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.
Don B Senior Member Joined: 27/09/2008 Location: AustraliaPosts: 190
Posted: 10:17pm 05 Oct 2008
Copy link to clipboard
Print this post
Just to throw in my two cents worth on this discussion.
I think that much of the conjecture on the value of capacitors can be explained by the Wikipedia entry on maximum power transfer http://en.wikipedia.org/wiki/Maximum_power_theorem.
Essentially, it says that the reactance component of an alternator (winding plus leakage reactance) needs to be countered by a corresponding external capacive reactance to achieve maximum power transfer.
It is important to note that both the inductive and the capacitive reactances can only balance at one frequency, or shaft RPM. Unfortunately, as the frequency increases, the inductive reactance increases, and the capacitive reactance decreases, (and vice versa)and the balance is lost.
It is also important to note that, when the inductive and capacitive reactances are equal, resonance occurs (when there is a driving AC voltage). As the AC voltage source and the inductance are in series inside the alternator, I imagine that the only practical connection for capacitors would be in series to achieve series resonance, and hence maximum current.
On the matter of measuring the inductive reactance of an alternator, one way is to set it up in a test bench with the winding shorted by an ammeter. The speed is then brought up to a value where the winding current does not exceed the alternator rating. If the ammeter is then removed and the open circuit voltage measured at the same speed, the alternator impedance (vector sum of the winding resistance and reactance) is the open circuit voltage divided by the short circuit current.
It is also important to know the frequency at which this test is made, and this will be the number of magnet pairs multiplied by the shaft revolutions per second. From this information plus the winding resistance (measured with the alternator at a standstill), the winding reactance (and hence the inductance) can be determined.
I am sorry, but I have insufficient time at present to add the formula necessary to determine the inductance, but it is quite straightforward.
Then there is also the matter of the 50% efficiency that applies when maximum power is being extracted???
Some food for thought.
Regards
DonB
Don B
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 01:53am 06 Oct 2008
Copy link to clipboard
Print this post
Hi Gill,
The AC across the caps, gives an indication of how much work is done by the caps. Useful info to know when power rating needs to be known.
I have to agree with you re a parallel combo. I only see a more defined peak with a combination, but no real improvement.
Hi herb,
The voltage boost is designed to give power where none would previously have been available. The variable nature of wind enables some techniques that may be slightly less efficient to still give a net gain. These techiques would not necessarily be applicable to hydro, or veg oil ICE, etc.
Gordon.
PS my testing shows that a bottom end voltage booster is more effective than parallel caps.
become more energy aware
"The capacitors are a series configuration and then there is a parallel configuration as well across the phases[Delta]. It does not seem to matter which side of the series caps, the parallel caps are, on the simulator."
I had previously done some tests with 200uF NP caps across each phase and noticed that with no load connected to the 3 phase bridge, there was a considerable load being created by the caps alone across each phase. As usual with my old lathe this was noticed by a significant drop in rpm so that is why I then decided to place the parallel caps after the series caps as depicted in the cct. I suspect that the circulating current through the caps and the voltage across the caps could be measured but with an RMS meter that I don't have.
Cheers,
9c12m
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 01:23am 18 Oct 2008
Copy link to clipboard
Print this post
I have had the opportunity to measure and record the AC current in the stator. The current does flow during the whole cycle, not just when the voltage is above the battery threshold.
I present 2 storage oscilloscope[DSO] pics of measured AC current during similar wind/rpm with and without series caps.
This is with caps
7393
this screen is the mV across the current shunt. the AC period was 4.7mS.
This is without caps
this screen is the mV across the current shunt. the period was 4.4mS, so it was a slightly higher rpm.
This is the same current shunt with very similar wind conditions. There is a marked difference in the measured outputs.
The capacitors appear to provide a lot more than just power factor correction or reducing leakage inductance. The current waveform to the battery was pure DC and I had trouble locking the DSO to any signals.
The caps seem to work in conjunction with the rectifiers to provide current gain. The exact mechanism I hope will be debated and a better understanding gained.
An interesting observation is the variation in the shape of the current output between positive and negative transitions with or without capacitors. The reference of the DSO was connected to the rectifier side of the current shunt. This is a delta wiring of the stator.
Gordon.
Edited by GWatPE 2008-10-19become more energy aware
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258
Posted: 09:58pm 21 Oct 2008
Copy link to clipboard
Print this post
Hi Gordon
Am I seeing a dc shift in the waveforms ? I take it you used dc coupling on the scope.
I do not subscribe that the main effect of caps is to counter leakage , its tied up with advancing the rotor and making better use of the rotor mmf IMHO.
It is very complex but for all practical purposes can be thought of as a means of increasing rotor mmf hence generating more emf at lower rpm this of course translates into current as we feed a fixed voltage.
I believe the Dc shift is the battery effect if you could repeat with a resistor load it would be interesting.
Also I assume the delta is closed. In this case the delta will circulte currents in itself to try to establise a Sine wave ie shorts out all odd harmonics.Hence current is swopped from phase to phase.
Try a separate phase connection (Jerry connection )
Herb
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 11:09am 25 Oct 2008
Copy link to clipboard
Print this post
Hi herb,
The current was measured by the voltage drop across a 350mm length of 2mm copper wire, as a shunt. I do not believe there is a voltage shift in the waveform as such, as there was no external reference for the measurement. When current flows one way, a voltage is developed in proportion to it according to the resistance. When current flows the opposite direction, the voltage is in an opposite,[negative] direction. This is what was seen on the oscilloscope chart.
I measured 605W today into the battery from the std 100S stator in delta with the caps.
I decided to check the AC current and measured a peak of 24.6A at an AC mill frequency of 425Hz. The ACV could have been in excess of 80V between phases. I was unable to confirm this with a measurement. I did record increasing AC output with increasing Hz. Power to the battery maxed at 605W. It would seem that a lot of power is resonating within the inductor/capacitor arrangement. This power may be available with additional transformer coupling on each output. I suspect the current would not be in phase with the voltage. The stator has not burnt out yet, but the caps did rise 25degC above ambient after many hours of punishment.
I will be testing a 7phase with caps soon.
Gordon.
PS edit:
The mill is unable to be wired in jerry rig, as I only have 3 slip rings. Delta wiring is on the stator. 8m wiring to the caps at the bottom of the pole, near the rectifiers and switching components. Edited by GWatPE 2008-10-26become more energy aware
Bryan1 Guru Joined: 22/02/2006 Location: AustraliaPosts: 1344
Posted: 10:37pm 25 Oct 2008
Copy link to clipboard
Print this post
Hi Gordon,
Well mate with it blowing a gale here yesterday I did a test with my genny, I dis-connected the 350 watts of PV so I could see the amps off the genny without the PV input. Now not long after I did see that the 20 amp shunt was topped out so yes I reckon I was seeing 500 watts + out of it.
Cheers Bryan
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 10:34pm 28 Oct 2008
Copy link to clipboard
Print this post
I have a second inverter connected to the F&P mill now. The AC before the caps is high enough to directly feed a 48V inverter when rectified. I have both a 24V battery and a 48V inverter connected to load my mill now. The battery gets charged first and then excess power is fed directly to the AC in my house.
My initial testing indicates higher output power is achieved with caps on a 24V battery load than if the mill was direct connected to a 48V battery load. There appears to still be a loading mismatch with the mill directly connected to the load without caps. I hope to test with caps and boost maximiser into a 48V load next.
Gordon.
become more energy aware
Bryan1 Guru Joined: 22/02/2006 Location: AustraliaPosts: 1344
Posted: 08:52am 29 Oct 2008
Copy link to clipboard
Print this post
Hi Gordon,
Eh mate what do ya reckon about 'Jerry Rigging' the 100s, meaning bring all 6 wires down the tower 216uf of non-polarised caps on each wire then rectified thru 3 seperate 35amp bridge rectifiers. Would this in your opinion be a better configuration or would each pair of caps cancel each other out. I do have a 6 wire cable I made up for my tower so if you think this could be a goer I can try it this weekend.
Cheers Bryan
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 09:29am 29 Oct 2008
Copy link to clipboard
Print this post
Hi bryan,
There should not be a problem with jerry rig. You will need to check if polarity is a problem.
If there are circulating currents from the delta connection resulting from a not pure sinisoidal output, then more current is in theory able to be produced in jerry rig. The extra diodes may reduce the diode loss slightly as well, with the reduced individual diode currents.
I did check the output waveform with the oscilloscope. The waveform is a close sinewave, so a benefit may be hard to see in jerry rig.
You will need to speak to the wind gods. I have not seen the long range forecast yet.
Gordon.
become more energy aware
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258
Posted: 01:52pm 29 Oct 2008
Copy link to clipboard
Print this post
Hi Gordon Bryan
The Closed Delta is removing harmonics ( distored waveform )hence the reason you see better sinewave. Infact your scope shows a sinewave plus the even harmonics that cause asymetrical distortion. With Jerry rig the odd harmonics will show up, could have some power available in them better in batteries than heating windings. Hard to say how much . The grid uses delta to get a better sinewave but here as we are rectifing at any rate wave form is not so critical we might as well have power.
Herb
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 12:11am 30 Oct 2008
Copy link to clipboard
Print this post
Well Bryan, the ball is in your court.
My weather station is saying wind and possible rain in the next day or so. You may want to rig tonight if you havn't done so already. I have redone my test rig. I am making a new higher voltage boost cct, so I hope to look at a twisted pole stator and cap interactions with a battery load and direct inverter load some more. A grid connected-grid feeding arrangement helps. I don't mind recycling current this way at the 1kW level. It is difficult to justify RE testing if the power is wasted in a dump load.
Hi Herb,
The harmonics you mention, I have not seen with the oscilloscope. The twisted pole fingers has smoothed the output waveform on my mill.
I noticed that direct rectification of the AC into a battery reduced the output. It seems that the voltage clamping, and resultant highly non linear current profile affects the stator performance in a negative way. The harmonic interaction within the stator to produce a better output waveform may allow capacitors to work better, with a net power output gain.
In the jerry rig, the inductor capacitor relationship will be as an independent unit. This will not be affected by the interactions of the outputs from the other phases. The output seen by the battery load will be warts and all so to speak. The capacitors will need to handle the potentially more peaky waveform. It is possible that non damaging heating in the stator may be transferred to damaging heating in the caps. Is this a possible outcome?
Gordon.
become more energy aware
Bryan1 Guru Joined: 22/02/2006 Location: AustraliaPosts: 1344
Posted: 07:32am 30 Oct 2008
Copy link to clipboard
Print this post
Hi Gordon,
Mate I'm that busy at work I aint had time to buy some 35amp bridge rectifiers, tomorrow sounds just as bad as I have a breakdown gearbox to fix and untill it's finished theres no 'beer'o'clock' friday night . Anyway I found some 20 core wire I had for years out back so hopefully it's not too far gone and if so I'll have enough wire to set my genny up on the pad I made a few years ago and go 20 metres up. Really I do need to get another 100 series F&P so if Dennis is around . One goal I'm doing is NOT twisting poles and leaving the stator standard My opnion on twisting is that there isnt enough wind around for charging and your blades are pvc crap time to think new site and decent blades.
Cheers Bryan
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 11:51am 11 Nov 2008
Copy link to clipboard
Print this post
Hi Bryan,
I can sort of see where you are coming from with the pole twisting comments. I personally think that a windmill should be given every opportunity to generate and allow harvesting of power to a battery, or other load. Some locations do not allow a better placement of the mill. My testing indicates the power levels to be in the 0-40W range that seem to be gained by twisting the poles. I do agree that a larger rotor is a better way of harvesting power at the low windspeeds though. I had a 3m rotor on my mill, but I had to reduce the dia to 2.4m and adjust the tail boom as the mill could not be electrically braked at the larger size. Strong sustained gusts could overcome the drag of shorted windings and the mill would effectively free run. Was easier to reduce the blade and mod the tail boom, than add a mechanical brake. I still have a spare set, 3m dia blades to experiment with on the 7phase.
On the subject of experimenting.
Has anyone else with a F&P mill, any feedback on success/failure with series caps?
Gordon.
become more energy aware
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 09:09am 12 Nov 2008
Copy link to clipboard
Print this post
I can interpret a lack of comments as
Existing F&P owners
1. have not tried series caps.
2. do not intend modifying existing setups so refer to point 1
3. do not need additional battery charging capacity, so refer to point 1
4. don't have access to capacitors so refer to point 1
5. are concerned about a voltage safety aspect so refer to point 1
6. do not believe results presented so refer to point 1
7. are confused about connections so refer to point 1
8. are worried that their existing setup will be less efficient so refer to point 1
9. probably a host of other reasons so refer to point 1
I may appear a bit cynical.
I know there has been a lot of interest in this aspect from the number of visitors/members viewing this thread and I would have hoped that in the time [almost 4months] since Bryan and I presented the results that at least one other F&P windmill owner may have tried series caps, and would have an opinion either way.
For those readers worried about efficiency, I should point out that a windmill application should not be confused with that of say a fossil fuel or other controlled power input application. A windmill has no comparisons to a motor driven test rig or a hydro system, where efficiency is important as there is control over the power input. The wind is not controllable[yet] and there are many inefficiencies not controllable. There are aspects that affect efficiency like exact tsr and perfectly loading the alternator, windmill furling angle, rotor angle to the wind etc.
I have a F&P windmill system pretty std minimal mods, that was as well matched to the load as many others and I saw a 100% power output increase without resorting to late furling. I don't care if the windmill consumed an extra 20 watts of wind energy to produce the extra 15W of output power.
I will step down from my soapbox now and see if I get a response from rattling the cage.
Gordon.
become more energy aware
Don B Senior Member Joined: 27/09/2008 Location: AustraliaPosts: 190
Posted: 09:05am 13 Nov 2008
Copy link to clipboard
Print this post
Hi Gordon,
I have been away since my (only) post on this matter on 6th October. Again, I think that a quick look at the Wikipedia item on maximum power transfer explains what it is that series capacitors achieve.
Series capacitors counteract the reactive voltage drop within the alternator. If the optimum capacitor value is used, the effective impedance of the total series circuit of alternator plus load is reduced to the sum of the alternator winding resistance plus the effective load resistance only.
When the capacitive reactance of the series capacitor equals the alternator inductive reactance at any particular shaft speed (and hence frequency), maximum current will flow. This is, of course, the condition for resonance at that shaft speed, or frequency.
It should also be noted that, ignoring the internal resistance of the capacitors for the moment, including series capacitors does not reduce the efficiency of the total circuit, as there is no energy dissipated within an ideal capacitor (or in the alternator inductive elements). There is, however, a loss in the resistance of the wire connecting the capacitor to the alternator when the resonance currents are flowing, but it should be minor.
In the real world, capacitors do have some internal resistance (which is why they can get warm), but the net effect of using them should be a real increase in current, and hence increaseed power transfer to the load.
I also can't see why you can not connect your series capacitors after the bridge rectifier, which means that you should be able to use electrolytic capacitors, which are easy to obtain in large values.
The real problem is being able to change the value of the series capacitance to maintain resonance as the shaft speed (and hence the frequency) changes.
Maybe, it would be possible to use a large value of capacitor PWM switched in parallel with a smaller value of permanently connected series capacitor to obtain a sort of variable value series capacitor?
The PWM switching could perhaps use some sort of maximum power tracking arrangement to maximise the load current?
This is all untested conjecture, as I don't have an F & P rig up and running as yet, but it is food for thought?
DonB Don B
davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499
Posted: 10:16am 13 Nov 2008
Copy link to clipboard
Print this post
Just a thought . . . perhaps adjusting the load on the F&P to maintain a relatively constant RPM (at the RPM where the caps resonant out the inductive reactance of the coil system), maybe using some form of MPPT as you suggest.
Or do you also have other optimisation problems, i.e. that of blade speed (tip speed?) versus actual wind speed?
Gill Senior Member Joined: 11/11/2006 Location: AustraliaPosts: 669
Posted: 02:05am 14 Nov 2008
Copy link to clipboard
Print this post
G'day DonB,
You read my mind. I thought the same thing a month or so ago. So I designed a circuit and built it. Testing and evaluation is currently under way. Back into it again this weekend. Herb will be pleased to hear I'm using fixed powers instead of fixed speeds for the tests.
Edited by Gill 2008-11-15was working fine... til the smoke got out.
Cheers Gill _Cairns, FNQ
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 12:18am 16 Nov 2008
Copy link to clipboard
Print this post
Hi Don B
I am afraid that capacitors can only pass AC current. DC current only charges a capacitor, or if reversed, discharges a capacitor. There is only DC current after the rectifier.
Hi Gill,
I hope those heatsinks are only for show.
On the matter of switching, or PWM switching of capacitors. The natural tendency of the series capacitor is to pass more power as the frequency increases. This is the relationship the windmill needs to have to follow the wind energy. Adding switching is complicating the natural behaviour. BTW the voltage should increase with frequency. The current should increase with the increased voltage. The series capacitor is the third component to form a cubic power relationship. There should be no need to add complications when the natural relationship is already there.
WRT my comments earlier,
I had hoped that someone else had actually tried a series capacitor arrangement on the AC. From the lack of evidence either way, I gather that F&P owners have not tried the simple series cap arrangement and have no data either way.
Before trying to think of other ways to make things more complicated, maybe the straight series capacitor arrangement should be tried first.
Gordon.
PS. I notice that there have been thousands of viewings since my last comments, with no added data either way from any previous testing, posted.become more energy aware
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 01:18pm 17 Nov 2008
Copy link to clipboard
Print this post
Hi herb,
I can see where you are coming from, but I have already increased the output current from approx 9.5A to 21.5A, without changing blades, or rotor, or number of stators, or stator wiring, or furling or load voltage. My mill produces power proportional to the wind energy until furling.
I have twisted the pole fingers to aid low windspeed operation and hum reduction. This same stator was used in the before and after measurements.
I will need a very good reason to look at any other changes to this mill. I am quite content to find more ways to use the extra energy produced. I think my fridge, computer and lights are using this power at the moment.
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.
Print this post
7393
. Anyway I found some 20 core wire I had for years out back so hopefully it's not too far gone and if so I'll have enough wire to set my genny up on the pad I made a few years ago and go 20 metres up. 
Really I do need to get another 100 series F&P so if Dennis is around 
My opnion on twisting is that there isnt enough wind around for charging and your blades are pvc crap time to think new site and decent blades. 
