You might know that we currently have a sole 320W solar panel. Initially I just wanted to check if it was worth buying a solar panel. While there is a lot of talk about the many benefits of solar panels, I was not so sure. Will the panels really last a long time like they claim. Will the efficiency reduce quickly as the panel ages. Will the panels really produce the rated power because they are tested in ideal laboratory conditions of 1000 W/m2 irradiance, but in real life I was not sure if we get so much sunlight etc. Well, after owning the panel for over 2 years now, I can tell you that most of my doubts have been clarified. So I was wondering if I should add another panel.

Even as I was building the project with one solar panel, I was forward looking and made sure everything that I purchased can be reusable incase I decide to upgrade to more solar panels. For example, the solar charge controller I purchased can handle up to 40 A solar current or the equivalent of 1000 W. I purchased it with the hope that I can add two more 320 W panels in the future if I wanted to. Likewise the cables that connect the solar panel to the solar charge controller and the wires from solar charge controller to the battery were are bought with at least 30 A capacity.

Even the MC4 connecters I bought were 3-to-i connectors (see picture below) so I can add two more panels in future. The plan has always been to go for 1 KW panels. You can read all about it in my earlier post. Even the flooded lead acid batteries I purchased were C10 batteries. Given that I have two 150 AH batteries, each, I thought I could simply charge each of them with 15 A power if I have 30 A coming from all the three solar panel at full power. Yet, I cannot add even a single panel with my current setup. How did things go wrong? Well, it’s my stupid brain.

3-to-1 MC4 connector

I was completely ignoring the fact that I am connecting the 2 batteries in series, not in parallel. So each battery is not going to receive 15 A out of the 30 A total current generated. Instead both batteries will be charged with 30 A. While that is not out of the spec for lead acid batteries, the health deteriorates faster when charged at much higher current than their rated C10 current. I want my batteries to last as long as possible. Still, I should be able to manage at least one more panel right? So two panels producing 20 A and assuming some 5 A load in the house, the batteries will at most charge with 15 A. There is a problem with that too.

You see, the software I wrote tries to optimize the health of the batteries while simultaneously trying to use as much solar power as possible and that leads to some tricky situations. But first, let me explain what I mean by optimizing for solar power and battery health. If you want to optimize battery health, you want to reduce number of charge cycles, and more importantly not discharge the battery too deeply. A lead acid battery that is discharged to 20% capacity and then charged to 100% has fewer charge cycles as compared to a battery that is discharged to 80% and charged to 100% before it degrades.

In addition, lead acid batteries like to be fully charged all the time. So if I keep the batteries always fully charged and only use solar power to run the household load, then any excess solar power generated by the panel is wasted because it cannot charge the battery any more and the power has no where to go. So what I do in my software is to let the battery drain up to 80% before the sun starts to give full power. Then the solar panel can power the house and any left over power can be used to charge the battery. All good so far.

The only problem is that I do not know before hand if there is going to be enough sunlight that day. Basically what happens is that my software disconnects the grid early in the morning around 4:30 am or so and lets the house run on UPS + batteries. By around 10 am in the morning, the batteries are down to about 80% discharged state. From there on, they start charging if it is a sunny day. However, if it is a cloudy day, there won’t be enough solar power to handle the load in the house. So the UPS + batteries continue to deliver power thus discharging the batteries to below 80%. My software detects this and connects the grid.

At this point, the grid is delivering power to the house and also charging the batteries via UPS. The charging current is 10 A. There is no setting on the UPS to reduce it (you can increase it if you like). Now lets say the cloud cover suddenly disappears, then the solar panel will also deliver 10 A at the peak. So my batteries will be charging at 10 A from UPS and another 10 A from solar. As a result, I cannot add any more solar panels. If I just add one more 320 W panel, the charging current will be 30 A under certain situations. So I did not want to take a chance.

But all is not lost though. This problem arises because I was going with a hybrid approach. I already had the UPS and batteries and I wanted to add a solar panel. The solution was a hybrid solar charge controller which charges the batteries with the solar panel power while not having any control on the UPS. If I remove my UPS and the hybrid solar charge controller and instead replace them with a solar charge controller with in build UPS like this one, then the problem will be solved.

Not sure if I will be doing all those changes as on date, but may be some day, when I need to add more panels, I will have to do it. For now it seems like the over spec’ed cables and devices are not being utilized to their full potential. These are some of the issues in project planning that one should get used to I guess. Imagine if you are a big business, and making such mistakes can cost you.