My Wind Generator Was Useless. Until NOW! (MPPT)
video description
Date: 2025-02-27
Related videos
Comments and reviews: 20
kriswillems5661
As mentioned before turbines are not solar panels. The voltage and the current have an influence on the rotation speed and thus the wind flow. The mechanical power input is not constant when you change voltage and current. I would follow the following approach: measure the wind speed by letting the turbine run without load (measure the voltage. Ideally you have a separate small turbine for wind speed measurement or a pitot tube, like in airplanes. For each wind speed, find the optimal voltage and current by stepping through the current range (you can use optimization techniques here. This process might take an hour, as you need to wait for mechanical stabilization after each step (A solar controller is just too fast for this. Then memorize the setting that gives most power. You'll need to redo the measurements many time to get more accuracy. After a while the controller can learn the optimized settings, and the can interpolate the V and I settings based on the measured wind speed and the data tables you collected. When measuring wind speed with a separate turbine you could use polynomial curve fitting to try to predict the behavior of wind over time, and adjust your setting a bit ahead of the actually windspeed increase or decrease.
reply
As mentioned before turbines are not solar panels. The voltage and the current have an influence on the rotation speed and thus the wind flow. The mechanical power input is not constant when you change voltage and current. I would follow the following approach: measure the wind speed by letting the turbine run without load (measure the voltage. Ideally you have a separate small turbine for wind speed measurement or a pitot tube, like in airplanes. For each wind speed, find the optimal voltage and current by stepping through the current range (you can use optimization techniques here. This process might take an hour, as you need to wait for mechanical stabilization after each step (A solar controller is just too fast for this. Then memorize the setting that gives most power. You'll need to redo the measurements many time to get more accuracy. After a while the controller can learn the optimized settings, and the can interpolate the V and I settings based on the measured wind speed and the data tables you collected. When measuring wind speed with a separate turbine you could use polynomial curve fitting to try to predict the behavior of wind over time, and adjust your setting a bit ahead of the actually windspeed increase or decrease.
reply
MrQuick999
Put Capacitors between battery and wind turbine or solar panel. For Wind: Batteries can't charge as fast as wind can get for a strong wind turbine but Capacitors can fill with it then bleed into battery when wind comes back down below max battery charge rates. If you have unstable wind where you are where its not a constant max unless a storm = $$$. For solar: Get panels more than you need and use the same trick for clouds or other things causing problems the capacitors can act as a buffer for more stability. < Volt limiter for capacitor to batteries also one way blocking diodes are good. You know why lol.
Personally prefer Cold Steam Tesla turbine systems. Reason: Heat with anything you can burn and cold steam drops heat requirement down enough for Hyper Thermophilic bacteria on a high density compost nutrient drip to run using human waste. < Hyper Thermophiles can multiply double every 24 hours in their optimal conditions which lower Thermophiles get ready for them with their own heat nutrients. No sun, no wind, and as long as Humanity has life then Thermophiles have fuel. Gonna be perfect for space habitats. Also Picks up soil from Mars and lets it blow in the wind No food and cold No care. They sleep.
reply
Put Capacitors between battery and wind turbine or solar panel. For Wind: Batteries can't charge as fast as wind can get for a strong wind turbine but Capacitors can fill with it then bleed into battery when wind comes back down below max battery charge rates. If you have unstable wind where you are where its not a constant max unless a storm = $$$. For solar: Get panels more than you need and use the same trick for clouds or other things causing problems the capacitors can act as a buffer for more stability. < Volt limiter for capacitor to batteries also one way blocking diodes are good. You know why lol.
Personally prefer Cold Steam Tesla turbine systems. Reason: Heat with anything you can burn and cold steam drops heat requirement down enough for Hyper Thermophilic bacteria on a high density compost nutrient drip to run using human waste. < Hyper Thermophiles can multiply double every 24 hours in their optimal conditions which lower Thermophiles get ready for them with their own heat nutrients. No sun, no wind, and as long as Humanity has life then Thermophiles have fuel. Gonna be perfect for space habitats. Also Picks up soil from Mars and lets it blow in the wind No food and cold No care. They sleep.
reply
bartowl
I hope you know that mppt for solar needs to work completely differently than wind ideally for wind you would need external wind speed sensor because at high winds you need to let-off load to let the wind turbine to climb on to right rpm and only then increase load. You basically have a wind curve pointing to most optimal rpm for given wind speed and then you need to adjust load to stay within that rpm. It is a bit like cascade of 2 pid controllers. for solar you typically just have one stage. Measuring rpm can be done by tapping ac voltage from before rectifier and detecting zero crossing. Typically a 3 phase rectifier bridge and properly set minimal voltage for turning on load should already be way better than those el-cheapo mppt drivers sold with turbines. But if you really want to harvest every single watt things get hairy the problem is that without external wind sensor you will never know if you are not loading the turbine too much and it just cannot spin-up to handle such or even greater load. And if course there is the entire gust handling and brake/safety thing in stormy winds i personally never met an optimal wind controller. You have my fingers crossed!
reply
I hope you know that mppt for solar needs to work completely differently than wind ideally for wind you would need external wind speed sensor because at high winds you need to let-off load to let the wind turbine to climb on to right rpm and only then increase load. You basically have a wind curve pointing to most optimal rpm for given wind speed and then you need to adjust load to stay within that rpm. It is a bit like cascade of 2 pid controllers. for solar you typically just have one stage. Measuring rpm can be done by tapping ac voltage from before rectifier and detecting zero crossing. Typically a 3 phase rectifier bridge and properly set minimal voltage for turning on load should already be way better than those el-cheapo mppt drivers sold with turbines. But if you really want to harvest every single watt things get hairy the problem is that without external wind sensor you will never know if you are not loading the turbine too much and it just cannot spin-up to handle such or even greater load. And if course there is the entire gust handling and brake/safety thing in stormy winds i personally never met an optimal wind controller. You have my fingers crossed!
reply
ZoeyR86
You don't need any of that use an open source bldc controller like the Vesc or Odrive atach to the phase windings, with some software tricks and independent wind sensor you can make it kick start it's self in low winds and use mppt to drive regen braking this skips the diodes and uses the mosfets as an ideal diode aray, you can also use the back emf to get accurate rpm once calibrated. Odrive has independent brake resistor support as well so you can load dump power.
The hardware is fairly easy to do any of this, but the software will be a little tricky to nail down. You almost want to make the mppt tracking flip flop extremely aggressive so it can watch input voltage and current and have it bounce between the sides of the peek at a crazy rate north of 1khz with a pwm north of 30khz and save the inductors for the output filters you can just treat the source as the inductor it's a big coil of wire lol
reply
You don't need any of that use an open source bldc controller like the Vesc or Odrive atach to the phase windings, with some software tricks and independent wind sensor you can make it kick start it's self in low winds and use mppt to drive regen braking this skips the diodes and uses the mosfets as an ideal diode aray, you can also use the back emf to get accurate rpm once calibrated. Odrive has independent brake resistor support as well so you can load dump power.
The hardware is fairly easy to do any of this, but the software will be a little tricky to nail down. You almost want to make the mppt tracking flip flop extremely aggressive so it can watch input voltage and current and have it bounce between the sides of the peek at a crazy rate north of 1khz with a pwm north of 30khz and save the inductors for the output filters you can just treat the source as the inductor it's a big coil of wire lol
reply
d614gakadoug9
If you really want the maximum power from the generator the best solution is probably a 3-phase (assuming that's what the generator produces) active power factor correcting MPPT system. That isn't a trivial thing to implement. It may, however, be the case that the generator output is soft enough that simple capacitive filtering, which usually would produce horrendously bad power factor, isn't too terrible. The higher the capacitance you use, the worse the power factor.
Don't forget that you may need to add additional circuitry to keep the maximum speed of the generator in check. You can't do simple things like shorting the windings unless their resistance is high. If you short a winding with low resistance you are dissipating next to no power (current may be high but voltage is nearly zero. You will need a shunt regulator that dumps power into a dummy load.
reply
If you really want the maximum power from the generator the best solution is probably a 3-phase (assuming that's what the generator produces) active power factor correcting MPPT system. That isn't a trivial thing to implement. It may, however, be the case that the generator output is soft enough that simple capacitive filtering, which usually would produce horrendously bad power factor, isn't too terrible. The higher the capacitance you use, the worse the power factor.
Don't forget that you may need to add additional circuitry to keep the maximum speed of the generator in check. You can't do simple things like shorting the windings unless their resistance is high. If you short a winding with low resistance you are dissipating next to no power (current may be high but voltage is nearly zero. You will need a shunt regulator that dumps power into a dummy load.
reply
TobiKellner
Hi,
there is something I am struggling with:
Say I want to maximise energy from wind or solar, using an mppt tracker, and then I want to convert this to a specific voltage, say for charging a battery.
My intuition would be that these are two separate circuits - one for getting MPPT out put from the wind or solar generator, at whatever voltage the optimum power point sits at, and another circuit (e. g. a buck circuit) that then takes whatever the output voltage from the MPPT is, and converts that to the desired output voltage.
Yet it seems that there is only a single switching circuit that does both tasks. I don't get this conceptually - how can a single switching frequency always guarantee both that the wind/solar source is operating at the maximum power point, and that the output is at the desired voltage
reply
Hi,
there is something I am struggling with:
Say I want to maximise energy from wind or solar, using an mppt tracker, and then I want to convert this to a specific voltage, say for charging a battery.
My intuition would be that these are two separate circuits - one for getting MPPT out put from the wind or solar generator, at whatever voltage the optimum power point sits at, and another circuit (e. g. a buck circuit) that then takes whatever the output voltage from the MPPT is, and converts that to the desired output voltage.
Yet it seems that there is only a single switching circuit that does both tasks. I don't get this conceptually - how can a single switching frequency always guarantee both that the wind/solar source is operating at the maximum power point, and that the output is at the desired voltage
reply
d614gakadoug9
If the total power available from the generator is moderate (say two to three hundred watts maximum, a flyback converter may be the easiest solution. A flyback can deliver output voltage above or below the input voltage even with the same number of input and output winding turns. Only a single active switch is required. The non-isolated equivalent topology is the inverting converter, which might be quite workable. The inductor (so-called transformer) in a flyback converter gets quite large at high power, especially if the voltage range at either input or output is large.
[edit] I would be inclined to experiment with a flyback converter that charged the inductor to just short of saturation then allowed it to discharge to zero current each switching cycle. This is, of course, requires variable frequency.
reply
If the total power available from the generator is moderate (say two to three hundred watts maximum, a flyback converter may be the easiest solution. A flyback can deliver output voltage above or below the input voltage even with the same number of input and output winding turns. Only a single active switch is required. The non-isolated equivalent topology is the inverting converter, which might be quite workable. The inductor (so-called transformer) in a flyback converter gets quite large at high power, especially if the voltage range at either input or output is large.
[edit] I would be inclined to experiment with a flyback converter that charged the inductor to just short of saturation then allowed it to discharge to zero current each switching cycle. This is, of course, requires variable frequency.
reply
evanmayer744
What's fascinating is that nearly every brand of solar generator (Anker, Ecoflow, Bluetti) has implemented a well optimized, high efficiency buck/boost MPPT controller on their units. These controllers often have very sophisticated self-monitoring, and automatically adjust for a wide range of DC sources. So convenient!
As a solar/battery system builder myself, I find it incredibly infuriating that standalone modules with this tech do not exist, or at least are very rare and obscure. A buck/boost MPPT controller with fast active tracking would be an absolute game-changer for 48V systems, especially portable systems that benefit from versatility. This tech has already proven massively beneficial on the solar generator market, so why hasn't someone come along to fill the gap in the DIY market
reply
What's fascinating is that nearly every brand of solar generator (Anker, Ecoflow, Bluetti) has implemented a well optimized, high efficiency buck/boost MPPT controller on their units. These controllers often have very sophisticated self-monitoring, and automatically adjust for a wide range of DC sources. So convenient!
As a solar/battery system builder myself, I find it incredibly infuriating that standalone modules with this tech do not exist, or at least are very rare and obscure. A buck/boost MPPT controller with fast active tracking would be an absolute game-changer for 48V systems, especially portable systems that benefit from versatility. This tech has already proven massively beneficial on the solar generator market, so why hasn't someone come along to fill the gap in the DIY market
reply
user-hk3ej4hk7m
You could use an unregulated or fixed ratio dc-dc converter to boost the voltatge and then pipe that into the MPPT converter. Since the dc-dc converter is not regulated, it will drop voltage the same way the wind turbine would, allowing the MPPT controller to see the voltage changes and position itself on top of the MPP.
These DC to DC converters can be hard to find, I know that Vicor has some for automotive, but a simpler alternative I think would be to play with the feedback path on any of these cheap DC-DC converters to see if you can get a proportional output. In theory it should be possible to just use a fixed duty cycle boost converter and go from there.
reply
You could use an unregulated or fixed ratio dc-dc converter to boost the voltatge and then pipe that into the MPPT converter. Since the dc-dc converter is not regulated, it will drop voltage the same way the wind turbine would, allowing the MPPT controller to see the voltage changes and position itself on top of the MPP.
These DC to DC converters can be hard to find, I know that Vicor has some for automotive, but a simpler alternative I think would be to play with the feedback path on any of these cheap DC-DC converters to see if you can get a proportional output. In theory it should be possible to just use a fixed duty cycle boost converter and go from there.
reply
mikael5914
Please make sure to integrate a diversion (dump) circuit for when the wind turbine is spinning but no current is flowing because scary things happen at that point! Basicly you need something to slow down (or stop) the turbine if the battery is fully charged and the charge controller or BMS prevent a current from flowing. I once created a very simple circuit with a comparator, mosfet and mosfet driver that switched a dump-load into the circuit once a certain voltage is reached. There are also much more sophisticated ways of doing this but please just make sure you have some diversion. I have experienced what happens when you don't and it's not pretty!
reply
Please make sure to integrate a diversion (dump) circuit for when the wind turbine is spinning but no current is flowing because scary things happen at that point! Basicly you need something to slow down (or stop) the turbine if the battery is fully charged and the charge controller or BMS prevent a current from flowing. I once created a very simple circuit with a comparator, mosfet and mosfet driver that switched a dump-load into the circuit once a certain voltage is reached. There are also much more sophisticated ways of doing this but please just make sure you have some diversion. I have experienced what happens when you don't and it's not pretty!
reply
OmgSaySomething
This mppt thing is waaay over mistified. I put a boost and buck converter in serial, powered from solar, the other end charges a battery. Two modules, it works for years, no magic - solar ranges from 15. 45V, I have a 24V battery, so boost is set to 30V, then the buck to the battery charging voltage 28V. When solar is more than that, only the buck works when less, then the boost starts workig. The end result is 28V constant as long as there is enough solar power to keep the boost-buck functioning, always drawing the maximum power from the panels. Works great for smaller batteries and no need of an arduino and unnecessary science at all: P
reply
This mppt thing is waaay over mistified. I put a boost and buck converter in serial, powered from solar, the other end charges a battery. Two modules, it works for years, no magic - solar ranges from 15. 45V, I have a 24V battery, so boost is set to 30V, then the buck to the battery charging voltage 28V. When solar is more than that, only the buck works when less, then the boost starts workig. The end result is 28V constant as long as there is enough solar power to keep the boost-buck functioning, always drawing the maximum power from the panels. Works great for smaller batteries and no need of an arduino and unnecessary science at all: P
reply
greatscott
When I first came in to the MPPT world, I was trying to figure out what to do with a remote repeater (2m US Repeater system) battery issues. One of the other hams working on the system ask why we didn't use an MPPT. To tell the truth I was unaware of this technology. I told him we did not have budget when we put this system in for extras. Which was true at the time. MPPT was new about a year after we did out install of the new system. Today we have a really cook system in place. I know more about it, but can and most likely need to know more. So thank you for the insight to a different chip and will let you know what we get.
Peace
reply
When I first came in to the MPPT world, I was trying to figure out what to do with a remote repeater (2m US Repeater system) battery issues. One of the other hams working on the system ask why we didn't use an MPPT. To tell the truth I was unaware of this technology. I told him we did not have budget when we put this system in for extras. Which was true at the time. MPPT was new about a year after we did out install of the new system. Today we have a really cook system in place. I know more about it, but can and most likely need to know more. So thank you for the insight to a different chip and will let you know what we get.
Peace
reply
NowInAus
Without a large inertial mass on the turbine you're fighting rapid changes in output, which makes your power maximising hard. So why not try some kind of crazy Scott mechanical solution such as having weights on the turbine axle that travel up a thread as the speed increases! Then you could use the increased rotary speed to raise rotational inertia and store energy, which you could return to the system over a longer time period when wind speed fluctuates. Max out and smooth mechanical energy so you're not using ridiculous curve riding in the electrical donain. Seems like the sort of thing you'd do!
reply
Without a large inertial mass on the turbine you're fighting rapid changes in output, which makes your power maximising hard. So why not try some kind of crazy Scott mechanical solution such as having weights on the turbine axle that travel up a thread as the speed increases! Then you could use the increased rotary speed to raise rotational inertia and store energy, which you could return to the system over a longer time period when wind speed fluctuates. Max out and smooth mechanical energy so you're not using ridiculous curve riding in the electrical donain. Seems like the sort of thing you'd do!
reply
wesdblack
I was wondering about this project as I had not seen an update in a while. I was also thinking about whether the direct-drive (no gears) system was part of the problem, perhaps like pedalling a pushbike on too low a gear: spinning fast yet little actual headway / speed. I believe many motors / generators have a 'sweet spot' RPM at which the magnets excite the coils best, yet the turbine also has its own sweet spot RPM and also torque at any given wind speed. It may be that gears are needed to match them up better, perhaps an efficient constant velocity transmission (CVT. Worth a look at!
reply
I was wondering about this project as I had not seen an update in a while. I was also thinking about whether the direct-drive (no gears) system was part of the problem, perhaps like pedalling a pushbike on too low a gear: spinning fast yet little actual headway / speed. I believe many motors / generators have a 'sweet spot' RPM at which the magnets excite the coils best, yet the turbine also has its own sweet spot RPM and also torque at any given wind speed. It may be that gears are needed to match them up better, perhaps an efficient constant velocity transmission (CVT. Worth a look at!
reply
Centar1964
If you just wanted heat a fixed resistance should track the power point of the wind turbine fairly closely, for electricity I have used cheap buck/boost converter pcb's from aliexpress, pick one that you can adjust the max voltage and max current and adjust to your battery and wind gen capabilities. They seem to track the power point fairly good. This could be used for the low power periods and for the high power periods a relay could bypass all of it and directly connect the wind turbine to batteries so you don't need a huge boost/buck converter.
reply
If you just wanted heat a fixed resistance should track the power point of the wind turbine fairly closely, for electricity I have used cheap buck/boost converter pcb's from aliexpress, pick one that you can adjust the max voltage and max current and adjust to your battery and wind gen capabilities. They seem to track the power point fairly good. This could be used for the low power periods and for the high power periods a relay could bypass all of it and directly connect the wind turbine to batteries so you don't need a huge boost/buck converter.
reply
rpontonjr
How about simple mechanical storage
Use the wind to slowly lift a magnet, then drop the magnet through a coil when it reaches height. The power output will always be relatively consistent, or zero. Or lift a bunch of ball bearings that then turn a turbine as the fall from the top.
With your EE mind, maybe you could do something with just capacitors i. e. Instead of trying to harvest extremely low voltage energy directly into a battery directly, charge up a capacitor and then harvest it all at once.
reply
How about simple mechanical storage
Use the wind to slowly lift a magnet, then drop the magnet through a coil when it reaches height. The power output will always be relatively consistent, or zero. Or lift a bunch of ball bearings that then turn a turbine as the fall from the top.
With your EE mind, maybe you could do something with just capacitors i. e. Instead of trying to harvest extremely low voltage energy directly into a battery directly, charge up a capacitor and then harvest it all at once.
reply
hansdegroot652
Hi I see projects online using solar power to keep a few 18650 lithium batteries charged. But since I am in a colder climate I cant do that since charging lithium when it is freezing is a bad idea. I am looking for away to power the part that reads temperature and switches the solar power to the charging part only if its not to cold. But at sun rise there is a bit of a period there is not enough light to power an arduino I can't seem to find a solution for this maybe if you get bored
reply
Hi I see projects online using solar power to keep a few 18650 lithium batteries charged. But since I am in a colder climate I cant do that since charging lithium when it is freezing is a bad idea. I am looking for away to power the part that reads temperature and switches the solar power to the charging part only if its not to cold. But at sun rise there is a bit of a period there is not enough light to power an arduino I can't seem to find a solution for this maybe if you get bored
reply
roboman2444
Better yet to take lots of measurements over the course of a week or so at various current loads, generate an optimal load/windspeed table that you then interpolate from. Or maybe also optimal windspeed/generator RPM.
You could have a small P&O adjustment algorithm running on top of that to generate a correction table too, which could be used to fine-tweak the main table over time.
This may be the best solution for your DIY arduino setup.
reply
Better yet to take lots of measurements over the course of a week or so at various current loads, generate an optimal load/windspeed table that you then interpolate from. Or maybe also optimal windspeed/generator RPM.
You could have a small P&O adjustment algorithm running on top of that to generate a correction table too, which could be used to fine-tweak the main table over time.
This may be the best solution for your DIY arduino setup.
reply
greatscott
Could you simply put a pre-boost controller or even the BQ25750 before your MPPT controller
Playing around with adaptive load control for your wind turbine would make some really interesting videos!
The issue here and I'm sure many places is we often go from very little wind to very high winds with not much in between. This winter we had a lot of wind but it's almost too much for currently widely available smaller systems.
reply
Could you simply put a pre-boost controller or even the BQ25750 before your MPPT controller
Playing around with adaptive load control for your wind turbine would make some really interesting videos!
The issue here and I'm sure many places is we often go from very little wind to very high winds with not much in between. This winter we had a lot of wind but it's almost too much for currently widely available smaller systems.
reply
RayleneCawood
Want wind: go high.
Want high; Go Winged Blimp Kite. (tethered by HV twin flex)
Blimp = high in low to no wind. (onboard H2 production by electrolysis)
Winged = high in high wind where a std tethered Blimp is blown down to the ground.
Use a Dihedral, Swept Back Wing and a trailing turbine blade for natural stability, rather than the much ado! It's all very complicated! programmer, and expensive computer control BS.
reply
Want wind: go high.
Want high; Go Winged Blimp Kite. (tethered by HV twin flex)
Blimp = high in low to no wind. (onboard H2 production by electrolysis)
Winged = high in high wind where a std tethered Blimp is blown down to the ground.
Use a Dihedral, Swept Back Wing and a trailing turbine blade for natural stability, rather than the much ado! It's all very complicated! programmer, and expensive computer control BS.
reply
Add a review, comment
Other channel videos
