Picaxe + MPPT + schematic + wind +code??
- Thread starter domwild
- Start date
premelec
Senior Member
I don't think the MPPT unit cares where the power comes from BUT you need to discern if you have false maxima in the system - off peak peaks under certain conditions... I've got a .pdf paper on this somewhere I'll try to find ref to... unless you are running pretty small power units it can get quite complicated. If you have monotonic increase and decrease around the peak power then you simply vary the pwm a bit and measure power and go again and find the peak... _while_ this process is going on you lose some power - however these losses can be quite small if you don't have to adjust very often. That implies load and wind being fairly steady for most of the time... Perhaps this helps your thinking on the subject! I hope it doesn't confuse you...
BTW the function of the MPPT can be considered as impedance matching - to get max source to load power transfer... the PV source is close to constant current whereas the wind generator could be lower source impedance variable voltage depending on the generator form being used - does it have a regulator etc... you might consult www.homepower.com if you haven't already...
Edited by - premelec on 16/12/2006 17:57:20
BTW the function of the MPPT can be considered as impedance matching - to get max source to load power transfer... the PV source is close to constant current whereas the wind generator could be lower source impedance variable voltage depending on the generator form being used - does it have a regulator etc... you might consult www.homepower.com if you haven't already...
Edited by - premelec on 16/12/2006 17:57:20
BeanieBots
Moderator
I've had a quite a bit of experience of MPPT for PV arrays, but as you have found there is very little out there for wind turbines.
The biggest difference is that a turbine MUST NOT EVER be run off load. This adds further complication depending on its use. If used for battery charging, the excess power MUST be dumped into a load resistor. If feeding an on-grid inverter however, it is probably safe to assume that your little turbine is not going to over-volt the grid but a dump load should still be employed as a safety device should the grid go down due to circumstances beyond your control.
Short of that, the same priciples should apply but with a noteable time lag on voltage drop when increasing load current.
should be quite a challenging project.
The biggest difference is that a turbine MUST NOT EVER be run off load. This adds further complication depending on its use. If used for battery charging, the excess power MUST be dumped into a load resistor. If feeding an on-grid inverter however, it is probably safe to assume that your little turbine is not going to over-volt the grid but a dump load should still be employed as a safety device should the grid go down due to circumstances beyond your control.
Short of that, the same priciples should apply but with a noteable time lag on voltage drop when increasing load current.
should be quite a challenging project.
There are some amazing bargains on ebay, eg 3 metre diameter for $1700Australian which makes MPPT less cost effective. Choices - build it strong/heavy/expensive to survive storms, build it light/big/smart (eg active furling/pitch control), or use commercial units with smart things like blades that flex in high winds. Only one thing counts in the end - $ per output watt. But there are many lies out there in the wind turbine market - eg 1 metre turbines rated at 600W. Realistic average watts are 30W/sq metre.
So the best guide is $/swept area. Next question - is it worth the cost of MPPT or is it better to get a slightly bigger turbine?
I can answer this in more detail, but first, are you running DC into a battery bank, or single phase/3phase grid intertie?
So the best guide is $/swept area. Next question - is it worth the cost of MPPT or is it better to get a slightly bigger turbine?
I can answer this in more detail, but first, are you running DC into a battery bank, or single phase/3phase grid intertie?
Dr-Acula and many others,
Thanks for that. I am going to charge batteries (36VDC) as I have a 36VDC UPS to convert to 240VAC. I do not wish to harvest power below the cut-in speed when the alternator reaches 36VDC and starts charging. From the little I know, it could be a Buck circuit I am looking for to better match the output to the 36VDC battery bank.
Regards,
dom
Thanks for that. I am going to charge batteries (36VDC) as I have a 36VDC UPS to convert to 240VAC. I do not wish to harvest power below the cut-in speed when the alternator reaches 36VDC and starts charging. From the little I know, it could be a Buck circuit I am looking for to better match the output to the 36VDC battery bank.
Regards,
dom
You may well be able to harness some ideas from the great DIY wind site => www.thebackshed.com <A href='http://www.thebackshed.com/Windmill/default.asp ' Target=_Blank>External Web Link</a>. The guru author, Glenn (a.k.a "Gizzmo"
,is based in Mackay (Queensland-Australia) & is very Picaxe orientated too.
MPPT (Max. Power Point Tracking )insights => <A href='http://www.solar-electric.com/charge_controls/mppt.htm ' Target=_Blank>External Web Link</a>
Edited by - manuka on 17/12/2006 19:36:42
,is based in Mackay (Queensland-Australia) & is very Picaxe orientated too.MPPT (Max. Power Point Tracking )insights => <A href='http://www.solar-electric.com/charge_controls/mppt.htm ' Target=_Blank>External Web Link</a>
Edited by - manuka on 17/12/2006 19:36:42
Good links from Manuka. No reason a buck converter won't work. I presume we are talking low power here - a few amps at most. The open circuit voltage of the turbine isn't stated, but if the buck converter is, say, 90% efficient, one might have 4 states - open circuit <36V, directly connected to load 36 to 40V, buck circuit with variable load 40V+ and then shutdown in high wind. What sort of buck converter are you looking at?
Dr-Acula,
Hope to get 10 Amps max. Have given the tough nut of the buck converter to an ex AT&T "Mr Analogue" as I do not have a black belt in electrickery and have stated that I can handle the code and schematic for the PWM section. You are perfectly correct with the stages. The alternator is a Fisher and Paykel washing machine motor well known to Oz and Kiwi windpower enthusiasts.
Thanks for the Gizmo hint; Glenn has helped me already quite a lot. His existing code and schematic on his site covers RPM measurement, load shedding and windspeed measuring and I am well aware that MPPT is the holy grail of windpower and even experts like Glenn, Amanda, etc. are struggling with the MPPT problem as far as I know from the Fieldlines forum.
As this is the Picaxe forum it might just be a case of this problem having been solved by someone with the Picaxe as the pwm command allows very convenient adjustment of the pulses up or down. I have seen it being used in the desulphation of batteries.
Code and schematic might just be available - I am dreaming again.
Hope to get 10 Amps max. Have given the tough nut of the buck converter to an ex AT&T "Mr Analogue" as I do not have a black belt in electrickery and have stated that I can handle the code and schematic for the PWM section. You are perfectly correct with the stages. The alternator is a Fisher and Paykel washing machine motor well known to Oz and Kiwi windpower enthusiasts.
Thanks for the Gizmo hint; Glenn has helped me already quite a lot. His existing code and schematic on his site covers RPM measurement, load shedding and windspeed measuring and I am well aware that MPPT is the holy grail of windpower and even experts like Glenn, Amanda, etc. are struggling with the MPPT problem as far as I know from the Fieldlines forum.
As this is the Picaxe forum it might just be a case of this problem having been solved by someone with the Picaxe as the pwm command allows very convenient adjustment of the pulses up or down. I have seen it being used in the desulphation of batteries.
Code and schematic might just be available - I am dreaming again.
Like premelec says, this is impedence matching. If the alternator happily produces 100V at 10A in a good wind, and were connected directly to a 36V battery it would only produce 360W instead of 1000W. There is certainly some advantage in MPPT.
1) I presume there is shutdown in high wind - preferably mechanical like auto furling. Trying to shut down by dumping to load resistors may end up melting the coils in a storm.
2) The picaxe can do the smarts and it will need some inputs. The alternator can charge a biggish capacator - measure the volts on that with a resistor divider. Also measure the current being drawn from this capacator via a differential amplifier measuring the volts on a 0.1ohm resistor. RPM may or may not be needed - I'll think more about that.
3) Picaxe would: - measure input volts, measure amps, multiply to get watts, increase load a bit via buck, wait a bit, measure watts again, if went down then next time, decrease load, and if went up then next time increase load.
4) Buck converter might need an analogue input to control it. I'm thinking that the pwm from a picaxe won't be fast enough to control it directly. I have used standard 50Hz power transformers for buck/boost circuits and the picaxe can control these directly as the timing is 7-10ms. But a 1000W transformer is very big. Maybe better to ask your friend to design a buck running at 10Khz with an analogue input.
5) The buck converter is going to be the complicated bit. Maybe the schematic of a PC power supply might be a place to start. Others might be able to help here...
1) I presume there is shutdown in high wind - preferably mechanical like auto furling. Trying to shut down by dumping to load resistors may end up melting the coils in a storm.
2) The picaxe can do the smarts and it will need some inputs. The alternator can charge a biggish capacator - measure the volts on that with a resistor divider. Also measure the current being drawn from this capacator via a differential amplifier measuring the volts on a 0.1ohm resistor. RPM may or may not be needed - I'll think more about that.
3) Picaxe would: - measure input volts, measure amps, multiply to get watts, increase load a bit via buck, wait a bit, measure watts again, if went down then next time, decrease load, and if went up then next time increase load.
4) Buck converter might need an analogue input to control it. I'm thinking that the pwm from a picaxe won't be fast enough to control it directly. I have used standard 50Hz power transformers for buck/boost circuits and the picaxe can control these directly as the timing is 7-10ms. But a 1000W transformer is very big. Maybe better to ask your friend to design a buck running at 10Khz with an analogue input.
5) The buck converter is going to be the complicated bit. Maybe the schematic of a PC power supply might be a place to start. Others might be able to help here...
BeanieBots
Moderator
Controlling any switcher from a PICAXE is quite straight forward. All you need is enough information about the power supply to identify the feedback circuit. Then, instead of connecting the input to the normal divider circuit for voltage control, connect it to a comparitor (low gain configuration, say X100). One input to the comparitor is the original feedback signal, the other is the smoothed DC PWMout from the PICAXE. The PICAXE can now control the voltage demand of the switcher.
You might need to play with the bandwidth components to make it more simple but just a little more 'slug' is usually all that's need.
Edited by - beaniebots on 18/12/2006 18:07:21
You might need to play with the bandwidth components to make it more simple but just a little more 'slug' is usually all that's need.
Edited by - beaniebots on 18/12/2006 18:07:21
Thanks for further replies. Stopping in high winds I would do via furling BUT have seen electronic stopping in an Air-X wind turbine ad. Load shedding is done when the batteries are full.
Yes, Dr-Acula, this is how I would program the logic, too:
Calculate watts and increase PWM pulse length and see what happens with the watts.
Increase/decrease pulse length constantly to feed the max of Amps into the batteries.
Once batteries are full, dump load usefully.
Yes, Dr-Acula, this is how I would program the logic, too:
Calculate watts and increase PWM pulse length and see what happens with the watts.
Increase/decrease pulse length constantly to feed the max of Amps into the batteries.
Once batteries are full, dump load usefully.