solar tracker

[PG1995]

Thank you, NG.

But I'm interested in some kind of graph/chart which shows the combined effect of the solar tracker and MPPT on the efficiency. I couldn't find it. Perhaps, you could use some better search phrase. Thanks.

Regards
PG

 

[NorthGuy]

But I'm interested in some kind of graph/chart which shows the combined effect of the solar tracker and MPPT on the efficiency.

You can use formulae that I posted to create a graph in Excel.

Then, if you build your own, you can collect empirical data and graph them along with a theoretical graphs.

 

[PG1995]

Thank you, NG.

I will do it some other time. I don't know how to do it in Excel and at the moment really have limited time. Anyway, thanks.

Regards
PG

 

[steveB]

Pg, you can do plots in Matlab pretty easily. Note that business's people and engineers need to be able to generate plots/graphs given data and/or formulas. It is very rare to get plots in a form that is correct for what you are trying to do. But making them exactly as you need them is relatively easy. You spent two hours searching for something you could have made yourself in 15 minutes.

 

[PG1995]

Hi

Do you think the the solar tracker shown in this video will work fine? The circuit diagram is given here? Thanks.

Regards
PG

 

[steveB]

I can tell you what bothers me about this setup. It looks to me that the sensors are set to trip based on light level with comparitors. But, their description says that the system will turn to match the sensor light equally. Also it seems that the sensors will see equal illumination at any angle. There is no shader between the sensors and in the video they make it turn by illuminating one sensor. Perhaps I'm just being stupid and missing the point but it looks to me that the system will not work as needed.

 

[PG1995]

Thanks, Steve.

I understand your point about the shader between the sensors. Perhaps we can do a little modification. And it's good you replied here so that others can also see where the problem lies.

Regards
PG

 

[NorthGuy]

I agree with Steve. As one of the options, you can install the sensors on the side of the panel. This way the panel itself will work as a shader and nothing will cast a shade on the panel. If the panel has an aluminum frame (as most good panels do), it will be easy to mount sensors this way.

Also, it is a horrible idea to put a tape accross the panel and on the edges. Even a small local shading of the panel will decrease panel production dramatically. Panel should be completely exposed and any shading avoided at any cost.

The video looks more like a demonstration of the principle rather than a working product.

 

[PG1995]

Thank you.

I agree with Steve. As one of the options, you can install the sensors on the side of the panel. This way the panel itself will work as a shader and nothing will cast a shade on the panel. If the panel has an aluminum frame (as most good panels do), it will be easy to mount sensors this way.

I'm finding it hard to mentally picture the configuration you are talking about. How a panel can work as a shader.

Besides the shader issue, do you find the setup correct? Do you think it will do as a school project? steveB told me that as it is shown the tracker has fast response time and in his opinion it should have slow response time. But I have seen some videos where the trackers are not just fast rather super fast. Thanks.

Regards
PG

 

[NorthGuy]

I'm finding it hard to mentally picture the configuration you are talking about. How a panel can work as a shader.

Typical panel frame looks like **broken link removed**. Imagine a sensor mounted on the side of the frame. The sensor will see the light to the right of the panel, but the panel will block all the light from the left side to the sensor.

Besides the shader issue, do you find the setup correct? Do you think it will do as a school project? steveB told me that as it is shown the tracker has fast response time and in his opinion it should have slow response time. But I have seen some videos where the trackers are not just fast rather super fast. Thanks.

I think this is better determined by experimentation. From the pure light viewpoint, when cloud cover changes rapidly, you want tracking to be fast to reflect all changes quickly. However, what you gain in panel production could be less than what you spend on the rotation. Therefore, it's worth rotating only when difference between sensors is relatively high. This should produce very slow motion following the sun most of the time, but if there's a sudden change, it might be a good idea to react fast.