 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 : Electronics : maximising wind generation
| Author | Message | ||||
Haxby Guru  Joined: 07/07/2008 Location: AustraliaPosts: 423 |
Hello all, I am old to the picaxe and electronics world but new to wind turbines. I read a MPPT question from Domwild on the picaxe forum and I would like some input on why my idea will/won't work. Here is my post on the answer unedited. I am interested in this groups comments: Hi Domwild, So you say that the two inputs to the calculation are the RPM of the turbine and the instantaneous power being used. The output of the calculation will then control the PWM of the batteries to be charged. Right? So the aim is to regulate the charging of the batteries while keeping the power output of the turbine at its maximum? OK here is another way I would tackle it: Strap your wind turbine to your car and on a still day, drive your car at 10km/h, 20km/h, 30km/h .....up to 100km/h You will need a variable load on the turbine. For each constant speed multiple, vary the variable load and note the point at which maximum power is produced. Then write down the RPM of the turbine at that point. Now you know what RPM the turbine will have to be running at to produce power at maximum efficiency for any given wind speed. Next, when the turbine is mounted on a pole, all you have to do is have a separate wind measuring instrument mounted somewhere on the same pole. To generate at maximum efficiency, the problem is then very simple. main: check wind speed check turbine RPM lookup table is turbine RPM lower than best efficiency? If yes turn off charging (this will increase RPM) if no, turn on charging (the load will decrease RPM) Goto main The faster this program runs, the faster the PWM. But even a dead slow 10Hz PWM would be fine. When the load is off, energy is being built up in the turbine in the form of higher RPM. Now for the charging circuit. You mention you are using a Fisher and Paykel washing machine motor: If you don't re-wire it, it will give you a relatively high voltage, somewhere around 100 to 350V DC. This is great for a reasonably long run of cable back to the shed. Most people like to fiddle with this and re-wire to 12 or 24V. I wouldn't bother. You just need to find a way to drop this to a useable voltage once at the batteries. Well it just so happens that a computer power supply is very efficient, has a SMPS built in, is cheap, and can work on a DC input of around 90 to 350V. (check the lable) The first thing a SMPS does is run 240V AC through a bridge rectifier and then to a capacitor. So DC won't hurt it. Now if you need more than 12V then you can stack multiple power supplies together. But there is a small modification you need to do to do this safely. (I will elaborate if you like) If you have a 36V pack, you will need 4 computer power supplies. Wire the 4th power supply between the 5V output and 12V output (7V). This will give you a total of 43V which is about 14.3V per 12V battery. Close enough. Then select your favourite mosfet to pulse the current when required. Is this not doing the same thing, without complex maths? OK so I didn't take into consideration the effect that the temperature of the air will dictate its density and perhaps alter the maximum efficiency at a particular RPM. This might make the lookup table only work at one particular air temperature. But I dont know the maths to work out whether this is a negligible effect or not. |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Interesting approach. Seems reasonable to me. I could see some issues driving around with the mill on a vehicle. Perhaps you can get away with that in the outback! That method would give you a controlled test environment and one that tests the whole system. One comment on: *** and can work on a DC input of around 90 to 350V. (check the lable) The first thing a SMPS does is run 240V AC through a bridge rectifier and then to a capacitor. So DC won't hurt it. *** shouldn't that be 90 to 350Volts DC. And in this case "the first thing the SMPS does is run 90-350VAC . . .". Not sure what the comment about "So DC won't hurt it". At this point in the circuit you will have pulsating DC across the cap (after the bridge rectifier). IMHO, your method would appear to remove a lot of the "hit and miss" in getting the most out of the whole machine. But then again I am relative beginner in this game as well. Good luck |
||||
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Haxby, Your approach is the only one I have seen from a pic person who does not want to try and do the impossible. Dom's current discussion is tilted towards failure as I see it because the feedback he intends to use will affect itself and cause it to oscillate or stall or both. At least you are trying to use real world information to make informed decisions, rather than feedback affected decisions The only operational problem I have with this approach is getting accurate figures on wind speed that the blades are seeing. Even if the wind is say 20 kph, the blades may be not looking in the right direction (furled, partially furled, false seeking etc). If we put the wind speed device not only on the same pole, but also of a HAWT type (maybe a computer fan without the stator (iron loss problems), but with the hall effect chips), pointing the same direction as the blades are, we stand a chance that this may work. This may even mean building it into the front of the hub (interesting engineering). Davef, If you can sift through this post http://www.fieldlines.com/story/2007/10/18/16953/116 you will get a better idea of psu behaviour, and what you can achieve in voltage range on the front end. Dinges notes that at low voltage it may not start until a higher voltage, but will run down to a lower input voltage after start. This is to do with how the bootstrap circuit gets the power over to the low voltage side to power the 494 pwm chip. It relies on a single turn in the primary to give some positive feedback to generate a wave train to get the whole show on the road. If the dc input to the switchers is too low, the feedback is not enough to create oscillation, and so the 494 does not get woken up. Once running though, the 494 makes it's own voltage until the input is just too low (usually about 50vac without the doubler turned on). Changing the fets to higher voltage and the caps as well, will extend this range still further. The ripple on the input caps gets less problematic as the voltage rises. The turnon of the switchers happens about 30000 times a second, and so can easily change pulse width to track the 50 hz ripple if it needs to. Higher voltage usually means less turn on time, more dead time, and less chance of running out of usable voltage to drive the tranny before it switches off anyway. The halfbridge topology seems not to mind a bit (lot) of ripple in this application. In fact if we use lower uf caps, we can actually get an improvement in power factor as we use more of the sine wave up. I have had some units with almost no caps on the input, and the thing has to track as much of the sinewave as it can. This stops it from just using the top of the sinewave, and forces it to use more of it. Isolating the psu secondaries above ground is necessary if you dont want short circuits from them when you series them up. A rubber "o" ring under the board where the screws hold it to the chassis is all that is needed to stop it grounding. ..........oztules Village idiot...or... just another hack out of his depth |
||||
| Haxby Guru Joined: 07/07/2008 Location: AustraliaPosts: 423 |
Thats good feedback, thanks. I was wondering if I was missing something basic. I had another idea: With regard to the problem of cogging in the F&P motors: Use the same external wind sensor to sense when there is enough wind for the turbine to be operating. Then if the blades are not spinning due to cogging, the circuit can briefly apply power to the turbine to get over the first step. I am collecting F&P motors at the moment. After I have a few I will make up some blades and start experimenting. Early days yet. I wonder if the density of air with respect to temperature will be of any significance. If it is, it can always be incorporated into the equation with an external temperature sensor and separate lookup table. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
If you can design a wind sensor that properly reflects the power available to the blades in whatever orientation they may be at the time, you can get up to all sorts of tricks I guess. For the decogging bit: The simplest driver I have used for this kind of thing is three fets, three hall effects (come with the motor anyway and already mounted), and 9 resistors. It is crude but will work to start the machine. Short runs only I suspect as it is brutal and barbaric..... but works for short periods with poor efficiency. (I would use a current limiting resistor as well) Otherwise a proper 3 phase dc switching circuit to do the job would be a serious project. ........oztules Village idiot...or... just another hack out of his depth |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
MPPT paper Interesting for both MPPT and PFC. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Davef, A fairly fulsome discussion was had here: http://www.thebackshed.com/Windmill/FORUM1/forum_posts.asp?T ID=1111&PN=1 I think... good on a test bench, probably failed on a real turbine in the real world. As we haven't seen anything more from this in the real world for a few years, it may be safe to assume that it was a fizzer. Gill pointed to "News is that MidNite Solar(former design engineers with Outback) in Arlington USA are soon to test a beta version of a wind MPPT the 'MidNite Clipper'. Looking forward to any reports on how it goes" http://www.midnightsolar.com/ These boys are smart cookies who have been chasing this angle for some years with lots of experience in the wind world with outback and inverters to suit (12-48v). They are about to unveil a working system on an obscure voltage turbine (250vdc), which makes me feel that they haven't got a turnkey model for normal voltage machines yet. One assumes that it is not as easy as the paper you refer to suggests. Their press release on 20/6/08 is: Classic controller "The MidNite Classic MPPT controller will be demonstrated in Custer Wisconsin at the MidWest Renewable Energy Fair. One 250VDC controller will be connected to a wind turbine and one 150VDC unit will be connected to a Solar array. In addition to the Classic, the MidNite Clipper will be demonstrated. Industry Beta testing will commence late Summer. This combination of Classic and Clipper is sure to change the small wind industry in how battery charging is done." Now they know full well that a normal voltage version would sell well. I can only assume that the beta version is not all things to all people, and working at this impedance should give them lots of flexability... so I don't have confidence at this stage that they have the real answer yet. This is probably a compromise between what they wanted and what they could get to work realistically. ......oztules Village idiot...or... just another hack out of his depth |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
oztules, Thanks for the heads-up. I was just given the link to this article today at work. I'll check out the previous discussion. Cheers, davef |
||||
herbnz Senior Member  Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Hi All At last some realistic approachs. I have been working on old computer supplies for some time yes you can modify easily by breaking into the feedback cct modify the output circuits. At present I am pushed for time to develop much but will watch this thread with interest. What we need is ideas on boost and/or buck ccts so so more people can build SMPS that can be controled then control ideas can be tried by all Herb |
||||
| domwild Guru Joined: 16/12/2005 Location: AustraliaPosts: 873 |
Haxby et. al., Thanks for the comments. From comments by others I have always known that MPPT and PWM go together. Then I have found this 770 kb abstract of a brazilian Master Thesis (listed in Haxbie's post), which once again uses PWM to drive a SEPIC buck/boost DC/DC converter for battery charging of a small 400W mill. Pls look at this article how this power engineer solves the problem and also fixes up a power factor problem at the same time. My knowledge of power electronics is insufficient to argue for or against the ideas presented in this abstract, however, my crude pseudocode example in a previous post makes the valid point that PWM changes the pulse width based on the power produced and tries to find a sweet spot. My knowledge of low-pass first-order filters is nil and Fieldlines, Picaxe UK and this forum have not been able to help me so far to understand the abstract, where such a filter is used to get at the instantanious power of the mill to drive the PWM. Taxation as a means of achieving prosperity is like a man standing inside a bucket trying to lift himself up. Winston Churchill |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Dom, I will try to help you to understand (in what I feel is a flawed thesis) what they are on about with this first order low pass filter stuff. A first order filter is simply a filtering network designed to halve the power of the signal for every octave/harmonic/doubling of frequency which the signal's frequency varies from the center frequency. A low pass filter is one where the filter favours the low frequencies, and attenuates the high frequencies. ie it passes the low frequencies well, and attenuates the high So, a first order low pass filter is simply a low frequency filter that passes low frequencies well, but attenuates higher frequencies at the rate of halving their power for each doubling of frequency. This is a pretty sloppy filter, but enough for this purpose... what purpose. They want to know what the voltage is, but with all the hash of the pwm and inductance switching harmonics, diode non-linear noises and harmonics, the thing we are trying to measure is partly clouded/obscured by electrical noise. As a test, try to measure the frequency of a mill once cut in is achieved with a multimeter that measures frequency. It works well right up until cutin... then it all goes to hell. This is what they are trying to avoid Luckily, most of these noises are well removed from the very low frequency of the emf being generated, and so they use a low pass filter to pass only the low frequencies through, and attenuate the higher frequencies at the rate of halving their respective power levels each time the frequency doubles. This simply allows the frequencies they are interested in to STAND OUT amongst the crowd so to speak, and get a measurement mostly removed from the hubbub of all the switching noises which would skew the measurement and so render it useless (or near so). So for all the fancy sounding words, what they mean is,: First we need to measure the voltage. It is clouded in electrical noise, so we need to filter out the hash so we can read the voltage clearly. We only need a fairly loose filter for this (first order) because the hash is well removed from the signal we are trying to get at. Now we can use a resistor capacitor network like you may have heard Flux refer to for trying to measure RPM by measuring the frequency. Same problem, same fix. Use an attenuating filter. You can use a inductor/capacitor circuit (good for low frequencies in your bass loudspeaker cross overs... see you have heard of low pass filters before)..... and I think you get the drift by now. A simple analogue coil meter will do the same job without the filters because it is protected from the high frequencies by the mass of the moving parts. You pic people will have difficulties designing filters good enough to stop getting noise. If you remember Sparwebs data log graphs, they had more spurious graphing points than was useful. I had trouble trying to guess where he was drawing the line. This was from noise, and measuring when the waveform wasn't where it was supposed to be..... all becomes noise as far as the program is concerned. Before you get completely hooked on this thesis, study figure 13 carefully. If a wind turbine can behave like that then physics has all changed. The simulated turbine jumped from a steady state 10ms to a steady state 12 m/s, and held the new rpm for 4-5 seconds regardless that the pwm was fiddling around doing, before it caught up. To consider this as anything more than wishful thinking mppt is beyond anyones wildest dreams. The intervening 4-5 seconds, the turbine would have oversped to hell, and the next readings would be fanciful. This algorithhym cannot work. Look at how it climbed the graph on the right. It looks sensible... nice wobble trying to sniff out the correct pulse width to match the new levels,.... until you realise that the graph on the left says power input remained rock steady while the mill tried to find the new power level. In the real world this thing would be completely unpredictable. It would wildly girate between overspeed and stall. This is only a Masters thesis, it is just to show he knows how to approach and present a research paper. Not that he has got it right. Try your own code, and work with it. If you are lucky, you may get it to approximate something useful, but these fellows didn't. PFC is a byproduct of continuous mode booster topology. The current and the voltage are in phase. If you look inside higher power switching supplies, they may well have a single/dual fet boost stage before a H-bridge/half bridge etc final stage. This is to placate the power generating authorities. ie present a near perfect power factor. Here is a lad who plays with plenty (8-10kw) pfc booster: http://www.stevehv.4hv.org/12kwPFC.htm Any inverter airconditioner/heat pump will have a boost circuit before the H bridge output stage (yes 3,4 or 5 KW boost converter)... all to improve the power factor of the compressor. The sepic isn't special in this regard.... even though it appears that way in the thesis. Has this helped clear up the low pass first order filter and pfc thing up? .........oztules Village idiot...or... just another hack out of his depth |
||||
 Print this page |
