Wind Turbine & Enphase IQ7 system integration success

WebPower

To update the buck power supply trials, I now have a 30-96v power supply rated at 400W or so. It seems to be able to handle the turbine. We haven't had a big wind storm yet and I'll use it the next time there is one to see how it does in high winds. The turn on voltage is about 29.5v and the output voltage is about 22.9. So it's working and will give much better power output if windy enough. It allows the turbine to spin faster and produce more power. However, in my urban environment with trees and structures I don't get enough steady wind at a high enough velocity to make power consistently. So for low wind cases, you will actually produce less power than straight from the 48v turbine since when I'm observing the turbine its running in the low to mid-20v range which doesn't produce any power. I'm going to install a manual switch initially to be able to cut in the power supply if we have a good wind storm. Otherwise, I'll keep the turbine directly into the circuit without the power supply since that cut-in voltage is 16v or so. I may eventually have it automatically switched in by monitoring wind speed. But that's going to take me getting the anemometer up and running. I've modified it to have only one magnet on the hub and have to get it calibrated. Hopefully sometime this summer I'll have it mounted and running via a raspberry pi. I'll modify the above circuit diagram to include the buck power supply. I'm thinking I might install another turbine at a family home in Iowa where it's windy using this circuit. Then see the results. But that will be a ways off. In the meantime, I'll keep fine tuning my operation.

Here is a video of the power output of the turbine with a buck power supply added. This is one of the ones that smoked after about 5 minutes due to high input voltage (I think).

https://forum.solar-electric.com/home/leaving?allowTrusted=1&target=https://app.box.com/s/u156ubseeqnbihcms1kkty5aymua01ku

WebPower

The revised diagram with the buck power supply added to the turbine output

WebPower

To update on the buck supply, it does work but in low winds the cut in voltage is too high to produce any power. The turbine has to get up to 29+ VDC before the supply kicks on. Lots of my wind is under that so I miss production. There is a guy on an Arduino programming site that has a wind turbine power converter that is boost and buck combined. It operates as low as 10v. The posts are 2-3 years old so I'm not sure if he still sells the boards.

https://www.hackster.io/philippedc/a-wind-turbine-mppt-regulator-with-an-arduino-uno-783462

But here is another design that may also be of interest:

https://mattgadient.com/mppt-sucks-for-wind-making-my-own-wind-turbine-charge-controller-with-an-arduino/

Then also a UNT masters thesis on a combined solar and wind MPPT:

https://digital.library.unt.edu/ark:/67531/metadc1062827/m2/1/high_res_d/ASSAAD-THESIS-2017.pdf

Will I build one? Maybe. The UNT provides the most code and might be the where I take from to do the mattgadient version. Right now I have a switch installed to manually toggle between the buck supply and direct from turbine to the IQ7. The UNT paper discusses cutting in higher loads as the wind speed increases. Something I might want to do. Light winds are forecast for the next two weeks which will limit my being able to test switching in the power supply during higher winds.

Pat

WebPower

An update after about a year of running the system. I can say my system is working smoothly. No more glitches with the microinverter. Its been pretty much hands off this year once I finalized the low voltage cut out and ensured that I wouldn't smoke anything. I put the 500W turbine up about 35 feet and wasn't getting much production due to leaves on the trees. I've since lowered it to 25 feet or so and it's below most of the trees in the area. It's getting more production at the lower height. For higher winds that I don't get that often, the buck supply will produce considerably more power. I've been dragging my feet in getting the anemometer mounted and working so I don't know the wind speed crossover point where the turbine reaches 29.7v or so when the buck supply starts generating power. I should have more time in the coming year to work on that.

When considering this system, you have to be in a pretty windy area. Here in Northern Virginia where I am, it's only really windy in the late fall to late winter when the leaves are gone. I've only produced about 1KW total on the turbine this year. The last three weeks I've produced the most power by leaving the buck supply in circuit all the time. I also put slightly larger blades (longer and just a tad wider) on the turbine to see what that does for production. I'm going to leave those on for a while to get some data on power production at the lower height. This week  was a little windy and you can see the power production at night. Not much of course, but it proves if you were in a windy area you could push wind generated power to the grid 24/7 with an enphase grid tied system.

A couple of weeks ago was cloudy a couple of days but windy for bits overnight on the 9-10th.

BB.

Very nice documentation of your wind-power harvest... that is 1 kWatt*Hour ( 1kW???) total of energy harvest for the entire year?

Comes back to location, location, location, plus a reliable wind turbine system as a requirement for "significant" wind energy harvest over time.

Do you have plans to install your system in a better location (clear of trees/wind breaks) in the future? We had one member here that had significant (high winds) during the dark winter months in Minnesota(?) area where wind could make a very significant contribution during the dark winter days and long nights.

-Bill

WebPower

Yeah, so all you readers out there, my average electric rate is $0.12/KWh. So I saved 12 cents so far this year! If you're going to install a turbine, make sure the area is windy. That's why I call this my hobby, I'm not making any money at it!  HAHA. Here's some pics looking south toward the turbine (right above the church looking workshop), and looking north and east from the front of the workshop. Typical suburbia setting.The turbine is above the roof of the workshop.

I'd like to install this in Iowa where there is lots of wind but my brother there doesn't have an Enphase system, nor solar of any kind. To do it would require an Envoy and one microinverter as well as a turbine and the interface circuitry. Probably at least a $700 endeavor. I'll keep making tweaks to my system so that I can collect more data. Like putting a watt meter on the solar panels supplying the turbine circuit. Then I could subtract the solar and bias power from what the envoy reports to get what Enphase says I'm supplying to the grid. I have shunt resistors installed for a raspberry pi power meter but haven't worked on the software yet for it. That way I could plot the actual power supplied by the turbine instead of just seeing what the power meter reports.

They're forecasting 25-40 MPH winds in two days. We'll see how much increase in power production the Envoy shows.

Pat

WebPower

There has been a noticeable improvement in power production with the turbine lowered and the leaves off the trees. 650W of generation in the last month. WooHoo...(sarcasm :-) ) .

For those of you who keep asking what if you put batteries instead of the super capacitors in the circuit. Well, I've now got 4 spare AGM batteries so I'm going to put them across the DC side of the diode bridge and see where power production goes. I should have it hooked up by the end of December. I need to put a low voltage cutout in the circuit to make sure the microinverter doesn't kill the batteries. I'm going to leave the super caps in place as well as the switching power supply as my system is set to run below 26 volts. I've done some initial tests allowing the turbine to free spin up to over 40v then put it onto the circuit. It appears to produce more power, and it causes the switching power supply to run higher than 29.8v.

The purpose of the test is two fold:

Will the turbine produce more power with it running at design voltage? Most likely.

Will the microinverter operate properly when hooked up to battery power? I think so. And does the low voltage cutoff cause any problems? It shouldn't.

The rest of the circuitry must remain as the microinverter must be kept in at least standby mode and I need to avoid the unpleasant and funky non-production mode I've had trouble with if It doesn't start producing power at a nominal 23 vdc input.

So stay tuned and I'll update when its up and running.

And Have a Happy Holiday folks!

Pat

WebPower

The modified circuit adding a 48v battery bank is as follows. I've received the low voltage cutoff and will be installing it soon. The only deletion are the super caps. Everything else is needed due to the operating peculiarities of the IQ7s. I'm going to use a CZH-LABS 48v low voltage cutoff. Initally set to 51.2 on and 48.4 off. 4x12v AGM batteries in series for the battery bank. I might have some wind tomorrow and will try to install this today. Stay tuned.

WebPower

As expected, the IQ7 just pulls power from the batteries until the low voltage cutoff disconnects the batteries from the circuit. Stubbornly, there was no wind to speak of yesterday. Weathermen and their forecasts... Next opportunity is Tuesday. Got over 700W out of the stored energy in the batteries though.

Most likely the time delay relay and the 26v/20v low voltage cutout are not needed if you use batteries. The circuitry would also probably need to be tweaked to take into account a 48 volt operating level if you didn't use the switching power supply to drop the voltage down to 24v from the turbine. It would also require a change in the solar panels supplying the voltage bias during the daytime. The IQ7 gets clamped to the battery voltage negating any real MPPT action. This IQ7 is a 295W one. The buck power supply is not in the circuit contrary to my diagram above. A work in progress trying to figure out some glitch that's going on with it.

Pat

WebPower

In summary:

If you use batteries instead of the super capacitors, the system will run fine. You'll need a low voltage cutoff for the battery part and you don't need the switching power supply at the turbine. You shouldn't need the other low voltage cutoff or time delay relay. Those are in my circuit because I'm operating the IQ7 at a much lower voltage. You still need to provide a bias to keep the IQ7 in standby at least. You should have a higher bias than my 16v if you are using 48v batteries. Otherwise you will cycle the IQ7 on and off due a sudden voltage drop when the batteries reach the cutoff voltage. That's what I'm seeing in my circuit. The IQ7 is pulling more current than my power supply can handle (R1 limits the current) so the voltage momentarily drops below the shutoff voltage of the IQ7. It drops to around 7.8 volts, so it causes a restart on the IQ7. Without the buck power supply at the turbine and no super caps, the system runs during the day at about 38v with no wind. The Solar panels are allowed to operate at a much higher voltage.

Will the turbine produce more power with it running at design voltage? Yes.

Will the microinverter operate properly when hooked up to battery power? Yes.

And does the low voltage cutoff cause any problems? Not if your bias supply is at a high enough voltage.

And it's back to location location location for me. Not much wind and not that high of wind when I get it means less than ideal operating conditions. I need a multi-stage supply to allow both low and high wind power production. If its low wind, then derating a 48v turbine to 16v or so makes sense, while I'd like to allow it to run at rated voltage in higher wind. The system as I have it now, doesn't allow the turbine to run at a higher voltage if windy.

That's something that I'm actively thinking about. I have a couple of ideas and will detail results if I get something like that implemented.

Pat

Mocher

We had a 11Kw Solar system with a 10P battery system using Enphase. We live in West Texas where we get a ~360 days of sun and 15mph winds on average. I am wanting to put up a wind turbine from https://windandsolar.com/missouri-freedom-48-volt-mppt-kit/

This system will generate 2000W. I am looking to tie in with a system similar to yours but will be connecting into the Enphase controller instead of the grid. Any suggestions @WebPower?

I also found this video that helps out: 
https://www.youtube.com/watch?v=wvnSh4iX110&t

WebPower

Hi. That's using an old version of the enphase. I have not put IQ's in parallel so I don't know how they will perform. Most likely not ideally as the MPPT's will compete. There are some articles about it saying it ends badly. He hints at one issue where he uses the battery maintainer to provide the bias voltage. You also have the issue where it's windy and you loose grid. You need to dump the turbine load. Super capacitors will only take so much energy. When they're full, you need somewhere to send the turbine power. That's why I have a couple of safety circuits in mine. Also my IQ7 gets into a weird state and won't produce power so you need to restart it automatically. All these things are in the notes I've written.

I'm working on installing an MPPT wind charge controller that was designed by a guy in France and is Arduino based. I'm putting the parts list together right now and will have the board populated in the near future. It can handle a 500W turbine. I'm going to see if that improves my turbine power production as right now I'm derating the turbine with my nominal operating voltage much lower than the turbine rating. The chinese switching power supplies are not reliable and I've smoked the 96v one now. I'm back to a 17vdc operating voltage running straight from the turbine to the super capacitors as shown in the 12/27/21 circuit diagram. Hopefully this MPPT will work out and solve the inefficiency problems.

Stay tuned...

Mocher1

I am going to try to keep the enphase on separate busses so that they will not compete for power. It will be like the picture attached.

Mocher1

Also, sorry for the name change for some reason it will not let me into my other account. I had to register a new account and have not been able to reset the previous account.

WebPower

That will be an interesting experiment. If you haven't bought the Midnite yet, I'd test the isolation between a couple of IQ8s. They don't like to share a solar panel array with another MPPT controller, you get a ground fault error. I have not tried two IQ's in parallel yet. I just bought an IQ7 off ebay and will try that this week hopefully and see what happens. With a 2000W turbine, you're going to have to share the Midnite's output between IQs. I've tried to isolate with buck and boost power supplies but haven't been successful yet. Still trying to find a reasonably priced isolating power supply to test out. Your diagram doesn't isolate each IQ unless the voltage controllers are isolating. Most are not, especially most that are on AliExpress and Amazon.

You still need to realize that the hysteresis of the IQ's as they turn on will cause you issues with delayed turn on from off condition since you don't have a bias supply to keep them in standby mode.

That's an expensive setup. Hope you got lots of wind. :-)

And remember, the IQ8's output 240VAC not 120VAC. The enphase controller in your diagram is an encharge I'm guessing? Otherwise the design isn't quite right if its an Envoy controller. An Envoy only monitors and reports, it does nothing to the power generated.

Pat

Mocher1

I realized something was wrong in my image, you cannot use voltage like below 41 volts. The more accurate measurement is Ah (Amp hours). If you can have a monitor device that can check it then you can determine the battery capacity. Amp hours is like the size of your gas tank.

@WebPower
Yes we have an average of 15mph (7.5 m/s) winds. We do have times where we are getting 50 mph gusts. No obstructions any where around for miles usually supplies lots of wind.

WebPower

Biggest issue you might have in the long run is making sure the turbine doesn't over produce, as in what if it's really windy and you get one or more IQx's fail/stop working. You'll need some kind of fail safe circuit to put the turbine power into a dump load. It gets complicated if you're using more than one IQx.

And you can get power out of the turbine below 41v. It just might not be power at the voltage level you need. The turbine produces power the minute it starts spinning. It's just most efficient at its' rated voltage. And at low speeds you are at low voltages. That's the problem with my circuit, it clamps the voltage of a 48v turbine down around 16.7v. So I don't get nearly the power output the turbine is capable of. A true MPPT controller for a wind turbine would be constantly adjusting it's output power based on the input voltage/current curve. Hopefully this MPPT board I'm populating is going to fix my problem and allow my turbine to operate where it's designed to.

pat

Mocher1

@WebPower I understand that about he dump load, but that is what the turbine control board is for. It has a Midnite controller and dump load.

WebPower

The IQ7's I have that I've tested all have a start up of 22+Vdc. It would be interesting to see how what you are proposing works. I'm skeptical that the Midnite will work without something else taking care of the weird conditions I experience with my IQ7s. You need to make sure if you drop AC and the IQs turn off that you dump the turbine load. And then if the IQ's dont start up you need to disconnect the DC input. Also, you need to put a bias voltage on them to make sure they stay in standby mode. If you don't it can take anywhere from 30sec to 2 minutes for it to turn on and accept/use the turbine power.

Mocher1

That is great advise.

WebPower

yeah, simple in theory, more involved once you actually try to implement it. Take a look at the results from my IQ's in parallel experiment. It will simplify your setup as you don't have to provide all those engagement controllers for different voltages. Just put the IQ's in parallel. In theory, three or four or any other in parallel should work the same as what I saw with two. In practice what I see with providing the IQ's with more power than they are rated is that they will just max out at their rating. Of course there is some maximum current/voltage that will smoke it but if you put enough in parallel you shouldn't have the issue as long as you have a little headroom on the rating of the combined system when compared to the power output of the turbine.