In the past, when solar panels were not cheap, many people (even large energy companies) used solar trackers to ensure that their panels were always pointed at the sun to ensure they collected every watt of energy. However, due to the sharp drop in panel prices, it is no longer economical to install complicated machines to track the sun. But all solar power plants still track something else, called the maximum power point (MPP), which ensures that even fixed panels can be optimized for power generation.

Although the small MPP tracker (MPPT) can be used for a $200 solar charge controller, which is very suitable for small off-grid settings, [ASCAS] aka [TechBuilder] decided to launch an open source version at a much lower price because most of these units The cost is spent on research and development rather than on the actual components themselves. For this reason, the method he uses for MPPT is basically the same as any commercial unit, called synchronous buck conversion. This uses a specially configured switch mode power supply (SMPS) to match the panel's power output to the optimal power point very quickly under any given conditions. It is even suitable for many different battery configurations and chemical compositions, all of which can be configured in the software.

This construction is very extensive, going deep into electrical theory and design choices. A notable design choice is to use the ESP32 on the Arduino because of the higher resolution available when performing analog-to-digital conversion. There is even a lengthy lecture on inductor core design, of course all the content of this project is open source. We have also seen ESP32 used with MPPT before. Although it is a little less refined, it is still very interesting.

Thanks [Sofia] for the tips!

Written for this project is great, very worth reading..

This is amazing, the documentation is a master class. Even if you don't build an MPPT controller for this in the end, you can guarantee to learn a lot of very useful things from this article! I am so sad!

*Equally daunting-low-key editors make this mistake time and time again-*This happens when a person has no time but power.

I like hourly x WattHours or hourly x AmpHours.

Ah... Ampere’s acceleration! !

Hope I/we can scale down for a smaller system. They need it more. RV etc.

There are some very elegant solutions, such as BQ25570, which can collect every joule from a small solar panel, buffer it in a supercapacitor, and then pulse the battery. Very suitable for low-power solar projects. Check the CJMCU-2557 evaluation board. You get what you pay for, a plug-and-play solution.

A good time to try to build one of them. It worked well until the phase angle of the control was reversed and started to send power from the battery back through the controller to the power supply. A lot of smoke.

It was not found that the current cannot be reduced all the time in synchronous mode, without the risk of pointing in the wrong direction as a buck converter. A 24V battery is used as the power source and a 12V battery is used as the load. Therefore, once reversed, the current cannot be prevented from reversing. Then once the circuit is cooked, there is nothing but the fuse to prevent the two batteries from pouring the magic smoke release agent into the circuit from each side.

No, he handled it (read the description, it's clear), look at the Q1 resistor. Basically, this MOSFET is placed in the opposite direction (so when it is turned off, its internal diode is in the wrong way, it does not conduct electricity, preventing battery current from flowing back to the solar panel). It is turned on by the MCU through an isolated DC/DC converter (so there is no ground loop here, with a floating mosfet like this). This is a good design, I don't know if it works in practice, but it seems to work in theory.

Can't this be completely avoided by using blocking diodes? I thought it was normal?

Great, I want to build a psu mode version, but I see it already exists! Will this become a low-voltage micro-inverter? This is what I want to do, connect multiple panels through these units, and connect the battery directly to my load through a parallel bus. I should make it all shade-tolerant and avoid giving back shadow panels because they can collect a few watts of power. Any ideas?

Very good, I am planning to install some solar panels on my roof. Since I am in the direction of SEE-NWW, I want to know whether it is more economical to install est-west (because the converter has 2 mppt) or SEE only. I plan to consume most of the energy myself, because the DIY route makes discount resale almost impossible (although possible, but below the market price, it is almost meaningless). So I need to know whether it is fun to trade longer production time for maximum output (because I have a lot of test power supplies available). So I might be able to connect two cells to the pseudo mppt tracker and record for a period of time to check both cases.

Panel direction. East, Southeast, West, Northwest

At present, I have also set up some panels for experimentation. If you are using what you made (watts!), then please check which times of the day you use electricity and adjust the angle of the panel for each consumption period.

In addition, some battery storage (when you use an electric kettle, it will say that you need more power in the short term) or you can pour the excess power into a thermos.

My brother uses his washing machine when the sun is out and his panel is generating electricity.

I have two main sinks: water heaters and pool pumps. The first one requires peak power, but the latter is more expensive in summer. In view of the fact that the pool pump must be operated when the solar output is high, it is a natural sink. The auxiliary sink is an air conditioner. Although it is not mandatory where I live, it is a good thing when we cannot manage the heat wave through passive processing.

I am considering placing my own (directional) power meter next to where the power supply enters the house (clamped at the end of the meter). Then, when the power input from the grid becomes negative (I generate more power than I need), my power receiver (hot water in my case) has a wireless link and they start to consume the matched power To keep it almost zero.

In order to do this with an immersion heater that may normally consume 4kW, the controller will contain a "dimmer" circuit similar to a triac, so it only provides enough power to balance it.

If I start to use something at home, such as a kettle, oven or shower, of course the immersion controller will automatically turn off until the redundant draw disappears.

In your case, you can draw a lottery based on the time/year on the power receiver, for example, a swimming pool pump only runs in summer and a water heater runs in winter, but naturally only when there is excess power.

CC BY-NC-SA 4.0 — A sad license for such a good project. At least FUGU firmware is truly open source, CC0-1.0.

Replace a resistor and the permit is no longer applicable.

This is not how it works, this is not how it works.

"The old lady posted a photo on her'social media wall' because a smarter friend was watching..."

This may apply to CC BY, but it definitely does not apply to "-SA", ShareAlike.

Let me guess, the "non-commercial" part?

Why? Do you want to get paid by commercializing this person's design?

No, I avoid participating in projects like this: no use, no help, no contribution, no donation. This is sad because this project looks very good, except for the license.

The maximum rating of the upper limit is 5K-10K hours, which is far less than 10 years (the break-even point of solar energy). Waiting to merge power.

The (electrolytic) cap can reach a life span of 10 years or longer on a regular basis. The key is temperature. Every 10°C below the specified maximum value. temperature. Life is doubled.

"The rated life is measured by the capacitor manufacturer during the stress test and is usually shorter than the life of the capacitor in the application."

Keep temperature, voltage, and ripple current within the parameters, and it will last longer.

I expect the ripple current in this application to be very high. But yes, over-specify the upper limit (higher voltage, 105C, etc.) and seek the manufacturer's highest service life offer (20 or 30,000 hours).

I didn't intend to build one of them, but as always, I absorbed some valuable tidbits. Thanks!

I have a lot of respect for anyone who is willing to take the time to provide information like this person.

I think "Maximum Power Point" is a very long slide presentation.

https://battledecks.wordpress.com/guide/

Hackaday: A notable design choice is to use the ESP32 on the Arduino, because the resolution available for analog-to-digital conversion is higher.

A separate ADC (@03:50) is used, and it is mentioned at least twice in the video to improve accuracy and reduce noise.

At first, I was a bit unhappy about the editing speed of this video, but it was full of surprising information, and it still lasts 18 minutes, and it was made in this way deliberately, if you need more, you can click to pause Button digest time.

OK! Esp32's ADC is very noisy, so this sentence in the article makes me confused/worried.

Does MPPT require a lot of mathematics? In fact, this is not the case. As long as the output voltage is limited and the output current is monitored at the same time, a simple comparison can solve the problem. You can change the PWM up and down to select the one with the highest output current. The real power is not important, we are just optimizing the maximum power. In theory, a simple op amp compares the instantaneous output current with the smoothed historical (low-pass filtered) current and can tell you whether the newly selected PWM ratio is better or worse than the one you just removed. It is indeed possible to construct an MPPT using only the analog part.

Do not. You are close, but there are too many shortcuts. The maximum current will be when the solar panel is short-circuited.

If you are only limiting the acceptable voltage, but still only optimizing for current, then it will still track the lowest voltage you set because this will produce the maximum current.

But any stupid microcontroller can multiply the voltage and current to optimize the maximum power, and then swing the PWM slightly to optimize a reasonable set point.

It can also be done with an analog multiplier, but the flexibility of the microcontroller allows you to easily add various control algorithms. It doesn't have to be fast. The fastest may even be the appearance of some clouds on a partly cloudy day, or the rare event that someone (or animal) casts a shadow on the solar panel.

Increased safety measures may require faster action, but for basic MPPT, a slow swing around the set point may be sufficient.

You are measuring the wrong current. Ignore the input current and measure the output current, now you only need to optimize the maximum value, no mathematical operations are required.

Main inductance 16 ga for 35A?

With 16 turns of the coil at 600 AT or 35 A, the core drops to 50%.

The description says 30 laps. So the magnetic flux of the magnetic core is reduced by 50% 20A

It seems small to me.

What do you think of everyone?

For the 3D printing masters here, what caused the inconsistent gaps between the lines on the top surface of the print shown in the title photo of this article (maybe the first layer actually printed?)? The reason I asked was that when I printed with PLA, I got this in my own prints. The lines in my PETG prints look more uniform.

In my experience, a line defined on the first layer like this means that the nozzle is too high. PETG usually benefits from a slightly higher first layer, so this may explain why it drops better without adjusting the hardware. You can try to adjust the nozzle height to find the right medium, or perform Z offset for each material in the slicer, so that the printing effect of PLA will be slightly lower than that of PETG.

Please be kind and respectful to help make the comment section great. (Comment Policy)

This website uses Akismet to reduce spam. Learn how to handle your comment data.

By using our website and services, you explicitly agree to the placement of our performance, functionality and advertising cookies. learn more