How to Upgrade a Solar Power System
Upgrading a Solar Power System
It was time to upgrade the solar power system installed for the shop almost two years ago. I’ve been happy with original system, but it needed expansion as soon as funding permitted. I also wanted to apply what I’ve learned over the last two years. Upgrades are needed because although I can run the shop when it’s sunny, the original system fell a little short on cloudy days.
Also, the battery capacity isn’t quite enough to power the small air conditioner after sunset — and I like to work in the evenings. Finally, I wanted to change from lead-acid to lithium iron phosphate battery chemistry. Lead-acid golf cart batteries are relatively cheap and easy to work with, but they have some limitations and — they’re a little messy.
A source of cheaper solar panels
Fortunately, I had a couple of windfalls. First of all, I learned of a company in Arizona, Santan Solar, that sells used solar panels. These panels have been removed from commercial installations after 10 years of service, and have a one-year guarantee. Since solar panel output declines at the rate of about 1% per year, we might reasonably expect a 10-year-old 250-watt panel to produce about 225 watts.
The incentive for this deal is the price — $50 per panel when I purchased in mid-summer of 2021. Since a new 250-watt panel was about $225, the deal was appealing. The only catch: you have to buy a pallet load of 25 panels. They have to be shipped by truck, adding about $320 to the cost. Still not bad for a net cost or $62.80 a panel. Upon delivery a couple of weeks later, I tested several of the panels using a Solar Panel Tester. They all came within the expected 200-225 watt output. Just fine for my purposes.
Lithium Iron cells at surplus prices
The second deal came from a friend who owns a surplus business in Omaha. We’ve known each other since high school, and he was interested in helping. He had purchased nearly 200 industrial excess LiFePo cells and later decided they didn’t fit his inventory mix. Brand new 60 amp-hour cells originally used for Segway repair.
When he offered to sell them for $12.5 per cell, I couldn’t refuse. That’s about 20% of the new price. The catch here was that they’ve been in storage for about 2 years. Since LiFePo chemistry is very stable and the batteries are expected to have at least a 10-year lifespan, that didn’t seem too bad. I now had a cache of the two most expensive components — panels and batteries — for my upgrades.
We need a new charge controller
I decided adding eight more solar panels to the array was the easiest way to start, and would address the main problem — adequate power to run lights and small power tools on cloudy days. Ultimately, I would like to expand the array to a total of 24 panels, but doubling it to 16 panels would do for now.
It had also occurred to me that wiring the new and existing panels into strings of four instead to two would improve the overall efficiency. Unfortunately, the MPPI charge controller that was included with the original inverter/charger would only work with two panels in series or a maximum of 75 volts. I was going to need a new charge controller.
Victron’s 5 Kilowatt model
A survey of available charge controllers yielded only a few models that could take up to the predicted 154 volt panel input. It also needed a substantial throughput: with 24 panels, it was theoretically possible to produce 6 kilowatts of power at peak sun. This narrowed the field considerably to just one model: the Victron MPPT 250/100 TR.
It could handle input voltages of up to 250 volts and produce a charging current of up to 100 amps. That wouldn’t quite get us to 6 KW, but it would probably handle a more likely 5 to 5.4 kilowatts — since solar panels rarely generate their full rated output. That accounted for the panels producing closer to 225 watts each. At just under $1000, it’s a little pricey, but Victron has a good reputation and this device gets good reviews. It is also bluetooth controllable and collects and stores statistics regarding the power generated. Victron has a great iPhone app that works with all of their devices.
It was easy to install, although they recommend it be mounted on a heat-safe surface. There is a massive heat sink on the back of the unit, and apparently it can get hot. I ended up mounting it on the wall closest to my battery installation in the shop on top of a piece of sheet steel. I also had to change out one of the analog meters to allow for the higher panel voltages.
I immediately noticed an improvement in the amount of power generated. It was probably only on the order of 5-10%, but hey, every watt counts. The higher panel voltage means that the current flowing from the panels is cut in half. That, in turn, means lower voltage drop along the cables from the array to the charge controller (about 50 feet away). Less power lost as heat in the wire. I also suspect that the MPPT algorithm in the Victron is simply better than the one in the Chinese inverter/charger. the entire inverter/controller/charger from China cost just over $1000. You get what you pay for.
Adding another bay to the solar array
With the charge controller upgraded, I could build the next bay of the solar array and mount eight more panels. I won’t restate the process here, since it was almost identical to the first bay of 8 panels. You can read about “How to Mount a Solar Panel Array” and “Framing” in separate posts on this site.
Results of the upgrade (so far)
So how did we do? Great! The analog metering I originally installed turned out to be a little optimistic on power generation, so I can only estimate the peak power output of the original array at about 1600 watts. With the new charge controller and additional panels, I’ve seen a peak power generation of almost 3600 watts — reported by the Victron charge controller app. I’ll take it. Solar panels rarely generate their advertised peak power and then only briefly in mid-day on a cloudless day at noon. Even a little dust on the panels can reduce their output.
On the more pragmatic front, I now have enough power to run about 150 watts of light and other incidental equipment on cloudy days. Meaningful solar collection starts earlier in the day and ends later in the evening. So for the first part of our shop solar upgrade, I’d say we did well. In the next post, we’ll talk about designing the new battery systems.