Wind turbine charge controller prototype/shield

From reply #12 mgshigtech informs us

I am an EE as well as a software engineer

I am making the assumption EE generally stands for electronics engineer, so you are an electronics and software engineer, is this correct?

In your lengthy reply # 23 you state

....transistors just eat voltage.
The only thing they are good for is controlling the current draw rate.

I hope you had learnt that transistors are current devices and that they don't eat voltage.
In the context we are talking about I think you mean to say that transistors dissipate power, and that you are eluding to their possible in-efficiency in such a circuit, is this correct?

In your long reply #23 you assert the following

A wind turbine needs mppt where a pv array doesn't

I wonder how you arrive at such conclusions?

You go on to write the following

There is one more issue that you didn't notice here, and forgot to mention

Who is it that you are referring to when you say 'you'?
And is it not presumptuous to say that whoever you is, actually forgot to mention something else?

Rather, would it not be better to have said something along the lines of 'Oh, and I thought of another issue and also something else to add as well'

Then

If this is how we are doing it, then we are just DIYers, not open source.

So, I am again assuming reading this that you think DIYers and open source are mutually exclusive and that there is not point to being a DIYer?
When you read back what you have written, I wonder if you see the level of your opinionated views?

It appears to me you have a dislike for transistors, for 12 volt systems and even to some extent the thoughts and designs of others.
Again it appears to me you make a lot of assertions but very little to explain in relevant detail any explanation to provide further understanding.
Though you state a liking for class D (audio) amplifiers several times.

I could continue...

It seems being an electronics and software engineer today is something quite different from how I became one.
I will apologise here now in advance if mgshightech feels offended in anyway by what I write, but I simply write in response to what I notice.

On a more positive note, I do notice some hint that mgshightech does have a keenness in wanting to engage in developing systems, though I am unsure in what specific direction that interest is at present, as he/she talks about arcing relays to radiation showers from Russia.

I wish to be able to communicate in a discussion where there is an openness, a mutual respect of the interests of others.
A nurturing of a pot of like minded people who do want to be 'DIYers' and who do embrace open source concepts.
Hopefully what I write may provide him/her (mgshightech) with a little prod to think about these things.

I may elect to bow out of this thread as unfortunately for me it has lost relevance.

Paul

@rockwallaby

a transistor turned all the way on, which is what we do with pwm and/or mppt can be considered a concoction of resistors and diode drops. The voltage coming out the business end of it will be lower than the voltage going in. So, you are welcome to word that by saying that they dissipate power. It makes little difference to me. High current, hot mosfets turned all the way on will tend to have a d-s drop of 1-2 volts. Of course, you can add more transistors in parallel to partially counteract this problem at a cost. That's what the datasheets say, and it isn't a lot different for bipolars. Well, losing voltage can push you away from maximum power point and cost you additional optimizational losses, so, for a a 30 volt panel for example can comfortably charge a 24 volt battery quite optimally. If, however, we subtract 1.5 volts from the 30 because of transistor losses, 28.5 volts remain. This will still comfortably charge a 24 volt battery, but if we then account for IR drops, we may be cut down to 27.5 or 27 volts. To charge a 24 volt battery up to its top at the 20 hour rate requires very close to 28.2 volts, so the 27.5 is 0.7 volts shy. The panel has to make up for that by reducing its current flow in order to push the voltage up higher. So, we lose both voltage and current. We can choose the panels that output a couple of extra volts. Of course, that can be done, again at a cost. So a system of multiple relays does not have much of any d-s drop, but they don't pwm very well. So the obvious solution to that problem is to have charging module options that use multiple pv arrays, turning them on separately so as to provide differing current flows for different charge stages. This is a good idea anyways for relay systems as relays tend to have current limits of around 60-120 amps. So, even in a 48 volt system, we are talking about 3-5 percent power loss, which for a 5 kw system amounts to around 200 watts, which can cost $300 or so to replace.

OK, so in truth, the relay strategy is not optimal for all designs. It sucks for long distance cable runs because you have to provide battery voltage at battery current wheras mppt systems can raise the voltage and have smaller IR loss. So other strategies need to be cared for in a flexible system. Also, if you know a good way to cut the voltage drop across transistors down to 0.2 volts or something, then of course, I would be very interested. So, yes, for 12 volt systems, mppt or pwm could easily (not necessarily) cost 10-15 percent of the power up front. For 48 volt systems, this is of course less of a concern. I am shooting for 48 volt (dual 24 volt poles) but a flexible system will allow for people to choose their favorite charging system.

Hmmm. regarding the question mark of being ee/cs. I guess I have been in enough degree contests to prefer to avoid that route because what we need more than to compete is to get along somehow. But let's just say that I'm pretty sure one of the driving forces behind arduino is open source strategy. So, when DIYers show up. This is of course beautiful. It will be more beautiful if we can work as a group to accomplish a public project. There should be plenty of people around to work on that. I can get our guys together and try to hit up kickstarter for public funding for a public project, but I'm reticent for a few reasons (that may melt away). So, one way or another, it is our intention to accomplish some strategic technological additions to the open source community that are designed to help re-balance the social flow. If you prefer for society to never evolve or improve, I suppose that is your right.

So, I'm going to tell you this because I guess I have to, but the reason why all the other stuff got into the talk is because I am a systems theorist who has spent some decades examining the current social structure. There is a pathway to overcoming some of its current weaknesses that lies down the lines of the concept of balance. Everything is connected to everything in the real world, and small deeds can have large results. ... or they can peter out and be forgotten in the infinity of time. (butterfly effect) So, I didn't want to get too deep into that kind of stuff because this isn't that kind of forum. Nevertheless, systems and complexity theory have a great deal to say about what we aught to be focusing on. So I have hoped for people to see the potential advantage of providing a top class solution to the dc power distribution problem .. open source. Of course, if people don't care about that, then I hope that they care about something else in relationship to the development of a top class mppt/charge controller architecture.

Let's see ... 1938, Otto Hahn discovers two kinds of fission, thorium Thorium fuel cycle - Wikipedia and uranium. Thorium is far far safer to fission than uranium, but you can't make bombs with it. So, I am going to try to respect the topic of the forum and not drag you any deeper into that. There is enough info. in the wiki page for anyone who is curious, except that I don't think it mentions chinas development of thorium fission reactors. OK, so what it boils down to is that I have a big picture in my head. I realize that some people don't like big pictures. I'm not accusing you of that. Feel free to make your own choice.

The arcs in relays as they open/close are of course definitely on topic. Arcs destroy relay contacts and can render charge control dysfunctional, potentially even dangerous. So if we want to use the relay option for cases where it is optimal, then we have to study this issue out.

I don't know just what sort of people are here. As my nameplate says, I am a noob on this site. I must decline to further discuss egrouphub in detail on this site because it is not on topic, and I wish to respect arduino.cc as I wouldn't be thrilled if people got too far off of topic on our site either.

Thank you for your comments. I hope others feel more positive about it. :slight_smile:

So, in case anyone is noticing, I have been on this project for some several months already. I have examined the curves, data of parts, etc. found a lot of ideal electronic parts and created some strategies, but I don't expect to out-think the arduino community by myself as that would of course be ridiculous. :slight_smile:

And class d amplifiers just save 1/2 to 3/4 of the power that is spent by other amplifier classes. This is a big deal if you are paying for all of your power up front. They can be powered directly by dc power sources, saving us the expense of the power supply and the energy it wastes humming at 60 hz all day, and the space it takes up and the shipping weight. So it's just another way of reducing system cost, but It is also not directly on topic. I just pointed out that if we can reduce the cost of multiple parts of a power system, we can cut the entire system cost by as much as approx. 50%, which is huge. The price of solar power dropping is huge. It's huge again if we can use it more efficiently (i.e. use less of it). And there are a lot of ways to do that that are improvements over what solar installers are currently setting up. (yes, I have discussed with solar installers and have a good idea as to what they are currently doing.) I'm not asking anyone to get involved in that kind of stuff right now, just trying to paint the picture a little wider in an attempt to help people to feel more motivated. :slight_smile: Looks like it backfired and got smoke all over my face. I'm reminding myself of Ronald Weasly and his wand. LOL.

OK, it has been a pleasure.

mgshightech

Okay,
Now you tell what you want from us.. :grin: I am also a guy like you to work collaboratively ]:D..

@ Khalid U R awesome dude. :slight_smile: I think "We got work to do" Hopefully some others agree on that.

BTW I failed to answer one of the questions. He asked why a wind turbine requires mppt while a solar panel doesn't. The answer is this. If a wind turbine is trying to charge a battery, the turbine will be spinning at different speeds. Both dc and ac motors produce different voltages at different speeds. The higher the speed, the higher the voltage. OK, so as long as the output voltage of a turbine is below the battery voltage, there is no charging because the battery is pushing harder at the electrons than the wind turbine. In fact, the system will most likely require some kind of diode to prevent the battery from dumping current into the wind turbine. .. Not what we had in mind... OK, so the rpm rises, the voltage rises the voltage reaches battery voltage and a tiny bit of current starts to flow. The voltage rises to around 2 volts more and a ton of current starts to flow. (btw I was keeping track of the wind turbine generation faq on alt.energy.renewable around 1995) .. So a ton of current starts to flow and it acts as a barrier to the increase of the wind turbines speed because the wind turbine can't produce the energy demanded by the current flow. So it slows down (or refuses to speed up). Well, when the turbine is spinning slowly, the force on the blades may rise a little, allowing for more current, but the energy output of the turbine will be proportional to force (torque) multiplied by angular speed. So, now that the turbine is blocked at a particular rpm from spinning faster, it can't collect the energy that is available because the speed can't rise. The solution to this problem is mppt. The mppt algorithm watches the current flow/voltage in tandem and scans the current/voltage dimension for the combination which produces the most power. When that optimum in power is produced at a voltage (respective turbine speed) above the battery voltage, the algorithm buck-converts the higher voltage down to battery voltage with higher current and presto ..... more power :slight_smile:

Now, let's talk about the solar panel. The solar panel is basically an arrangement of P-N junctions. They convert light into current. Basically, a 2 electron volt photon is incident onto the panel and it jumps an electron over a two volt junction gap producing 2 electron volts of available power. If more 2 ev photons are incident on the panel, then more electrons jump the 2 ev junction gap, but they are still only at 2 volts potential because that is all the energy available in the 2 ev photons. We arrange a bunch of these in series, to raise the voltage up to something that can match our batteries. Now, what happens when more sunshine hits the panel is that because of quantum mechanics above, the voltage output from the panel doesn't really change (not much anyways). In stead, the panel becomes capable of sourcing more current.

Now, not all photons are red 2-ev photons, so this makes the dynamic a little more complex. If we raise the voltage required to charge the battery, what happens is that the 2 ev photons are close to being the weakest of the bunch. There are also infrared photons, but the panels will not be designed to receive them because receiving them means shrinking the junction gap voltage and hence the cell voltage, and hence the cell power. ok, if we raise the voltage demanded of the panel, The red 2 ev photons will discover that the voltage on the junction has risen above 2 volts and they can no longer push the electron over the gap, so we lose current. However, there are higher voltage photons in the mix. Visible light contains half the energy of sunlight, and the panels are tuned to be most receptive to particular photon energies. So, what happens as we demand more voltage from the panel is that the higher energy photons can still jump the gap. Visible light is a concoction of photons between 2 and 4 electron volts. The cells are probably not particularly sensitive to 4 ev photons (blue) because they just can't engineer the silicon to be optimally receptive to all photon energies. Next, keep in mind, that if a 4 ev photon is incident on the panel, it can jump only one electron over the gap which is operating at around 2 ev. This means that the extra 2 ev in the photon are wasted.

So the designers of the panel are caught in a conundrum of shrinking the gap and receiving the IR photons, but wasting the extra energy of the higher energy photons and the alternative of raising the gap energy to receive the full power of the higher energy photons, but excluding the use of the lower energy photons. So you can see why pvs have efficiency problems. Recent attempts to produce panels that can split the energy of a single high energy photon into two and jump two electrons across the gap have been successful. However, these cells/panels are not ready for production and may never be.

So the way this boils down is that panels produce only slightly more current if short circuited (like maybe 25% more) then at voltage ranges from small up to their designated maximum power point, their output current is nearly constant and it begins to drop off as we demand more voltage from the panel than it can produce. So there is this sweet spot of about +- 6% of the panels output voltage where it produces very close to its maximum power. If we demand voltage below that, then we waste voltage. If we demand higher voltage, then the panels current sharply drops off. It's better to demand lower voltage than higher voltage because if you raise the voltage required of the panel as much as even 15% above its maximum power point, you will discover that it outputs very little power. Most of the photons incident on the panel will be unable to make electrons jump the gap.

Once again, the photon energies in sunlight are pretty much the same no matter how much light you are getting, so the panels output voltage pretty much stays the same. It just becomes more current-capable.

So, if we charge a 24 volt battery with a 30-31 volt panel in a relay configuration, the panel and the battery are closely matched. MPPT is useless because there is no spare voltage to dc-convert down, and PWM is bad because the pwm transistors will absorb the last volt or so, causing a mismatch between the panel and the battery. So, if we use a 35-37 volt panel in stead, then MPPT is slightly useful (get us maybe 5-7% advantage). PWM is perfect for the 35 volt panel, but becomes wasteful for the 37 volt panel. So, the next thing we do is we say ... we hate line losses and purchasing a lot of copper, so we raise the panel voltage up a lot higher by putting them in series. Then, MPPT really shines because it transforms the extra voltage into extra current. Basically, MPPT controllers can be considered to be dc-dc converters. PWM and Relays do not do this. So the answer is ... if you are running power a long distance (say a couple hundred feet) then you want an mppt controller, and the more distance you run, the more voltage you want the mppt controller to accept. Midnite solars highest voltage mppt controller accepts 250 volts and dc-dc converts it down to battery voltage. Now, when matching panels to the mppt controller, you have to consider the panels open circuit voltage which will be some 10-20% higher than its maximum power voltage. Failing to do this could cause a failure when the controller turns the current flow off and the panel output voltage rises to its peak. We will basically risk crashing through the dc-dc converter transistors. So, what this means is that for the 250 volt controller, you want to run a maximum power voltage of around 180-190 volts.. so you choose your panels to do this.

BTW, crashing through the mppt converter transistors is a high risk for wind power because the voltage output is so drastically variable. So, if you are using wind and MPPT, then you are wise to include a voltage clipping stage that will cut the voltage down to MPPT maximum if it goes higher.

also, nother btw, working as a team does not mean that we have to do what we don't want to do. We can combine our efforts into a larger project by making only small adjustments in what we do. I.e. suppose we want to do a 12 volt charge controller or a 12/24 volt charge controller of a particular type. You can still do exactly that. Only thing required to hook this into a larger project is to comply with a group defined standard for module/system interfacing. That's it. Now what we created becomes part of a configurable, flexible system. People can use our charge control module, or they can use someone elses charge control module, and all we had to design was just one module and leave the rest of the work to someone else. Someone else can worry about the control panel code, the vt100 interfacing library, the OOP classes and subclasses that drive the modules etc. (I already have a vt100 library on another platform that can create guis on a vt100 panel) let's see ... nother small issue ... if we don't want to go through the expense of adding a vt100 panel to a cheap controller, we can run a vt100 emulator on a computer and spare the panel. :slight_smile:

OK, maybe I am pesky. I guess I am just really excited about this project.

@rockwallaby ... some of our other guys are working on a scada system. But they have been a little slow. I found a pre-written php web based scada system.. actually there are a couple of them that I found. But one of them is pretty much written up in french, which loses me. It is our intention to do the same kind of scada work you are talking about as another component of the project. :slight_smile:

Hi mgshightech,

That sounds interesting, about what you say regarding SCADA systems.
I have worked with industrial SCADA systems for many years, programming them to work with industrial PLC's.

Funny you mention that it is written up in French, as I am just on a French website checking out some things to do with motorcycling for when I head back there in June this year for some months. You could use something like Google translate to help maybe?

What I have been working on over the past months is something that appears like a SCADA system, but to work with small micro-controllers like Arduino's.
There is still much for me to do in regards to developing it to how I intend, and yes, progress has been slow for me as well.

There are two basic approaches I have used;
One is where the Arduino is the server, serving up data to the client, I do this using json formatted data. It is nice and efficient.
Second is where I have the Arduino push up data in json format to my SQL database on my hosted site. Then the host server simply pulls data out of the SQL database and gives it to the client when requested.

The second approach is where I am moving toward as it allows for better through put of data if there are many clients connected.
The Wiznet ethernet for Arduino can only support four simultaneous connections according to the data-sheet.
By having the Arduino push data up to a hosted site gets around this problem and allows for many other possibilities.

The Arduino essentially is the component which gathers sensor data and provides any needed output control, such as PID loop controllers and so forth.
The client side (browser) is mostly made up in javascript and I currently use Backbone.js as my framework to keep things as neat as possible.

Another active member on this forum, Graynomad, is developing a very nice piece of hardware and the software to go with it that might also interest you.
It is called ArdweeNet and it is designed with networking in mind, using RS-485. What Rob is developing I believe slots quite nicely into this area of control.
Maybe Rob might come in at some point if he is interested to do so.

I would be interested to learn more about what you have found in terms of that php based SCADA system if you get a chance.

Regards,
Paul

mgshightech:
He asked why a wind turbine requires mppt while a solar panel doesn't. The answer is this.

The Wikipedia article about MPPT is wrong, then? You ought to correct that.

I wouldn't be so bold as to assert that the wikipedia article is wrong. It is only in special cases that other methods are just as good. If you are aware of the special cases, you can take advantage of them to your benefit. If you are unaware, then mppt away.

sorry I haven't been around .... flu ... I have been quite miserable and am still sick.

@rockwallaby ... I thought I named those packages for you. Somehow I don't see th post here, so I am going to repost them. (at least it seems that way to me) the package names are stantor (french) and seer_2 or s.e.e.r. 2 (english) both available on sourceforge. I would try the seer2 package first. Just look them up on sourceforge. I'll be back around as soon as my body is ready for it.

look here: Unbound Solar® – Solar Products & DIY Advice by Our Experts

you will see that the these curves show that the maximum power voltage doesn't change much as the sunshine level drops. The wikipedia article shows curves with a more dramatic alteration, but all the curves I have seen for market panels look like this. (above) So, what you do is you figure you're going to lose a volt IR in transmission, and that if your remaining (max power point) voltage at max sun is around 6-8% too high for the battery, you will not lose much power because that is within the range where you will only lose around 3% of your power. You'll lose that much power by forcing it through a transistor. So, then your panels output voltage can drop as much as 12-14% and you will still be within range to only lose 3% of your power. The stats of panels show that the output voltage won't drop more than that. There are things helping you by the way. When it gets cold and your panels output more voltage, your batteries require more because their internal resistance rises. When it gets hot and your panels output less, your battery requires less. When your panels output current drops and the voltage drops a little as well, guess what, the battery doesn't ask as much voltage because it's internal resistance is causing a smaller IR drop because of the smaller current. .. So we are going to do tests to verify this stuff, so stick around if you are waiting for the test results.

So if you pick just the right voltage of panels for your condition, you can happily charge with relays and get as much as any other method. Unfortunately, only maybe 20% of panels lie in the proper voltage range. The rest are going to require fussing. There are multiple things that drag you out of this ideal zone. If you are covering more distance, you may consider using relay charging with 18/35/70 volt panels to offset the IR loss of the cable, or, you can go high voltage/mppt. Relay charging with 19/37/72 volt panels would only make sense if you were charging a long ways away from your panels, and for panels in the low voltage range like 14/28/56 volts... well, if you want them to reliably approximate maximum power, you are going to have to series/overvolt and mppt. Consider the 14 volt panel you lose a volt in IR losses, now you're at 13, and you can only trickle charge. If you put a blocking diode in there, you are dead. In my opinion, forget the blocking diodes, just have your controller turn the panels off at night. Relays suck power for their coils, but only 1-2 watts for 60-120 amps of current capability. That is a lot less than losses in 60 amps of blocking diode. (like 20-30 watts)

Also, I don't think mppt controllers are generally designed to boost ... assume they can only buck and supply them with extra voltage.

I'm going to try and use relay charging. We'll place the panels close to the charging station and use 2 guage cable. I have an mppt controller, so I'm going to make a comparison and cough up real data.

mgshightech:
If you put a blocking diode in there, you are dead. In my opinion, forget the blocking diodes, just have your controller turn the panels off at night. Relays suck power for their coils, but only 1-2 watts for 60-120 amps of current capability. That is a lot less than losses in 60 amps of blocking diode. (like 20-30 watts)

Take another look at Tim Nolan's project to see how he uses mosfets for the blocking. More generally, he really does a great job of explaining his circuit and the various sections of it.

I agree with you on dropping the buck/boost for simplicity. Assuming the panel/turbine and batteries are matched well the batteries won't care too much if you overvolt them a little so long as you keep the current within bounds (or is that saying the same thing?). An enduser can always pop on an external module for buck/boost anyway.

What we are going to do is define a module interface that can accomodate multiple types of charging. And I think the first modules to be designed will not use buck/boost. Later on we may move to design some buck/boost options.

I looked at Tim's use of a fet to block reverse current. For small charging currents at low voltages, this could genuinely be an efficiency boost over using a diode. I'm not sure why he needs an extra fet for that job, but it's probably some confusion about the nature of mosfets that is causing me that. I looked at the specs of his transistors. One of the issues I have discovered is that VDS for mosfets is temperature dependent and apparently tends to rise with temperature. The specs I looked at for his transistors did not show VDS figures for differing temperatures. Actually it just showed single RDS ratings that I assume you are supposed to use to figure your VDS. so I doubt that the specs are telling the whole story.

Nevertheless, I'll make half an apology to those paying attention for shooting down mosfets. for 12, maybe 24 volt charging, it looks like it is not particularly hard to keep transistor voltage drop below 0.4 volts. However, for higher voltage charging, it appears that RDS has to rise because the transistors have to be different higher voltage models, and it could easily become difficult to beat a blocking diode with a mosfet when doing 48 volt charging at high current. Another reason for not apologising though is that if Tim is right, and you need an extra set of fets to act as current blocking, then that doubles your voltage loss. (ouch) Relays can simply be turned off. They don't need any reverse current blocking.

So I started a writeup describing some module interface definitions. This writeup is not finished, and is open to discussion. It is here: http://egrouphub.com/wiki/index.php/Charger_Modules I'll be looking over Tims design for clues as to how the interfaces may need adjustments, but I think I have most of the necessary details right so far.

OK, so here is how I am picturing the deal. Everyone who writes a significant piece of code or designs a significant module up front will have rights to everyone elses designs who also participates in this fashion. When that is done, and we have reasonable solutions for a fair set of modules, we may wish to close it off for newcomers. However, We will publish the interface definitions and make them publicly available so that we don't become a hindrance to the technology. Ultimately, the goal is not just to make money, but to cut down the cost of solar power systems. This is extremely important for humanities future. We will examine the locations of the people who are originally involved in the engineering and try to make tentative global dividing lines as to marketing territories that give people the space nearest to them. There should be a lot of new, unconquered turf down this alley because the market has not yet properly responded to the drop in the cost of solar panels, so there should be a lot of options for doing new, related projects.

:slight_smile:

short discussion of the values of two-pole power systems

http://egrouphub.com/wiki/index.php/DC_Power_Distribution#The_multi-voltage_problem

OK, I got the picture with the mosfet reverse conduction. I was a little confused because of my experience with vlsi design where mosfets are symmetric. Looks like power mosfets are not symmetric, and include a reverse conduction "body diode", so they can't block any current in reverse. So he places the mosfet in in reverse so that it can block in reverse and then turns it on at the same time he turns on the other mosfet. This obviously forces a double voltage drop. In his case, both drops together are probably under 0.6 volts.

RDS for power mosfets from the IR datasheet:

small transistors size of Tim nolan's (to-220)

v RDS MaxI
60 .028 30
100 .077 17
200 .18 10.5
250 .28 7.9

Larger TO-3P

V RDS MaxI
60 .014 62
100 .055 41
200 .085 23
250 .14 18

the IRFIP150 (100 volt .055 41 amps) is around $3 a piece
You can see that as soon as we jump out of the 60 volt transistor range, the voltage loss rises rapidly. For the IRFIP150, it is quadruple the loss of the 60 volt version. Just passing 40 amps through it will cause an over 2 volt drop and incur the transistors maximum power dissipation of close to 100 watts. To get this loss under control we will have to have 4 of the things. ($12) That will only drop the voltage drop to 0.5 volts which doesn't beat a diode. And that is only a 40 amp charger ... (whewsh) To get the same efficiency for 60 amps (a reasonable size charger) ... 6 transistors..... $18 + heatsink + silicone padding/screws (probably reducible for large quantity orders). A 40 amp relay is $5. A 60 amp relay is $5 (I got them for $4). A 120 amp relay is $12 at aliexpress.com .. .. and the relay solves the reverse blocking problem for free, and without voltage loss. So you can see that if your panels match your batteries, relays have the clear advantage for higher voltage charging. Their problem is that they don't pwm well.

Panels have open circuit voltages considerably higher (like 20% higher) than their maximum power voltage, so charging a 48 volt bank will require an open circuit voltage of 75 volts to do pwm and something closer to 150 or 200 to do mppt. You need some spare voltage on the transistors for protection against inductance, etc. On the other hand, the battery helps to give you that spare voltage. To do mppt for a system that can accept panels during the day and rectified 120vac at night, you need the 250 volt transistors. Just 10 amps per transistor will incur 1.4 volts of loss .. 80 amps ... 8 transistors, probably $25 + heatsink bla bla bla, and if you want to block, use a diode or a relay cause your transistors are imposing a lot of loss. To get your transistor loss down to diode level, you have to triple your transistor count ... (24 transistors) .... ouch. Oh woops ... I forgot something... to get the 80 amps out, you may only need 40 amps in for mppt, That sounds like it could halve the transistor count, but it really doesn't because the 40 amps will be bunched up into half of the duty cycle, making it 80 amps and causing the full voltage drop half the time. As the voltage drops down, the duty cycle rises, causing the same voltage drop, but for more time. So there is some efficiency advantage at higher voltages because the voltage drop is smaller compared to the total voltage. It shows up in less loss because of a shorter duty cycle.

Obviously, to move a lot of power at 12 volts, we are talking about really high currents, like 100s of amps (400 for a 5 kw system). So you can see how the 12 volt solution works great for a small system, but the low RDS doesn't save you if you want to move much power. To move 200 amps for 2.4 kw, you have to have 10 60 volt transistors at 20 amps each, incurring a 0.28 volt loss which doubles to 20 and 0.56 for a reverse voltage blocking system. So that's about 5% power loss. To do it with relays, we are talking 4 relays at $5 each, free blocking and no appreciable loss. Now if we put the four relays into a 3 module system. first module does 100 amps, second and third do 50 each. The controller detects the current flow of each module, and for multi-stage charging, it simply turns off the modules it doesn't want to run in stead of pwm-ing it. Only thing remaining is to match the panels to the batteries, and you'll still be ahead if your panels have an extra 0.5 to 0.7 volts.

A 24 volt system may still be able to use 60 volt transistors. This creates a competency window for transistors. We can basically cut the percentage power losses of the 12 volt system in four, or only cut them in two and halve our transistor count. Unfortunately, we can't mppt up to very high voltages this way, but simple pwm power losses will be relatively low.

So it appears that transistor/mppt charging is only competent
a:with small, low power systems
b:systems where we want to raise the voltage to avoid distance IR losses
c:when panels of the proper voltage are more expensive, demanding use of higher voltage, cheaper panels
d:when we are too clueless to match our panels to our batteries
e:maybe 24 volt systems that use low resistance transistors and don't mppt up to over 50 volts.

and it seems to me that d: is probably the most common of these conditions

If anything I am suggesting itches anyone, yeah, just please let me know. :slight_smile:

This project has attracted a group of its own. Go here to find the people in cooperation:

Are you still working on this project? I have built a wind generator and was interested in what others have done to control it with an arduino.

I am looking at something similar, but to include MPPT from both solar and wind. I say wind, because aI am sure that there is a best-power speed for the turbine at any particular wind speed. As I don't have a wind tunnel, that will take quite a bit of tuning. The plan is to maximise that acquired power from both the PV and turbine (1kW possible from each), charge batteries, and when excess power is available, dump power to a water heater when the output voltage exceeds 27.5V (24/28V battery electrics). Power dump will go from zero to 1kW progressively over the next 1V, into a 0.15 Ohm (or thereabouts) resistive heater. This will require 3 buck converters operating relatively independently, so I may use 3 separate Arduinos to drive them.