Lithium Battery Upgrade 6: Up the Creek
On a boat cruising forum I follow, one of the members has a tagline of something like, “A house is but a boat so poorly built and so firmly run aground that it would be useless to try to refloat it.” This post is about solar boating and if you are not into boats, just remember that a solar cruising boat is just a solar house that has a motor to move it around. There are lessons here for landlubbers.
I decided that Sun King needed a good test and the only way to give it one is to head to the river and stay gone for a few days. Between rain and a dam release in Alabama, the Escambia has been at flood stage for a while. When it is low, there are snags that can be difficult to avoid, but you can see them. And avoid them. At flood stage you just hope you clear what you can’t see. Part of the mission is to scout conditions for a single day run from Alabama down to Pensacola Bay. After this trip, only one segment needs to be explored.
In review, batteries have been upgraded from 350 amp-hours (ah) of Flooded Lead Acid (FLA) batteries to 640 ah of Lithium Iron Phosphate (LFP) batteries. The usable power is much greater with the LFP. The LFP batteries are delicate and require Battery Management System (BMS) controllers. These are computer controlled switches, basically to prevent too much power going in or out. It turns out that having TWO BMSs running side by side makes for some interesting interactions.
If you took two FLA batteries, one fully charged and one half charged, connect them in parallel. The full one would drain into the puny one. It is like a sewage system, except that it is electrons running downhill. However, the LFPs are not doing anything close to that. My Port battery decides it is going to run things while the Starboard battery naps. I don’t seem to have any say in this, but the motor and lights are on and the coffee machine works, so Num Me Vexo?
The Starboard battery stays full. Also, curiously, the Port battery does not recharge if I take a break. The Bluetooth connection between my phone and the BMS indicates that the charge switch does not even turn on until capacity drops below 75%. There may be some method in this madness, as it reduces the number of cycles and prolongs the lifetime of the battery. At my age, this is not really a factor for me, but if that is how it wants to run, then fine.
The solar power is not wasted when underway, though. If the sun is up, I may look at the app and see that the battery is only providing 5 amps of current while the charge controller is kicking in 20 amps to run the motor at normal cruise power. If I take a break and really pull some power with Mr. Coffee or the microwave THEN the Starboard battery decides to chip in a bit and then cut back when the heavy lifting is done. The BMS boards have a setting for maximum power out. The batteries can EACH deliver over 300 amps and I do have the BMS dialled back, but the one battery can still run the galley. After 4 days, the Starboard battery was only down 8% and not charging, while to port was near to 50%.
Long story short, I successfully ran upstream against a raging current to Pollard, Alabama, and ended the trip back in Florida. Everything did pretty much what it was supposed to do, even if it did things in unexpected ways.
Now we get to some things that need work. The solar panels should have been producing more power than the motor was burning. Not much more, but some. There was a known issue with the charge controller. This was one that I had. I loved my old ones, but they did not support LFP batteries. If you use a FLA charge cycle on Lithium, after the peak voltage stage, the charger effectively just cuts off, wasting power. With some controllers you can just switch off the solar for a moment and the controller thinks it is night. Turn the solar back on and it is a new day and you get some more solar power. Doable, but not acceptable.
Another thing, 3 60 cell modules in series have a Voc (no load) of around 105 volts. That far away from the nominal 26 volts of the battery the controller is not really very efficient and it runs hot. Watts used to heat the charge controller don’t do you a bit of good. I was getting a max of around 24 amps from the solar.
Finally, there was an MP4 connector that was running warm to the touch. Not good. That would have to go.
After studying various controllers and my situation, I decided to go to a fourth panel to run in pairs. Single panel power production from a 2+ 1 configuration did not look any better than 3 in series, so I added a panel. I had considered several options, but re-installing one of the original 6 panels would be neat and easy. I removed the stern canopy, made of wood and canvas, chopped a meter off the end, restretched the canvas and relocated the anchor light. It is not so stylish as I would like, but I am planning a full canvas do-over and will construct a more shapely canopy frame at that time.
Ok, lower voltage from solar (70v) and add a second controller for redundancy. What could go wrong? Well, the second charge controller looks the same, but it is an old firmware revision, so it won’t run LFP. Fine. The controller is beefy enough to run the two sets in parallel. First indications don’t show a great improvement in power, but a peek up top shows that mowing my yard has gotten the panels filthy. I am not greatly concerned.
I still want to run the module pairs on independent controllers. 2 reasons. One is the redundancy factor. The other is that if I open the hatch, which is one of the panels on slides, I have a huge imbalance that could cause hot spots. The controller I have is not ideal because it has a fan that can suck humidity and salt air through. I have been on the lookout for sealed controllers like the ones I used to have.
Victron is the usual go-to for boaters, but they are expensive and I don’t find just the right one. John has some good gear in stock, but none of those are right. I LOVE looking at controllers on Amazon and Ebay, with their convoluted Chinglish descriptions and wild specifications. I will write sometime on translations and what to look for, but the basic advice for those sources is Caveat Emptor, or let the buyer beware. I did find one that shows promise and have it on order. It will have to wait.
In the meantime, I am making preps for another outing of at least a week, or for however long the food and water last. There are friends to visit, shipwreck archaeology and a party at the beach. I am having to carefully resupply after a raccoon got into my food locker, ate the food, broke my bowl, slobbered on my spork and stole my life jacket for its store of fruit bars. It is always something. –NC