm Guru  Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Top Stuff Gordon , I can now visualise what you were explaining to me , and I hope to put it into practice as soon as my repairs are done and I just finnished the #2 quad chassis this afternoon . I take it that with 48v you have 4 x 12v connections 2 x 24v , 1 x 48 with an extra set of caps and rectifier for the 48v high end / RPM . I may as well be the one who asks the question as probably the most electronically challanged ? The formular for working out the cap size Z=1/2piFC , what does it mean ? I take it Z is the cap size but do not have a clue on the rest ... I see what you mean about getting a total amps/watts read on this set up ,I suppose it would have to be on the DC feed to the grid inverter to get an overall output , matching one mill against the other appears difficult with one batt bank , no doubt there will be a way somehow.....Phill PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Use your logger to measure the current and voltage on a leg of the 3 ph AC, and calculate that in your spreadsheet/ graph (and multiply out for the 3phase) .........oztules Village idiot...or... just another hack out of his depth |
||||
GWatPE Senior Member  Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi phill, the Z is the impedance the F is the AC frequency the C is the capacitance the pi is 3.14159265........ There will be a bit of experimenting required to get the best results. rpm/frequency and voltages will need to be approximately known. The power could be calculated from measurements of the AC mill voltage and current in 1phase leg, converting to a DC form and multiplying out. The alegro sensor will produce AC output, that will need to be rectified, An opamp can do this. The piclog will be able to record if the signals are made compatable. The difference to the normal piclog will be that the power recorded will be widely varying voltage and current. The power will not necessarily be output power, as this will include power that is harmonically circulated by the caps. I will need to think about this some more. I think that if the batteries are replace by a DC capacitor array, that is then connected to the battery, that the DC volts, and amps can then be used. Gordon. become more energy aware |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Hmm, Perhaps a bench test of current/voltage in the leg versus all the battery connected / other DC inputs will give us a reasonable error reference..... and the diode losses... and the cap temp losses..  .. I thought this was supposed to be simple?
.......oztules Village idiot...or... just another hack out of his depth |
||||
Don B Senior Member Joined: 27/09/2008 Location: AustraliaPosts: 190 |
Hi Gordon, Regarding the alternator "real" power output, this will, of course, be the classic E X I X Cos Phi, (where Phi is the angle between the voltage and current). If the Allegro chip that you mention is the one that I am thinking of, then it only measures current, leaving the volts and the angle determination (plus the AC) still to be grappled with. I have not explored what is available, but I would imagine that someone makes a chip that simultaneously looks at volts and amps, measures the angle, and outputs something equivalent to watts. I assume that something like this must be at the heart of the "smart" solid state electricity meters. If you are measuring the current in one phase only, and a phase to neutral voltage, then the total power will be simply the power in one phase times the number of phases that you have (three for the F & P, and four for your special alternator??). If you can't access the neutral, then determining the total power output from a phase current and a phase-phase voltage becomes more complicated, even if you have a "watts" chip. Unfortunately, few, if any, wind generators would have an extra slip ring to bring out the star point neutral just for convenient power measurements. As you suggest, it is probably much simpler just to sum the DC outputs (in watts)from whatever loads you have attached at the time. I imagine that a Picaxe micro could be set up to do this. At least this way, you don't have to wrestle with angles. Regards Don B |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi oztules, The caps only seem to heat, at the really high power levels. Sustained wind peaks that produce the 600+W, on the single stator setup. This is wind gusting above 40kph, sustained for 8 or so hours. The caps were good, even after the 2 days punishment, where the mills produced 8.5kWhr. I suspect that the chinese controller that was attempting to shut down the mill at the peak levels, was the real culprit. This effectively caused a parallel resonant cap short to the mill. This will be eliminated with my new dual voltage sensing diversion load protection system, that will load before the caps. As far as recording power goes, I will still measure the 24V component in my system. Any additional peak power that passes directly to the 48V system will be a bonus, that unfortunately will not be recorded, sort of a hidden bonus. Hi Don B, there is 4 quadrant multiplier [special opamp] that can perform the power calculation in real time. I have used one on DC, but not AC. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi All, An important consideration that should be noted, is that the capacitor coupling arrangements are charge pumps, and as such, these should not be left unloaded. An unloaded system, even if the mill is only producing a few ACvolts, has the potential to still produce hundreds of DC volts on the output over time. Due care should still be taken and treat all connections as lethal. This could however cause a problem with a grid connected system while the inverter was still connecting. It is assumed that if the capacitors are implimented, that a suitable overvoltage protection is installed as well, and this was functioning. Gordon. PS edit: I will place this comment at the start, and in strategic places through this thread. become more energy aware |
||||
| KiwiJohn Guru Joined: 01/12/2005 Location: New ZealandPosts: 691 |
Gordon, could this hazard be handled by bleeder resistors across the caps? |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi kiwijohn, the hazard only exists if there is capacitance across the output rectifiers, and the ccts are disconnected from the loads while in operation. The bleed resistor would have to be capable of dissipating the full power of the capacitors cct, and this would defeat the purpose. The solution is to just incorporate overvoltage protection into the loads, and turn OFF the mill before working on it. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I have had a look at the cct again, and my original thoughts on capacitance being required across the rectifier outputs is wrong. The waveforms are all in phase, so the rectified voltages will be additive. This does make the cct less problamatic as a charge pump. The unloaded voltage of the mill with caps will still be doubled, with the doubling arrangement. A 48V system could still potentially have many hundreds of volts across the outputs, if tested unloaded, in high winds. This is because there is no where for the power to go and the unloaded emf will be approx 2x the loaded voltage, and the doubling effect will double this again. I know Bolty has been testing his chinese mill with this cap arrangement. It will be good to hear how his system is going. I have tried to look at current and voltage phase relationships in the alternator. I have to current sense with an isolated sensor. I will be looking at a sensing coil on one output leg and this way I will have DC isolation. I can then just use AC coupling to measure the voltage on the same output wrt ground. I should then be able to use a dual channel CRO, or DSO to see phase relationships with and without caps. This should highlight any power factor changes. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Gizmo, I hope this does not cross any forum etiquette. I know many TheBackShed readers frequent other forums and there are many other windmill RE readers on these forums who may not be reading this thread, so I have posted the capacitor findings on the fieldlines forum, with a mention of TheBackShed. I have decided that the potential benefits of using capacitors for the average joe on their windmill is something that I should quickly pass on. I have been persuing an electronic form of windmill MPPTing for years. The boost ccts I developed [analogue and digital] are complicated. I am surprised at how well this new system works. This AC coupling system will still require voltage limiting[regulating]. I see no reason why many kW systems would not benefit as well with reasonably modest components. Gordon. become more energy aware |
||||
Tritium Newbie  Joined: 05/08/2008 Location: United StatesPosts: 1 |
Gordon, Hope you don't mind but in your thread on otherpower there was a question where you posted the schematic so I posted a link to page 15 of this thread where you placed the schematic. It took me 2 days of reading to find it. Thurmond |
||||
imsmooth Senior Member Joined: 07/02/2008 Location: United StatesPosts: 214 |
Gordon, I have a test bench where I can vary the RPM and measure voltage and current. Is there any specific testing you would want done? |
||||
Bolty Regular Member  Joined: 03/04/2008 Location: AustraliaPosts: 81 |
SUCCESS! Thanks to Gordon's fantastic passive component voltage doubler circuit and help from him, we have been able to dramatically improve the output from my 1Kw Chinese turbine. He had always thought that it would be difficult to increase the output at the top end using capacitors as he has on his F&P. This has probably proven to be correct, as the impedance of the Chinese alternator is very low. This means that the values of the series capacitors would have to have been very high. Using series capacitors of 235 uF in each leg of the 3 phase supply leads worked OK at small currents, but presented too high an impedance at higher power, thus limiting the throughput of power. It also meant that my electronic controller could not brake the turbine at excessive wind speeds, due to the lack of loading through the capacitors. Runaway is not something that I was keen on! The other reason why this does not work well on Chinese turbines is that they are very low frequency compared to F&P. At maximum power my turbine operates at around 35 HZ. This is an order of 9 times or so lower than F&P. This would mean massive capacitors of around 5000 uF would need to be used if the top end boost was to work! Where the clever bit with capacitors and the Chinese turbine is in allowing a DC isolation between circuits, and yet still allowing the AC to pass. And all of this is done without active (unreliable) circuits. To illustrate the vulnerability of active components, this exercise has cost me 8 blown Mosfets in my controller. (The controller is a simple short circuit across the bridge rectifier to act as an electronic brake when the voltage exceeds 75 volts) Replacing the mosfets was a trivial exercise compared to the benefits that I have ultimately gained. (The reason why the mosfets blew was due to placing capacitors…as mentioned by Gordon in a posting above…. across the outputs of the voltage doubler rectifiers. When the overvoltage brake activated, the large capacitors were dumped into the mosfets…BANG!... all part of the learning and discovery process…. Those people that have never made mistakes, have never done anything!) So a description of what Gordon proposed and that has worked so well is 2 separate parallel circuits from the turbine 3 phase leads to my Latronics controller circuit. 1. The 3 phase leads from the turbine go through the normal bridge rectifier into the DC input of the controller. This is exactly as per normal. 2. Also from the 3 phase leads from the turbine are 2 sets of 3 phase rectifiers wired as a voltage doubler (as per Gordon's published circuit above) This also connects to the DC input of the controller. So at low wind speeds the voltage getting to the controller, goes through the voltage doubler, giving power at much lower wind speeds. At higher wind speeds, as the current increases, the majority of the current passes through the normal rectifier. However it does appear as though some extra power is being supplied by the voltage doubler path, but I am unsure of how much. The benefits of this are quite dramatic. Prior to fitting this circuit, my turbine had to be spinning at 13 Hz to feed the grid connect inverter. Now with the extra parallel voltage doubler circuit, the turbine connects at 7 Hz. By observation what this means is substantial. Instead of the turbine spinning in low winds, producing nothing, I now get meaningful extra power generated below 13 Hz. This is hard to accurately quantify with my measurements, but an AC current clamp meter in the inverter feed to the mains suggests between 50 to 100 watts extra power under 13 Hz. From 13 Hz up, I suspect that this extra 50 to 100 watts may persist for part of the way up to the maximum. It may even be the case that the extra value of power being generated actually increases. It is very difficult to compare these sort of readings in fluctuating winds in the field. Yesterday my wind turbine generated 4Kwh of energy to the grid connect inverter. Admittedly the wind was averaging 5 m/s . However the big benefit is the almost continuous generation of useful energy at very low wind speeds. If I canget an extra 50 watts for periods when there is very light winds, that will accumulate to be quite significant over a year! Bolty |
||||
| Bolty Regular Member Joined: 03/04/2008 Location: AustraliaPosts: 81 |
Sorry about Word placing the last listing in narrow columns! Hey Gordon, I am amazed by this magnificent circuit. Not only does it give me more power, but I have now discovered 2 substantial fringe benefits. 1. Because the voltage doubler places double the voltage onto the controller, for a significant proportion of the time, the main rectifier is reverse biased, as it has a lower input voltage than is placed on it's output by the doubler circuit. This means that when the grid is disconnected and there is no load on the turbine, the voltage doubler circuit current pumps the voltage output up to the brake voltage of the controller of 75 volts. Then when the brake is automatically applied by the controller, the braking current pass through the doubler circuits and not through the main bridge rectifier (as the main rectifier is reverse biased) This has the advantage of being a softer brake, as the brake current is reduced by the impedance of the series capacitors in the voltage doubler circuit. This gently retards the turbine bringing it rest over about 10 seconds. During that time, the controller actively pulses the brake application, which sounds like about 10 times a second. In contrast when the wind increases rapidly, the brake is applied more firmly, as obviously it is now braking through the forward biased main rectifier. 2. Due to the softer braking, the back emf generated by the braking current is now well under 100 volts. I have observed reading from 81 to 92 volts. BTW I have confirmed that the source of the high voltage (over 100 volts) across the controller capacitors is coming from the turbine or controller ciruitry. I can be sure of this, because I had the inverter totally disconnected during these tests! How the back EMF is being generated, I am unsure, as the controller capacitor bank is disconnected from the turbine during braking. (For those who are wondering my Latronics controller has approx 60000uF capacitor built into it to smooth the voltage going to the grid-connect inverter. This capacitor bank is disconnected from the input ciruitry by a bank of Mosfets, before another bank of Mosfets clamp the input and hence brake the turbine. Triple bonus all round for me for caps on Chinese mills! Everyone needs to go for caps on mills! |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Bolty, The capacitors seem to be working a treat. I think that the point of the capacitor sizing for the boost arrangement is not clear. The capacitors used are 470uF 180VW in the back to back arrangement as pictured on the scematic on p15 of this thread. There are 12 caps used, and the 2 extra 3phase bridge rectifiers. Gordon. PS edit: Hi imsmooth, there will always be room for testing. I am unsure if a testbed can reproduce what happens on a windmill. This may be useful in a micro-hydro comparison. A testbed cannot be used to simulate how blades work in changing tsr conditions that occur with normal alternator loading in wind conditions. The high end power may be the best area to test. If you record a power output v rpm into a battery load, and then add series caps. Record the changes. There may be a reduction in power at low power levels. This aspect is made up for with the voltage doubling arrangement. You had a linear output with rpm into a resistive load from prior testing. I would expect the output curve to bend up with increasing rpm with caps added. Start with 470uF caps at 200V rating for the F&P you have. This is a start. become more energy aware |
||||
| fillm Guru Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Great stuff Bolty and Gordon , Sounds like you have had a major win/win , I would like to try one of those latronics controllers one day on my set up , do you still use batteries with your set up in conjuction with the latronics controller and the capacitor set up ? PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I have covered one of the many possible F&P configurations, and bolty has confirmed benefits with a 1kW chinese mill. I may still consider testing on my AxFx mill and other F&P rewires. The capacitor application and suitability for particular windmill systems will most likely be debated either way. I have read discussions on other forums and there is much debate and sub text discussion. The results that I have presented are based on measured data. It is possible that the improvements are mainly related to the blade matching. The alternator may be benefiting on the sideline. The typical outputs presented for F&P mills seemed to be in the 250-300W region for a single stator/ferrite windmill system. I had seen higher power outputs presented for micro hydro at proportionally higher rpm. I doubt that the F&P is really as bad an alternator as is suggested on other forums. I have recorded the 300% output improvement on my mill from typical performance that had been obtained on a variety of windmills. It is probable that F&P RE windmill enthusiasts have accepted the typical outputs, as the loading reigime used just limited output currents. The Amp*turn has been used to describe this limit. I have suggested that a F&P was typically overdriven[used to overcome cogging]. The reduction in cogging on my mill with the same rotor has enabled me to explore windmill loading. It is possible that the winding and battery loading combinations typically presented to allow a low windspeed mill cutin has severely compromised the top end power harvesting. The windmill has typically been overloaded. I selected a F&P winding configuration that is normally used on a 24V battery with DC coupling. This same winding with AC coupling, the output is completely changed. It is possible that the DC coupled system just completely mismatches the windmill system. The AC coupling allows the mill to effectively operate at twice the voltage, with the associated benefits. The capacitor doubling arrangement recovers lost power at the bottom power levels, with this higher normal cutin rpm, by simply shifting the windmill operating loading levels. The charge pump effect still occurs when the slightest capacitive loading is present. This may cause problems with solid state devices used on the load. I have yet to measure the exact power levels involved, so the jury is out if a bleed resistor will take care of it. This thread is probably going to need some simplifying. I am having trouble with the large number of pages now present. Gizmo, it may be time to look at how this may be presented in the articles section. Gordon. become more energy aware |
||||
| Bolty Regular Member Joined: 03/04/2008 Location: AustraliaPosts: 81 |
Hi Fillm I do not use batteries, just straight from the Latronics controller into the PVE1200 and then the grid! As for Gordon, that is a different story. I would not know where to begin to describe his system. The situation with capacitors is not related to the Latronics inverter. Gordon has shown that capacitors benefit EVERY wind turbine. |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Thanks Bolty, I do not think this thread could benefit from a lengthy description of my setup. I may be tempted to do an article in the future. This could collate a lot of stuff in other posts into one location. Gordon. become more energy aware |
||||
| Page 10 of 25 |
|||||
 Print this page |
